High Voltage Forum

General electronics => Electronic Circuits => Topic started by: petespaco on December 23, 2018, 05:49:46 AM

Title: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on December 23, 2018, 05:49:46 AM
I have been experimenting the the popular "12-48 Volt 1000 Watt ZVS Induction heaters" for a year or two and have many videos and web pages describing that work, as many of you may know.
   More recently, an improved version has become available and is being sold by well over a dozen vendors.  Instead of one pair of IRFP260N's driving the coil, they doubled up on them and added 2 fans on top.  The also added 50% more capacitance to the  tank circuit and increased the number of turns of the output coil from 7 to 10.  These tank changes should lower the oscillation frequency which is good for many things that people would want to heat.

  Just recently, I ordered one of them, a complete kit, called "Combo 2" by seller Aliexpress.
I had to wait quite a while for an analog ammeter and it’s here now.  I still haven’t connected it all up yet, but I have spent a lot of time unraveling the various  descriptions and pseudo instructions that I see on the vendor offerings.  These units are seldom soldi n as complete a kit form as the one the I got from Aliexpress.
Some vendors call them 1800 watt, some even call them 2500 watt, but I’m pretty sure they are all the same.
  Yesterday I put up a new web page to help new buyers understand what they just bought and how to set it up with a minimum of disappointment.
I tell you all this, because I’d like to get this link to others, since I am already getting inquiries anyway.
There are two web pages referenced here on this cover page:
 https://spaco.org/Blacksmithing/ZVSInductionHeater/ZVS1800Watt/1800-2500WattZVSInductionHeaterNotes.htm

Both pages are pretty messy, but, due to the apparent popularity of these devices, I wanted to get something out there.
Time will tell.

I will add to that page and to my youtube channel once I get my system up and running.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on December 23, 2018, 05:35:44 PM
You got my attention with this, there has for a long while been small ZVS drivers available, but these "high" power seems more interesting as the price-tag is getting closer to what you could build a similar unit for yourself.

I looked at "1800W 40A ZVS Induction Heating Board" on ebay and it seems to be the exact same unit as your 1800W model, on some pictures for the 2500W model it looked like it was 120mm fans on top and it had a 3rd heat sink in the middle, so just from the pictures it must been built bigger for then higher power rating.

I placed a best offer on one of them and will see how cheap I can get one from China, as I got a few different "cheap" power supplies that I wanted to test on a ZVS induction heater before telling the world about it is a possibility :)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on December 25, 2018, 09:17:31 PM
Good eye, Mads.
   Yes, I now see the extra heat sinks and additional pair of gate resistors.  So now it looks like 3 Mosfets in parallel on each side.
And I also now see that the 1800 watt model  has 90mm fans, versus the 120mm fans for the 2500 watt model..
I can only say that I did some  sloppy investigation!
I'll already updated  my pages to reflect this difference.

I will look forward to your comments when you get your board.
By the way, we have all heard the warnings about NOT connecting these zvs heaters directly to a switched mode power supply and then plugging the power supply in.  They say that the slow start characteristics of the SMPS cause it to start slowly and cause the mosfets to lock up.
And I am sure that is generally true.  But the server power supply that I got with my Aliexpress heater (Emerson 48-2900U Switched mode Power Supply)
doesn't even turn on until about 7 seconds after power us applied.  I will probably set my system up to power the supply first anyway, just to be safe.

(a bit red-faced),
Pete Stanaitis
   
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 02, 2019, 01:54:03 AM
Hello, Mads.
I hope I am not getting ahead of myself here, but I see a new issue with both the 1800 watt and the 2500 watt versions of these ZVS heaters.
  It has to do with the little buck converter that runs the fans, and in the case of the one the I bought, the pump.
Here's what I have to say:
---------------------------------------------------------------------------------------------------------------------------------------------------------------------
Important note if you intend to run your heater on 48 Volts:
You can NOT use the little power supply (buck converter) for the fans! You will have to provide a separate 12 volt supply at a couple of amps.
That is because the maximum rating of the regulator chip on the buck converter board is 40 volts. Don't ask me why they did that, but, hidden in the Chinglish, is a warning that operation at voltages higher than 36 will "burn" the fans and/or pump.
---------------------------------------------------------------------------------------------------------------------------------------------------------------------
I'm getting closer to powering the thing up.  I checked the no load current draw of the pump and its only about 300ma, but I will need to connect it to the actual system and fill it with water to see how much current it will draw,  Then, same for the fans.  Then find or make a simple power supply for them.

Have you noticed this situation, and what is your feeling about using that buck converter?

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on January 03, 2019, 12:09:11 PM
Thats a pretty important detail you discovered there, thank you for sharing.

I am still waiting for my mine to arrive, hopefully I wont get hit by the Chinese new year in February as ebay has it stated as shipped off on 25th December 2018. But still it says Estimated delivery: Tue, 08 Jan - Tue, 19 Feb
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: flyrod on January 05, 2019, 03:46:29 PM
What is the chip on the DC-DC converter?  I'm guessing it's a LM2596.  It could be a high voltage version of the same thing: LM2596HV which would take 60v or so.  Does it look like this:?

(https://i.ebayimg.com/images/g/V2oAAOSwKb1bPyOK/s-l1600.jpg)

From here:

https://www.ebay.com/itm/123560506283

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 05, 2019, 05:27:23 PM
Thanks for your question, flyrod.
  Yes, the chip IS the "LM2596HVS" version.
That makes the Chinglish warning about not using the converter when the input is over 36 volts even more confusing..
When I looked up the datasheet a couple of days ago, i googled "LM2596" and got this:
http://www.ti.com/lit/ds/symlink/lm2596.pdf
I didn't look beyond the first few pages when I found the max input voltage to be 45.
I even found another datasheet when it was spec'd at 40:
https://www.onsemi.com/pub/Collateral/LM2596-D.PDF

It didn't occur to me that to investigate the "HV".  Sorry, I hope I didn't cause too much confusion.  I guess I was biased by the warning that I mentioned earlier.
http://hmsemi.com/downfile/LM2596HV.PDF

  By the way, my buck converter doesn't have the pot.  There's a fixed resistor in its place.

I did fire the system up yesterday and it worked, BUT the idle current (current flow with nothing in the work coil) is at 26 amps.  I think it should only be about 6 amps.  It IS oscillating. When I put a graphite crucible into the work coil the current does increase to 40 amps.  Frequency is about 77Khz, by the way.
I have attempted to contact the vendor through aliexpress.  We will see how that goes.

For that initial test, I did add a radiator and  unplugged the buck converter and powered  the fans and pump from a separate supply, just in case.

Sorry for jumping to conclusions,
Pete Stanaitis
---------------

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: flyrod on January 06, 2019, 12:34:15 AM
Can you tell what the gate drive is on your new board?  Is it the same pull-up resistors as on the little boards?

How are you measuring 26A?  Could you be using a different shunt that needs calibration maybe?  If 26A is accurate, that's a lot of power going somewhere.  Does anything besides the work coil heat up quickly?  For example if one of the caps is defective, 1000w should pop it in short order and make the problem clear.  A shorted turn in the work coil could also explain it.

Good luck in your experiments.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 06, 2019, 07:11:25 PM
i just checked the current by adding a direct reading ammeter in series with the shunt meter.  The readings are the same, within a couple of amps, since the scale on the new meter is fairly coarse.
  Haven't yet dismantled the board to look at circuitry, or to scope the gates.  The way this thing is made, you can't see inside much because the 4 heat sinks cover the sides and the fans cover the top.  when I do dismantle it, I probably should bring out a couple of test points.
I was hoping that I wouldn't have to dig into the thing at all.
No feedback from the Chinese vendor yet.  Is it Chinese new year right now?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 07, 2019, 08:17:44 PM
Today I examined the heater by eye and with an ohmmeter.  Can't yet see any reason why the idle current should be so high.
   Circuit components are identical to the 1000 watt board, Just twice as many.  Except that there are 9 of the 0.33 ufd capacitors and a 10 turn 60+ mm ID work coil.
Stumped.
   I did see one video where the guy was testing an 1800 watt board with an ammeter in the ckt.  That's where I got MY "expected" 6 amps idle current from.  That and the fact that my 1000 watt heaters do about the same.
  I may try a smaller diameter work coil just to see what happens.
But I am still hoping for some sort of feedback from the vendor.  Chinese new year for 2019 (year of the pig) apparently does not start for a few weeks yet:
Jan 28 to Feb 19.
  I now see that I can get the gate signals from the bottom of the board, so I can look at them once I get my courage up.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 10, 2019, 05:33:48 PM
Yesterday I added a short lead to each gate of the Mosfets so I could scope them.  I didn't change anything else, but when I fired it up, the idle current is only  about 14 amps instead of the 26 amps that I had previously!  Then, when inserting the crucible, the current went up to  36 amps instead of the previous 40 amps.
 The gate signals are all pretty good looking square waves and they are all the same.
Why the change???---  I don't know, but all I did was to solder those gate leads in place.  Bad joints???

The aliexpress vendor didn't seem to pay any attention to my problem description.  They emailed me back to say:
"friend , can you tell us in aliexpress message ?"
Aliexpress customer service, by the way, seemed very concerned when I chatted with them about the issue earlier.  But all they did was to pass my concerns on to the vendor.

I did make a short video of the the above and may put it up on youtube as a troubleshooting aid.
If anybody who is viewing this message has one of these 1800 watt or 2500 watt ZVS heaters running, please tell me what YOUR idle current is.

Thank you,
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 11, 2019, 05:10:39 AM
Well, it is finally up and running properly.
 Drawing about 4 amps idle current (still using 48 volt power supply), then goes to about 35 amps with crucible fully inserted.  Now need to mount a fan on the radiator and do a few other things.  Hate to say it, but I am not certain what caused the difficulties.  Later in my troubleshooting, I discovered that two adjacent turns of the work coil were touching, but separating them didn't help.  I took the fans off to get at the coil so I could insulate it with high temperature fiberglass sleeving.  When I put it back together, it works fine.
Maybe a cracked foil on the board?  Don't know for sure.

Next steps if it continues to work properly:
  Shoot for max current of about 40 to 45 amps.
Place fire brick below coil
  Try some steel  (to max out current) and to see how quickly it heats things to 1200° C.
1" Sq.
3/4" pipe  (pipe heats real fast since its a one turn "coil"
Melt some copper.
   Make a crucible cover. 
   Find/make tongs for pouring.
   Make a simple open face mold?

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on January 14, 2019, 12:22:07 PM
My first impression of the 1800 Watt IH is not great.

- The whole PCB is bent in a curve due to the fans+mount is bigger than the holes for it on the board.
- One of the input inductors had broken off the PCB from rough shipping.
- Connector to the fan controller was destroyed and broken off the board.

Returning the product is more expensive for me than its worth, so I filed for a refund and seller only gave me a laughable 5$ refund out of 50$. For that he will earn a negative review. For others warning, I bought from "Gowin Electronic" ( http://www.ebay.co.uk/usr/gowin_electronic?_trksid=p2057872.m2749.l2754 ): https://www.ebay.co.uk/itm/1800W-40A-ZVS-Induction-Heating-Board-Module-Driver-Heater-With-Heat-Sink-Kit/312211105813?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

Shipping was fast, seller communicates fast aswell, but they are not giving up any money without a fight.


I will repair and test it later.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: the_anomaly on January 14, 2019, 01:15:53 PM
I have also just purchased one, Yosoo 1000W by huhushop on Amazon.  Bought it Sunday morning, got it delivered Sunday evening.. man I love Amazon haha.

Looks ok except one of the FETs is not fully secured to its heatsink; the screw is only part way threaded in.  Also no thermal compound, Mads does your unit have thermal compound on the FETs?

As long as I can use it to heat stuck bolts on this outboard engine I am repairing I will be happy.  I've been tempted to build Uzzors PLL induction heater http://uzzors2k.4hv.org/index.php?page=pllinductionheater1 (http://uzzors2k.4hv.org/index.php?page=pllinductionheater1) but I took this gamble to hopefully save time and money. 

One thing I've been warned about using ZVS induction heaters is you need a power supply with a fast turn on transient otherwise oscillation may fail to start and resulting in both FETs turning on at the same time.  Probably best to connect the power supply to the induction heater through a relay so you can turn on the power supply, let it fill its output filter caps and then activate the relay.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 14, 2019, 05:08:38 PM
Mads-  Sorry to hear about your shipping issues.  I did get one or two of the 1000 watt units where one lead of  the 470 5 watt resistor had broken off the board.   The buck converter terminal on your board is different from any I have seen.  That makes me wonder how many actual manufacturers are in this game, and how much actual thought goes into each one's design variations.

the_anomaly-  I assume you will need to have some flexible way to get power and water cooling to your work coil if you plan to heat stuck bolts.  Please keep us/me informed on how you do that. 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 16, 2019, 04:26:38 AM
It's working fine now:
/>
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: the_anomaly on January 17, 2019, 02:01:36 AM
Nice video Pete!  I am actually not planning to use water cooling, probably 12-14ga solid copper wire.  I am hoping that using it only for a minute at a time will be sufficient to get a 7/16" bolt red.  Hopefully this will be a short enough duty cycle to keep the coil from melting.  If I do need water cooling I was planning to go to the auto store; they usually stock 1/8" copper tube for oil pressure gauges.  Since space is tight for my application, I need to use a smaller coil.

Is that a South Bend lathe in the background?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on February 18, 2019, 12:13:02 AM
Thanks for the kind words.
  I should have gotten back to you sooner.
I don't think the coil will melt in 30 seconds or so, but it will transfer a lot of heat back to the standoffs that it is connected to.
When you say "get a 7/16 inch bolt red", do you meant just the head or a whole bolt?  And, how hot is "red"?   If it just needs to glow a low to medium read color to the naked eye, you should be okay,  But if you need it to be bright red or orange. it can take quite a while, once the bolt's temperature passes the "curie point", becoming nonmagnetic.  That's because the heating rate will slow considerably.
  if you plan to use a work coil of less than 7 or 8 turns, please let me know how it works out. 

That's an Atlas 10F lathe in the background.  Wish it was a South Bend, but it works for me.

Pete Stanaitis
---------------   
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on March 07, 2019, 01:26:47 PM
I finally got around to get the IH out of the box and repair it, here is part 1 of a series of videos on this IH.

The future parts will be power supply building and testing.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 07, 2019, 03:51:16 PM
Hello Mads.
  Putting the standoffs on the bottom is a good Idea.  I wish I'd have thought of that.

It is interesting to see how many slightly different implementations of these devices exist.  Your device is the first one I've seen where the work coil is oriented horizontally.
On my 1800 watt unit, there was no heat sink compound, but they did have piece of thin, white material sandwiched between the Mosfets and the heat sink.  I thought about replacing that material with heat sink compound, but so far it hasn't been a problem.  I did blow up one Mosfet, but that was my own fault for paralleling one gate with too low of a resistance.  I knew better, but did it anyway.  Such is the life of a novice!

By the way, I just powered  up my 2500 Watt ZVS induction unit and did a youtube video of it yesterday.
It is here:
/>You will see my messy method of getting the water away from the board.

Pete Stanaitis
---------------

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on March 09, 2019, 08:28:48 AM
I built my first experimental cheap power supply and it was a great success!

12 laptop chargers put together for 40V 51A.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: flyrod on March 10, 2019, 04:09:25 PM
I did blow up one Mosfet, but that was my own fault for paralleling one gate with too low of a resistance. 

So do these boards use the same pull up resisters for gate drive?
 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on March 11, 2019, 08:22:21 AM
5 minute stress test at full power. 1800 Watt peak input power.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 11, 2019, 03:17:35 PM
Flyrod: "So do these boards use the same pull up resisters for gate drive?"
If you mean "do they use the same gate drive circuit as does the 1000 watt unit?"
Yes, for the 1800 watt unit that I have, at least.
The same 470 ohm 5 w wirewound resistor, 10K resistor, 12 volt zener  and FR307.

If you mean: "do all 4 gate drivers use the same pull up resistors for gate drive?"
The answer is still "yes".

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on March 15, 2019, 08:49:01 AM
A new 2200 Watt power supply made from server power supplies, and with steady cameras :)

Test of it in a long induction heater run is being edited and will come online in a few days

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on March 18, 2019, 10:35:04 AM
/>
This test is being run at 36VDC and everything stays cool over a course of 10 minutes, I would say it can run stable at this power input. The 5 minute test at 40VDC showed resonant capacitors heating up.

Test is powered by the "2200 Watt server power supply for induction heating"
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 18, 2019, 06:11:07 PM
That sure was a successful test.
    Is the "glowing" color that we see on the video that same as what you see in person?  Do you do anything special with the white balance of your video camera?

I am starting to work on establishing operational upper frequency limits on these devices.  I wonder how high they can go and still work okay.  Today, I certainly exceeded that number, so now I will have to "back up" to see where the real tripping point is.
So far, I hypothesize  that the data I gather from the 1000 watt unit will transfer directly to the 1800 and 2500 watt devices, since their gate circuits appear to be identical.  (Having said that, and seeing that two of three gate circuits are in parallel, I WILL have to verify that some day.)
By the way, I am using a cheap inductance meter for these measurements.  I think it is this one:
https://www.ebay.com/itm/LC100-A-High-Precision-Digital-Inductance-Capacitance-L-C-Meter-Tester-SMT-Clip/253959986062?hash=item3b2131db8e:m:mOpTSkV_q1oIM9pHKxVlPfw

Here's the record of this morning's test:
Objective:
  Determine the lower work coil inductance range for safe, useful operation of the popular 1000 Watt to 2500 Watt ZVS induction heaters.

Date: march 18, 2019

Subject for testing, March 18, 2019:
Testing a 10" U shaped work coil on the 1000 watt unit.

Method:
Make up a "one turn" coil to use as a worst case situation.  This coil is simply 25" of 1/4" OD copper tubing that is bent into a "U" shape, with about 3/4" between the arms.

Summary:
My initial conclusion is that the oscillation frequency observed during this test is somewhat beyond the safe or useful upper limit for this design.
  I hypothesize that the lower limit for work coil inductance is about 1.0 microhenry.
-----------------------------
Notes:

First I checked out the system using a "stock" 2" ID 6 Turn coil.  It ran, as it normally does,  at  about 6 amps of idle current.

Then I removed the 6 turn coil and replaced it with this "U" coil:
The measured inductance of this "U" coil is about 0.240 micro henry.
At present, the coil is actually several inches longer than 10.  I call it a "10 inch coil" because, once fully shaped, it would have a work area about 10 inches long for heat treating knife blades.
It has a spacing between the tubes of about 3/4".

For reference, these 1000 watt ZVS heaters come with coils that produce resonant frequencies between about 103 kHz and about 83 kHz.  These coils have measured inductances of about 1.05 to 1.38 micro henry.

Results:
The system  DID oscillate, but at about 154 kHz. 
And the gate waveform degraded significantly from the almost square shape it has with the "stock"  6 turn X 1 1/2" to 2" ID coils.  The signal almost looked like a sine wave with a slightly flattened top, both the leading AND the trailing edges.

  The idle current went to 16 amps.  I think this extra 10 amps is caused by the degraded gate signals causing the Mosfets to be running in linear mode for a good share of the time.  Also, the capacitors heated rapidly within a minute or so of "power on".


Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on March 20, 2019, 12:41:22 PM
That sure was a successful test.
    Is the "glowing" color that we see on the video that same as what you see in person?  Do you do anything special with the white balance of your video camera?

It seems I have rather good UV filters on my phone and DSLR, as you actually see a red glowing work piece in the video and not a purple glow. It does however not filter it all, it does glow more bright on the video than in reality and I guess you have the same experience.

I am starting to work on establishing operational upper frequency limits on these devices.  I wonder how high they can go and still work okay.  Today, I certainly exceeded that number, so now I will have to "back up" to see where the real tripping point is.
So far, I hypothesize  that the data I gather from the 1000 watt unit will transfer directly to the 1800 and 2500 watt devices, since their gate circuits appear to be identical.  (Having said that, and seeing that two of three gate circuits are in parallel, I WILL have to verify that some day.)
By the way, I am using a cheap inductance meter for these measurements.  I think it is this one:
https://www.ebay.com/itm/LC100-A-High-Precision-Digital-Inductance-Capacitance-L-C-Meter-Tester-SMT-Clip/253959986062?hash=item3b2131db8e:m:mOpTSkV_q1oIM9pHKxVlPfw

Here's the record of this morning's test:
Objective:
  Determine the lower work coil inductance range for safe, useful operation of the popular 1000 Watt to 2500 Watt ZVS induction heaters.

Date: march 18, 2019

Subject for testing, March 18, 2019:
Testing a 10" U shaped work coil on the 1000 watt unit.

Method:
Make up a "one turn" coil to use as a worst case situation.  This coil is simply 25" of 1/4" OD copper tubing that is bent into a "U" shape, with about 3/4" between the arms.

Summary:
My initial conclusion is that the oscillation frequency observed during this test is somewhat beyond the safe or useful upper limit for this design.
  I hypothesize that the lower limit for work coil inductance is about 1.0 microhenry.
-----------------------------
Notes:

First I checked out the system using a "stock" 2" ID 6 Turn coil.  It ran, as it normally does,  at  about 6 amps of idle current.

Then I removed the 6 turn coil and replaced it with this "U" coil:
The measured inductance of this "U" coil is about 0.240 micro henry.
At present, the coil is actually several inches longer than 10.  I call it a "10 inch coil" because, once fully shaped, it would have a work area about 10 inches long for heat treating knife blades.
It has a spacing between the tubes of about 3/4".

For reference, these 1000 watt ZVS heaters come with coils that produce resonant frequencies between about 103 kHz and about 83 kHz.  These coils have measured inductances of about 1.05 to 1.38 micro henry.

Results:
The system  DID oscillate, but at about 154 kHz. 
And the gate waveform degraded significantly from the almost square shape it has with the "stock"  6 turn X 1 1/2" to 2" ID coils.  The signal almost looked like a sine wave with a slightly flattened top, both the leading AND the trailing edges.

  The idle current went to 16 amps.  I think this extra 10 amps is caused by the degraded gate signals causing the Mosfets to be running in linear mode for a good share of the time.  Also, the capacitors heated rapidly within a minute or so of "power on".

Thanks for doing these experiments with lower inductance work coils and the resulting frequencies and currents. As expected it heats up a lot from increased switching losses due to more time in the linear region, but also it did not explode! I had expected it to burn to ground from your U turn coil :)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on March 21, 2019, 07:09:05 PM
Hello everyone!
Another induction heater enthusiast here... I recently purchased the 1800 w unit and was going through all the info everywhere (that spaco.org website was of enormous help, many thanks!)
But anyways... there are several things that I am not so certain about.
First of all, seems like everyone is using switching power supplies (servers or that crazy combo of a dozen laptop power bars=) ... I looked up the price + shipping for the server supply and shipping costs are more than the unit. I started to think, what can I power this with, all the high current supplies are quite expensive.
I got an idea to use one of those "voltage converter" units (europe/us) which is essentially a 1:2 transformer depending which side you connect to. So it should half our 110V into like 55V which is close enough. I am thinking of connecting a diode brigde/cap to that to make it a nice ripply DC and then use a zener to regulate final output to no more than 48 V. Does that sound sensible? As I understand, the ZVS unit will only draw as much as it needs, depending on what is going on the the work coil.
The other thing is, my fans are connected to the board, there s no controllers or anything in the vicinity of the connection - . Debating whether to use a separate supply with them or not if I go for 48 V. They even ziptied the wires to the posts.
When I get all the parts, I will make a youtube video using all this=).
 



 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Hydron on March 21, 2019, 07:51:23 PM
Unfortunately that's not really a viable way to build such a supply. Firstly those voltage converter units are normally autotransformers, with no isolation from mains input - you have to treat it just like you would mains (regardless of voltage) so very unsafe for a job like this. Secondly any linear regulator on the output would have to be very large to deal with the current.

As for the availability of server power supplies, they do sometimes come up cheaply when an older server is scrapped and sold off in bits. As you said though, shipping can hurt. If you can't find something else suitable then a few large lead acid batteries in series could be a candidate at least for use in testing.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on March 21, 2019, 11:04:48 PM
Sadly, I already purchased the unit=( and only then came accross this forum. Oh well, I can return it or...
I could'nt find any pictures of the insides, but a closer look at the ebay posting - it looks like it's a torus inside. Hard to tell from the picture. Maybe there is an option to modify it into a proper transformer with prim and sec windings.
Regarding zeners, I am thinking to wire about a dozen of them in series which should crank up the tolerable wattage to about 14W. Also possibly immerse them in something cold if needed.

Well , I understand it all sounds like not a great idea, but since the unit is being shipped I can at least peak inside and maybe ship it back, and stick to SLAs or something.

I knew this is going to be an adventure...
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on March 22, 2019, 10:56:16 AM
Hi Peter and welcome to HVF

I got a good collection of server power supplies and have more than I need. I can ship 2kg for 15 Euro, it might take 2 packages to ship 3 of them, but send me a PM with your offer for 3 power supplies that would give you 36V at 60A, you could also tinker with them to adjust them up to about 40V in total.

But generally you are not going to get good results with a Royer IH when using a weak power supply, this oscillator will eat all the power it can get and will just pull down the voltage of the power supply if its too weak, which can lead to failure if the voltage drops under 10VDC due to linear switching/uncertain gate state.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: RocketScienceSmurf on March 22, 2019, 11:25:19 AM
Hello everyone!
Another induction heater enthusiast here... I recently purchased the 1800 w unit and was going through all the info everywhere (that spaco.org website was of enormous help, many thanks!)
But anyways... there are several things that I am not so certain about.
First of all, seems like everyone is using switching power supplies (servers or that crazy combo of a dozen laptop power bars=) ... I looked up the price + shipping for the server supply and shipping costs are more than the unit. I started to think, what can I power this with, all the high current supplies are quite expensive.

You could also do like I did and just take an old welder. Voltage and current ratings are often close to what these inductions heaters use.

https://highvoltageforum.net/index.php?topic=32.0 (https://highvoltageforum.net/index.php?topic=32.0)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 22, 2019, 03:46:16 PM
You might also try rewinding some 1000 watt Microwave Oven Transformers  (MOT).  You could rectify their output.  I don't know how ripple free the power to these induction heaters needs to be, so I don't know how much capacitance it would take to smooth it.  But, if you did use lead acid batteries, they  would be a good ripple reducer, I'd think.    Then your "MOTs"  would simply be acting as a battery charger.
   I would suggest simply going the MOT route with a lot of capacitance  (just under the amount that would trip the mains breaker on power up), but, if you have to buy them, you might as well simply buy the server power supply anyway.
  Depends on how much tinkering you want to do.
  I tried to find the unit that you (badpeter) ordered on Ebay.  I think I found it, since it was the only one I saw that had the fan wires zip tied to a post.  I didn't see any place on the back side of the board where   the buck converter parts might be located.  But it could be hiding in there someplace, I suppose.
  Was it this one?:
https://www.ebay.com/itm/1800W-ZVS-Induction-Heating-Board-Module-Flyback-Driver-Heater-Tesla-Coil-Fan-US/123420945506?epid=2291469563&hash=item1cbc762c62:g:D2UAAOSwFnxbwEuo
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on March 22, 2019, 08:19:21 PM
Microwave is an option that was considered, if I find one laying around, definitely ripping out the transformer. Not too certain how much work i'd need to do winding the thing to the proper voltage though. I got lots of caps, so can add quite a bit of capacitance. And yes, that looks like the unit I got, except it was a different seller. Board is covered with finish so no way to trace anything. And strange words on the back "Is strictly prohibited". I guess we'll never know what is...=)

Re server power supplies, I am potentially interested. Will send ya a message!

By the way that voltage converter unit that is still in shipping is rated for 3 kw, so it should supply enough power even for a bigger unit. Here i am limited by my 40 amp breakers in the apartment. Will have to connect it to the outlet on the stove, if i decide to use it at all.
I will test everything and run a dummy load trying to get 40 amps at correct voltage, to see that it is capable of producing that.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 23, 2019, 01:50:47 AM
MOT:  There are plenty of youtube videos out there abut rewinding MOT's.  Good that you have lots of capacitors.

Work safe and "keep one hand in your pocket" when working with high voltages.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on March 25, 2019, 03:45:43 AM
I received my "donut" and I think I will keep it for now! (picture).
It is a center tapped auto transformer, nice thick wire. Halves the voltage as expected. I  DC-fied the output via a bridge and recorded the following:
no load=57v,
49.1V at 5.2A  (258 va)
41.4V at 11.4A (465 va)
this is far from the advertised 3000VA, but i think V*I can still get a bit better with higher current, towards the middle of the range. I cannot however measure higher currents at the moment to confirm that.

What i can do is unwind some of the wire to make it exactly 48 volts instead of 55. But seeing how I kind of get 48 v at 4 amps anyways, do I even need to do that? maybe a bit of overvolt is alright.   

Once i get the proper high current diode bridge from china, I think can fire up the induction board!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 28, 2019, 04:27:17 AM
Even if you get 12 or 15 amps at about 40 volts, you have to subtract about 6 amps for idle current , leaving only 6 to 9 amps to do heating of the work.  That isn't very much.  That's only about 240 to about 360 watts.  Something doesn't sound right .
  I would also be concerned about using an autotransformer without an isolation transformer in between.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Hydron on March 28, 2019, 09:34:40 AM
Running direct from mains is extremely unsafe for something like an induction heater where you may contact the work coil by mistake.

There have been plenty of other suggestions that are much safer, go for one of those!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: RocketScienceSmurf on March 28, 2019, 11:00:02 AM
Even if you get 12 or 15 amps at about 40 volts, you have to subtract about 6 amps for idle current , leaving only 6 to 9 amps to do heating of the work.  That isn't very much.  That's only about 240 to about 360 watts.  Something doesn't sound right .
  I would also be concerned about using an autotransformer without an isolation transformer in between.

Is the efficiency and idle current constant? Ie, if you have 6A idle current, is the "wasted" energy constant even when you insert a work piece or is it possible that the efficiency actually increases when there is something in the coil to transfer energy into?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 28, 2019, 01:29:52 PM
Quote
Is the efficiency and idle current constant? Ie, if you have 6A idle current, is the "wasted" energy constant even when you insert a work piece or is it possible that the efficiency actually increases when there is something in the coil to transfer energy into?

No, I don't think the "wasted" energy is constant.    What I do know is that, when you place a work piece into the work coil, the components on the board get hotter, so there's actually MORE "wasted" energy at that point.  And, I don't account for it very well.
  You ask a good question.
  I don't know how I would measure  "EFFICIENCY increases".
At least with these ZVS heaters, there's a lot going on when you insert something into the work coil.  With the current draw from the power supply increasing and the frequency changing, a little or a lot, depending on the material that the work is made of.  Then there's the heat extracted from the components on the board by the fans, and, of course, the heat radiated by the work itself, some of which goes to the work coil cooling system and some into the air.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: DICKEYBIRD on March 30, 2019, 01:42:45 PM
'Morning all & Happy Saturday!  Newbie here. :)

Today I'm working on a PS for my recently purchased 1800W unit.  After seeing Mads videos, I obtained some HP DPS-750RB PS's and am performing the isolation mods as I type this.  I understand everything except for what to do with the ground pigtail that was captured under the of longer of the board mount screws.  Should it be clipped off & discarded or reinstalled under the nylon insulator?

I noticed on another forum that it's recommended to leave one of the series connected PS's un-modded (not isolated) but Mads modifies all 3 in his video.  It looks like either way will work but is one preferred over the other?

(Edited to add another query)
All the mosfet mount screws were completely loose (.5mm gap between it and heatsink!)  Is the white "paint" on the face of the mosfet enough of a heatsink compound or should I add some?

...and thanks for what you guys do here!  I have nothing to offer electronically but I am a hobby machinist & and lifetime tinkerer. :)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 30, 2019, 06:04:30 PM
I would not add any more heatsink compound.  Although most heatsink compounds are non conductive, make sure that none of it can touch the Mosfet leads, just in case.   If I understand the heatsink compound  "science" correctly, its only task is to fill in any tiny gaps between the device and the heatsink, to promote maximum heat transfer.
  Certainly do snug up the screws.
By the way,  I use a pair of "Snap-on 105 Ignition pliers"  to loosen and tighten the Mosfet mounting screws, since they are so hard to get at in the 1800 watt and 2500 watt ZVS units.

Capacitor safe working temperature-  I found this:
"The normal working range for most capacitors is -30°C to +125°C with nominal voltage ratings given for a Working Temperature of no more than +70°C especially for the plastic capacitor types."
  Even 70°C  (158°F)is well over the temperature where you can hold your hand on the part.
It would be good if some of you can read the temperature of the capacitors during extended runs to see how hot they actually get, and to see if the temperatures stabilize during a run.
In my case,  The capacitor temps do seem to stabilize close to acceptable values:
 [ Invalid Attachment ]
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on March 31, 2019, 09:46:10 PM
'Morning all & Happy Saturday!  Newbie here. :)

Today I'm working on a PS for my recently purchased 1800W unit.  After seeing Mads videos, I obtained some HP DPS-750RB PS's and am performing the isolation mods as I type this.  I understand everything except for what to do with the ground pigtail that was captured under the of longer of the board mount screws.  Should it be clipped off & discarded or reinstalled under the nylon insulator?

I noticed on another forum that it's recommended to leave one of the series connected PS's un-modded (not isolated) but Mads modifies all 3 in his video.  It looks like either way will work but is one preferred over the other?

(Edited to add another query)
All the mosfet mount screws were completely loose (.5mm gap between it and heatsink!)  Is the white "paint" on the face of the mosfet enough of a heatsink compound or should I add some?

...and thanks for what you guys do here!  I have nothing to offer electronically but I am a hobby machinist & and lifetime tinkerer. :)

I just left the little wire hanging out in the air, not connected, so I might aswell just have cut it off, they seems to work just fine without it.

You are right that one power supply can be left un-modded, the one that sits in the negative end of the chain, I am however not sure if that would put a limit on how many you can put in series, but I would personally not try for higher than 60 VDC with 5 in series. Solely based on assumptions about creeping distances on the PCB being relatively low for a 12V product.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Hydron on March 31, 2019, 10:23:26 PM
I'd personally ground (by not modifying) the supply in the middle of the chain - that way each end is only half the voltage with respect to earth. This would not work if you had to put a non-differential scope probe on the IH though.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: DICKEYBIRD on March 31, 2019, 11:51:05 PM
Thanks fellows for the learned advice!  I'll clip those wires since it will be less likely for my Klutzy self to short out something important.  I think I will leave 1 of the 3 PS's un-modded...that fits my laziness perfectly; only one more to go!

The project is coming along pretty well. I was able to remove the original MOSFET screws & replace them with M3 SHSC's so that they could be torqued with a cut-down allen wrench without de-soldering any parts for access.  The screws on the DC terminals were replaced with SHSC's as well so they could be tightened properly.

I "dumpster-dived" a dead automotive battery charger at my day job recently.  After stripping out the old dead Chinese electronic parts, I am now installing some (hopefully not-so-dead) new Chinese parts.  I procured a small, cheap auto heater core and am installing it behind the louvered side panel on the charger case with a couple 12V PC fans for cooling, a small 12V pump & a 1 gallon can of distilled water.  I should have everything needed for the IH in one neat package that can be rolled around the shop & stored easily when not being used.

I'm scrounging around now for some wire & switches to finish it up next weekend & if it works I'll post up some pics & maybe a video.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on April 01, 2019, 04:02:35 AM
^Looks like everybody's got different struggles=)
So over the weekend, I put everything together and it works... (numbers in my previous post were completely nonsensical due to me forgetting about RMS vs Pk thing). I was confused, but hopefully not anymore. 

The torus has 184 wire turns which makes it about under 1V per turn. It was converted into a regular transformer, tapped on 54th coil which gave the correct 50 V when made DC (idle). I hooked up the unit and the voltage dropped to about 38V at 6 amps. I have put bolts and coins in the coil. They do heat up to orange glow. Small Items seem to draw no more than 10 amps.
I also inserted a graphite crucible which draws larger current and voltage drops to 20V.
Fans seem to be alright, and are rated for 24 V by the way.

Next steps would be - getting a larger, longer coil to go around the whole crucible. I think I might go for proper switching power supply since I found a good deal on ebay.
I cannot understand why transformer voltage drops so much with even small-ish current - what I read on the internet regarding transformers' voltage regulation issue sounds a bit contradictory. It should be no more than 5% and that is only when transformer is saturated. Definitely not the case in my case.

Pictures attached=)
(https://highvoltageforum.net/proxy.php?request=http%3A%2F%2F%5Battachment%3D1%5D%5B%2Fattachment%5D&hash=0c82d6fed55acc164f4acdd9cd8f1251675d9e42)
(https://highvoltageforum.net/proxy.php?request=http%3A%2F%2F%5Battachment%3D2%5D%5B%2Fattachment%5D&hash=dbe476cda4102b1e3917d625310626af5e8b8815)


 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 01, 2019, 04:08:02 AM
Glas to see that things are coming along, DICKEYBIRD.

For what it's worth,  I just uploaded my next, and possibly last video about my adventures with the 2500 Watt ZVS Induction Heater.
It is here:
/>
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: DICKEYBIRD on April 01, 2019, 09:03:01 PM
Great video Pete, thanks for posting!  From what you've discovered so far, do you think  it's possible to heat smaller pieces to high temps without making a smaller work coil?  Or do smaller pieces just take longer with less power?  The intended use of my 1800W unit is the hardening of O-1 tool steel stock -  (approx. 1/4" to 1/2" dia, 1" to 2" long)  Maybe a "focusing" coil or device to position inside the existing coil? 

Again, thanks for your enlightening videos!
Milton
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 01, 2019, 10:20:41 PM
Re: Heat treating small pieces pf O1 tool steel:
I think pieces as small as you describe can easily be heated to the required temperature:
"Heat Treatment. The heat treatment requires O1 steels to be slowly preheated to 649°C (1200°F) and then heated at 788-816°C (1450-1500°F). Then these steels should be held at the same temperature for 10 to 30 minutes and finally oil quenched."

That temperature is just about the curie point anyway, so current should not be dropping much.  Note the pre-heat conditions, too.
As far as the time at temperature, as you probably know, this relates to getting the heat all the way to the center of the material.  So, with pieces as small as you mention, you'd only need a couple of minutes to do the job.
  Another way to handle pieces that require higher temps is to put them into an insulated graphite crucible  You should have no trouble getting over 2000°F  with steel that way.  In one of my 1000 watt videos. I get a few hundred grams of 1" diameter steel to over 1200°C by insulating below, the sides, and above the work coil.
  So, there are many "ways to skin the cat".

These comments are from a practitioner (me). not from a scientist.  Curie point varies with carbon content and with content of other metallic alloys.   Required hardening temperatures vary, too, with alloy levels.
 
 i have even considered using a hollow graphite tube to help concentrate the heat.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 14, 2019, 12:23:14 AM
New video-   Work piece size versus coil size:
/>These principles apply for all sizes of ZVS induction heaters, as far as I know.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: DICKEYBIRD on April 14, 2019, 03:58:34 AM
Very informative video Pete; thanks for taking the time to put it together & post it!  I'm itching to get mine going but unfortunately I haven't had time to work on it much lately.  A neighbor brought me a some more machining jobs to do for his company in my spare time.  The ol' hobby machine shop can always use the influx of cash!

Where did you find the braided fiberglass sleeve you have on your work coil?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 14, 2019, 03:00:18 PM
Fiberglass sleeving:
https://www.ebay.com/itm/1meter-Manhattan-F240-2AWG-6-68mm-240-C-Fiberglass-high-temperature-sleeving/161643155584?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

They sell it my the meter.  I usually buy 3 meters at a time and, so far, they send me one whole 3 meter length.  Which is a good thing.
In the past, I have used 6mm sleeving instead of that 6.68mm stuff, but it's a bit harder to thread onto the  1/4" tubing that I use for coils.
If you look for "high temperature fiberglass sleeving" you will get lots of choices.

By the way, I have written and "filmed" lots of stuff for the 1000 watt ZVS induction heater that relates to the 1800 watt and 2500 watt models.
Go here to see it:
https://spaco.org/Blacksmithing/ZVSInductionHeater/1000WattZVSInductionHeaterNotes.htm

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 16, 2019, 01:20:34 AM
I just assembled an Excel spreadsheet of data about the latest induction coil power transfer video, along with some data from earlier tests.  I am sure to add to this spreadsheet as new information dictates.  It is probably not the best presentation of the data, but I try graphing the results in a few different ways to make some sense of it.
It is here"
https://www.spaco.org/Blacksmithing/ZVSInductionHeater/ZVS2500Watt/Work Coil current and frequency change with size and shape of material.xlsx (https://www.spaco.org/Blacksmithing/ZVSInductionHeater/ZVS2500Watt/Work Coil current and frequency change with size and shape of material.xlsx)

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: DICKEYBIRD on April 18, 2019, 04:32:55 PM
Thank you Pete; I have some on the way now.  That should check one more thing off the list of magic smoke releasers on this project! ;)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on April 18, 2019, 10:13:06 PM
Great video and excel sheet Pete, thank you for sharing. I also have some fiberglass hose in the mail, however long its going to take from China, but then I will box up the heater with the 3 server power supplies.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on April 22, 2019, 04:07:07 AM
I finally made something that works...
Here is my somewhat new setup with all previous mistakes accounted  for (I had no idea what I was doing previously, hope that has changed haha)
a.k.a the guide to using unregulated power supply for maximum benefit!

1. I moved all my operations to my bench connecting directly to the stove outlet via a proper thick cable for high amps.
2.I finally found the sweet spot where to tap the transformer accounting for voltage sag. The idea is to overvolt it just enough so that when the unit is connected to IDLE coil voltage drops to fifty volts (the voltage that the unit is rated for).
3. While doing so, I burned both 24v fans, so now I am using external fans from a pc PS.
4. I managed to melt aluminium in about 50 mins, with a power output between 1700 to 1500 w, using my diy kiln enclosure from another project, for heat insulation. I can measure the power with ac meter from ebay which is super handy. It can only be used on the transformer AC side. Still waiting for shunt dc meter to arrive for unit -side measurements.
5. The setup is stable and nothing is too hot. The transformer is output is taken as two secondaries connected in parallel, sent to a pair of fifty-amp rectifier bridges, with heatsink immersed in water.   
6. I made a couple new coils. The five turn work coil makes the unit run between 40-50 kHz, as it is seen on the oscilloscope. As I understand, smaller coil means higher current, higher frequency of oscillation and potential higher energy draw.

I think I can finally do copper and cast something useful when the THERMAL blanket and the sleeve arrives.

(https://highvoltageforum.net/proxy.php?request=http%3A%2F%2F%5Battachment%3D1%5D%5B%2Fattachment%5D&hash=0c82d6fed55acc164f4acdd9cd8f1251675d9e42)

(https://highvoltageforum.net/proxy.php?request=http%3A%2F%2F%5Battachment%3D2%5D%5B%2Fattachment%5D&hash=dbe476cda4102b1e3917d625310626af5e8b8815)

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 23, 2019, 02:55:40 AM
It's good to see that you have your system working, badpeter.
   I still think I will get some linear power supplies going one of these days.

Over the last few days, I have been preparing a copper melting video and I finally, just today, got it up on youtube.
It is here:
/>

We may be able to compare results when you get your DC current meter.  I think you should monitor DC voltage too.

I suggest that you monitor Mosfet gate signals carefully as you use "smaller" coils that increase tank circuit frequency substantially.  I think you will see that the gate signals round over more as the frequency goes beyond some certain point. This means that the Mosfets are spending more of their "on" time in the linear region which is a very bad thing.  Yes, this condition will cause more current to flow, but that extra current will be heating the Mosfets not the work.  Also, since you are using a linear power supply, you may see the voltage drop well below that 50 volts as you increase the load.  This in itself has a negative effect on the gate signals.
 See the December 8, 2016 entry that is about half way down this page:
http://spaco.org/Blacksmithing/ZVSInductionHeater/1000WattZVSInductionHeaterNotes.htm

Pete Stanaitis
---------------


It is Here:
 
/>
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: DICKEYBIRD on April 24, 2019, 05:24:05 AM
One comment & one question please:

1) For anyone that's building a power supply using 3 or 4 of the HP server PS's like Mads showed in his video...go ahead & isolate all of them as he did.  I read on a couple posts elsewhere that it's a good idea to leave one of them unmodified.  I tried that & it didn't work.  The last one in the series would initially turn on then shut down after a few seconds.  I then did the isolation mod & all is good.  Whew, no magic smoke was released!

2) Question:  Since there is an excess of power available with the 3 (or 4) x 750W PS's, would it be OK  to run my radiator 12v cooling fans and water pump from one of them?  I haven't measured the total power usage but I'm thinking about .5A total for the 2 fans & maybe 1 to 3A for the w/s washer pump I'm using.  Is it unwise to do it that way?  A noob question for sure! :-[
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 25, 2019, 04:57:18 PM
Mads-  You mentioned that you will soon "package" your induction heater.  I will be looking forward to the method you use.
Just recently, I see that the Chinese seem to be stepping up their game with the 1000, 1800 and 2500 watt ZVS induction heaters:
https://www.ebay.com/i/163558914286?rt=nc&var=462944781862&_trkparms=aid%3D222007%26algo%3DSIM.MBE%26ao%3D2%26asc%3D20160908110712%26meid%3D521ef284b2c7416bb345b6aa93d150c5%26pid%3D100677%26rk%3D6%26rkt%3D19%26sd%3D163558914286%26itm%3D462944781862

They appear to be simply packaging the parts into a nice looking box, and maybe adding a DC SSR?

The price of these units is about half the price (or more in some cases) of the LH-15A  7.5 KW Induction heater that is being sold by many vendors:
https://www.ebay.com/itm/15KW-30-100-KHz-High-Frequency-Induction-Heater-Furnace-LH-15A-110-V/182826308639?epid=17031092066&hash=item2a914c141f:g:nYUAAOSw9TdZ4A~z
By the way, this one is the lowest priced unit I have seen.  There is one caveat----  you need a TIG welder type of water cooler for it.

dickeybird:
Regarding your question about using "excess" power to run your fans and pump---       I suppose that would work, but I prefer to take the load off of the power supply. 
  It is all too easy to stick something into the work coil that causes a high current flow.  If you watched my recent 2500 watt videos, you may notice that at least once I inadvertently mentioned the current going over 50 amps when I put a large piece of pipe to far  into the work coil.
  Of course you will need to get a buck converter if you do use the heater's power supply if your fans or pump run at lower voltages.  In my case, so far at least, my fans and pumps all run at 12 volts so one 3 amp buck converter or one external power source may work well.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 01, 2019, 09:05:38 PM
 Congrats on the successful melt!  I saw another individual on YouTube melt about 1.5 pounds in approximately 20 minutes using 1600 watts and a similar setup.  He gives little if any info so not sure about any details of his test.  That is about the same time it takes me to melt same amount using my small DIY gas fired foundry furnace.  Do you think your melt times would be similar if you did 1.5 pounds?  If you can poor 1/2 pound of copper after 7 minutes and if melting charge time is proportional then about 20 minutes would be a good estimate?   Thank you for all the info you have put out there.  Good to know that the scientific method still exists - great stuff.

Just had another thought.  The 2500 watt units are so cheap now maybe we could buy two, then rearrange the coils on each such that they both wrap around one crucible thereby theoretically doubling the power- your thoughts?  Not an engineer, more of a DIY idea-type guy.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 02, 2019, 05:10:31 PM
Re: "---time to melt 1.5 pounds of copper----in 20 minutes?---":
I don't know for sure, but I think it would not take the whole 20 minutes because part of the time is taken up by heating the crucible itself.  But I don't know exactly how it would go, since 1.5 pounds of molten copper  would pretty much fill that crucible right to the top.  As you can see, I filled the crucible as full as I could get it and it melted down to only about 1/3 the depth of the crucible.  That means that I'd have to continually add bits of copper as it melts.  That could make things worse because I'd be uncovering the crucible more often, but it might be better because the unmelted bits of copper might keep the current up a but higher.

Re: doubling the 2500 watt heaters:
I have been asked that question before.  Personally, I don't think it would work.  I'd be afraid that the magnetic fields would interact in a bad way.  But, maybe they would sync up together?
One other issue is that you would still have to obtain an additional power supply, cooling system, etc..

I hope that some of the smart HV/Tesla guys on this forum will comment on this.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on May 02, 2019, 09:57:40 PM
Mads-  You mentioned that you will soon "package" your induction heater.  I will be looking forward to the method you use.
Just recently, I see that the Chinese seem to be stepping up their game with the 1000, 1800 and 2500 watt ZVS induction heaters:
https://www.ebay.com/i/163558914286?rt=nc&var=462944781862&_trkparms=aid%3D222007%26algo%3DSIM.MBE%26ao%3D2%26asc%3D20160908110712%26meid%3D521ef284b2c7416bb345b6aa93d150c5%26pid%3D100677%26rk%3D6%26rkt%3D19%26sd%3D163558914286%26itm%3D462944781862

They appear to be simply packaging the parts into a nice looking box, and maybe adding a DC SSR?

The price of these units is about half the price (or more in some cases) of the LH-15A  7.5 KW Induction heater that is being sold by many vendors:
https://www.ebay.com/itm/15KW-30-100-KHz-High-Frequency-Induction-Heater-Furnace-LH-15A-110-V/182826308639?epid=17031092066&hash=item2a914c141f:g:nYUAAOSw9TdZ4A~z
By the way, this one is the lowest priced unit I have seen.  There is one caveat----  you need a TIG welder type of water cooler for it.

I will reuse a old cabinet and other stuff I already got in my workshop, I will see what I can house it in, without buying any new parts, recycling for the future!

That is indeed getting cheap for boxed heaters with warranty, it is almost like they want everyone to own one :)

Just had another thought.  The 2500 watt units are so cheap now maybe we could buy two, then rearrange the coils on each such that they both wrap around one crucible thereby theoretically doubling the power- your thoughts?  Not an engineer, more of a DIY idea-type guy.

The issues here would be proximity heating between the two coils, that they simple induction heat each-other and they would properly interact on each-other in a destructive way if they are out of phase from running different frequencies. I am not too sure on this, but I doubt it will end up well with so little control over the switching that we have in a Royer oscillator.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 03, 2019, 04:20:42 PM
Do you think we will ever see them make a 4k bare bones unit?  I see they have a 6k unit in fancy box with overload, meters and pump (no power supply) for $900 shipped, but as far as I know no one has tested it yet.  maybe because it would overload many people's homes?

DIY version:

/>
I want that.  array of big caps may be a better method, though more expensive probably. 

this MIT EE student gives a bill of materials and an unreadable schematic for a 3k with some good tips:
https://www.instructables.com/id/30-kVA-Induction-Heater/
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: shortyg83 on May 03, 2019, 09:37:49 PM
Glad I found this post it has a lot of useful information.
I ordered the 2000W ZVS model. It should arrive today.
It has a built in display that will read voltages, temps etc. And says it has protection built into the unit.
I bought it was a power supply that works on 120v or 240v. I live in the USA so for 240V I will have to hook the power supply up via split voltage, which should be the same as single phase 240 in other countries.
I will report back with some results once I get some testing done.

https://www.ebay.com/itm/ZVS-Induction-Heater-Heating-2000W-45A-Module-Power-Supply-3000W/273107970557?epid=11016480565&hash=item3f9680e9fd:g:A4MAAOSwG1Vb7Nvk
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 06, 2019, 10:43:50 PM
Interesting 2000 watt ZVS induction heater module.
I see that "they" are producing many new variations lately.
It's nice that this one has the LCD readout for several variables.
"Protection" has to be a big plus.  It is great that the system can tell you which Mosfet has failed.
I am a bit surprised to see that it has only one fan and that the fan doesn't blow on the capacitors that are closest to the work coil.
I also note that the capacitors in the video linked on the item's "for sale" page are very different from the ones shown in the item link itself.
Lastly, I wonder why the bolt in the work coil didn't seem to glow during that 5 minute video.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: shortyg83 on May 06, 2019, 11:39:34 PM
I made a quick video just to see if it worked before the final setup.
It heated up a knife I am making from an old file in just a few seconds glowing red.
The video on the ebay listing is before they changed the device. The ones they are selling now look like the same from the ebay pictures and from the video I am linking.
I have no idea how the bolt didn't glow. I see in the sellers video the screen never reads more than 550Watts. It could be the bolt was too small or didn't contain the right metal content.
In my first test it got up to over 1400w pretty quickly and the metal was glowing. I will make more videos once I get everything in better posistion.

I will not that I need a bigger water container. Even in my quick video the water running through the coil got quite warm so I want to get a tank that holds at least 1 gallon of distilled water.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 07, 2019, 12:54:12 AM
Great first demo!

It sure did heat that blade!
Looks like you have a good start toward a nice enclosure.

  I use radiators with fans to keep the water cool.  We have a local on-line auction house that has various kinds of radiators from time to time so I get then pretty cheap.  An old automobile interior heater should do a good job.
I did some real basic calculations on water heating:
            
How much will the water heat up?            
assume an idle current of         6   amps
Assume half goes to the coil         3   amps and the other half goes to Mosfets, caps, etc.
   power supply volts      48   
      Watts   144   
   Btus @ 3.4/ watt      489.6   
            
            
One btu raises the temperature of one pound of water one degree F            
One gallon of water weighs ~         8   pounds
Temp F rise per hour, degrees:         61.2   with NO load.  (No heat loss to the surrounding area)

The current distribution is just a guess, but just to have some idea.
You probably know this, but it is important that your cooling water goes INTO the bottom of the coil and comes out the top.
I have boiled water more that once by getting this wrong!

Pleae let me know if I got the calcs.  wrong.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: shortyg83 on May 07, 2019, 04:50:37 AM
Great first demo!

It sure did heat that blade!
Looks like you have a good start toward a nice enclosure.

  I use radiators with fans to keep the water cool.  We have a local on-line auction house that has various kinds of radiators from time to time so I get then pretty cheap.  An old automobile interior heater should do a good job.
I did some real basic calculations on water heating:
            
How much will the water heat up?            
assume an idle current of         6   amps
Assume half goes to the coil         3   amps and the other half goes to Mosfets, caps, etc.
   power supply volts      48   
      Watts   144   
   Btus @ 3.4/ watt      489.6   
            
            
One btu raises the temperature of one pound of water one degree F            
One gallon of water weighs ~         8   pounds
Temp F rise per hour, degrees:         61.2   with NO load.  (No heat loss to the surrounding area)

The current distribution is just a guess, but just to have some idea.
You probably know this, but it is important that your cooling water goes INTO the bottom of the coil and comes out the top.
I have boiled water more that once by getting this wrong!

Pleae let me know if I got the calcs.  wrong.

Just out of curiosity how would the direction of water flow through the coil change how well the water cools? Is it so any steam that may exist escapes upwards?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on May 07, 2019, 10:59:10 AM
I want that.  array of big caps may be a better method, though more expensive probably. 

I got some large induction heater caps for sale, real bargain to the size of them :)

https://highvoltageforum.net/index.php?topic=56.0
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 07, 2019, 02:45:24 PM
Re:Just out of curiosity how would the direction of water flow through the coil change how well the water cools? Is it so any steam that may exist escapes upwards?
Answer: If the water flows into the top of the coil. it can simply run down to the bottom coil and out to the "drain" without filling the coils.   When I change coils, I have to get all the bubbles out.  This flow orientation also helps to get ALL the bobbles out of the system before I turn the power on.
I did this wrong on one of my earlier "1000 watt" videos and you could see steam and water jumping out of my radiator's expansion tube. 
The MAIN idea is to eliminate the possibility of any steam forming in the first place due to an incompletely filled coil.

Mads:  Please remind  about the specs and prices for your capacitors----
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: DICKEYBIRD on May 07, 2019, 05:22:14 PM
Great info, thanks all!  Never even considered the water direction issue; thanks Pete.  Due to enclosure space issues, I'm using a 1 qt. distilled water tank for the aluminum auto heater core fitted with 2, 12V muffin fans and a water pump.  I'm hoping that will be enough but if it isn't, I'll add an additional external 1 gallon tank.

I'm using this cool little thermostat to read coolant temp & switch on the fans when needed: https://www.amazon.com/gp/product/B019I3YCFS/ref=ppx_yo_dt_b_asin_title_o01_s00?ie=UTF8&psc=1 This speed control for the pump: https://www.amazon.com/gp/product/B007TH4EN6/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8&psc=1 and this "wattmeter" to monitor power: https://www.amazon.com/gp/product/B0753DPC2D/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8&psc=1
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on May 07, 2019, 07:58:39 PM
^wow so much activity here!
Well, I managed to melt my first copper! so yay...
I rebuilt the PS with a powerful fan underneath, using the original box. It looks very neat.
But most importantly, insulation does miracles.

The data is the following:
845 g of copper melt in a graphite crucible (so-called "3kg-crucible" on ebay)
in about an hour with an output of just under 1kW.

the question is, how do I draw more power? I cant make the coil any smaller because of crucible+insulation being quite thick. I can have more turns. I was thinking, what if I do TWO coils, meshed together (not touching), turned the same direction.
Would that double the drawn power(some sort of principle of superposition)?

I plan to make aluminium bronze and cast a nice golden-looking dagger=)

Here are some pics:   
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: shortyg83 on May 07, 2019, 09:14:05 PM
So for some reason certain types of metal appear to be causing the induction heater to draw to much power and go into a protect mode. It is a 2000w heater but I am using a 3000W power supply and I was assuming it would only draw up to 2000W. But I put a piece of 440C bar stock in and if I move the bar through the coil at a certain speed where say length that is currently in the coil hasn't been heated at all yet the heater starts drawing 2200 then 2300 Watts and ends up going into protect somewhere just over 2300W. If I move the bar stock through slowly where it has time to heat smaller sections at a time it works fine.

Is there anyway for me to limit the power from my power supply where it can only goto say 2200 or so watts? It currently outputs 53.5V at 56.1 A DC.

The heater is 24v-65V with a max Current of 48Amps. I don't think these heaters like being run below 48V though.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 07, 2019, 10:15:30 PM
badpeter:
  You need thinner insulation so you can make a coil the more closely fits the crucible.
Like this:
https://www.ebay.com/itm/INSWOOL-2300-PAPER-Flexible-Refractory-Paper-1-8-x-50-partial-rolls-also/131815404577?hash=item1eb0cf7421:m:m03iqra7mYUl9GLx0gPJ9rA
 what is your Idle current with that coil?  I see that the turns are pretty far apart.  That should reduce its inductance and make it run at  a higher frequency which may not be helpful, since you are primarily heating graphite, not copper.   You may also have a higher idle current so the net current for power transfer to the load will be less.
It appears that you are only heating about half the crucible.  That can't be helpful, can it?   Why use a 3 kg crucible and not a one or two kg crucible? Or get one that is already insulated with a hard ceramic shell such as the one you have seen my use.

I think everyone who get serious about this stuff should invest in an LC meter like this one which costs USD$10.00:
https://www.aliexpress.com/item/Digital-LCD-Capacitance-meter-inductance-table-TESTER-LC-Meter-Frequency-1pF-100mF-1uH-100H-LC100-A/32829243371.html
Then you can make coils and be able to predict how well they will work, while not overtaxing your heater.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 07, 2019, 10:30:58 PM
shortyg83:
"for some reason---":  The reason is that various alloys suck up more power than other do.  Shapes that fit the work coil more closely also tend to soak up more power.  That's usually a good thing-----  more power transfer!
  No individual coil works for everything.  If you are serious about these devices, you will have to learn how to make coils to match the work.  Many of my earlier videos show these effects.
  Usually, ferrous metals (those containing iron) soak up more power than do non-ferrous metals.  That is true up to the point where they start to glow a medium red color, where, by the way, they also become non-magnetic.  This is called the "transformation temperature" or "Curie Point".  At that temperature, one of the major factors that makes them heat, goes away.  That's due to the cessation of hysteresis as the magnetic property diminishes.  You saw that effect as you moved the stainless steel piece through the work coil.  (440C contains approx.  82% iron and the rest is alloys).

Link to youtube channel videos for the 1000 watt heater which has several relevant videos:
https://www.youtube.com/playlist?list=PLVxWen9M87dBhvInCQ-3pELWdxCM3XH4I

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 07, 2019, 11:38:21 PM
Would insulation on the outside of coil rather than inside be better?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on May 07, 2019, 11:51:20 PM
Peterspaco:
I might try the thinner insulation, but the one I have is quite amazing. May be losing some efficiency with a thinner one.
Idle current: looks like i forgot to record it with the latest setup. will have to get back with that.
Crucible wrapping is one thing i didnt think of. Will try that next!
I hope to cast bigger things so even a 3kg crucible is too small. I made a whole sword with aluminium when I ran my propane mini kiln!

 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on May 07, 2019, 11:55:50 PM
hightemp:
If insulation is outside then water will definitely boil in the coil not to mention that the water be stealing a lot of useful heat away. The previous setup with bricks was somewhat like that, before I had the fiber insulation
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: shortyg83 on May 08, 2019, 01:24:14 AM
hightemp:
If insulation is outside then water will definitely boil in the coil not to mention that the water be stealing a lot of useful heat away. The previous setup with bricks was somewhat like that, before I had the fiber insulation

They make crucibles that come with ceremic sleeve insulation so you don't have to use all that. You can probably find one to fit your coil.

https://www.ebay.com/itm/Induction-Melting-Furnace-Graphite-Crucible-with-Ceramic-Shield/273657494243?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649&fbclid=IwAR0jaVvHhOtCpPofcWgwc7rDLkak0ZXkLETXEp42kzXw-Gjk7o_rHoHgLv4
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 09, 2019, 05:31:34 PM
Annealing Brass Cartridge Casings with a short (2- layer X 3 Turn) coil:
I just put this video on youtube last night:
/>
Many ammunition "reloader" guys ask about this, so I thought I'd try a coil design that might work.  (I am not a relaoder guy),  I see that some already do use these small induction heaters to do this task, but I am not sure they are running their systems in a sustainable way.
So my focus is to make sure that Mosfet gates turn on rapidly by using coil designs that keep the oscillation frequency to  acceptable (upper) limits.
 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 09, 2019, 08:09:24 PM
Peterspaco:
I hope to cast bigger things so even a 3kg crucible is too small. I made a whole sword with aluminium when I ran my propane mini kiln!

badpeter, keep up the good work, very curious to know how much copper you will be able to melt.  I am also interested in melting larger quantities (1-5 lbs esp.) knifes, swords, parts, etc.  but have not yet purchased this 2500W unit because AFAIK no one else has reported that they can melt more than 500 grams, at lest repeatedly.  So probably this design just does not have enough power to melt larger quantities (maybe it can melt a pound or two max if tweaked properly?)  Peterspaco has the best tests and has confirmed repeatable 500 kilogram melts with excellent results.  I'm hoping that someone shows what the practical limit of the melt size is, especially for copper casting alloys.  Obviously at some weight,  it will not melt a given quantity no matter how long you run it, and if takes too long (say 1 hour?) then the metal will probably pick up too much oxygen causing porisity, etc.   Will it melt 1,2, 3 lbs. in 15 minutes, 30 min, 60, never... ?   Peterspaco total power output is only approx 1600 watts when melting copper so we are already close to maxing out the power.  There is some headroom there - approx 700W that is still available - however this power is not being used because this unit can not output anymore amps to the crucible/copper load.  Possibly, if the crucible was a heavier gauge steal then it might utilize unit's full 2500 Watts, however, I don't think that you would ever want to melt copper in steal because it would absorb way too much iron causing embrittlement, poor properties.  I know very little about induction heating so tweaking the board to run hotter (larger caps,etc.) is beyond me.  For all I know it is not even possible or if so, prohibitively expensive.   
   We do have other alternatives.  Larger, though much more expensive induction furnace.  I did contact a regular industrial supplier of induction furnaces here in the states and was quoted $5000 for their smallest unit that can melt 4-6 lbs of copper in 5 minutes).  When I said that was way out of my budget he basically hung up on me.  Though he did say theirs are built way better.  A custom sized DIY propane/gas furnace will easily melt 5-30 pounds.  Resistance furnace is another option. Those 2-3 kilo gold/silver melting units that sell for about $300 - however I wonder if they are also under-powered for melting more than even a kilo or two of copper?   The thing that draws me to both electric induction and electric resistance melting is the noise factor - not bothering my nice neighbors.  Another possibility is some type of custom DIY arc furnace.  An arc type furnace may actually be the most efficient and quickest method for my size melts, though again the commercial units are prohibitive and there is even less DIY info on them.   Eventually, I will decide on probably one of these three electrical methods.  With limited electronic knowledge and resources the decision for me will boil down to good proven info from DIY sources and of course economics.  Opinions on the best electric method for small size melts of 1 to 5 lb size melts?

Arc, Induction, or Resistance???
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 10, 2019, 05:13:21 AM
I just tried melting some more copper with the 2500 watt unit.
It went well despite my making a newby mistake by putting to much cold metal into the molten copper in the crucible.
The video is here:
/>
Interestingly, the current dropped more than in previous melts as the copper first melted.  I think I have some ideas about that, which will require further testing, one of these days,  But, even at 24 gross amps, (18 amps net) and 48 volts or (864 watts),  the full  crucible sure did melt well.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on May 10, 2019, 04:16:15 PM
Hightemp:
Well, the way I see it, power output will determine how quickly you heat things up and WHERE the thermal equilibrium is reached - aka power IN starts being equal power OUT.
It is also determined by losses. Even if energy is pumped in at a slower rate, but you somehow can magically improve insulation, you will reach equilibrium at higher temperatures. The other thing to note is, exponentially more energy is required to raise something to a degree at higher temperatures than at lower temperatures (relative to outside temp that is). That fact was especially evident with my other-other pottery kiln which lights up an LED when SSR is on so it is easy to see the difference.

You can reliably melt about 3 kg s of copper with about 800 W. I will be doing further tests + casting via lost PLA method soon ish, will have real numbers. It is funny how i started thinking like a scrapper a bit - always checking if there is some copper laying around somewhere haha.
peterspaco: I nearly doubled the coil so it covers full crucible. Power stayed nearly the same, even decreased a bit. Frequency has halved. I think the only way to increase output is to change diameter that can't be done currently with my setup.
To kind of summarize it (correct me if i am wrong) but MORE power is drawn if we have:
Lesser coil diameter
More material inside the working volume
Higher frequency
Less turns in the coil

 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 10, 2019, 11:51:23 PM
I just tried melting some more copper with the 2500 watt unit.
It went well despite my making a newby mistake by putting to much cold metal into the molten copper in the crucible.
The video is here:
/>
Interestingly, the current dropped more than in previous melts as the copper first melted.  I think I have some ideas about that, which will require further testing, one of these days,  But, even at 24 gross amps, (18 amps net) and 48 volts or (864 watts),  the full  crucible sure did melt well.

Congrats - Job Well Done - Thank you so much, you are the best !!   That was a significant amperage drop compared to the last melt, approximately 30% net amps drop, at least towards the end (18 net vs 26 net amps).  The only difference is the extra charge, I think (9 ounces vs 18 ounces)?      It is confusing why these units can not maintain amperage throughout the melt.  I'd almost think that with the crucible filling with metal that the amperage would if anything go up, but then again, induction heating is mostly Greek to me.   badpeter is going to be doing some similar size melts and casting in molds, so his tests may confirm this, if they have not already?  Great stuff guys - love it!   Foundry fumes can be nasty so if you don't already have some kind of exhaust system, you may want to add one.  I use a simple box fan cranked on high with same size heppa filter attached to the back.

Correction: Not sure, but I think brass shot would be a better "thinner" than pure zinc since zinc would instantly vaporize at copper melt temps.  Brass shot would give off a lot of zinc fumes as well but might have a better chance mixing if "dunked" under the melt.  Aluminum bronze is being used for swords and such cause it is one of the strongest copper alloys even without high tech heat treatment.  But the ratios need to be more precise and may be more difficult to cast because of shrinkage.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: flyrod on May 11, 2019, 04:59:43 PM
Annealing Brass Cartridge Casings with a short (2- layer X 3 Turn) coil:


Thanks for sharing your results.  This is my version for annealing brass:

Not a valid vimeo URL

This is off topic, but to whom it may concern:  I'm having difficulty with this site.  I get a lot of 403 errors when clicking on links.  I can get 2-3 page loads before I get blocked.  I can still load pages through a proxy, but then I can't post.  To post I have to remember what I wanted to post from the previous day and do that first, after a couple attempts because "Your session timed out while posting. Please try to re-submit your message." For me, this really detracts from the experience...
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on May 11, 2019, 08:34:31 PM
Mads:  Please remind  about the specs and prices for your capacitors----

100$ each or we can make a deal if you can use more than one.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 12, 2019, 12:18:58 AM
Flyrod:
Looks like you have the process well in control!

How far down the casing does the anneal go?  ( I have heard that you don't want to go any farther than the just the neck).
I'd think you'd want to quench in water, but I guess many folks do as you do. 

How about some numbers:
-System power capability
-Net Current to the work
-Temperature (I see things glow a LOT, but with white balance problems being what they are, I can't tell.)  What temp ARE you aiming for?
-Frequency

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 14, 2019, 06:05:25 PM
^wow so much activity here!
Well, I managed to melt my first copper! so yay...
I rebuilt the PS with a powerful fan underneath, using the original box. It looks very neat.
But most importantly, insulation does miracles.

The data is the following:
845 g of copper melt in a graphite crucible (so-called "3kg-crucible" on ebay)
in about an hour with an output of just under 1kW.

the question is, how do I draw more power?


Badpeter, any luck on getting more power?  Wonder what would happen if you added another one of the donut things and put in parallel?   The extra 500 or so watts may greatly reduce your melt times and the metal may be cleaner - less time to pick up oxygen, dross, etc.   Petespaco gets between 32 and 38 amps when melting about 9 ounces, however on the last larger 1.2 lb copper melt test the amp output dropped all the way to 26 amps or 1250 watts, after all the metal was added. 
  I have not yet decided on a PS.  Possibly a MOT rated at 1250 Watts or more.  At 50 volts that is about 25 amps.  Problem is getting 60 turns when manually rewinding magnetic wire that is heavy enough to handle say 32 amps all day long.  Petespaco amazingly got about 41 turns producing 31 volts on a small test MOT here on YouTube.
 Potential problems deal mostly with rewinding - choosing correct wire gauge (ie single a 14 gauge wrap or ganging multiple gauges in parallel, say three 18 gauges together, also scaring the wire while wrapping and expense of magnetic wire. Curious, how much did you pay for those donuts?   Anyhow, hope that you eventually show us some more of your bigger melts and some cool castings. 

NOTE:  My MOT ideas are based on assumption the 2500 ZVS works best at 48 voilts.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on May 14, 2019, 07:41:48 PM
Hightemp: I think the only way to get more power is lesser-diameter coil and more material/graphite in that diameter. I don't think I can alter my current setup much anymore.
My donut is using two secondaries in paralel to help with distributing high current over twice the wire as well as for cooling reasons (more turns take up more area of the donut). Adding things in paralel will do nothing. To draw more power - something has to consume it (aka work piece). It cannot be forced.
I have made some ingot molds outta plaster+sand (50/50) as well as a mold for a british ww2 thin dagger off a 3D printed piece from Thingiverse. I also printed a sphinx figurine to test out lost PLA method. When i did that, someone suggested purchasing a roll of water soluble PVA filament and just use that to print with! that be so much simpler.
Re power supply, I see that ebay has so much more to offer now! I would've never gone with the donut if these options were available for a comparable price for high power regulated psu (around 90$). the diy donut transformer journey was a long one and I do not recommend it.

Re MOT - seeing how transformers drop a lot in voltage when connected to load, I overvolted mine when load free such that when idle zvs is connected voltage drops back to around 55 volts. (overvolt was on the order of 15 V or so - quite a bit). This killed the fans by the way, connect them to external supply if you go that way). 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 14, 2019, 08:35:28 PM
 I am amazed that you melted almost 2 pounds (845 grams) on less than 1000 watts.  Did you have much metal loss, and dross on the 1 hour melt?  I wonder how petespaco's power would drop if he tried 2 pounds on a bigger coil/crucible - maybe similar to yours - or about 1 kilowatt power when fully melted?  What is inside diameter of your coil and crucible's approximate area - or is 6 kilogram capacity for copper or gold?  Thanks for the fan tip on overvoltage for mots.

I did some simple two part plaster casting years ago using metal casting plaster, basically what you are using, sand and plaster.  Make sure to dry them completely, regular kitchen oven works fine.  Pouring when molds are hot 450 degrees helps with mold filling.  Did not try any lost wax casting but the water soluble patterns sound very interesting.  If you can dissolve/drain pattern within mold without loosing pattern detail and without contaminating mold so that high temp burnout is unnecessary then it would simplify the process considerably.  I'd be very interested to know how that turns out for you.  Have you poured the 3-D figurine, or knife molds yet?  Please keep me posted !!

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on May 14, 2019, 09:58:28 PM
There wasnt much dross, copper was clean, mostly from av cables. Coil has a diam of 12 cm and something like 16 turns.
Here is my youtube channel where I melt some aluminium with my other diy propane kiln (sword and ingots in another one). When I finally get to make an induction video will post that on there too.
/>Also, on unrelated note, the epic power plant tour vid may be of interest to high voltage enthusiasts... but thats a different story...
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 14, 2019, 11:28:43 PM
There wasnt much dross, copper was clean, mostly from av cables. Coil has a diam of 12 cm and something like 16 turns.
Here is my youtube channel where I melt some aluminium with my other diy propane kiln (sword and ingots in another one). When I finally get to make an induction video will post that on there too.
/>Also, on unrelated note, the epic power plant tour vid may be of interest to high voltage enthusiasts... but thats a different story...

Thanks for info and for sharing your video.  Sword looks beautiful -congratulations!   I encourage you to finish and put on your mantel.  Possibly to eliminate air holes you need more metal to melt to allow for a riser and/or a gate, not sure though as I have never casted anything that big?  Correction on prior post -you are using a sand type mold, different than the "plaster of paris" type mold I used where you mix equal parts of water to very fine sand/plaster, pour on pattern then let it set so it is rock hard.  I know little about sand casting so don't have any other tips but there are many videos on youtube about sand casting.  Luckygen1001 is one of the better ones to check out.  He does lots of aluminum, bronze, zinc and cast iron with fantastic results.

Wow, you are using huge coils -almost 5 inches and still getting descent power.  That is very interesting as I still have an old salamander graphite crucible about that size -may have to get it out of the moth balls.     If you do 3-D printing those could go hand in hand with these induction melters, giving you an unlimited source of intricate patterns. Most of the patterns I made were machined from wood using basic woodworking tools (drill press, router, etc. - the old fashion way).
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 17, 2019, 10:57:17 PM
hightemp1:
  Just a  couple of small but important corrections:

Quote
however on the last larger 1.2 lb copper melt test the amp output dropped all the way to 26 amps or 1250 watts, after all the metal was added. 

Should read:
however on the last larger 1.2 lb copper melt test the amp output dropped all the way to 18 amps NET, or about 864 watts into the work, AT THE MOMENT THAT THE COPPER MELTED .
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 18, 2019, 10:48:59 PM
I've been doing more with work coil design lately.  I've just published a somewhat lengthy (sorry about that!) youtube video showing me laboriously producing a 2 Layer 3 Turn coil.  Its part of my attempt to heat brass cartridge cases faster, while keeping the ZVS heater board safe and able to run continuously without failure.
Anyway, to make it easier to follow those "exploits", I also just put up a webpage dedicated to the work coils themselves.
It will be updated regularly for the next few months at least.
It's at:
http://www.spaco.org/Blacksmithing/ZVSInductionHeater/WorkCoilsForZVSInductionHeater.htm
Any pertinent coil-related videos will be (and already are) linked to on that page.

Pete Stanaitis
----------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: flyrod on May 19, 2019, 04:04:12 PM
Its part of my attempt to heat brass cartridge cases faster, while keeping the ZVS heater board safe and able to run continuously without failure.

Here is a thread about doing this on a reloading forum:

http://forum.accurateshooter.com/threads/induction-brass-annealer-redux.3908353/

It looks like most of those guys are also using the cheap chineze boards too.  There is not a lot of electronics expertise over there, but they've got things working through trial and error and people sharing their results.

With a better gate control I've been able to run 400kHz without problems.  This works well with small, thin cases in small coils.  The idea is to quickly heat the neck of the case without heating the head.  A small coil allows you to do this, otherwise you just heat the whole case and ruin it.  The neck acts like a gasket, so you want it soft.  The head has to hold the pressure, so you want it hard.  If you soften the head it becomes dangerous because it can "KB" and vent gas and metal particles towards the operator. 

These guys have a commercial induction annealer and a lot of technical info published on their web site on brass hardness, etc. if you want to get into the details:

http://ampannealing.com/



Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 19, 2019, 06:40:41 PM
Yes, I have been to both those sites.  I am certainly not planning to re-invent the wheel, just to understand these particular ZVS units better.

Re:
Quote
With a better gate control I've been able to run 400kHz without problems.  This works well with small, thin cases in small coils.
Believe me, I appreciate the "better gate control" comment.

I have read that the Mosfets we are using are good to at least 1 MegaHz.  So it's the design of the gate ckt in these induction heaters that's the problem.
And I am not about to monkey around with them.
 
What voltages and currents can you handle at 400kHz?
How much net current or power is going into the casings when you anneal them.
You said "small thin cases".  is 7mm small?  What is "big"?
Can you comment on work coil sizes that you use?

As we all know, it is said that higher frequencies work better for transferring power to nonferrous materials.  But, really, how big a difference DOES it make?  For me, I think these heaters could go up to about 115 kHz or 125 kHz or so which could be about a 25% increase from  to ~90 kHz where I am running the 1000 watt unit right now.   But is it worth it?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 20, 2019, 02:57:46 AM
hightemp1:
  Just a  couple of small but important corrections:

Quote
however on the last larger 1.2 lb copper melt test the amp output dropped all the way to 26 amps or 1250 watts, after all the metal was added. 

Should read:
however on the last larger 1.2 lb copper melt test the amp output dropped all the way to 18 amps NET, or about 864 watts into the work, AT THE MOMENT THAT THE COPPER MELTED .

Thanks for correcting/clarifying.  Can you or anyone else explain why the measured amps and power dropped so much from start to finish when melting copper, especially since there is no iron involved?  Seams like the more copper we melt the less power these heaters put out?  Also, can a large amount of copper, say 2 pounds in a crucible, somehow be arranged to get the full 2500 watt rated watt output using these 2500 ZVS heaters and still have enough power to melt a kilo of copper?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 21, 2019, 06:12:18 AM
For hightemp1:

My latest copper melting test video will be up on youtube  in about one hour after this message appears.
Hopefully it suggests that there may be a way to get more copper melted in a reasonable amount of time

It will be here:
/>
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 21, 2019, 06:12:55 PM
Okay, hightemp1, here's the answer:
https://www.academia.edu/23024259/TRANSIENT_NUMERICAL_ANALYSIS_OF_INDUCTION_HEATING_OF_GRAPHITE_CRUCIBLE_AT_DIFFERENT_FREQUENCY

Just kidding, but this paper has a lot of useful information, especially after most of the mandatory academic math is over with,  at about Figure 4.

The main take-away that I got is that  getting that taller crucible that you mentioned (and, somehow insulating it with a THIN material)  and making a work coil with enough turns to encircle it,  might really help to transfer more power.  This should lower the frequency, which should increase penetration depth, if I got it right.

Again, a caution:  If you do this, you really need to be careful not to overtax the system the first time you try it out.  Insert the crucible slowly into the work coil.   

Let us know how it works out.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 23, 2019, 04:07:17 AM
For hightemp1----
  I am moving your last private message to the "public" area, since I think that most of the issues you bring forth are  of general interest to others.  I hope you don't mind----:
Your latest comments:
[/quote]Regarding melting copper, I seem to recall you really need a lot of power as you do not have the advantage of good heating before reaching curie-point, as there is is no curie-point for copper as its not a ferromagnetic material. So your only hope is lots of insulation to keep the heat in.[/quote]
You are correct.  There is also no "curie point" for graphite.

Call me a Doubting Thomas but .....   Knowing what I know about these heaters I don't want to be the 1st person to try and cast something useful, other than a ingot.  Ingots can and should be cast at lowest temp possible for best results.  Small intricate castings on the other hand require superheat temp of 200 degrees in excess of melt temp, else metal freezes/poor detail, etc.  I assume higher temps are exponentially difficult and that is why no one has tried/succeeded AFAIK.
I have been thinking about this.
I do have several sheathed thermocouple type K temperature meters that are capable of measuring temps up to at least 2200°F.
I have used them for years to measure hearth temperatures for  my woodgas to electricitly projects.
It would be easy to stick one into the molten copper to check temperature.  I would turn power off when I do this so the ferrous material in the T/C sheath wouldn't effect accuracy of the reading.  There is a bit of a concern that the molten copper would "poison" the T/C sheath, but I have a few older ones that I could sacrifice if it's true.
  Once, a long time ago, I sold several hundred sheathed thermocouples  (made by a friend of mine) at a time to Boeing to be used in particular foundry practice.  They dropped the T/C into the molten bath each time they wanted to know the temp.  They'd get one or two sample readings before the sheath melted.
  I don't think we'd have the same problem with copper though.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 23, 2019, 05:51:09 AM
Some of those quotes were from Mads, buy yeah, no problem, I'm grateful again - you're the best!  I've heard many pyrometers are not very accurate at these high temps, but most are consistent so if you have a good pour with proper filling then you can repeat measured temp.  Color tests is also difficult and varies depending on day/night, clouds/sun.  Generally a very bright yellow boarding on white is supposed to be around 2200, I think.  Not sure where one needs eye protection so please know/understand or just use protection if you plan on testing and please be careful with hot metal Pete. 

I'm thinking if these 2500 watt heaters don't cut it for casting bronze, then scaling these up somehow, or using another design that puts out maybe 4 to 10KW - the most my home will handle safely. One of my previous post shows a link with bill of materials for a design that accepted 8-15 KW input, though I did not see a schematic.  The lower range could use 220 volt but the upper ranges needed 3-phase power lines -that would be prohibitively expensive.

Curious too, as to why I don't see people casting smaller aluminum items with these, or at least some of the easier to cast, Zamac zinc alloys.  Just my biased opinoion, but I think everyone should possibly have one of these in the kitchen ;)  Just speculating on copper for now.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 23, 2019, 04:56:10 PM
Don't know much about pryometers, but it may be large in relation to the copper load so preheating it with a torch somehow so that it does not cool the melt too much may help.  Those copper ingots can be easily cut & reused.  You may want to pour the molten metal load into water, this makes copper shot, not sure, but that may be better for next reheating melts.  Manufacturer recommendation of not topping off crucible (75% Max-?) will be safer and probably extend your crucible life?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on May 23, 2019, 09:54:18 PM
I finally got to do the casting yesterday... ;D ;D ;D
So I can confirm that aluminium melts in about 15 minutes and copper indeed takes just an hour (my setup is unchanged from the specs that i mentioned).

Also I found out that 50/50 plaster/sand mold, pre treated in an oven DOES withstand thermal shock and doesn't explode like I worried. Great for ingots and other things. The dagger turned out crappy due to air bubbles - i forgot to make the air holes. I think I can use this mould one more time, although I kind of ruined it getting the piece out.

Now my power got even lower to a ridiculous 570 W (wtf!) I thought hard about it and realized something... crucible is withering away. It is much thinner now. Dont know where all that graphite is going but every firing i get slightly less and less power. This got to be the explanation.
The amount of metal inside seems to not affect power drawn. Looks like the amount of graphite is determining factor.

Now I am hearing people add borax to purify the alloy. Will have to do that for cleaner result.
I recorded my experiments, just got to cut the vid!= )
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 23, 2019, 11:20:08 PM
Badpeter, good stuff - cut that sucker - if you post a descent pic of the casting we might be able to tell exactly why it did not turn out (is surface finish & air bubble similar to last pour)?  Crucibles are by nature disposable especially if you have long melt times, and have lots of dross; also, using same crucible for different alloys can screw up castings easily - how many melts did you get out of it?  I used to cast mostly silicon bronze and it did not require a flux, only a good skimming before pour.  Other copper alloys did require both a flux and a skim, not sure about aluminum, though I heard many just skim and pour, as well.  Also not sure on this either, but borax/fluxes may contribute to shorter crucible life due to their corrosive nature. 

How many volts/amp did you start with on this last pour, how many amps did you finish with, and how much aluminum did you melt? 

I know copper can be melted at 2000 degrees with these, everyone is doing it - just looking for confirmation that the required super-heat temps of 2200 can be obtained with a pound or two of copper using these heaters (call me skeptical).

Update: possibly induction swirling melting action has eroded the soft graphite - no clue here???   Careful you don't get a pot of molten metal somewhere not wanted !!!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on May 24, 2019, 08:53:50 PM
Finally! :)
The dagger didn't turn out pretty but the rest works okay. Hope it answers some questions!

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 24, 2019, 10:31:35 PM
I would never have believed it, but there it is right before my eyes.  A Kilo (?) of Aluminum Bronze cast in one hour using less than 1KW with these ZVS heaters.  Metal actually looked pretty fluid and did a descent job of filling the mold looking very sword-like - CONGRATULATIONS AND THANK YOU !!!

I am only guessing but here is what may have happened.  Aluminum was not skimmed and produced most  of the dross (prepare AL/Bronze ingots beforehand - they will be cleaner). Dross was not removed before pouring or was incorporated in melt (lower part of casting (blade) was filled with drossy metal because drossy metal is lighter and filled the mold first.  Upper part (handle) actually looks very good.  The air hole on the handle is normal shrinkage occurring at the thickest part (either pour directly at thickest part or have a separate riser there).  Other possible cause of the air hole is a mold breakage or mold halves were not clamped hard enough(did you notice the metal leaking out at the end?) Metal should be melted as quickly as possible so getting a 2500 watt PS or better would speed things up -not sure by how much.  Do you know the composition of the aluminum, I think most people use pure aluminum without other impurities, not sure though.  Wish I could analyse further but my better half is waiting for me.

Pete and others may have questions & suggestions too.  Regarding PS. coil, electronics, etc. they will have to advise, but again, I am amazed the power you got out of it.  Granted, you could have melted that same amount of alloy using gas in about 15 minutes with forced air, but the point as far as I'm concerned you have proven that bronze casting can be done with these heaters?   
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 24, 2019, 11:16:04 PM
badpeter:
  I think you did pretty good on that pour, for a first attempt.
But---
  You need a riser that acts like a "mold sprued cup" above your sprue.
https://www.researchgate.net/figure/Sprue-cup-shape-in-casting-mold_fig1_325985131

 This gives you a reservoir for material so it can fill in as the metal starts to freeze.
Also, you need a couple of vents in the top of the mold along its length to let gases out. This will also allow the metal to flow faster to the far end of the mold and reduce the pressure so the mold is a bit less likely to leak.  Finally, when those vents start to fill and/or overflow, you know you've got  enough material.

 With something that (relatively) small, you might need as much as 30 to 40% extra material  to fill that mold sprue cup and to make up for any leaks.

I am no experienced caster, but have done some aluminum, bronze and iron casting, being coached and guided by those who are.

hightemp1: The T/C's that I often use have sheath diameters of about 1/8" and I recommend preheating them with a propane torch just before inserting them into the molten metal for small melts.

OT:   Several friends and I have been doing iron ore to wrought iron smelts on an off for the last 15 years or so.  In this case, only the slag actually melts.

Also, when making blister steel, we toss broken used glass into the crucible.  It melts and sits atop the steel, preventing oxidation.  I think this will work well for copper, but probably not for aluminum bronze, due to its low melting temperature.

---Just a shade tree mechanic, looking on,
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 25, 2019, 01:38:12 AM
OTH, I did not realize that aluminum bronze had such a low melting point - 1900 F and such a high pouring temp of 2300.   Perhaps the temp was not hot enough to melt all the copper and mix homogeneously so the lower part (blade) was more aluminum and upper was more copper.  The molten metal did look more orange than yellow -what color did it look like to you?  The test ingot looked kinda striped - silver spots and gold spots indicating poor mix of copper and aluminum - did you do any kind or strength or hardness test?  Induction heating is supposed to mix the metal for you - or maybe not and you just had to mix it yourself ?  Have to think about it some more.  ???

Update: Just noticed in your description that you have 12% alum and no iron - I think most aluminum bronze normally contains 1-5 percent iron -   I would think without any iron the melting and pouring temps would go down from 1900 and 2300 making it easier to cast, not sure how much though.  Badpeter, what are all your opinions:

 pouring temp(color),
weight of metal load,
manual skim/stir before pouring or not ,
did it look like there was a lot of mixing of molten metal from the induction currents(some say this mixing can look like boiling),
ingot tests,
cause of problems in handle & blade area,
did metal get hot enough for proper mixing to occur,
using more metal for pouring sprue/risers/vents,
drossy or clean metal ?
 

Forgot to mention that the video itself was very professional - congrats there too.  8) 
Can't wait for Part 2...........



Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on May 25, 2019, 05:06:14 PM
So...
Casting: there were 3 problems with moulds: lack of air holes, difficulty aligning two sides, significant crack between two sides. I will redesign the way the mould is, hopefully 3d print something for the mould to address those issues. Also I hope it can be reusable. Riser can be increased too.
I still have one more test to run with lost PLA method, will see how that turns out.

Metal: It melted completely, but I didn't stir it enough and didn't skim. I think convection doesnt do much.  Will try borax as a lot of people seem to use it. It is weird that if you have 7% Al you get that golden shiny color. Try 15%... and it starts looking like it is all Al, althought most of it is copper! In other words, I have put too much Al in this alloy. resulting alloy is still quite strong though.

Power: something is going on with this! With the same setup, every run has less and less output. I tried a quick melt yesterday, max power decreased by another 100 w. I dont know whats going on except my theory with crucible getting thinner and thinner. Imagine if I could get all 1800 w out of this! I'd be able to melt iron (and add it to my alloy). Will have to wait until new crucible from ebay, if power goes high again, I d say that be the proof. Rather inconvenient that crucibles have such a limited life. I wonder if there are any other options beside graphite... 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 25, 2019, 07:08:23 PM
So...
Casting: there were 3 problems with moulds: lack of air holes, difficulty aligning two sides, significant crack between two sides. I will redesign the way the mould is, hopefully 3d print something for the mould to address those issues. Also I hope it can be reusable. Riser can be increased too.
I still have one more test to run with lost PLA method, will see how that turns out.

Metal: It melted completely, but I didn't stir it enough and didn't skim. I think convection doesnt do much.  Will try borax as a lot of people seem to use it. It is weird that if you have 7% Al you get that golden shiny color. Try 15%... and it starts looking like it is all Al, althought most of it is copper! In other words, I have put too much Al in this alloy. resulting alloy is still quite strong though.

Power: something is going on with this! With the same setup, every run has less and less output. I tried a quick melt yesterday, max power decreased by another 100 w. I dont know whats going on except my theory with crucible getting thinner and thinner. Imagine if I could get all 1800 w out of this! I'd be able to melt iron (and add it to my alloy). Will have to wait until new crucible from ebay, if power goes high again, I d say that be the proof. Rather inconvenient that crucibles have such a limited life. I wonder if there are any other options beside graphite...

Badpeter, I would redo the mold.  When plaster casting I always had "fins" at the parting lines especially with multi-part molds - seldom had leakage (ill fitting mold) though. Looks like you have some kind of hybrid sand/plaster type of mold.  How much water do you add to the mix (very little = sand, lots = plaster), also sand requires no baking to expel the water.  I used to get a ready-made metal casting plaster in 20 lb bags from a local aluminum plaster metal casting operation - worked fine for smallish copper alloys but since plaster of paris burns at 2200 F I could never reuse the molds.  Not very familiar with sand casting mechanics though most use it today - materials are cheaper than plaster, sand is naturally porous so venting is less of an issue, and resulting castings have better physical properties than plaster castings because the metal cools faster in a sand mold. Like plaster molds, sand molds can never be reused either, AFAIK.

Found this cool Foundry Manual produced by the Navy that has a lot of good practical info - it's huge but has lots of hyper links:
https://maritime.org/doc/foundry/index.htm

Crucible question is good one, unfortunately I don't know if the quality of graphite induction crucibles varies or not, but I do know graphite is very soft and I suppose very restive, hence the high temps we are getting.  I know gas fired salamander crucibles are more durable and less "chalky" therefore maybe less susceptible to induction erosion currents.  Wonder if they conduct as much heat or not, and if they available in cylinder form. 

How many melts have you had with that crucible?
What melt size  2,3 or 4 pounds did you have? 

Approximate color and what color changes if any did you notice that occurred from copper melting to right before you pouring?
These colors are a guide only:
Light orange 1800
Yellow 1920
Light Yellow 2010
White 2190*


*2190 give or take a few degrees is where I would cast smallish silicon bronze castings and most all other bronzes - white color.  Not sure what temps to cast this hybrid Al/Cu that does not contain iron as I could find no exact info on this alloy.  Normally Al/Cu with 3% iron, small castings pour at 2400 close to cast iron temps. What difference not having 3% iron makes I can only guess that 2200 or white hot would be best pouring temps for your 88/12 alloy.

To increase convection stirring action you might want make the coil thicker,shorter, wider-spaced???  Peter said that on his 1.2 pound melt the metal was boiling.  It was not actually boiling (4644 F is boiling point of Cu) -- he may have had too much convection stirring which I believe is also not good (too much turbulence caused oxidation/dross).  Petespaco, you don't by chance remember achieving white hot temps with the bigger melts do you?

You are on the right track I think and trust one way or another you will do it.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 25, 2019, 07:58:29 PM
Quote
Peter said that on his 1.2 pound melt the metal was boiling.  It was not actually boiling (4644 F is boiling point of Cu) -- he may have had too much convection stirring which I believe is also not good (too much turbulence caused oxidation/dross).  Petespaco, you don't by chance remember achieving white hot temps with the bigger melts do you?

Okay, let's call it "bubbling", for lack of a more acceptable term.
I didn't notice any difference in radiated color between large and small melts.  They all "bubbled" a short while after melting. 
As far as videos go, the white balance setting and circuitry of still cameras and camcorders varies so much that what you see on a "screen" is NOT predictor of what the eye is seeing.  And, of course, different eye/brain combinations "see" color differently, too.
  I hate to have to dig into lens filtering, but might just give some a try  some day, to be able to show what the melt actually looks like.
You guys might want to look at the melting points for various alloys  of aluminum.
I just looked up a binary eutectic table for aluminum, just for starters.  the eutectic of aluminum and  copper is only 1018°F, for instance.

Also, badpeter,--- I noticed a current shunt in your latest video, but I don't see any wires connected to it.  Did I miss something?  Are you reporting mains current or are you reporting current into the heater driver board?  And, are you reporting gross or net current?
 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 25, 2019, 08:14:23 PM
I just tried melting some more copper with the 2500 watt unit.
It went well despite my making a newby mistake by putting to much cold metal into the molten copper in the crucible.
The video is here:
/>
Interestingly, the current dropped more than in previous melts as the copper first melted.  I think I have some ideas about that, which will require further testing, one of these days,  But, even at 24 gross amps, (18 amps net) and 48 volts or (864 watts),  the full  crucible sure did melt well.

Peter, Just watched your video again.  At one point you said the copper was boiling.  I think that may have been induction stirring of the metal because copper boils at 4644 degrees.  This stirring action produces what is called a meniscus where the top of the melt raises in center and lowers at crucible edges.  Sounds like there may have been  some pretty vigorous stirring, to the point of bubbling?  Stirring is good for alloying to a point, but too much, I would think, causes excessive oxidation/dross.  So maybe a lower frequency would be better for melting (longer or thicker or wider-spaced coil)?  BTW, how is your crucible holding out?

Noticed again that when you dumped in the extra copper the noise level dropped by what sounded like a factor of two.  I thought maybe you blew out one of the fans, but that would have been impossible since you have them hooked up separately.   I think you said the extra load caused that.  So some parts from the heater caused the noise, like transformers make noises for some reason, so noise is emitted from the coil, and load is like a single turn transformer that varies with load/sound somehow?  Just curious again.

Also wondering if you noticed a significant color change from your melt temp of 2000 (light yellow) to pour temp?  2190 is white hot and is where I used to pour small bronze alloy castings.



Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 25, 2019, 08:48:41 PM

You guys might want to look at the melting points for various alloys  of aluminum.
I just looked up a binary eutectic table for aluminum, just for starters.  the eutectic of aluminum and  copper is only 1018°F, for instance.

Also, badpeter,--- I noticed a current shunt in your latest video, but I don't see any wires connected to it.  Did I miss something?  Are you reporting mains current or are you reporting current into the heater driver board?  And, are you reporting gross or net current?

Ok, that is a low eutectic so I take from that nice tidbit of info, that maybe proper mixing took place shortly after everything melted and melting point was far lower than I thought.  So evidently, there is a huge gap between melt temp 1100 and possible target pouring temp of 2200.  Should have been a rainbow of color changes from melt to pour?

Good question on the amps - no clue here but hope to know.

Update: Badpeter, not sure on alloying, but I think you are supposed to melt the metal with the higher melting point first then add lower melting point metal???  Also, I think you should make some ingots of Al/Cu first, whatever size(s) best fits in your crucible.  For smaller melts, you could pour a load into water producing Al/Cu shot.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on May 25, 2019, 10:40:21 PM
Molds: I use 50/50 plaster / sand mix, also 2 plasters for 1 water. Found information suggests that pure plaster cannot stand heat shock very well so sand makes it softer (although weaker) so it doesnt crack. Oven treatment is super neccessary so the mould doesn't explode on contact with hot metal from steam forming throughout. (it dries for couple days too before that but it is not enough)
Pure sand casting doesnt have this issue, but it is a chore to work with. I ve used it in the past for my other propane vids.
Transformer would hum only slighlty louder on overload. The massive fan under it drowns any possible sounds. It is not a part of the system in any way and connects separately to mains power.
The shunt does nothing at the moment. Crappy ebay meter burnt out and i left the shunt until I get a new one. As I explained in the description, to measure DC current just divide power (blue meter) by DC voltage (orange multimeter) (aka 700w/46v=15a dc.) So I have about 15 amps dc going in the induction unit. It can also be calculated the long way by utilizing turn ratios on transformer and ac current on mains, but thats the long way.
The hot metal is yellow to white, and is quite bright. It even shine through thick insulation. It loses brighness quickly when crucible is extracted.

So basically now I am going to wait until i get another different crucible from ebay (also ammeter). If i see increased power with new crucible, it will be conclusive that graphite is responsible for the most power draw. 
Meanwhile I may cut the existing crucible in half and plug a small coil, see how that goes. 
 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 25, 2019, 11:46:32 PM
Molds: I use 50/50 plaster / sand mix, also 2 plasters for 1 water. Found information suggests that pure plaster cannot stand heat shock very well so sand makes it softer (although weaker) so it doesnt crack. Oven treatment is super neccessary so the mould doesn't explode on contact with hot metal from steam forming throughout. (it dries for couple days too before that but it is not enough)
Pure sand casting doesnt have this issue, but it is a chore to work with. I ve used it in the past for my other propane vids.
Transformer would hum only slighlty louder on overload. The massive fan under it drowns any possible sounds. It is not a part of the system in any way and connects separately to mains power.
The shunt does nothing at the moment. Crappy ebay meter burnt out and i left the shunt until I get a new one. As I explained in the description, to measure DC current just divide power (blue meter) by DC voltage (orange multimeter) (aka 700w/46v=15a dc.) So I have about 15 amps dc going in the induction unit. It can also be calculated the long way by utilizing turn ratios on transformer and ac current on mains, but thats the long way.
The hot metal is yellow to white, and is quite bright. It even shine through thick insulation. It loses brightness quickly when crucible is extracted.

So basically now I am going to wait until i get another different crucible from ebay (also ammeter). If i see increased power with new crucible, it will be conclusive that graphite is responsible for the most power draw. 
Meanwhile I may cut the existing crucible in half and plug a small coil, see how that goes.

Using plaster, OK.  If you cut 2 keys in each half before pouring the other half you should not have alignment problems. For a two-part mold it seamed like I always did 3 halves, the first half was disposed of and only used to scrape a good parting line to limit parting flash lines.  Not sure what ration of plaster to sand I was using and not sure of other additives they used, but the molds were very strong after air dried and still pretty strong when bone dry though much lighter.  You can skip the air drying and just oven dry vented between 300 and 450 timing it such that molds are 450 all the way through when ready to pour.  Yes, they must be bone dry - witness the king of random video - he pours into a wet plaster mold and it sounds like popcorn popping. :o
 
Yellow to white so maybe between 2000 and 2100 - Nice.  Not sure on your alloy but that may or may not have been the correct pouring temp. Eye protection is necessary somewhere thereabouts?

I just paid $50 for a 3000w PS though I have no clue if it will work.  It will be four 750w server PS's wired in series.
/>Would like to hook up so that input power is switchable between 12,24,36 &48 volts giving at least some sort of way to adjust power for whatever that is worth. Basically I want more power for melting, not ways to adjust to lower power.  Adjustable may have some unforeseen use, but not still sure if I can get it running stably with all four in series.  I have seen similar HP power supplies with 1200w @ 12v so if you put 4 in series you'd get 4800 watts, maybe useful somewhere down the road.  Or maybe 2 of these 3000w supplies everyone is using could also be combined in some way giving 6000 watts - now were talking!!  Supposedly, they work in series fairly easily, though Mads says in the video that large smoothing caps may be required to balance things out?  I guess I'll find out one way or another.  Also, no clue clue if hooking em up in parallel if possible?

That is a big crucible so maybe that's a good idea.  May have some life yet for testing at least, but be very careful with worn crucibles, they are fragile enough and even more so when 2000 degrees plus. Not sure what happens when a pot of molten metal exposes a coil but know it will burn right through most everything else.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 26, 2019, 12:26:49 AM
hightemp1"
Quote
Noticed again that when you dumped in the extra copper the noise level dropped by what sounded like a factor of two.
The change in noise came from the power supply.  When the current increases to some specific point, the fan's speed is increasesd and vise-versa when current goes down.  And that's what you heard.   My  comment at that time wasn't all that accurate, was it?

Nailing down colors to exact temperatures:
  I am not sure that all metals glow with the same colors at the same temperatures. Some googling does not help to clear this up.
But,  even the amount of backlight can have a huge effect on any particular metal.
That's why a blacksmith shop is often dimly lit.  When I am demonstrating the heat treating of fire strikers to the public, outdoors, it's tough to see the curie point, unless I sorta hide the part under the forge.  Even then, I miss it occasionally.
But just to give you an extreme example. when forging titanium, It gets unbelievably white-white-white in a short period of time.  I have never seen any other metal do that even when forge welding or watching them pour 15 or 20 tons of steel.  And, again, people have widely varying color perception abilities.
  I suppose I better get back out there with a thermocouple.  I am also going  to fool around with home made camera filters, but don't hold your breath.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 26, 2019, 12:55:15 AM
hightemp1"
Quote
Noticed again that when you dumped in the extra copper the noise level dropped by what sounded like a factor of two.
The change in noise came from the power supply.  When the current increases to some specific point, the fan's speed is increasesd and vise-versa when current goes down.  And that's what you heard.   My  comment at that time wasn't all that accurate, was it?


Noise Level:  Your comment was plenty accurate, I'm just, you could say "electrically challenged".

TC & Filters: screw the TC and Filters -- I'd just like to know what you see, with proper eye protection, when you melt a pound of copper for half an hour, but I expect, and rightly so, to not be holding my breath.    As I said before, I am in no hurry at all and I am just amazed how gracious you are with sharing your time and knowledge.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 26, 2019, 01:56:42 AM
So...
Casting: there were 3 problems with moulds: lack of air holes, difficulty aligning two sides, significant crack between two sides. I will redesign the way the mould is, hopefully 3d print something for the mould to address those issues. Also I hope it can be reusable. Riser can be increased too.
I still have one more test to run with lost PLA method, will see how that turns out.

Metal: It melted completely, but I didn't stir it enough and didn't skim. I think convection doesnt do much.  Will try borax as a lot of people seem to use it. It is weird that if you have 7% Al you get that golden shiny color. Try 15%... and it starts looking like it is all Al, althought most of it is copper! In other words, I have put too much Al in this alloy. resulting alloy is still quite strong though.

Here are some links from that foundry manual on curing casting defects:
https://maritime.org/doc/foundry/part3.htm#pg149
https://maritime.org/doc/foundry/pg157.htm
https://maritime.org/doc/foundry/pg159.htm
https://maritime.org/doc/foundry/pg161.htm
https://maritime.org/doc/foundry/pg163.htm
https://maritime.org/doc/foundry/pg165.htm
https://maritime.org/doc/foundry/pg167.htm
https://maritime.org/doc/foundry/pg169.htm
https://maritime.org/doc/foundry/pg171.htm

The color effect is such that some metals dominate color much more easily, I don't understand that either.

By weight, I think you want 88 percent Cu and 12 percent Al

 A primitive hardness test is just nailing it with a screwdriver and comparing marks with soft copper ingot or other metals you have lying around. Stength is putting it in a vice, if it does not brake after a few good wacks, your good to go.  Bending without weakening may be ok depending on use - just means it is malleable.


Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on May 26, 2019, 04:38:40 AM
FYI, gypsum (the active ingredient in plaster) dehydrates in two stages: first to hemihydrite around 170°C, and then to anhydrite around 700°C.

Actually, looking it up, it seems it goes straight to anhydrite around 170°C, at least that's what the text claims (https://wfs.swst.org/index.php/wfs/article/download/231/231).  But the amount lost is just less than the total.  But more than hemihydrite.  It's like it's... quintahydrite, as it were (i.e., 1/5th of an H2O). The final total is plausibly correct (the exact figure is 20.93% water in stoichiometric gypsum).  So now I really wonder if this paper is missing something, or if it has to do with material purity (it was only "90% pure"), or if no one really knew the truth about gypsum dehydration and just kind of went along with it all these years because, who cares it's just gypsum, right?

Anyways-- I've never had problems when heating it to dull red hot.  But I've always seen exactly the bubbling you describe: not violent, but still too much for the material porosity and venting to handle without leaving bubbles.

FWIW, I usually use 1:1 to 2:1 sand (fine or sifted sand preferred) and plaster, for simple casting investment.

Protip: when the metal has fully solidified, drop the mold in a bucket of water.  The plaster will spall off, freeing the casting in no time. :)

Plaster-based investment is also reusable, just dehydrate in the oven to get back to the active hemihydrite form (hardens when water is added).  Needs to be smashed up finely, of course.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 26, 2019, 05:16:48 AM
FYI, gypsum (the active ingredient in plaster) dehydrates in two stages: first to hemihydrite around 170°C, and then to anhydrite around 700°C.

Actually, looking it up, it seems it goes straight to anhydrite around 170°C, at least that's what the text claims (https://wfs.swst.org/index.php/wfs/article/download/231/231).  But the amount lost is just less than the total.  But more than hemihydrite.  It's like it's... quintahydrite, as it were (i.e., 1/5th of an H2O). The final total is plausibly correct (the exact figure is 20.93% water in stoichiometric gypsum).  So now I really wonder if this paper is missing something, or if it has to do with material purity (it was only "90% pure"), or if no one really knew the truth about gypsum dehydration and just kind of went along with it all these years because, who cares it's just gypsum, right?

Anyways-- I've never had problems when heating it to dull red hot.  But I've always seen exactly the bubbling you describe: not violent, but still too much for the material porosity and venting to handle without leaving bubbles.

FWIW, I usually use 1:1 to 2:1 sand (fine or sifted sand preferred) and plaster, for simple casting investment.

Protip: when the metal has fully solidified, drop the mold in a bucket of water.  The plaster will spall off, freeing the casting in no time. :)

Plaster-based investment is also reusable, just dehydrate in the oven to get back to the active hemihydrite form (hardens when water is added).  Needs to be smashed up finely, of course.

Tim

Good protip -that should help with castings physical properties too.  Did not know about reusing either. 

I think lost wax castings require heating plaster molds to 1200 for complete removal of wax residue/mold contamination, and to accept smallish castings/prevent freezing.

The bubbling issue appears that Babpeter has resolved it since he has little mixing, while Petespaco has excessive bubbling.  Difference between two setups include less power, bigger coil/crucible.  My concern will be reducing/limiting convection with a max PS that is Petespaco's size 2500W and using a large crucible that is similar to badpeter's large 1 kilo capacity.  No clue if that is doable or not?

Or maybe Badpeter has not resolved it and you are saying this is what has caused all the porosity in the casting knife blade area?  Have you ever melted anything, skimmed melt and/or used flux, poured, then examined for porosity?  Have you ever heard of anyone using these heaters to cast copper alloys successfully, other than ingots full of pockmarks?

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 26, 2019, 01:11:08 PM

Anyways-- I've never had problems when heating it to dull red hot.  But I've always seen exactly the bubbling you describe: not violent, but still too much for the material porosity and venting to handle without leaving bubbles.

Tim

Or maybe Badpeter has not resolved it and you are saying this is what has caused all the porosity in the casting knife blade area?  Have you ever melted anything, skimmed melt and/or used flux, poured, then examined for porosity?  Have you ever heard of anyone using these heaters to cast copper alloys successfully, other than ingots full of pockmarks?


Page 22 on link confirms your observations. Even states that mains power is not even suitable for melting copper alloys.  Author concludes that medium frequency furnace is the solution.  :(

https://core.ac.uk/download/pdf/146466277.pdf

I think this puts the kibosh on my zvs project. Unless someone has somehow overcome this bubbling issue?   I am assuming there is no way to increase frequency of zvs heater by a magnitude factor of 10 ???  I saw somewhere that both the dental and jewelry use induction furnaces successfully.  Don't know at what frequency and if special inert atmosphere is being used or not?

Author also concludes that Arc Furnaces are not appropriate for Al and Cu melting, also due to violent melting, mixing, dross, air bubbles, etc.
That leaves resistance melting.  Not nearly as exciting but it is a path or way out.  Non-metallic heating elements were recommended for copper, metallic fine for aluminum.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 26, 2019, 05:11:34 PM
So...
Casting: there were 3 problems with moulds: lack of air holes, difficulty aligning two sides, significant crack between two sides. I will redesign the way the mould is, hopefully 3d print something for the mould to address those issues. Also I hope it can be reusable. Riser can be increased too.
I still have one more test to run with lost PLA method, will see how that turns out.

Metal: It melted completely, but I didn't stir it enough and didn't skim. I think convection doesnt do much.  Will try borax as a lot of people seem to use it. It is weird that if you have 7% Al you get that golden shiny color. Try 15%... and it starts looking like it is all Al, althought most of it is copper! In other words, I have put too much Al in this alloy. resulting alloy is still quite strong though.

Power: something is going on with this! With the same setup, every run has less and less output. I tried a quick melt yesterday, max power decreased by another 100 w. I dont know whats going on except my theory with crucible getting thinner and thinner. Imagine if I could get all 1800 w out of this! I'd be able to melt iron (and add it to my alloy). Will have to wait until new crucible from ebay, if power goes high again, I d say that be the proof. Rather inconvenient that crucibles have such a limited life. I wonder if there are any other options beside graphite...

Badpeter, from previous post you may have noticed that I stumbled onto something that may be causing your problem.  Thank you T3sl4co1l !!!   Industry does not use low frequency induction for melting of Cu or Al because of the "bubbling" that causes gross oxidation. Industry standard for melting Cu alloys is Medium frequency furnaces  using frequencies 10x and greater than what you are using to minimize oxidation buildup. I think we are out on a limb here, but if you would like to pursue I have some suggestions.  Keep in mind that this advice is coming from someone who is clueless about induction and only did some hobby casting years ago:

1. Get frequency as high as you can.  Someone said actually removing caps.  Peter and others here hopefully can advise you here on getting maximum frequency. I think you have the 1800w version (less caps) so less bubbling is occurring - a good thing.
2. Also, Peter suggested using a glass covered flux but was not sure if melting point was low enough.  Maybe plain old borax?  Again, my experience with fluxes is limited, so experiment, or ask around.
3. Al/Cu alloy may be more prone to dross pickup, not sure.  Some tin bronze or Phosper/Coopper variation(10% approx) may be less susceptible to oxidation.  Purchase some from your local foundry, also ask for some phosper-copper shot(deox).  I have found that they are usually pretty good about that, fair prices, and even their gates, sprues, risers and such are usually very clean. Also, use only one alloy per crucible.  If dead set on using Al/Cu maybe decreasing the Al a little.  There are some strong Cu/Zn alloys but I have no experience with them.  My favorite silicon bronze probably would not be hard enough for knives.  Don't be afraid to ask a foundryman what he would use.  If your objective is more art than function then do not use Al/Cu.  I had good luck with above Ph/Cu alloy in the past.
4. Add glass cover flux with metal? Skim metal before pour and add phosper-shot if using above tin bronze (approx. 1% phosphorous - not sure what percent Cu is in shot so you will have to calculate).  Pour in a mold designed for the least amount of turbulence as possible with a riser and vents and a larger sprue.
5. I was thinking about bottom pouring (remember handle of knife was pretty clean) but can not think of a safe way to pull that off without metal going everywhere.  Peter has experience with cast iron so possibly has a safe suggestion.
6. Power - the faster you can melt the better.
7. Hopefully, others smarter than me will chime in.

Again, not sure if this is worth pursuing or not.  OTOH if you want to be a renegade, maybe you can get a good cast, come up with something novel, have bragging rights about being the first on youtube to do it, etc.  Personally I'd like to see you do it but that is irrelevant.

Regarding other forms of electric melting.  Arc is not used for copper - same reason even more violent oxidation stuff happening.  Resistance using ceramic elements is used for copper and is a proven safe technology, but I'm sure is not perfect either but probably more doable, I think.  DIY designs are freely available.

Update: I recall hearing bear bottles (brown glass) have a lower melting point around 1800 though that may be too high also?





Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 26, 2019, 05:53:20 PM
hightemp1:
Thanks for digging up that link.  Very interesting.
We have a few  companies our area that use induction furnaces for aluminum and copper alloy melting every day.  In fact, we use one on a regular basis and another occasionally.

On to the subject at hand:
Quote
Page 22 on link confirms your observations.  Even states that mains power is not even suitable for melting copper alloys.  Author concludes that medium frequency furnace is the solution.  :
https://core.ac.uk/download/pdf/146466277.pdf

I think this puts the kibosh on my zvs project.
------------------------------------------------------------------------------
Note what  the author of the link actually calls  "medium frequency":
"2.2.3 Choice of furnace freguency
In terms of metal melting, core less induction furnaces may be classified
into
(a) Low frequency - up to 200 Hz and including mains frequency (50 Hz) and
triple frequency (150 Hz).
(b) Medium frequency - 200 Hz to 10 kHz. Most popular frequencies in
this range are kHz, 2 kHz and 3 kHz.
(c) High frequency - over 10 kHz. Some small laboratory furnaces use
crucibles at these frequencies but this is generally experimental
work."
 -------------------------------------------------------
Take a careful look at some of my earlier videos.
You will see that we are running at about 38 Kilohertz with the 2500 watt ZVS induction heater, which is already 'way higher than the scope of that paper!
By the way, I think the 1800 watt ZVS induction heater ran at about 62.5 kHz.
---------------------------------------------------
Quote
Page 22 on link confirms your observations.  Even states that mains power is not even suitable for melting copper alloys.  Author concludes that medium frequency furnace is the solution.
Here, there's a slight misinterpretation-
He is saying that mains FREQUENCY (50 or 60 Hz) ---  not "mains power" is not suitable----,
And again, "medium frequency", to him, and in 1989, is 200 Hz to 10 kHz.

To sum up, we are already running at much higher frequencies than discussed in at least the first 30 or pages of that paper.


Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 26, 2019, 06:27:40 PM
OK, Game on Again !!!

regarding oxidation,  all that bubbling still can not be good so good foundry practice is advised.

Suggestion:
 Badpeter may be seeing less stirring action on 1800W heater than Peter is seeing on the 2500W because there are less caps therefore more frequency on 1800 unit?  But 2500w unit may melt faster.   Wonder if there is a way to get best of both worlds by having a variable step frequency on the 2500w unit.  Start out by using full 2500 watts until melted, then have a switch on the caps that somehow cuts out four of the caps so the frequency goes higher and you get less stirring?   Would power go down to 1800w, or is this feasible?


Have I redeemed myself or not? 
quote/paraphrase from the movie Dumb & Dumber

Update: Just dawned on me that continually adding metal to initial small molten bath may sorta give a permanent semi-molten state thereby cutting down on bubbling as well - as usual not sure?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on May 26, 2019, 10:03:33 PM
I wouldn't be concerned about frequency, or stirring.  At least not until you're looking at aerospace alloys.

0. Use a thick, fairly conductive crucible (e.g., steel) to shield the melt from magnetic field.  (Marginal for aluminum, and not practical for cuprous alloys.)  Note that graphite has to be quite thick to provide much shielding.
1. If stirring causes oxidation, prevent oxidation (duh :) ).
1a. For cuprous alloys, use a cover of slag.  This should be mostly silica, lime and soda; you're really just going for a melted glass of modest viscosity.  Straight-up borax is too thin and will pour out with the metal.  (Add sand, alumina, clay, whatever to thicken it.  Add fluxes (borax, soda, lime..) to thin it.)
1b. Use a charcoal cover, or other reducing agent.  Using a graphite crucible to begin with, already helps a lot (graphite burns at red-hot temperatures, slowly but surely -- the thin blue flame is carbon monoxide burning off).
1c. Without any need for air in, or exhaust out, there is the option of using inert gas, or even vacuum. :D This is probably harder to pull off (e.g., you have to diffuse the inert gas into the porous refractory first, otherwise it's not doing much), or difficult in general, and expensive (vacuum hardware, plus feed-thrus to handle everything under vacuum?), but absolutely a possibility, something to think about in the long run maybe.
2. If stirring causes hydrogenation, just do a degassing step.  This is normal practice with aluminum.  May be harder with cuprous alloys.  Maybe the hydrogen can be burned out, say by dropping some CuO into the melt (push the slag aside so it doesn't dissolve into that)?  Maybe that doesn't work with zinc present, dunno.  Inert gas sparging will work for both, in any case.
3. Stirring probably causes a lot of slag entrainment (in aluminum, aluminum bronze, etc.), which is a good reason to use a flux (for aluminum, NaCl+KCl eutectic is fine, or you can get commercial mixes that include some fluorides as well).  This may introduce more hydrogen, so again, consider degassing.

Regarding removing caps -- for the same power, you need probably proportionally higher voltage, or matching equivalent to this.  Keep this in mind as the ZVS oscillator output voltage corresponds to supply voltage, and is limited by component ratings.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 29, 2019, 09:40:16 PM
Just in case everyone isn't "burned out" on this topic yet,  I just uploaded my work coil characteristics spreadsheet today.

It details my own observations about several work coils that I have gotten with various ZVS induction heaters or that I have made myself.
It, like most of my stuff, is a work in progress:
http://spaco.org/Blacksmithing/ZVSInductionHeater/Work%20Coil%20Data.xlsx

By the way, just yesterday, I noticed that the inside diameter of the graphite  crucible that I have been using for all my copper melts so far has eroded by about half, from its original thickness.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 29, 2019, 09:50:22 PM
The link below gives a formula to calculate stirring.  My calculations show that the 1800W heater has far less stirring than the 2500 w heater by a factor of 10x, and the 2500w unit is insanely high.  I calculated that the 2500w unit has crazy off the chart high levels of 1250 while the 1800w unit is still very high at

I takes awhile to figure it out, but wondering if someone could independently verify my numbers when convenient for you?

https://www.foundrymag.com/meltpour/metal-stirring-coreless-furnaces/gallery?slide=6

Update: my assumptions were:
1800 unit had .75kw when melting started, @ 70000 hertz
2500 unit had 1.5kw when melting started, 35000 hertz

Even though they used different sized crucibles, I assumed they were the same, for comparison sake:  3.5" tall, 1.5" wide. 
Specific gravity of copper 8.5  ----  Resistance 1.75 best guess.


If these numbers are in the ball park then it will make quality castings with the 1800 unit very challenging. More significant though is without some type of modification to the 2500w unit, quality castings may be improbable.  Is it possible to lower stirring on 2500w heater buy a factor of 15x, or even 10x?  A major design change/mods of 2500w unit, to adapt for melting metal, may be in order and I am the least qualified person in the room to answer -  suggestions please ? ????

UPDATE: ABOVE CALCULATIONS WERE NOT CORRECT.  CASTINGS S/B DOABLE ACCORDING TO UPDATED STIRRING INDEX NUMBERS - RECALCULATED FURTHER DOWN.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 29, 2019, 10:07:53 PM

By the way, just yesterday, I noticed that the inside diameter of the graphite  crucible that I have been using for all my copper melts so far has eroded by about half, from its original thickness.

Nice work on the coils for 1K units -thanks. 

Tim turned us on to the fact that we start burning the shite out em at red hot colors.  I thought they did not start burning until much higher white temps were reached.  Again, don't know if the quality of china graphite crucible varies or not.  Tim advised thicker graphite, I think, but all the ebay stuff is only 5mm and I have not sourced any others - not sure they exist at reasonable price.   Adding some clay content would make stronger but would we still get the heat needed since clay makes crucible "cold"?).  I read average life on industry induction crucible is 30 heats, though industry crucibles may have varying amount of clay added for strength, and industry may also use other silicon carbide clay type that are more expensive.
Peter---how many "equivalent melts" can a forger get on a 99.95% graphite crucible?
Badpeter --how many melts did you get, did you crop off yours and do any more tests?

Maybe excessive stirring is also contributing to short life - do you notice any wave-like lines inside of crucible?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on May 30, 2019, 12:13:24 AM
You may find this of interest:
http://hamwaves.com/antennas/inductance.html

Of course it doesn't give loaded Q, but unloaded L and Q at least are nice.

From Q and applied voltage, you can calculate coil losses, and then idle current, and get an idea of what loaded current consumption will be.

Loaded Q is best guesstimated from ratio of enclosed areas, and typical values for materials.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 30, 2019, 03:39:39 AM
Guess there is a reason why I flunked physics.  I redid calculations, and got much better numbers:

FOR COPPER ON 2500 UNIT = STIRRING INDEX OF 125 = VIOLENT STIRRING
FOR COPPER ON 1800 UNIT = STIRRING INDEX OF 62 = HEAVY STIRRING

FOR BRASS ON 2500 UNIT  = STIRRING INDEX OF 62 = HEAVY STIRRING
FOR BRASS ON 1800 UNIT  = STIRING INDEX OF 31 =  LIGHT STIRRING

MAYBE THIS EXPLAINS WHY COPPER INGOTS TURN OUT SO BAD & WHY BADPETER SAW LITTLE STIRRING ACTION.

BETTER YET, THESE RELATIVE NUMBERS HELP PROVE THAT GOOD BRASS CASTINGS ARE DOABLE ON BOTH OF THESE HEATERS.  LOWER FREQUENCIES ON THE 2500w UNIT INCREASE STIRRING SO METAL CASTING WILL BE MORE CHALLENGING -- COIL DESIGN, GLASS COVERING, AND FOUNDRY PRACTICES SHOULD HELP (GLASS COVERING/FLUX WILL FURTHER LIMIT CRUCIBLE LONGEVITY ON FRAGILE GRAPHITE).  EACH ALLOY WILL HAVE A DIFFERENT STIRRING INDEX - COPPER ALLOYS ARE LESS CONDUCTIVE THAN PURE COPPER, THEREFORE LESS STIRRING TAKES PLACE.  SPECIFIC GRAVITY OF METALS ALSO PLAY A ROLE IN STIRRING, AND HELP EXPLAIN WHY ALUMINUM IS MORE DIFFICULT.  LIGHTER ALUM ALLOYS STIR EASIER, ALSO OXIDES/DROSS HAVE SIMILAR SG, SO DROSS GETS SUSPENDED IN THE SWIRLING BATH.  IN HEAVIER ALLOYS LIKE COPPER, MORE OF THE OXIDES FLOAT TO SURFACE, TO BE SKIMMED LATER. 

This is a clusterfuch - due to erroneous calculations I deleted a prior post and modified others (strikethoughs).  Please don't rely on unconfirmed stirring numbers, but based on Badpeter's knife casting and many other copper ingot castings, the calculated Stirring Index numbers on both Cu & brass seam reasonable.

SI formula clearly shows a direct correlation between stirring and a metal's castability, and proves that the ease of casting parts is affected by several variables including the alloy's resistivity, SG, and the melting unit's frequency, power, & crucible size.


Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on May 30, 2019, 08:00:08 PM
Just in case everyone isn't "burned out" on this topic yet,  I just uploaded my work coil characteristics spreadsheet today.

Not burned out at all! I read all the replies and learn a lot. I just do not have much to contribute with and are now also waiting for my 2nd try to buy fibreglass insulation from China :(
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 30, 2019, 08:17:32 PM
Just in case everyone isn't "burned out" on this topic yet,  I just uploaded my work coil characteristics spreadsheet today.

Not burned out at all! I read all the replies and learn a lot. I just do not have much to contribute with and are now also waiting for my 2nd try to buy fibreglass insulation from China :(

Mads, I just ordered the 2500w version and some crucibles (long/narrow 1.5 kilo size, no outer shell, the other is same one Peter is using with ceramic shell (approx 1# capacity) - both graphite).  Both from China so, yeah, about another month, at which time I will have probably forgotten about this project.  :(
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 30, 2019, 08:38:31 PM
I keep asking about modifying the 2500w board to get the higher frequencies best for metal melting, but no one seems to want to tackle the issue.  The design appears pretty simple so maybe modding is not too difficult; or, maybe because design is so simple modifications are more difficult?  Peter's coil notes in last post show the 2500w unit has more caps and I think advised me to use fewer caps.  How to safely go about doing that is Greek to me (if i remove existing caps do I need to add stronger ones - if i remove caps will i have to upgrade other components, add extra components, etc).  I have repaired TV, Stereos, by replacing bad caps, and measured bad transistors, fried diodes, under-spec'd voltage regulators, etc.  so comfortable on simple soldering end; but, as you all know, clueless on electronic circuits.  Peter's notes also show using narrow crucible/coils may increase frequency. Thought about using 5/16" coil but notes show this may have little if any effect. 

Any and all advice on modding these heaters to get higher frequencies more conducive to metal melting would be greatly appreciated.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 30, 2019, 10:17:29 PM
The issue of changing frequency on the 2500 watt unit:
   Sure, it is easy to do.  Just remove some capacitors.  Just make coils that have lower inductance.  Do just one or do just the other, or do both.  You can see how all this relates if you watch my 1000 watt videos and visit my 1000 watt pages.
  A while back, I spent time determining how fast the Mosfet driver circuit can get the Mosfets into full conduction  SO THAT A USER CAN DEPEND ON ALMOST 100% DUTY CYCLE.
I have done that to my own satisfaction.  The oversimplified (but nonetheless useful) number for upper frequency is about 120kHz.  Higher than that and out go the Mosfets.
  But almost ALL of the work that I have been doing lately is the learn the minimum coil inductances that will produce the desired power transfers without exceeding my 120kHz upper limit.
  I think i am pretty close to having this one "in the bag" too, at least from the inductance standpoint.   I won't even connect up a work coil with an inductance less than 0.7 microhenry to the 1000 watt unit.  And, for now, I wouldn't go less than that on the 2500 watt unit either, since, as far as I can tell, they have identical Mosfet driver circuits.
  However, just to add to the still  UNknowns, I am NOT confident that we can 100% transfer gate drive concepts directly to the 2500 watt unit since we have 3 sets of  drivers in parallel. 
  Lastly, for today, I am not sold on the REAL value of going beyond 120kHz anyway. 
Sure, I have heard the more power gets transferred at high frequencies, but this offset by thinner skin depths.  Oh yeh? And by exactly how much?  And, is it worth the effort, when I see that a WIDE variety of small-end commercial units seem to be quite happy in the 30kHz to 80kHz range.
   Lately I have been spending a lot of time trying to improve my video capabilities so I can do two camera shoots better and be able to show the bubbling copper better.  I feel that you guys may be getting too focused on stirring, so I want you to see exactly what I see.  To that end, I am now into camera light filters and white balance issues.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 30, 2019, 10:33:16 PM
The issue of changing frequency on the 2500 watt unit:

---Sure, it is easy to do.  Just remove some capacitors.  Just make coils that have lower inductance.  Do just one or do just the other, or do both. 

---Lately I have been spending a lot of time trying to improve my video capabilities so I can do two camera shoots better and be able to show the bubbling copper better.  I feel that you guys may be getting too focused on stirring, so I want you to see exactly what I see.  To that end, I am now into camera light filters and white balance issues.

Great, I will delve into that in due time!

Yea, I know, I'm compulsive about stirring.  Reasoning, correct or incorrect, is the lower stirring index # that I can get, the cleaner my castings should be; and, usagae of less flux that further damages crucible.  Funny how industry praises the advantages of stirring and before even doing an actual melt, I am already abhorring it.

Please don't go to the extra trouble on my account.  BTW - I think your low resolution videos are just fine.  You spend less time processing, etc and more time shooting pool, etc.(things that really matter!) If you feel extra clarity is required maybe adding a picture with link in video description is all that is needed.  Coming from someone who has never done youtube so, as usual, take with a grain of salt.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 31, 2019, 02:15:46 AM
The issue of changing frequency on the 2500 watt unit:

 However, just to add to the still  UNknowns, I am NOT confident that we can 100% transfer gate drive concepts directly to the 2500 watt unit since we have 3 sets of  drivers in parallel. 

 Lastly, for today, I am not sold on the REAL value of going beyond 120kHz anyway. 
Sure, I have heard the more power gets transferred at high frequencies, but this offset by thinner skin depths.  Oh yeh? And by exactly how much?  And, is it worth the effort, when I see that a WIDE variety of small-end commercial units seem to be quite happy in the 30kHz to 80kHz range.


 

Just by switching to some of the less conductive bronzes my figures move from brasses high stirring index of 62 to a medium SI in the low 50s).  Then, assuming we raise board's frequency by only 15kH (35 to 50) we now decrease SI to 40 or the high end of low stirring - closer to my target! 

If modding board is deemed necessary, do you think it would be safer to remove caps one at a time,  or make coils with lower inductance (there are two on board and just swap em)? 

Again, this is all my seat of the pants math that I frankly don't trust -- also, suspect you tire of me ambling on about these SI numbers - helping knobs like me has got to be demeaning - please know that we all so value your experience & kind advice.

Update: Unfortunately, as you increase ZVS heater power, the metal stirring goes up too.  Last calculations were based on power output of 1500 watts.  Assuming power level of 2000 watts (much better!), then frequency would need to go up even more from 35kH to 70kH, or doubling the present frequency.  So 2000 watts of power @ 70kH gives a descent SI of 38.  Does anyone think a 70kH MOD can be safely done - if so how specifically (please :))))would you modify the board.

Note: The power ouput on Peter's 1.2 lb melt test started at 2000kW and after all was melted power output ended @ 1250kW
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 31, 2019, 05:21:58 PM
Quote
If modding board is deemed necessary, do you think it would be safer to remove caps one at a time,  or make coils with lower inductance (there are two on board and just swap em)? 
My first choice would be to "make coils with lower inductance".  HOWEVER!!!!---  The two toroidal  chokes on the board are NOT the work coils.  They are there to decouple noise from the power supply.  It's the work coil that I would make with lower inductance.  That's what my work  coil data spreadsheet is all about.  I only list those two chokes  "FYI".
  I would NOT reduce the number of capacitors on the board.  All that tank circuit energy has to go somewhere and, with fewer of the same size capacitor, I'd thing they would heat up even more than they do now.  If you look at the way others build this kind of driver, you will see that they often attempt to spread that energy out by using many, many tank capacitors.  Of course in many commercial units, the capacitors are water cooled.  That's a whole different ball game.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on May 31, 2019, 06:26:01 PM
Quote
If modding board is deemed necessary, do you think it would be safer to remove caps one at a time,  or make coils with lower inductance (there are two on board and just swap em)? 
My first choice would be to "make coils with lower inductance".  HOWEVER!!!!---  The two toroidal  chokes on the board are NOT the work coils.  They are there to decouple noise from the power supply.  It's the work coil that I would make with lower inductance.  That's what my work  coil data spreadsheet is all about.  I only list those two chokes  "FYI".

Thanks again.  I'm just an ignoramus.  I will make the smallest diameter coil that will still fit crucible.  Taking the caps off board & water cooling them is only a last resort option for me. I now see that a Medium SI is actually industry recommended level for most copper alloys & we are there or close on many alloys without any mods on current setup, just like yours.  Metal exceptions being pure copper and similar good conductors like silver, sterling, gold, & pure aluminum.  You were correct again...I was making a mountain out of a molehill.   
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on June 03, 2019, 06:09:33 PM
Hopefully I will be able to contribute with some new insights once new crucibles and other stuff arrives!
I feel left out =)
Maybe I'll try something different meanwhile
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on June 06, 2019, 07:03:04 PM
Alrighty... while the next video is being cut I'll share these, hopefully new insights.

I received my new 2 kg crucible. I chaged the coil to a 5 turn coil. This time, I WEIGHTED the crucible before use so I can see how much graphite is lost. And, it seems, a lot! After an hour runtime a 234 g crucible lost 25 grams of graphite. And I can confirm with absolute certainty that power draw depends on amount of graphite - it was at a healthy 1083 W with newly inserted crucible, and by the end was about 970 w as graphite was lost. I will not be surprized if for my next melt I will start with those 970w. The amount of material in the crucible seems to have insignificant (although measurable) effect on power draw.
So this creates a practical problem. I do not want to pay 15-20$ for a new crucible every ten melts. Also, even bigger problem is that with a tiny crucible it is very inconvenient to remelt things you have casted - you d have to laboriously cut it up in small pieces.
A truly useful crucible should be at least the size of cookie tray ingots, and good for many melts.

So, my next R and D will be devoted to being able to manufacture crucibles with clay/plaster/sand/whatever mixed with graphite. There got to be a recipe somewhere. Or switch to steel container, but I have no idea where to get something with thick walls.
Another idea is two concentric pieces of pipe acting as a heat jacket. Loose graphite can be put between the cylinders and replenished as it burns out - to keep power output consistenly high.
But at that point, I might as well design a regular electric kiln that will not have those ridiculous problems with small impractical crucibles, constant power draw issues due to burning graphite and all other sorts of maintenance that come with the iduction setup.

Anyhoo, the new vid is coming!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 06, 2019, 10:18:08 PM
badpeter:
  I just watched your earlier dagger casting video again.
If I hadn't said it earlier, I really appreciate your willingness to share.
From it I can understand your concern about being able to use your induction heater for melting that much material. 
If you have no particular attachment to "art" of induction heating, then going with an electric kiln, as you suggest in your recent post,  sounds like an appropriate solution.

However----  (isn't there always a "however"?):
I noticed that quite a bit of heat was coming out through the insulation in that video.  Whereas, the hard ceramic insulating cups that I am using NEVER show ANY signs of glowing at all on the outside.  I wonder if conditions would change significantly if you were to obtain such a crucible of the size you want.

Also--- about the depletion of the graphite:  the carbon can't really oxidize if there's no oxygen, right?   I 'm sure we have all read about commercial melters using inert gases, but that would be hard to do for us, I think.
But, what if we used a tight fitting cover?  The existing oxygen would be used up pretty quickly and then oxidation would be minimized.  We would only open the crucible to add material.   I have seen crucibles that have such a cover.
  For my own system, I will make some covers.  A guy gave me a 2" X 4" X 24"bar of graphite several years ago, so I have plenty of material  to experiment with.

For one source of graphite, the farm machinery companies around here used to sell quart size containers of powdered graphite for lubricating corn planters.  It wasn't too expensive.  Might be one approach to your inner and outer steel shell idea.  Just pour in more graphite powder as (and if) it still depletes.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on June 08, 2019, 11:24:28 AM
Hi,

Do you think, there is an principal limitation to the ZVS Royer circuit, like efficency or maximum Power output, compared to more complex used circuits?
Except for the savety and that it's difficult to adjust your output power.

I mean with cooled mosfets in parallel you can achieve really high currents and there are some pretty nice ones out there, which can handle 600V with fast switching like the ixth62n65x2.

Furthere more I tested the 2000W 45A 20-70V induction Heater with the LCD panel.
It uses 4 (2 parallel) IRFP260 so you really shouldn't go over 70V, tried it and got an big arc. The Heatsink was glued on and removing it would proabely also destroyed them. Now im trying the ixth62n65x2.
After trimming the Shunt with tinn the current reading was within 5% acuracy beofre it showed an 3 times lower value. It has also an poti for limiting the current below 45A, if the treshhold is reached it switches off, so it's more a protection for the power supply.
It normaly tests the ouput and the Mofets before switching on, that's a nice feature.
Unfortunaly the unit doesn't show the frequency.
I could achieve a Power consumption of 3000W. If you want, you can make a Solderbridge over one of the two Shunt to bypass/reduce the current limitation. You maybe want some watercooler for the Mosfets the, to keep the original Mosfets and therefore the Resistance down.
I worked with frequencies from 40 kHz up to 200 kHz. Not sure how well the gate signals were or if theres a way to reduced the switchable Frequency. That was a task for the future.
I already orderd a new one and the new fancy Mosfets so hopefully I can compare them.

Another tipp, the used copper pipe shouldn't be to thick. If it's to thick you maybe reduce the resistive losses, but with the thicker tube you can't achieve strong magnetic field due to the larger volume of the coil.
I'm going to use 6mm with 4.1mm inner diameter. The next lower pipe commonly aviable would be 5mm with 4.1mm that's much less area. Bigger pipes aren't not woth the extra used space in my opinion. Most cheap 15kW heaters also uses 6mm pipe.

With kind regards Robert
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on June 09, 2019, 04:44:08 PM
Do you think, there is an principal limitation to the ZVS Royer circuit, like efficency or maximum Power output, compared to more complex used circuits?
Except for the savety and that it's difficult to adjust your output power.

And stability, because it's just a dumb oscillator and doesn't always start up at all, or at the intended frequency (more a problem for high frequency oscillators).

So yes, that's all.  Like how a car is an engine on wheels, with steering.  It's not like you need a throttle, or brakes, or a windshield. ;)

So, that's why we design more complicated circuits.  They're not very complicated really.  Tesla coil drivers have been made worse than what's needed for this.  (My typical controller circuit is implemented in about 200 components, maybe not something you really want to build a kit of, but not at all impossible.)

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 09, 2019, 06:50:35 PM
T3s---:
I don't have any recent experience with "the more complicated circuits", but, regarding the "ZVS Royer circuit", please tell me more about:
Quote
---doesn't always start up at all, or at the intended frequency (more a problem for high frequency oscillators).

I ask because I don't think I have had either of these problems.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 10, 2019, 12:44:07 AM
 ::)
Alrighty... while the next video is being cut I'll share these, hopefully new insights.

I received my new 2 kg crucible. I chaged the coil to a 5 turn coil. This time, I WEIGHTED the crucible before use so I can see how much graphite is lost. And, it seems, a lot! After an hour runtime a 234 g crucible lost 25 grams of graphite. And I can confirm with absolute certainty that power draw depends on amount of graphite - it was at a healthy 1083 W with newly inserted crucible, and by the end was about 970 w as graphite was lost. I will not be surprized if for my next melt I will start with those 970w. The amount of material in the crucible seems to have insignificant (although measurable) effect on power draw.
So this creates a practical problem. I do not want to pay 15-20$ for a new crucible every ten melts. Also, even bigger problem is that with a tiny crucible it is very inconvenient to remelt things you have casted - you d have to laboriously cut it up in small pieces.
A truly useful crucible should be at least the size of cookie tray ingots, and good for many melts.

So, my next R and D will be devoted to being able to manufacture crucibles with clay/plaster/sand/whatever mixed with graphite. There got to be a recipe somewhere. Or switch to steel container, but I have no idea where to get something with thick walls.
Another idea is two concentric pieces of pipe acting as a heat jacket. Loose graphite can be put between the cylinders and replenished as it burns out - to keep power output consistenly high.
But at that point, I might as well design a regular electric kiln that will not have those ridiculous problems with small impractical crucibles, constant power draw issues due to burning graphite and all other sorts of maintenance that come with the iduction setup.

Anyhoo, the new vid is coming!


Bads, come on -- cut that sucker - how did everything turn out !!!

On the narrow crucible problem.  Agree, cutting up castings, etc. to fit these narrow crucibles is a pain.  Suggestions include making and pouring your own alloys into self-made pre-heated ingot molds that are long & narrow.  Also, I have seen wide-body graphite crucibles, but I think stirring increases with width so a less conductive alloy may be advised?  I cut up pieces using a band-saw, or saws-all.

Making own crucibles and eliminating oxygen are both good ideas but may be difficult.  Clay bonded crucibles may not get the heat we want?  Possibly ramming  graphite within a concentric 1 piece u-shaped steal tube, then sealing top with high temp ceramic that hopefully seals off oxygen?  With copper alloys, you then have to apply some sort of hot face lining to the inner metal to prevent metal contamination - similar lining would be necessary for an all stainless steal crucible?  Obstacles include longevity of lining, metal, and O2 elimination, plus cost competition with graphite.

Dross/slag further limits crucible life so using alloys that don't require as much additives may extend crucible lives.

  Losing 10% or more each melt means that after 6 melts crucible is reduced by half and since the crucible is what is heating/melting our metal, you are correct this is a big problem. Assuming $21/crucible & 7 melts/ crucible life -that is $3 crucible cost per melt - approximations only - not good ?  Three dollar crucible cost would be approx 10x the energy cost (assuming 1000 watt for 1 hour ?).  I see that the alternative furnaces that use electric resistant elements may also be using 100% graphite so crucible life may not be much longer?  However, by nature I would think that clay-type crucible would be usable with resistance furnaces so crucible life may be 2-3 Xs longer  - but maybe heat time/Watts consumed go up ??  Fatter graphite, like 10mm thick or more would be nice but where to source economically?  Lastly, with the power limitations we have on our ZVS units, we are pushing the limits on max. quantity of metal that can be realistically melted.  Personally, I hope to be able to melt 1 to 3 pounds of copper alloys into useful castings weighing two ounces to two pounds.  That is more than 10 times what an acetylene torch can melt, and quieter/cleaner than a propane setup.  Alternative as we both know is resistance melting.  I did not think of or compare the two methods beforehand so hopefully induction will work out.

As you can see,  I mostly only have more questions to your questions.    :-\ 
To some extent, we are pioneers here, discovering, solving, understanding problems as we go.  Many here at not at all interested in the melting side of things and only want to create large sparks.  ::)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 10, 2019, 03:18:01 AM
This isn't really a "reply" to previous posts, but just a tiny bit of additional information  about the molten copper----

I just uploaded this youtube video about the looks of the surface of the molten copper:
/>
It's not my best work, but just another step toward better photography.
I also measure the temperature of the molten copper a couple of times.
I'm in the process of making several improvements that I hope will benefit those of you who haven't yet actually peered into a ZVS induction heater's graphite crucible with molten copper in it.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 10, 2019, 04:19:17 AM
Do you think, there is an principal limitation to the ZVS Royer circuit, like efficency or maximum Power output, compared to more complex used circuits?
Except for the savety and that it's difficult to adjust your output power.

And stability, because it's just a dumb oscillator and doesn't always start up at all, or at the intended frequency (more a problem for high frequency oscillators).

So yes, that's all.  Like how a car is an engine on wheels, with steering.  It's not like you need a throttle, or brakes, or a windshield. ;)

So, that's why we design more complicated circuits.  They're not very complicated really.  Tesla coil drivers have been made worse than what's needed for this.  (My typical controller circuit is implemented in about 200 components, maybe not something you really want to build a kit of, but not at all impossible.)

Tim

Just curious Tim, for you, how hard would it be to mod this board so amps could be approximately maintained throughout melting process? 

In Peter's bigger melting videos amps start out at about 38 and finish somewhere around 25.  That is a 33% loss or about 600 watts of power. ???
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on June 10, 2019, 05:24:58 AM
T3s---:
I don't have any recent experience with "the more complicated circuits", but, regarding the "ZVS Royer circuit", please tell me more about:
Quote
---doesn't always start up at all, or at the intended frequency (more a problem for high frequency oscillators).

I ask because I don't think I have had either of these problems.

Startup is an issue at very low Q factors.  Transistor gain is quite high so it's not usually an issue at common Q factors.

You might just see it with copper coils tightly fitted around steel pipe, or with lossy coils (say a stainless pipe coil used for process heating).

Frequency is a problem for this circuit for example,

(https://www.seventransistorlabs.com/Images/Fluo1.jpg)

which runs around 500kHz, but can lock into other modes in the 400-650kHz range depending on what load you have attached.  The problem is if there is extra capacitance at the load, in addition to what's on the oscillator board itself.  The connecting wires between oscillator and load form another resonant circuit, and the frequency response becomes much more complicated.  In this case, the transformer is a step-up and its secondary has a resonance near this frequency, even if I didn't put explicit capacitors on it.

As I said, you're less likely to have problems with this, at low frequencies where the capacitors are more likely to act together.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 10, 2019, 04:09:10 PM
This isn't really a "reply" to previous posts, but just a tiny bit of additional information  about the molten copper----

It's not my best work, but just another step toward better photography.
I also measure the temperature of the molten copper a couple of times.

Peter, what interested me the most in your last video me is after the melt & pour when the emptied crucible was reinserted inside coil, the amps shot way back up to 38.  This tells me that maybe very little graphite was lost in the approx 15 minute melt.  More curiously, is that the reduced amps from melt start to melt pour (38a to 25a, or 33%) simply seems to have more to do with the metal going from solid to liquid state - maybe how this self-oscillation circuit mysteriously works - can someone please explain to me why amps crash so much from cold metal to melted metal, or am I just missing something obvious again?? 

Oddly, on Bad's last reported melt the amps only went down approx 10% in an hour.  Not sure if amps would have recovered if he had reinserted the empty crucible into coil after melt.  So Bad's 1800w board appears to behave differently than Peters's 2500W board, but there is also different starting power input levels (1kw vs. 1.8kw), and different run times (15 min vs. 60 min) so comparison is fuzzier. However, possibly a huge difference in individual board behavior. Bad's smaller amp loss may be due more to crucible burning and Peter's larger amp loss may be due to board design, self-oscillation??

Also, the blue gas burning was clearly evident towards end of the video so I am thinking crucible burning is accelerated as crucible temp increases.  Blue flame may have just been some crude on copper too.

update: Lastly, the pryometer went up to about 2015f and then you said that it stopped climbing rapidly.  I suppose it could have gotten to 2200 but just would have taken progressively longer?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on June 10, 2019, 08:23:43 PM
When the load resistance is low (e.g. cold copper), it tends to reflect rather than absorb magnetic field; this reduces power consumption compared to a better matched load.

Whether current draw increases or decreases with heel size and temperature, depends on which side of best match you're running at.  Which in this case I think depends on the crucible thickness, and porosity (overall conductivity), as far as how much magnetic field it absorbs before reaching the copper within.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 10, 2019, 10:12:13 PM
When the load resistance is low (e.g. cold copper), it tends to reflect rather than absorb magnetic field; this reduces power consumption compared to a better matched load.

Whether current draw increases or decreases with heel size and temperature, depends on which side of best match you're running at.  Which in this case I think depends on the crucible thickness, and porosity (overall conductivity), as far as how much magnetic field it absorbs before reaching the copper within.

Tim

Cool, so if I got it right - the higher the metal temp the higher the metal resistance, then the higher the resistance the lower the power output of unit.  Thought there might be a relationship.  This could be my downfall since I would like to melt higher resistance alloys like silicon bronze that have very high resistances = not enough heat to melt.  Unfortunately, most of the good copper casting alloys have very high resistances.  Also, bad for me is the less resistance the greater the stirring.  The intricacies of induction melting are challenging.  Will try to squeeze all the power I can out of a 2500w unit but it may prove to be under powered for my size melts.  Please correct me if my resistance/zvs power relationship is wrong - thank you!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 11, 2019, 03:42:39 AM
There is very little increase in current, if ANY as the copper heats up (but before it melts).  The decrease in current that I report occurs at exactly the moment when the copper melts.  That has been my direct observation EVERY time I have melted copper.  And, the deeper the pool of molten copper, the more the current decreases.
  I don't see the "cold copper" phenomenon mentioned recently.
Also, since the addition of solid copper to the crucible does not seem to increase the current, I don't think the copper itself is participating to any significant extent in speeding up the melt.    It's radiation, convection and conduction that is transferring the heat from the glowing crucible walls, to the copper, in my opinion.
  ---So I will go "out on a limb" and say that I don't think the copper alloys that you want to use will present any worse of a melting problem than does the copper that I am using. 

Now for my hypothesis as to why the current drops just when the copper melts:  I think the molten copper actually "shorts out" the graphite as the copper makes intimate contact with the whole wall of the crucible.   

Lastly, for today, about stirring:
As you know, I am trying to be able to show exactly what is going on in the bottom of the crucible.
  I almost got to performing  my improved camera color-shift filtering test today, but--
A friend and I had a chance to inspect a pig iron smelter from the 1880's and we did do it.
Well worth the 20 mile trip.

  I hope we will all see that there isn't much "violence" going on in the crucible, after all.
I did get some graphite covers and additional insulations pieces cut and did a dry run to test for a 2 camera shoot, though.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 11, 2019, 08:29:11 PM

  ---So I will go "out on a limb" and say that I don't think the copper alloys that you want to use will present any worse of a melting problem than does the copper that I am using. 

Now for my hypothesis as to why the current drops just when the copper melts:  I think the molten copper actually "shorts out" the graphite as the copper makes intimate contact with the whole wall of the crucible.   

Lastly, for today, about stirring:

Absolutely, videos clearly show amps dropping once molten metal bath starts and peek when the pot is full (38a to 24amps, I think) - somehow the molten metal bath inhibits power getting to the graphite.  Now we kinda know why amps crash so much, but unless we can rig up some kind of controller circuit to force more power to crucible I guess just knowing exact cause doesn't do much good anywho.  Tim, stated that they are not really complex for him.  Adding two hundred components is way beyond my pay grade.  But possibly something less complex, that just gives a little more power would be sweet??

According to my SI calculations copper, silver, gold (the super-conductive elements) all have very high stirring indexes.  This has been confirmed by one of the retailers of these units, for what that is worth.

I would have liked to have seen the old-time pig iron smelt - make America great again! :(

update: Just got my crucible, like yours.  A little dissappointed how thin the graphite is (11mm top rim, and a little over 6mm body).  Also was hoping the graphite would sit flush with the ceramic shell, but instead graphite is 2mm higher so minimizing O2 to outer graphite would be more difficult.
Also, received the 1.5kg the other day.  No shell and the body is 8mm thick.  Thought about coating it with that high temp spray kit, but that stuff is so expensive and I've heard that because graphite is so slippery that it may not hold for very long.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 11, 2019, 08:57:37 PM
Sorta OT, but---
Quote
I would have like to have seen the old-time pig iron smelt - make America great again! :(

This smelter hasn't actually been in operation since about the year 1900.  We just checked it out for historical curiosity sake.
See:
https://www.youtube.com/edit?o=U&video_id=0zTf4WdYXc0

If you want to see us MAKE "Wrought Iron" from iron ore, go here:
https://www.youtube.com/edit?o=U&video_id=493AsqZ4P54

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 11, 2019, 09:15:00 PM
Sorta OT, but---
Quote
I would have like to have seen the old-time pig iron smelt - make America great again! :(

This smelter hasn't actually been in operation since about the year 1900.  We just checked it out for historical curiosity sake.
See:
https://www.youtube.com/edit?o=U&video_id=0zTf4WdYXc0

If you want to see us MAKE "Wrought Iron" from iron ore, go here:
https://www.youtube.com/edit?o=U&video_id=493AsqZ4P54

links did not work, for me?  Any ideas for a "simple" -more power" controller board add-on for these heaters?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 12, 2019, 02:13:23 AM
Quote
links did not work, for me?  Any ideas for a "simple" -more power" controller board add-on for these heaters?

Hmmm--- I just tried both links and they did work for me.  Just a simple "click" on each one and there I was.

I don't have any idea for a "more power" controller board to add on.  However, maybe you want to bit the bullet and buy the $1000 unit that is 7 KW.
https://www.ebay.com/i/221663967797?chn=ps
(This is just one of the many sellers)
  That ought to do it.  You will, however, need a TIG  water cooler for it and they run at least $400 more.

I still feel that it's a "heat in vs heat out" equation.  You only get 3412 btu's of heat for each 1 KW of electricity.  So, if you are losing more than you are putting in, the temperature quits rising, right?   There is still the opportunity for any or all of us to conserve more heat.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on June 12, 2019, 02:17:20 AM
Alright people, here it is...
Thank you for the valuable suggestions - I did not think of oxygen issue at all. I wonder how less the deterioration will be if I add either a ceramic shield or somehow make the whole thing happen in CO2 atmosphere (yes, for my kind of "lab" that be a bit of an overkill!)
I will probably entertain the idea of getting a drilled out STEEL crucible somewhere. Will need to be very creative with that. (Curie point a possible issue?).
Will do maybe one more melt/vid with graphite crucible, not to get too repetitive.
Anyways, the current vid is self-explanatory. lots of commentary and measurements given.
If I cleared PLA, it would have worked wonderfully.
Wonder if i should melt the sphinx (since its not great) or leave it as souvenier and recast my dagger with new material (i found a small source of copper recently!)

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 12, 2019, 05:36:48 AM
Alright people, here it is...
Thank you for the valuable suggestions - I did not think of oxygen issue at all. I wonder how less the deterioration will be if I add either a ceramic shield or somehow make the whole thing happen in CO2 atmosphere (yes, for my kind of "lab" that be a bit of an overkill!)
I will probably entertain the idea of getting a drilled out STEEL crucible somewhere. Will need to be very creative with that. (Curie point a possible issue?).
Will do maybe one more melt/vid with graphite crucible, not to get too repetitive.
Anyways, the current vid is self-explanatory. lots of commentary and measurements given.
If I cleared PLA, it would have worked wonderfully.
Wonder if i should melt the sphinx (since its not great) or leave it as souvenier and recast my dagger with new material (i found a small source of copper recently!)

/>

You know what they say Bads, third time is a charm - I vote for the dagger :)  Just remember, they are never perfect and will almost always need finishing of some sort.  If you like polished look a skilled finisher can do miracles but does take time and experience.  Personally I prefer the as cast look with minimum finishing, but you really need clean castings, good mold design, etc. for best results. 
Don't know anything about PLA but maybe it is similar to wax and needs a 1200f burnout cycle? 
Steel crucible may work, considering you'd like a 1-5 percent iron in your dagger; though, I really have no clue how much iron would be absorbed, and if too much is absorbed it may freeze.  Also, could get contamination of other elements depending on composition of steel?  I would think it would be physically safe, but Peter would know much better than me how strong various steel crucibles may be at 2200-2300 degrees and maybe recommend some type of steel. 
I really don't know how to tackle the burning crucible issue.  I previously mentioned some ideas but they are experimental only.  Go to go for now - thanks much for posting!

Update: Just thought of one potential problem with molding the dagger -- maybe the large thickness transition  between the handle and the blade.  I don't know the best way to design it.  Possibly, have the handle act as sprue, probably have more of a taper transition between the handle and the blade, and/or place a core in the handle so overall thickness will be more uniform?  This will of course be just one the many possible molding, casting, finishing, etc. problems you many encounter.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on June 12, 2019, 02:55:12 PM
Just a quick response from my side.

The higher the resistance of the material, the better it is for heating it. Steel for example, it starts with 60% power than rises to 100% due to increasing electrical resistance and goes down to let's say 40%. These are just estimations, based on my analogue 80a current meter.

So getting up with the temperature is for non magnetic materials in theory better. That would at least explain the overall low power consumption for these relatively high voltage.

For a steel crucible, I can say the following. The normal steel will get destroyed from the air/oxygen and the aluminium, starting at 700°C/1292°F. For example a 2mm thick steel crucible will be destroyed after around 10h to 20h completely unusable at 900°C. Steel with 25% Cr, 20% Ni will withstand the oxygen in this temperature but will still be "eaten" by the aluminium. There are some coatings which slow the process like "3M Bornitride Suspension WP"

http://technical-ceramics.3mdeutschland.de/en/products/3m-release-agents-and-lubricants.html#c880 (http://technical-ceramics.3mdeutschland.de/en/products/3m-release-agents-and-lubricants.html#c880)

Cost in Germany around 40€ for 1 litre.
They also have other coatings. But they only slow down the process. They are likely to crack and there is the crap happening again....

For controlling the real power of a ZVS Circuit, you can basically only lower the input voltage, for all the other stuff you have to change components on the board.

Best regards Robert
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 12, 2019, 04:05:28 PM
Badpeter-
  Maybe try casting the dagger vertically?  That way it would be certain to fill.  I think they do cannons that way.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 12, 2019, 05:02:38 PM
Quote
links did not work, for me?  Any ideas for a "simple" -more power" controller board add-on for these heaters?

Hmmm--- I just tried both links and they did work for me.  Just a simple "click" on each one and there I was.

I don't have any idea for a "more power" controller board to add on.  However, maybe you want to bit the bullet and buy the $1000 unit that is 7 KW.
https://www.ebay.com/i/221663967797?chn=ps
(This is just one of the many sellers)
  That ought to do it.  You will, however, need a TIG  water cooler for it and they run at least $400 more.

I still feel that it's a "heat in vs heat out" equation.  You only get 3412 btu's of heat for each 1 KW of electricity.  So, if you are losing more than you are putting in, the temperature quits rising, right?   There is still the opportunity for any or all of us to conserve more heat.

The error I get is "Oops, something went wrong" on a blank white page - maybe just my server?

On conserving more heat:  I will likely be using kaowool with the narrow 1.5 kilo crucible and may try thin layer of koawool with the smaller crucible/liner combo.  Problem with porous insulation is the contribution to faster burning of graphite.  Unless there are some trade secrets we are not privy to, short graphite crucible life  may be a necessary evil in trying to adapt these heaters to melters.  Also, of course, the extra space the insulation uses will further decrease BTUs to crucible because insulation pushes proximity of crucial farther away from coil so I may be slicing my kaowool to maybe 1/2" (use gloves and mask when handling wool).

On $1500 furnace -->  I can only dream :(. 

Bert,  excellent stuff!  My take is because metal is better conductor maybe heat will not be great enough to melt copper even with insulation?  If it could melt, then maybe a descent  alternative considering we are currently only getting 6-10 melts with graphite anyhow?  And, if copper alloys without aluminum were used maybe iron crucible life would significantly exceed a graphite one?   A simple temperature test I suppose would be to find metal tube (preferably stainless) surround it with insulation and see if white hot temps can be attained.
Other unknowns exist, including previously mentioned metal contamination effect (good or bad).



Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 12, 2019, 05:08:44 PM
Badpeter-
  Maybe try casting the dagger vertically?  That way it would be certain to fill.  I think they do cannons that way.

Vertical may cause too much turbulence.  I was thinking at an angle, not to steep, maybe 25 degrees with the blade edge pointing down.  Just a wild guess though.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on June 12, 2019, 07:46:41 PM
One approach could be a cheap steel pipe, weld a cover on the end and maybe use some of this coating.
The material costs are so small, if you have the welding stuff.

Keep in mind that using normal tools on really high quality stainless can ruin them.

I tested different pipe diameters and most of them easily reached the 2 kW power limit of my heater. The heater has a current and voltage sensor, so I can't go that high above the rating.
I can trick the unit by changing the components, but it's also nice to have a real power display without adding 50% or so on top.
I have a beefy 5kw adjustable unregulated power supply for 0-200V that comes handy, so it was no problem to get the highest possible power output with different shapes.

For example, I start with 30V with 1.5 kW then it rises up to the 2 kW and then it reaches curie temperature it goes down to 800W, and I turn the Voltage up to 60V and have my 2kw again.

These exact numbers depend on many things like the material, the coil, the geometry of the workpiece. Just some figures out of my mind, to get the point.
At the end of next week I can do some more tests.

If you use steel, you should really look the temperatures of the mosfets. I was able to get 50 A even at 30 V with good fitting steel tubes.

I think my coil has an inner diameter of 40 mm/ 1 5/8" something like that and 6 turns, runs with 157 kHz and uses an overall capacitance of 0,86 µF, which are rated for 100 A.

Robert
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 12, 2019, 08:45:31 PM
One approach could be a cheap steel pipe, weld a cover on the end and maybe use some of this coating.
The material costs are so small, if you have the welding stuff.

Keep in mind that using normal tools on really high quality stainless can ruin them.

I tested different pipe diameters and most of them easily reached the 2 kW power limit of my heater. The heater has a current and voltage sensor, so I can't go that high above the rating.
I can trick the unit by changing the components, but it's also nice to have a real power display without adding 50% or so on top.
I have a beefy 5kw adjustable unregulated power supply for 0-200V that comes handy, so it was no problem to get the highest possible power output with different shapes.

For example, I start with 30V with 1.5 kW then it rises up to the 2 kW and then it reaches curie temperature it goes down to 800W, and I turn the Voltage up to 60V and have my 2kw again.

These exact numbers depend on many things like the material, the coil, the geometry of the workpiece. Just some figures out of my mind, to get the point.
At the end of next week I can do some more tests.

If you use steel, you should really look the temperatures of the mosfets. I was able to get 50 A even at 30 V with good fitting steel tubes.

I think my coil has an inner diameter of 40 mm/ 1 5/8" something like that and 6 turns, runs with 157 kHz and uses an overall capacitance of 0,63 µF, which are rated for 100 A.

Robert

Wow, fascinating stuff, all way over my head.  My take on your variable PS, is since we don't have a sweet ass PS like that I'm thinking we could sort of simulate one??  Using graphite as an example: Peter's crucible uses 38 amps when new, but if it had more graphite (wider/taller) maybe it could use 60 amps?  But since we are limited to 50 amps on the 2500w unit we would have to gradually lower the crucible (instant variable amp/power regulator),  lowering it completely when metal is all melted, or when fully inserted and safely stays below 50 amps??  Granted this is a crude variable PS - problems include accelerated burning of lower part of crucible, rigging up a lowering device.

On stainless steel,  yea I tried to file it once and it practically filed my file. ???

I am not a welder but always wanted to try.  Also, thought about welding a SS tube but did not think welding would hold at 2300f, but then again I know nothing about welding either, but ignorance has never  stopped me from blabbing out ideas.  Not sure if same "dipping crucible" concept holds with steel because of curie stuff happening :-\ ?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 12, 2019, 09:14:43 PM
hightemp1:
Quote
But since we are limited to 50 amps on the 2500w unit we would have to gradually lower the crucible (instant variable amp/power regulator),  lowering it completely when metal is all melted, or when fully inserted and safely stays below 50 amps??
I think this is an excellent idea.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on June 13, 2019, 03:04:37 PM
Yes, exactly.  There is a best-load matching condition.

Too low resistivity, and there's no voltage drop in the work, the magnetic field reflects nearly perfectly, and frequency simply rises, instead of DC current being drawn.

Too high resistivity, and there's no current flow, the magnetic field penetrates nearly perfectly, and little DC current is drawn.

We know that, somewhere inbetween, current consumption is higher.  We don't necessarily know if we're on the rising or falling slope of that curve, or optimally on the peak, but we know definitely that there must be a peak, somewhere between these extremes!

This has been today's shop application of calculus (https://en.wikipedia.org/wiki/Rolle%27s_theorem).  Cheers. ;)

Likely, it just so happens that a full heel of molten copper is below the peak, while a chunk of steel above curie temperature, is above the peak.  Graphite of this thickness is certainly above.

Different people have different setups, landing above or below as well.

The positioning of that peak, is driven by the number of turns of the work coil, how close it is to the work, the characteristics of the power supply, etc.  Frequency affects skin depth, which affects the equivalent load that a given chunk of material reflects.

If you're below the peak, I would suppose, try adding another turn or two; if below, remove a turn or two.

This type of supply also delivers the most power into the lowest Q factor, and the lowest impedances.  Q can be lowered by reducing the distance from coil to work.  (You need some space for insulation, and loads like copper won't give a terribly low Q anyway, even at point blank distance; there's only so much room to push in this direction.)  Impedance can be lowered by reducing inductance and increasing capacitance.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 13, 2019, 06:46:24 PM
Here's my final test of watching the surface of the molten copper with better camera filters:
/>
Pete Stanaitis
----------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: badpeter on June 13, 2019, 06:47:12 PM
^^^Yesterday I found a steel valve cap just the right size, that probably can be used as a crucible. Can't wait to test it out! needs a bit of cutting which ill have to go around somehow. I wonder whats the draw is going to be. Due to its shape it cannot be taken out of the coil though.
Wish I had any welding tools or such, maybe Ill find someone who has that.
Dagger hopefully on Friday=) New patented "sunken" mould style will prevent metal from escaping on the sides. 8)

I wonder if it is a good idea to find optimal power consumption by shorting turns of the coil to quickly change its characterictics (easier than fiddling with caps) and watch how output changes.

peterspaco - satisfying to watch the molten surface! I may try something like this, my dslr has some manual settings for video, maybe I can get away without filters even. or use an old floppy to cover the lens!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 13, 2019, 10:22:12 PM
Here's my final test of watching the surface of the molten copper with better camera filters:
/>
Pete Stanaitis
----------------

-Melting action is indeed very clear now - nice!  Definitely more of a shimmering than bubbling. Maybe even more of a medium stirring rather than heavy stirring.
-I'm betting that is dross on the sides covering most, effecting some of the mixing while also highlighting it?  Wondering what the best way to skim it off would be? 
-2200 degrees Fahrenheit with very little preheat time is sweet! 
-Insulation was smoking starting off.  Extraction of fumes especially inside, sure would not hurt (again, I just use a hepa filter sized to my box fan cranked up on high).  Some sort of high temp spreadable ceramic on lid/insulation, hot face I think it's called, may reflect heat back into crucible better?
-Just weighed my new crucible similar to yours = 138.5 grams, 6.33cm.  Hoping for 10 melts?

Bads, nice score on the cap buddy!  Trying with some sort of brushed on high-temp ceramic, or as-is? 
-sunken mold sound interesting.  To cut a couple keys in plaster, I just used a small spoon to hog out a little plaster on first layer.
-no clue on shorting stuff - the word itself scares me.
-yeah,watching molten metal, kinda like sitting in front of bonfire - sweet!
-tomorrow I will post some before/after, cut-up pieces of a sprue and a gate that I purchased from a local foundry some 30 years ago. 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 14, 2019, 12:24:48 AM
I thought I'd wait until someone else mentions the "shorting turns" thing.
When I did that, (unintentionally, early on)  I always liked to have a bunch of extra Mosfets and gate components around.  And even some heavy copper wire to use for replacing evaporated pcb traces.  For me, half a second is tooooooo  looonnnggg!  But maybe your heater's circuit is different.
See:
https://spaco.org/Blacksmithing/ZVSInductionHeater/1000Watt12to48VoltZVSInductionHeaterTroubleshootingGuide.htm
for how and WHY I used to have to repair them!
I have seen people do that "shorting" thing with other units, but you'd better be sure that your driver  board itself has a current limiting function.  Just because your power supply may have current limiting is NOT good enough.  It's the degradation of gate signals that kills the board.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 14, 2019, 02:38:13 PM
Two identical sprues on top left.  Top right is a gate, I think.  No idea what end products were.  Notice the diameter 3/4" of sprues - they must have been large castings.  Pieces on bottom were cut to fit in our small crucible (the larger ones still might not be small enough - :()  First melt may be making ingots to fit my crucibles.  Alloy is silicon bronze and total weight is five pounds.  Took me 30 minutes to cut several pieces - much, much slower than cutting up wood, though my bandsaw is in dire need of a good tune-up.  Used a regular 6 tooth/in wood cutting blade at lowest speed (maybe 600 rpm?) and it worked just fine.  Still have not received my induction heater from China.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on June 14, 2019, 10:19:02 PM
I finished the circuit for the unit.

Printing on the Board:

2000W Rated Power
input voltage DC24V- DC70V
maximum input current <48A
ZVS2000S High frequency induction heating
8300515

As far as I can tell, the controller checks for a mosfet fault before switching the relay on.
Voltage, current a temperature are always checked.
You can add an extra external switch for short heating applications.
There's a potentiometer to limit the max current.

The controller does NOT know if the circuit is oscillating or not, but you can check your setup with a low current setting. In case something went wrong, the unit should open the relay faster, that's handy before start working with different coils/voltages or after a repair for obvious reasons.

If the coil inductance is way too high/low, it can detect a mosfet fault.
The components with an "X" are unknown.

Due to the applied voltage of 100V, the mosfets, transistors, optocoupler and the resistor "R15" blew up. This resistor is crucial for the proper working. I have to wait for the new one board to get the value to repair it properly.
If I had removed the heat sink, the mosfets were most likely also got damaged and since there was no information available.
I wanted to know, how high I can get with the input voltage.

The measured current from the unit was way too high, I shortened the shunt and adjusted the value with some extra tin.
With that method you can also increase the maximum power output relative simple.

The optocoupler for the fan/pumpe is always on.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 15, 2019, 02:00:16 AM
Thanks so much for putting up the schematic.
  I'm not too sure yet, exactly how the system "checks the Mosfets" before turning power on, though.

Re: your comment about max input voltage:
The datasheet for the IRFP260N says 200 volts max.  Somebody on this forum has said the you can expect to see pi X the input voltage across the Mosfet.
If that is true:
48                     55                       65  Volts In
3.14159            3.14159             3.14159
150.79632   172.78745   204.20335 Volts across Mosfet



This seems to imply that you might get away with 55 volts, but 65 volts would be over the top.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on June 15, 2019, 09:25:07 AM
I updated the schematic a bit (removed the 1k resistor after the relay) and there are two versions now.
One with the real component names and one for ltspice simulation with other components.

If the relay is open, the voltage over c3 and c4 goes up to 22V, after that the unit can activate the mosfets and recognize a voltage drop.
If the mosfets are shortened or the transistors are damaged the voltage won't rise/fall.

Since the 200V are the minimum rating, it should be OK to 70V.


Changelog:
-added diode name/diode model to the simulation
-added correct transistor model to the simulation
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 16, 2019, 12:03:50 AM
Quote
Since the 200V are the minimum rating, it should be OK to 70V.
I respectfully disagree.  On the datsheet that I saw, "200" was the ONLY number.  There was no entry at all in the "max" column.

Where do all the "uC" points go?
It appears to me that they would all originate at the output of the opamp, but I guess I just don't understand that kind of a schematic diagram.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on June 16, 2019, 09:38:15 AM
The µC stands for microcontroller.
I didn't bother to track the all the way back to it.
Of course this isn't the whole schematic since there is also an lcd display.

https://www.banggood.com/2000W-ZVS-Induction-Heating-Module-Board-Flyback-Driver-Heater-Good-Heat-Dissipation-With-Coil-Pump-Power-Adapter-Kit-p-1464491.html?cur_warehouse=CN

5V worth arguing? Rule of thumb is 4 times.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 16, 2019, 03:39:30 PM
Thanks for clearing that up and thanks for the link.  Now I understand that there's a lot more going on than I thought.
That's quite a package!
When I multiply 65V X 45A, I get 2925 watts.  I wonder why they only rated it as a 2000 watt unit. 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 16, 2019, 07:06:48 PM
Thanks for clearing that up and thanks for the link.  Now I understand that there's a lot more going on than I thought.
That's quite a package!
When I multiply 65V X 45A, I get 2925 watts.  I wonder why they only rated it as a 2000 watt unit.

Robert, 2.2kw was stable?  maybe making this a superior design & deal?  Input, 120 or 240, make any difference on output?  2 to 2.5kw area makes a 20 amp circuit on 120 volt doable.

Update:  Per Product Description - (Short Circuit/Overload: Input current is greater than 40A, voltage greater than 55V, enter protection).   If so designed, then should not go over 2.2kW, I presume.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on June 17, 2019, 08:01:10 AM
Thanks for clearing that up and thanks for the link.  Now I understand that there's a lot more going on than I thought.
That's quite a package!
When I multiply 65V X 45A, I get 2925 watts.  I wonder why they only rated it as a 2000 watt unit.

Robert, 2.2kw was stable?  maybe making this a superior design & deal?  Input, 120 or 240, make any difference on output?  2 to 2.5kw area makes a 20 amp circuit on 120 volt doable.

Update:  Per Product Description - (Short Circuit/Overload: Input current is greater than 40A, voltage greater than 55V, enter protection).   If so designed, then should not go over 2.2kW, I presume.


The heater goes in protection mode, if the voltage is below 20V, greater than 70V, current greater than 48A or the heat sink temperature over 65°C.
Even fast peaks are detected and trigger the protection.

The mains voltage doesn't matter at all, the input voltage of the unit is relevant. Btw my supply uses 3x380V.

Note that the idle loses increases with a higher input voltage, but also the heating capacity of the coil, it also increases the oscillating current through the coil. So the capacitors need to handle these extra current.
These caps are rated for 50A effective current, this is already reached at 54V input voltage and the standard coil.

ICap, peak=2*Pi*f*UCap,peak
UCap, peak≈Pi*Uinput             // measured Peak Voltages from multimeter could be off by 50% due to high frequency
ICap, effektive≈ICap, peak/1.414

This oscillation current through the caps shouldn't depend on the load.
With 3kw input wattage there could easily be a loss of 500W or more.

Pworkpiece=Pinput-Pcoil+cap-Pmosfet-Premaining circuit
Pworkpiece≈Pinput-Pidle-Pmosfet

The mosfet losses can be assumed to be relatively small, probably under 100W with that heat sink. These depend on the Input current/voltage and switching frequency.

The 48A is already a seriously big current that's why I bought some new mosfet with a higher voltage rating. A heating power of effective 2kw at the workpiece is most likely enough for most applications, but you can't always reach this with the standard configuration for specific shapes like bolt heads.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on June 20, 2019, 02:14:55 PM
Hello everyone,

I got around to test the new arrived heating unit. It came with a different coil (4 instead of 6 windings).  I have to correct my thought about the current limiting potentiometer. It's a bit strange, you can limit the current and you can trim the shunt resistor.Therefore, a real current meter is a have to. Maybe it's some sort of current limiting in the first 50% and shunt tuning with max current at the last 50% or something.

Here are the results from some testing:
Coil with 6 windings, 50 mm high, 45mm inner diameter
with no workpiece:
In 24,4V 45W; Out 74V peak on the caps
In 38V 90W; Out 112V peak on the caps
In 50V 185W; Out 152V peak on the caps
In 60V 250W; Out 188V peak on the caps
The voltage over the caps are pretty much spot on the PI times input voltage. Due to the higher oscillating current, the power loss increases also.

Heating up a workpiece:
Steel bolt 22mm diameter and 60 mm high
In 52V, used power 1600W at beginning, 2100W peak and 700W at around 900°C.

I maybe solder a copper pipe direct at the capacitor connection to cool them with the water too. A cooled heat sink for the mosfets would be great to, since the get a little bit warm at high currents.

The used resistor is a 5.1 Ohm, so I updated the schematics, the simulation is much more stable now.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 22, 2019, 06:10:01 AM
Picked up one of these crucibles. Price is not too bad. Not 100% graphite, but has some silicon carbide in it along with some clay, maybe.  Hoping it is more durable but kinda doubt it, though it is thicker than the graphite ones -almost 11mm.   Chipping may mean it has less bonding strength and is more prone to breakage.  Hoping it heats as well, seems to have a lot of graphite (feels slippery like ones we are using).  Holds a little over 1 pound of copper.  OD is 3" wide by 3" long so with insulation the coil would have to be wider, about 3.75" and about 6 turns.  ID is 2.2" but narrows to 1.5". 

https://www.ebay.com/itm/Silicon-Carbide-Graphite-Crucible-Tool-Metal-Melting-Casting-Cup-Shape-Furnace/254233218790?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

Description says: "Surface can be treated with antioxidant."  Sound like Chinenglish, not sure what it means.  Maybe this silicon/graphite type will accept some type of wash that could increase longevity?

Having some power supply issues so it maybe sometime before I actually melt anything, if ever.  :(
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on June 22, 2019, 05:21:58 PM
If that's the traditional combination of materials, it probably won't be very conductive, and the coating is a flux/slag material that tends to bead up on the surface and drip off (presumably can be refreshed with the right goop).

SiC is a modest conductor when heated, making microwave heating possible, and maybe induction still (but maybe with higher voltages/frequencies than are available here).

SiC-graphite crucibles can practically be quenched in water; they're very tough, as ceramics go. :)

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 22, 2019, 05:47:22 PM
If that's the traditional combination of materials, it probably won't be very conductive, and the coating is a flux/slag material that tends to bead up on the surface and drip off (presumably can be refreshed with the right goop).

SiC is a modest conductor when heated, making microwave heating possible, and maybe induction still (but maybe with higher voltages/frequencies than are available here).

SiC-graphite crucibles can practically be quenched in water; they're very tough, as ceramics go. :)

Tim

Maybe it will do induction with insulation?  If not, then this crucible may be relegated to melting lower temp metals and/or internal non-ferrous heat treating..

Microwave melting sounds interesting - wonder how long to super heat a pound of bronze in modified microwave with a typical 1200 watt microwave. 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: flyrod on June 22, 2019, 08:36:34 PM
I finished the circuit for the unit.

Thank you for posting that!  I think that explains the 6 small transistors I was seeing in the ebay pictures.  I tried to ask about it in this thread:

https://highvoltageforum.net/index.php?topic=167.0

So the 12v for the gate drive comes from a separate DC-DC board?

Also, what is the big capacitor that is not part of the tank?  Is this C8 in your diagram?

Finally, are there no bypass capacitors on the supply to Q3 and Q4?  If not, I would be curious to see what this point looks like on a scope.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on June 24, 2019, 04:58:39 PM
I finished the circuit for the unit.

Thank you for posting that!  I think that explains the 6 small transistors I was seeing in the ebay pictures.  I tried to ask about it in this thread:

https://highvoltageforum.net/index.php?topic=167.0

So the 12v for the gate drive comes from a separate DC-DC board?

Also, what is the big capacitor that is not part of the tank?  Is this C8 in your diagram?

Finally, are there no bypass capacitors on the supply to Q3 and Q4?  If not, I would be curious to see what this point looks like on a scope.


Hi there,

The first transistor switches the supply voltage for the gate drivers on/off, the second one switches the relay and the other 4 are the gate drivers.

There's an additional 12V power supply for the microcontroller, fan, pump and the gate circuit.

The capacitor C8 reduces the current/voltage spikes during the mosfet switching, reduces the impact of wiring inductances from the power supply and smooths the measured voltage for the microcontroller.

The transistors have no additional parts, there is no gate resistor, unfortunately.

I tested my new arrived "IXFX100N65X2-ND" mosfets and they didn't really work. During the ordering I didn't notice the slower body diode.
The heater didn't turn on at voltages over 40V or with a load and stalled at 20A current using. So I hooked up my Tektronix 2225 with my "very fine" 100:1 Chinese probes and measured the gate signal with the analogue differential add function. With no loads it was a near perfect square wave, not like the simulated one, but with increasing current the ringing was really visible.
Like 20V peak-peak at 20A. Could be a bad measurement setup, the Chinese probes or most probably the slow diodes.

Similar to this, just worse:
https://highvoltageforum.net/index.php?topic=648.msg4550;topicseen#new

As supposed here, it gets worse with the rising current.
https://www.mikrocontroller.net/topic/207864#2057621

Also the Voltage in the oscillator tank dropped at 20A to 60% from the unloaded one.

So I'm waiting for the real 260n mosfets, measure them with the scope and maybe buy some other mosfets with higher voltage ratings.
Something like these:
https://www.digikey.de/product-detail/de/rohm-semiconductor/SCT3030ALGC11/SCT3030ALGC11-ND/6204894

But I want to check the circuit with the original mosfets first, maybe they are also ringing, but with them it doesn't matter as much.

Greetings
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 25, 2019, 05:45:55 AM
On the melting side of things, I found some good melting/foundry links:

https://ia600205.us.archive.org/23/items/Handbook_of_Lost_Wax_or_Investment_Casting_Sopcak/Handbook_of_Lost_Wax_or_Investment_Casting_Sopcak.pdf

On molding here is an oldie but goodie.

https://books.google.com/books?id=c98mAQAAIAAJ&pg=PA88&dq=induction+furnace&hl=en&sa=X&ved=0ahUKEwjHvIjwu43iAhVCPK0KHR5XCtkQ6AEITTAG#v=snippet&q=frequency&f=true

A Dental Casting Manual of Sorts:

https://ia800106.us.archive.org/35/items/DTIC_ADA173766/DTIC_ADA173766.pdf

Dude who tours knife-making- Note the flatness of the swords and blades - Uniformity in thickness:

/>

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on June 29, 2019, 06:58:06 PM

0. Use a thick, fairly conductive crucible (e.g., steel) to shield the melt from magnetic field.  (Marginal for aluminum, and not practical for cuprous alloys.)  Note that graphite has to be quite thick to provide much shielding.


Tim

Too thick is bad as well.  This quy could not melt anything with a 20mm thick custom made crucible.  More power required to heat larger crucible so can't get to superheat pouring temps.

https://highvoltageforum.net/index.php?action=post;quote=4362;topic=530.120

UPDATE:  Correct link should have been:
/>At the end of video he realizes that 20 mm is too much of a load.


Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on June 29, 2019, 08:34:31 PM
Maybe we should split the topic, so that this stays for electronic discussion of the heater and the heating of metals, but the melting, casting and crucible discussion has its own thread?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on July 04, 2019, 06:19:51 AM
Pushing the limits with a small work coil for annealing brass cartridges.  For this video, I make and use the smallest diameter work coil yet for the 1000 watt ZVS induction heater.   Although I focus on annealing a brass cartridge casing very quickly, this video is really a next step in my ongoing series about work coil design and  work coil frequency limits for reliable circuit operation.
It is here:
/>
(I stopped just BEFORE melting anything)

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on July 07, 2019, 04:06:47 AM
Finally got my ZVS unit.  While hooking things up, I realized that my planned auto fuel pump is not designed for pumping water and would ruin the pump in short order.  Wondering what would make a good pump?  I scavenged a small radiator and a huge server fan that will be used for water cooling. Possible pumps I have thought of:  submerged aquarium pump, auto windshield pump - any other ideas, and I assume flow rates of above mentioned pumps would be sufficient?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: johnf on July 07, 2019, 04:48:48 AM
garden supply store for fountain pump or similar
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on July 07, 2019, 05:24:20 AM
While the garden pump mentioned by johnf might work okay, I suggest this one (or one like it):
https://www.amazon.com/Yosoo-Circulation-Brushless-Selling-Quality/dp/B00PXXJEHE/ref=asc_df_B00PXXJEHE/?tag=hyprod-20&linkCode=df0&hvadid=198097979185&hvpos=1o5&hvnetw=g&hvrand=17395343945830728741&hvpone=&hvptwo=&hvqmt=&hvdev=c&hvdvcmdl=&hvlocint=&hvlocphy=9019020&hvtargid=pla-348418857326&psc=1
  It's about USD $20.00, but is rated for hot water, which certainly will be an issue.
As I mentioned in my most recent video, I was pleasantly surprised at how much flow I got from it even when using the 1/8 inch OD copper tubing for the coil.  I also suggest that you try to minimize the "head" that the pump has to work through.  In other words, don't put the radiator  all the way down on the floor making the pump have to work real hard.  As the "head" gets higher, the pumping rate and pressure goes down.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on July 07, 2019, 05:28:04 PM
Subtopic:  Adding a (Cheap?) frequency meter (counter) to my induction heaters-

I just uploaded this page:
https://spaco.org/Blacksmithing/ZVSInductionHeater/MakingAFrequencyCounterForNoisyAndSineWaveInputs.htm

It's actually a story about what NOT to do, but it turns out okay.


OT for hightemp1:  That water pump that I recommended----  I would not consider it to be submersible.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on July 07, 2019, 08:19:33 PM
Pushing the limits with a small work coil for annealing brass cartridges.  For this video, I make and use the smallest diameter work coil yet for the 1000 watt ZVS induction heater.   Although I focus on annealing a brass cartridge casing very quickly, this video is really a next step in my ongoing series about work coil design and  work coil frequency limits for reliable circuit operation.

Pete, I love your videos, you really get deep into the topics, do a lot of experiments and share your results. Kudos to you Sir.

You video editing was also really good in this video, keep up the good work, it had a nice flow and I enjoyed it from start to end!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on July 08, 2019, 09:13:42 PM
Thank you for your kind words, Mads.

Now I want to talk about a "Cheap" Frequency counter for the ZVS Induction heaters:

I am certain that most people who get into this  "World of  Electrical Oscillations" do have frequency measuring equipment.

I have an old Tektronix 422 Oscilloscope that I use to roughly measure frequency by counting  the number of cycles I see on the screen at a certain sweep rate.  I also have an inexpensive tiny battery powered DSO-112 touch screen  oscilloscope that has the ability to give me a digital readout of the incoming frequency (if I push the right spot on the screen at the right time and have my magnifiers on).

  That's not much in the way of frequency measuring equipment, but it's enough for my BENCH needs at present.

  However, my projects are often scattered about my shop and I can't always get my bench scopes to the unit under test.  And, the touch screen of the little scope isn't all that handy for big fingers when I am in a hurry.
  I currently have two different ZVS induction heater systems.  My testing has made me realize that it is important, for several reasons, to know the tank frequency at any given time, so I decided to dedicate a frequency counter to each machine so I'd instantly and always know the tank frequency.
  To that end, I went through an all-too-lengthy process to choose/build/connect "cheap" frequency counters to my heaters.  (I even built two extra sets while I was at it).
  Here is a webpage that I just put up which takes you through the whole process:
https://spaco.org/Blacksmithing/ZVSInductionHeater/MakingAFrequencyCounterForNoisyAndSineWaveInputs.htm

Pete Stanaitis
--------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on July 09, 2019, 08:45:17 AM
Good thing that my multimeter can handle these frequencies, after making a new coil it's always interesting how the inductance is.

As I said earlier I had some ringing Problem with the IXTH62N65, because of there really slow body diode (typ. at 25°C 445ns).
As a comparison, the IRFP260N (typ. at 25°C 268ns) there's also some ringing but it's not consistent and gets better with heavy load.
I also tested some FGH80N60 (typ. at 25°C 61ns).

The FGH80N60 worked just fine but they got really hot.
So I ordered some IKW50N65 (typ. at 25°C 52ns) which should be fast and cool enough. I'm not sure how the relative slow turn off delay time will matter, but I'll see.

The next step for 200V input voltage would be some good capacitors, but there are the next problems.
With these high frequencies and "high" voltages of the standard 2000W heater model, the capacitors are already at there current rating with only 50V input.
So that's not a long term solution. Other similar capacitor like the used MKPH-R or Dawncap aren't that suitable. If the overall capacitance is too low the frequency rises, which is a bad thing for the slower IGBT. If the capacitance per capacitor is too high, the currents are way above the ratings.
So more small capacitors would be the way to go, but these won't fit more than four on the original board.
I could make a new board/circuit, but the start/stop routine works really good and I want to keep it (no need for turning off the big PSU...).

With 4 capacitors which cost 15 to 25 € I can also go with some seriously big stuff like this:
http://www.eectech.store/cs-30122-solid-state-high-frequency-film-capacitor-12uf-500vac-p0319.html
Which can handle the voltage, current and frequency. It's even smaller than 4 big film caps. With some cutting at the PCB it will fit nicely.

However, my previously statement with the 6mm copper pipe isn't true in that case and a10mm pipe is probably more suitable.
Since I have two units, I put 4 originals caps on the IRFP260N unit with 6mm; 1/4" connectors and the second one with the new cap, IKW50N65 and 10mm; 3/8" connectors.

That should power things up. The oscillating current should increase with the higher capacitance which lead to a smaller working coil, which should be better for screws and non crucible stuff, which I use it for.

There's also a mains powered updated version of this circuit:
http://www.joretronik.de/Oszillatoren/Oszillatoren.html
This can't hang up at start up and can't fail like the royer.
Since my PSU is limited somewhere to 4kw-6kw I don't see a need to go higher up in voltages/complexity right now.
I'm not that firm with high energy stuff at that high frequency. For a proper circuit you would need a start/stop routine, over current protection, a relay and so on.
I try to get this thing working, build a housing and that's it, since i have seen many "super big" induction heater projects which never finished...
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on July 17, 2019, 07:56:41 PM
'Morning all & Happy Saturday!  Newbie here. :)

Today I'm working on a PS for my recently purchased 1800W unit.  After seeing Mads videos, I obtained some HP DPS-750RB PS's and am performing the isolation mods as I type this.  I understand everything except for what to do with the ground pigtail that was captured under the of longer of the board mount screws.  Should it be clipped off & discarded or reinstalled under the nylon insulator?

I noticed on another forum that it's recommended to leave one of the series connected PS's un-modded (not isolated) but Mads modifies all 3 in his video.  It looks like either way will work but is one preferred over the other?

DICKYBIRD, how are these in-series server power supplies working out for you?  Supposedly, there are two 750 watt versions and one of them can not be DC isolated.  Which version are you using?

MADS, I just noticed from a prior post (page 2 or 3 of this thread) - that you may be assembling some of the units together for resale, and that you are waiting for fiberglass braiding to complete them.  Have you received the braiding and please update us on your progress?

I am also attempting to run four of these together but am confused on DC isolation and possibly burning them up if they are not DC isolated.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on July 18, 2019, 05:49:27 AM
I'd personally ground (by not modifying) the supply in the middle of the chain - that way each end is only half the voltage with respect to earth. This would not work if you had to put a non-differential scope probe on the IH though.

Hydron, where would you put the DC grounded power supply if you had four units in series (ie 2nd or 3rd position?).  My tests show that 3 of my 4 power supplies have the dc ground isolated from their individual cases.  When I put them in series with the grounded one in position 1 as MADs advises, then I get 50 volts when measuring all four DC rails to cases.  Don't know if this is best position or, frankly, if present configuration has the DC "floating". :-\    Any and all advice greatly appreciated.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: DICKEYBIRD on July 19, 2019, 01:06:44 AM
Hiya hightemp.  I did the isolation mod on all 3 of my PS's and got over 36v.  I tried leaving 1 of them un-modded as suggested elsewhere but it wouldn't work.

As far as how well they work under load for long periods, I can't say yet as my IH project had to be put on hold due to some paying machining work.  I did run a 36VDC motor with it for an hour or so & it worked fine.

I hope to finish up the project soon.  I can't wait to see it work! :)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on July 19, 2019, 02:23:29 AM
Hiya hightemp.  I did the isolation mod on all 3 of my PS's and got over 36v.  I tried leaving 1 of them un-modded as suggested elsewhere but it wouldn't work.

As far as how well they work under load for long periods, I can't say yet as my IH project had to be put on hold due to some paying machining work.  I did run a 36VDC motor with it for an hour or so & it worked fine. 
I hope to finish up the project soon.  I can't wait to see it work! :)

DICKEYBIRD, I read as you did that leaving one of the PSs grounded is best - like you I don't know alot about electronics and really don't understand why this method is better (other than supposedly it leaves you with floating DC ground intact on all of them and is safer for both equipment and humans).  Hydron has stated that putting the grounded PS towards the middle of a series connection may offer extra safety by lowering the float voltage apparently - I may try putting the grounded one in position 2 or 3,  but would prefer some expert advice beforehand?

The reason I ask which model you have is because another forum (an RC battery website) says Model pl-18 supposedly can be DC grounded, but PD-18 can not be (it will work for a while but eventually shorts itself out).  I have the pd-18 model and may have figured out how to ground it, but don't know a surefire test for a DC float series connection with one unit un-grounded.  Any advice from engineers or the likes much appreciated.

Tests Done so far:  All three adapted supplies individually test like they are DC isolated with multimeter (0 volts when case is probed to DC rail).  Also individually, each one powers a 12 volt car headlamp equally well.  When all 4 are put in series, in the order MADs advises, then I get 48 volts.  A Series DC ground test shows 53 volts on any of the rails to any of the cases (does not mater if all cases physically touch each other or none of them touch - still get 53 volts).  Am I correctly floating DC ground? What happens on full power if I accidently touch the case?

update: I HAVE ABANDONED THIS 48 VOLT DC FLOATING GROUND OF FOUR HP POWER SUPPLIES CONNECTED IN SERIES. FLOATING MEANS IT HAS NO DC GROUND AND ANY DEVICE OVER 50 VOLTS IS REQUIRED TO HAVE GROUND - TOO CLOSE FOR COMFORT & TOO MANY UNANSWERED QUESTIONS.  FURTHER, THIS HP MODEL HAS CONTRADICTING DOCUMENTION ON FLOATING PROCEDURES AND NO INFO ON GANGING THREE OR MORE PS. FOR AN ELECRONIC'S KNOB LIKE MYSELF IT IS JUST TOO RISKY.  BETTER TO SPEND HUNDRED DOLLARS ON SIMPLE ONE UNIT 3KW SUPPLY.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: DICKEYBIRD on July 19, 2019, 03:24:38 AM
I just checked & I have the-PD18's as well.  I sure hope they keep on working.  I tried one un-modded PS in several different locations in the series but nothing worked so I modded the 3rd one & all seems to work well with no voltage fluctuations, beeps, etc.  I think the motor I ran in the test pulled maybe 6A or so & the PS fans speeded up somewhat a couple times.  I can't answer your DC floating questions because I'm definitely not electrically knowledgeable...just a 71 year old auto dealer service guy & self-trained weekend machinist/tinkerer that surfs the web & youtube for electrical help.  Maybe somebody here can enlighten us?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on July 19, 2019, 03:36:44 AM
I just checked & I have the-PD18's as well.  I sure hope they keep on working.  I tried one un-modded PS in several different locations in the series but nothing worked so I modded the 3rd one & all seems to work well with no voltage fluctuations, beeps, etc.  I think the motor I ran in the test pulled maybe 6A or so & the PS fans speeded up somewhat a couple times.  I can't answer your DC floating questions because I'm definitely not electrically knowledgeable...just a 71 year old auto dealer service guy & self-trained weekend machinist/tinkerer that surfs the web & youtube for electrical help.  Maybe somebody here can enlighten us?

DICKEYBIRD, you can test to see if each PS is DC isolated, I think?  What I did was separated them completely, turned one on and check volts between positive rail and case - it should read zero.  Mine, after initial MOD, all read between 6 and 9 volts, if I recall correctly.  After isolating an extra daughterboard ground screw within the case, and the additional back DC ground screw to the case, then I got 0 volts on each one.   

Here is the link I was referencing:

https://www.rcgroups.com/forums/showthread.php?3137941-Converted-DPS-750-PSU-info_dump

Yes, we definitely need some enlightening.   MADS 15 minute burn test may have been insufficient,  and maybe during your test, the fans speeding up should not have occurred if motor load was constant?  I think MADS said in his video that the fans flucuated also but said they were the fans on the ZVS unit and that was normal.  Though he did say the PS cases were quite warm to the touch and water temps rose quite a bit but figured those high case temps were normal because the PSs were so densely power packed.

Also wondering,  when you say 'nothing worked when you left one PS grounded to DC in any series orientation', do you mean that you got zero volts out of all of them while in series, or that it would not power the motor properly, or did you mean something else?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on July 19, 2019, 11:19:45 AM
Hi to everyone,
My intension to use the original 2000W ZVS circuit with other Mosfets/IGBTs wasn't a success.
Furthermore, due to the non optimized control for the IGBT the new ones died again after around 30s of solid working with no obvious reason, no ringing at the gate or so. The build in fail-safe, which should open the relay in case of an over current, didn't work and the relay welded itself to a permanent on position.

After that I salvaged the few remaining parts.
I'm now going to use the circuit from Jörg Rehrmann.
I will have LESS fail-safe options, but these can handle the current and voltages regardless of the case.
It's basically just a mains fuse at the DC side, which won't like the DC, but it's cheap and will do the business and a large primary contactor, which is manually controlled. So I can decide, when it's too much current.
With my analogue ammeter I can assume the used power.

http://www.joretronik.de/Oszillatoren/Oszillatoren.html (http://www.joretronik.de/Oszillatoren/Oszillatoren.html)
Last one on the site. I'll keep the 18V control-voltage always on and switch the primary voltage, so the IGBT's won't come in strange half open states.
According to the designer, this circuit should always oscillate.

The Chinese board always switched off randomly even at 75% Voltage/current, that was a pain, it randomly said over current/voltage/temp....

So my inertia question:
Is the circuit capable for higher power level --> I wouldn't recommend it for more than 1,5kW.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on July 21, 2019, 11:18:06 PM
Work Coil Lead Extension:
Short video on Swaging Copper Tubing is here:
/>
  Sometimes, when making a work coil, I don't make the leads long enough and have to add several inches to one or both ends.  Sometimes it's just laziness. I often don't premeasure the stock.  I just start with the  remainder of the tubing roll that I got from the hardware store and then cut the new work coil off that roll when I'm done.
  So sometimes I need to make minor adjustments to the length of coil leads to get them going in the right direction.  One way to deal with lengthening leads is to simply buy brass unions, with either flare or compression fittings.  Of course they cost several dollars each.  One could also use copper sweat tube unions, but they are hard to find for small tubing.  I even tried finding some 5/16" tubing to use for a union, with no success, even at a big box store!
---And, I have to go to the store to get them when I need them.
  So, in looking for alternatives, I found a few videos of guys swaging the end of the tube with any one of various tools.  The swage simply opens up the end of one tube so the other tube slides inside by about  1/4" and then you soft solder the joint.
  Well, I liked that idea but didn't want to spend anywhere from USD$ 20 to well over USD$ 100 to buy a swage or a swaging set.  So I made my own, just for the 1/4" tubing size that I usually use.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on July 22, 2019, 12:00:18 AM
Pushing the limits with a small work coil for annealing brass cartridges.  For this video, I make and use the smallest diameter work coil yet for the 1000 watt ZVS induction heater.   Although I focus on annealing a brass cartridge casing very quickly, this video is really a next step in my ongoing series about work coil design and  work coil frequency limits for reliable circuit operation.

Pete, I love your videos, you really get deep into the topics, do a lot of experiments and share your results. Kudos to you Sir.

You video editing was also really good in this video, keep up the good work, it had a nice flow and I enjoyed it from start to end!

Mads, I agree completely. 
Peter, all of your videos are very thoughtful and helpful.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on August 03, 2019, 03:24:12 AM
Testing a large "Pancake" work coil.
   I just uploaded the test of my latest work coil.  It's a 10 inch diameter spiral or "pancake" coil.
It is here:
/>  There are lessons to be learned in this video, but I didn't find this coil to be as useful an everyday tool as I thought it might be.
I hope some of you  have some questions or ideas about this test.
At this point I think I have found the approximate range of frequencies that this 1000 watt heater can work successfully at.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on August 03, 2019, 06:35:43 PM
Wondering if a two dimensional coil would be any better.  And, for non-magnetic and/or thinner sheet metals, possibly wrapping work-piece around the outside or inside of cylinder shapes made of iron, graphite or high resistant silicon.  Also wondering if even heating of sheet metal would not be more suited to resistant ovens with fine temperature controls, and of course the use of simple propane or gas/ox torches that are commonly used.  But, induction heating is highly versatile, adaptable and efficient with correctly shaped coils so I would think heat treating would somehow work just fine.  Peter, try using it to boil corn or maybe a fish boil  ;)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on August 04, 2019, 11:51:14 AM
Probably just that the impedance is so high, from the high inductance.  Same overall dimensions but fewer turns, should be back with normal power levels I would think.

Peter, try using it to boil corn or maybe a fish boil  ;)

Back when I was working at an induction company, there were at least several lunches I ran on the machine in the proto lab.  Seems that a can of Chef Boyardee can't take much more than a kilowatt, two even with stirring, and so the heat time is about two minutes.  Just as you'd do with a bowl of the stuff in a microwave oven, oddly enough. ;D

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on August 04, 2019, 05:48:18 PM
Quote
Probably just that the impedance is so high, from the high inductance.  Same overall dimensions but fewer turns, should be back with normal power levels I would think.
Regarding "should be back with normal power levels":
   For this 1000 watt heater, 22 amps at 48 volts IS the "normal" max. power level.

Regarding cooking with it:
   I think I'd rather buy an Ikea HOB (Hotplate Or Burner, as the English say) for USD$ 88.00.  It is clear the we CAN heat the bottom of a ferrous pan pretty well.

I will probably run this coil on the 2500 watt unit one of these days to see if and how the results differ.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on August 07, 2019, 02:32:25 AM
Okay---  Here is the result of attaching the 10 inch OD "Pancake" work coil to the 2500 Watt ZVS Induction Heater and testing it:
/>
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on August 14, 2019, 03:42:34 PM
Hi everyone,

Last week I finished my own ZSV heater circuit.
After some many cooling problems and my first self made water coolers it worked great.
The circuit from Jörg Rehrmann works fantastic with no issues.
The capacitor from Eectech has 4.0µF 400Vrsm 700Arms and don't get that warm.
For cooling I have two of the Chinese water pumps in parallel, of course watercooled (they get really hot after some time) with a 30 litre canister.
With pancake coils the circulated water is around 30 °C warmer than the supply water.
With my nice case I can touch most of the heater without shocking my self.
I confirmed 5kW at 50A and 100V at the input. The water cooled IGBT are getting up to 80°C at 50 A, that's OK, not great but solid.
I plan to make some more professional water coolers and upgrade the pumps, the cooler the system runs, the more efficient it gets.

For me the 650V IGBT are more than enough and it was the right decision to not use the not so suitable 1200V IGBT.
The stainless steel is glowing at 1150 °C and was melting from behind.  :o
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on August 14, 2019, 07:05:57 PM
For Bert911---
  Good that you have your induction heater running now.
What applications for this induction heater do you have in mind?
Can you tell us the frequency of operation and inductance of that pancake work coil?

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Lane on August 15, 2019, 01:51:04 PM
Nice to find this place. I've been playing with induction on and off for a while; but just now trying to wrap my head around this beast. I thought it was strange at first that only 5% of the power was going in to melt a non-ferris metal; but then I ran the numbers of specific heat of graphite...

Then we can talk about mass of graphite, position in/around the coil, size, volume, exposed area, coupled volume, etc... I'm not at all shocked someone was able to melt a few kg with 1000W in an hour with good insulation. I've recently found some good references on painting a ceramic coating on (by yourself) to standard graphite crucibles though; that should help a lot to avoid oxidation. Heating for a full hour might require a different coil setup, loading method, etc.

I was curious about why bronze casting might be bad with induction... I presume melting copper first and then adding bronze would work just fine. Just might not be able to load the crucible full of scrap bronze up front?

I've got a lot to read; and a lot to learn, but sounds like a few of you have it under control here to some extent. My early experiments proved about 20 Wh consistently heated a 20mm crucible to copper melting temperate. Even as the crucible degraded (and current dropped); the 20 Wh metric was still sound. Look forward to comparing numbers on larger metal pours with you all. Hey Pete (you specifically know I'm not melting copper metal)!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on August 15, 2019, 06:17:55 PM
Hi everyone,

Last week I finished my own ZSV heater circuit.
After some many cooling problems and my first self made water coolers it worked great.
The circuit from Jörg Rehrmann works fantastic with no issues.
The capacitor from Eectech has 4.0µF 400Vrsm 700Arms and don't get that warm.
For cooling I have two of the Chinese water pumps in parallel, of course watercooled (they get really hot after some time) with a 30 litre canister.
With pancake coils the circulated water is around 30 °C warmer than the supply water.
With my nice case I can touch most of the heater without shocking my self.
I confirmed 5kW at 50A and 100V at the input. The water cooled IGBT are getting up to 80°C at 50 A, that's OK, not great but solid.
I plan to make some more professional water coolers and upgrade the pumps, the cooler the system runs, the more efficient it gets.

For me the 650V IGBT are more than enough and it was the right decision to not use the not so suitable 1200V IGBT.
The stainless steel is glowing at 1150 °C and was melting from behind.  :o
Good to hear that it is working nicely.

I have looked at the schematic, since Rehmann has it protected I assume it is quite good. Would you mind showing us yours, ie with parts?

You use a separate power supply of 18V (always on) instead of the R1,ZD1/C1 net? And 100VDC feed of course?

I guess the schematic lends itself to an LTSpice simulation or two...

I have a bunch of 48V server supplies, some NOS (2000W) and some legendary ESP120 (used) that I was thinking of series connection two for 100VDC (well 102,8 V nominally...) . I was just wondering how to protect them, I assume distance (ie cables, inductive decoupling by series inductances, over voltage protection (fast) )
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Lane on August 16, 2019, 11:22:45 AM
Just before my first near-catastrophic failure. There was a bright light in the wiring of the induction coil feed cable. Should have known better with those fat copper tails hanging around on top of the bad connection. Fact is; I am not sure what to use for a PCB connector with this type of 8 ga. stranded wire. This wired setup does help to isolate the coil/work heat from the PCB. Unfortunately those hookup parts that came with this 1000W unit were always junk.

Been running this for quite some time... At 24v briefly, then 48v; with a factory new chopped off 20mm (height) x 40mm (diameter) dense graphite crucible floating in the coil; it can draw in excess of 18A at 48v. As the crucible degrades it drops down as low as 10A. Still getting little more than 20Wh consumption when heating the larger crucible to melting temperature (though it's quite short and small in volume).

Tough to see; but the HAZ on those copper strands is widely varied. Some of them were glowing red hot after a very short heating session. I had to cut it off at 7Wh when I smelled burning. Time to refactor; not sad that I got a lot of milage out of this already (7,500 Wh and more than a dozen 20-40mm diameter crucibles).
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on August 16, 2019, 01:49:53 PM

Those standoffs on the PCB have worked okay for me for quite some time, but I do use 1/4" OD copper tubing coils to connect to them and  I water cool the tubing.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Lane on August 17, 2019, 06:21:45 AM
Thanks Pete. I looked at the 1,000W unit more closely with a clear head. There was a good amount of heat introduced to the end of the wire due to a poor quality connection where the wire met the standoffs. I cut off the end of the wire and checked the screws into the PCB (all still tight). Put everything back together, and it pulled 14A again with a very degraded (empty) crucible. So I'm back in production with the 1,000W unit.

On the other hand; I have a 3,500W unit showing up in the mail tomorrow. I don't know all that much about it at the moment. It was $35 though; so I figured it was worth an experiment or two. Even if I only use it for simple tasks it's a good deal at that price. I was able to find a very similar model that had a schematic in the manual. There appear to be some very minor differences between the schematic and this board; like the prescribed inductance of the work coil. The microcontroller appears to be socketed; so it might not be too difficult to start fresh with customized software down the road. In the mean time; I would be happy to have some level of power control and a timer. Definitely planning to use coils that aren't pancake shaped for these experiments. Looked over your work coil spreadsheet Pete, and it appears I may want to add capacitors to this unit? Hitting the specified 120µH coil inductance (20-30kHz operating freq.) might be hard going by the data you've published.

https://www.amazon.com/gp/product/B07M64PGH3

Also have a 48V 2,500W unit a few days out; but don't have any serious plans for that extra power yet. I simply need another unit online so I can allow one to be cooling while I continue with other hot work.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Lane on August 17, 2019, 10:45:26 AM
Earlier Self quote: pulled 14A again with a very degraded (empty) crucible. So I'm back in production with the 1,000W unit.

Figured I would post a bit more technical detail while I'm thinking about it; and can test things again. Here is a look at my collection of new and degraded crucibles. I get at least 20 full heating cycles (>2,000F) out of graphite crucible with a ceramic fiber insulation mostly insulating things (seems like the more the better). Those heats take between 1.5 minutes and 3 minutes depending on how used the crucible is. I can get closer to 50 total heating/cooling cycles (before the crucible turns to dust) if those are split between full melt temperature and lower temperature annealing cycles. I've read that a paint on ceramic coating might help prolong the life of off the shelf pure graphite crucibles.

The 2 oz. crucibles are on the right; and a selection of cut 40mm (largest), 6 oz. and 3 oz. on the left. Due to wall thickness and density differences; the cut 40mm, cut 6 oz. and uncut 3 oz. all weigh between 25-30 grams; while the 2 oz. is only 13.5 grams (when new).

This particular setup draws around 19A with a 30 gram graphite crucible partially inside a 3 three turn 2 layer 45mm I.D. coil. I could likely draw more current if I fully submerged the crucible; but I'm already near the upper limit of a 1,000W unit; so I tune current draw by moving the crucible up and down. I have tried an uncut 40mm x 40mm crucible at 44 grams; and it draws far too much current this way.

While these are small; the whole range of crucibles all used 20Wh of energy to hit the >2,000F metric. It's a bit less than 20Wh with a warm graphite crucible; and can be a hair more with a larger load of metal. While the lighter 13.5 gram 2 oz. crucible draws less current; it still required 20Wh to hit melting temperature. Clearly data isn't going to hold up with larger melts; but it's a starting point in terms of my observations this far. I rarely go over an ounce in terms of metal weight with this setup; so that is very much negligible in terms of heating power/time for me.

I'm curious to see what I can do with kilogram sized crucibles; but a quick check indicates that a 1kg crucibles weights 140-160 grams, 2kg 240-330? grams, 3kg 340-360 grams, 4kg 447 grams, 5 kg 690 grams. Those were the best ranges I could find in terms of weights from product listings at least...

Unfortunately I can't tell what frequency my current furnace is operating at; or what my coil inductance is. The LC100 meter I bought years ago arrived broken (displays an error when attempting to measure low range inductance). Waiting on a replacement to show up now that I actually need one. Still need to build a frequency counter display when I get around to it as well.

Obviously Pete's work has helped me a quite a bit in getting this far; thank you very much (again) for all that. I'll try to condense some of data collection to post up as I have time. Here is a quick look at a melt cycle for 2.900 grams of non-ferrous (group 11). Took only 15Wh, and 1:35 seconds even in a heavily degraded crucible. That old wiring might have hobbled some of my early data (too much energy usage for same work); but even so, breaking 2,000F in 1 min 30 seconds is totally decent either way in my opinion. Here is a look at the meter pre-melt (from cold temp), cold furnace; first start wattage/current, then furnace color and final Wh count.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Bert911 on August 18, 2019, 01:04:35 PM
So many posts  :D

What applications for this induction heater do you have in mind?
Can you tell us the frequency of operation and inductance of that pancake work coil?

For now, I'm using an induction heater for my diploma thesis, later I want to use it for metal working, losing screws, heat treating and cleaning. Heat works like a charm for removing grease or paint, no chemicals needed.
This coil runs at 86 kHz with 0,86 µH of inductance.
From my tests for pancake coils I noticed a drastic increase of loss in the inner winding the smaller the inner diameter gets.
I wouldn't recommend smaller than 25mm, better higher.

I've recently found some good references on painting a ceramic coating on (by yourself) to
standard graphite crucibles though; that should help a lot to avoid oxidation. Heating for
a full hour might require a different coil setup, loading method, etc.

I used some coating for stainless steel containers to protect it from the aluminium:
3M Boron Nitride Suspension WP

http://technical-ceramics.3mdeutschland.de/fileadmin/content/download/produktinformation-EN/PI_boron_nitride_suspension_wp_e.pdf

The coating works great but is very sensitive to mechanical scratches, but there it's undamaged it's doing a great job of protecting the steel.
At the damaged area, the 12 mm thick steel was eaten away during 4 hours!

I have looked at the schematic, since Rehmann has it protected I assume it is quite good.
Would you mind showing us yours, ie with parts?
You use a separate power supply of 18V (always on) instead of the R1,ZD1/C1 net? And
100VDC feed of course?
I guess the schematic lends itself to an LTSpice simulation or two...
I have a bunch of 48V server supplies, some NOS (2000W) and some legendary ESP120 (used)
that I was thinking of series connection two for 100VDC (well 102,8 V nominally...) . I
was just wondering how to protect them, I assume distance (ie cables, inductive decoupling
by series inductances, over voltage protection (fast) )

LTSpice isn't that good at simulating IGBTs, I couldn't get it to work and I don't think the simulations are representative anyway, since ringing and non linear inductances may occur.

The boom of this circuit is attached and can be loaded directly into Digikey. I used the MUR1560 diode in the circuit, which is totally over dimensioned, but the size of it bothered me the most.
I would recommend the MUR460 in the circuit and two MUR1560 at the input terminals.
I can't give any recommendations for the inductors since I don't know the value and the one's from Digikey aren't really suitable.

I have some normal reset able house fuses in parallel for some simple current protection, but if they switch, there could be a huge voltage spike, which should be covered with the MUR1560 Diode at the input.
I used a 12V power supply for the fan and the cheapest dc-dc converter from ebay to generate 20V, which is then reduced to 18V, not that necessary, but won't hurt anyway.

https://www.ebay.de/itm/3x-XL6009-DC-DC-Boost-Modul-Step-Up-LM2577-Schaltregler-Konverter-Arduino/252796314014?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

The fan and the control voltage is always on, it is safer to apply the control voltage before the main voltage for a safe oscillation.
I like to slowly increase the supply voltage to see how the current rises.
I also use a power relay to switch the transformers on and off but I have done this just up to 30A to reduce the inrush currents.
You could simply switch the supplies on and off.

Next week I can make some close-ups from the circuit.
Since the circuit components aren-t that expensive I will build a new version with 4 IGBT'S (two in parallel) with some better hose management.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on August 19, 2019, 09:08:12 PM
Back on topic for the 12- 48 volt induction heaters, for a bit----
Here are two recent questions from one of my "pancake coil" youtube videos that I answered.  Hopefully I got the answers correct:

1st Recent question:
Hi, Is Coil Voltage also 48V or more than that?
If Yes, is there difference between 2500W, 1800W and 1000W?

Answer:
Yes, the voltage across the work coil is a lot higher than that of the power supply.
I just measured the voltage across a 12 turn 2 inch ID work coil on my 1000 watt heater with a multimeter and it is 220 volts rms.  I also looked at it with my old Tectronix 422 scope and then remembered why I don't pay much attention to the voltage across the work coil.  When I connected the 10X scope probe and turned on the power, I got a 300 volt peak to peak signal (at about 65kHz) that was approximately sinusoidal.  But, after about 30 seconds, the scope probe became too hot to touch, the insulation on the ground lead melted, and I almost burned it up.  The scope probe circuit's inductance (whatever it was) tried to act as a parallel work coil, but with only tiny conductors!!!
  I don't think these readings would change very much for the 1800 watt or 2500 watt heaters, but I have not measured them.  I really don't need to do that for my applications.
  For what it's worth, the experts say that the voltage across the Mosfets will equal
  the power supply input voltage X pi.

2nd Question:(In reply to my earlier answer)
I want to measure frequency with my multimeter it's up to 250V. I guess can't do with it.

Answer:
If your multimeter reads rms AC voltage, it should be okay.
 If you live in the USA, check the voltage at a  power outlet in your home.  If it measures approximately 120 volts, then your meter is reading rms.  If this is true (which is  quite normal for a multimeter) then you will be able to read the voltage across the work coil.
As always, use caution when measuring voltages this high.

(By the way, I find the posts about 220 volt AC input circuits interesting, but I that other readers realize the real dangers of working with power line voltages and currents.)

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on August 22, 2019, 07:55:15 AM
Very few multimeters ca measure voltage in the 10 kHz range, due to bandwidth limitation. Not really necessary with Trms, as the resonance creates a fairly sinusoidal wave. Watch maximum input voltage too, typically they are still high impedance at a few ten kHz, but better check it.
Same goes for scope probes. Normally they catake full voltage rating up to a few hundred kHz

I wonder if your probe heating was due to the high magnetic field created eddy currents in your probe. In that case a multimeter probe, which is constructed of much thicker wires would go quickly?

And this all depends on how the field is, probe alignment etc.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on August 22, 2019, 09:47:58 PM
The problem with the melting oscilloscope probe ground lead is that you are grounding your inverter through your oscilloscope! To measure on inverter outputs you always need a differential probe or an isolation transformer for the mains feeding the inverter and keep it floating off ground. I always recommend this great video:
/>
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Lane on August 23, 2019, 09:29:02 AM
Bert911: Thanks for the tip about the 3M coating material. I'm still not sure if a coating will contaminate my work in some way; but I'll keep looking into that, and will try something as I continue this process.

Pete: You made some great observations about the coil voltage; I think I had seen the 3 x Pi voltage reference somewhere before. I would have been scared to connect to my scope, and I'm really glad you did it for the sake of science (as others have pointed out). One question I still have; is it safe to touch the coil when running? I avoid it of course; but I've always been curious. You don't have to try that yourself of course...

Yes; working with 220V circuits is something to be careful with; but it's commonplace outside the US. Here in the US one has to be even more careful because it's tapping two legs. There doesn't appear to be anything strange about this board though (nothing extra is going to go hot when connected). There is no physical power switch, and the input section is isolated from the rest of the circuit. I'm planning to put it on a two pole breaker for a power switch as a safety measure. After inspecting the schematic I posted previously, and comparing to this 3,500W unit's board; I noticed only a few minor differences. The capacitors on the output are larger, some components have a higher current rating, and a few other minor things to note. All the board component labels are different; but otherwise the values and connections are almost all identical. At some point I'll try to relabel the schematic based on this circuit for reference purposes.

I had to pick up some 100k thermistor fuses to connect to the boards temperature sensor ports. I'm planning to run them outside the furnace insulation so they don't see anything close to internal temperatures. I'm hoping thats enough to keep the controller from shutting down. Being power adjustable; I'm planning to start testing at the lowest setting once I'm ready to juice it up. Just waiting for a meter capable of measuring coil inductance. I also don't know what the output voltage for the fans is on the board; if it's 18v I'm going to have to order some new fans for long term use.

I went through a few theoretical coil designs. Two 20' spools of 1/4" copper tubing would create almost exactly a 120µH coil  if connected end to end without any shaping. A 50' spool makes some other designs possible as well. I got a 1kg and 3kg graphite crucible to work with. Also a larger 6kg A shape silicon carbide crucible which I haven't tested; but read should work with induction heating. 

The 48v 2,500W unit showed up here the other day, but I haven't had a chance to wire and mount it.

Has there been any testing of the limits in terms of coil frequency with 48V 1,000W or 2,500W units? Should I grab a 1,000W board to sacrifice and push those limits? Since I'm already working with larger coils, it's not much trouble (and smaller coils are easy). Looked at running solid copper for the 120µH coil; but even with furnace insulation I wasn't sure it would stay cool enough. An 1,800C infrared pyrometer showed up; but I haven't been successful measuring metal temperatures in the small crucibles yet. I get almost no heat detected through the ceramic insulation. Not sure about the emissivity setting though (and my graphite is almost powder so it barely glows at the moment).
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on August 23, 2019, 03:59:53 PM
Quote
One question I still have; is it safe to touch the coil when running? I avoid it of course; but I've always been curious. You don't have to try that yourself of course...

I have touched the coil many times, but I would not recommend it.  After all, were see a few hundred volts across it, right? But, I am not dead.
I do touch the coil with "one hand in my pocket" from time to time just to verify that the water is cooling the work coil properly.  Occasionally, I have touched both coil leads at the same time WITH THE SAME HAND and I do feel an uncomfortable "tingle".
  I still think it is a DUMB idea, particularly since I already do have a temperature meter measuring water temperature on my radiators.
  I would NEVER attempt that if I had a board that was powered by 220 volt mains, even though the output ckt is "isolated".

Thanks for the scope "heads up", Mads.
  I should have realized that. 

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on August 23, 2019, 04:07:38 PM
Quote
Has there been any testing of the limits in terms of coil frequency with 48V 1,000W or 2,500W units?

Yes.
I have not found a lower frequency  limit yet, but lower seems to be better for the Mosfets.  Things still work well down to about 26 kHz.
It's the upper limit that I have spent most of my tine evaluating.

See my data and my opinions here:
https://spaco.org/Blacksmithing/ZVSInductionHeater/WorkCoilsForZVSInductionHeater.htm
--And click on the "Work coil, Observed" spreadsheet.

Pete Stanaitis
---------------

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on August 23, 2019, 05:04:39 PM
The problem with the melting oscilloscope probe ground lead is that you are grounding your inverter through your oscilloscope! To measure on inverter outputs you always need a differential probe or an isolation transformer for the mains feeding the inverter and keep it floating off ground. I always recommend this great video:
/>

Exactly how is the inverter connected to mains PE/GND, which the scope is? If it is galvanic ally connected to mains it it could be lethal to touch the coil.
I have understood that the inverter I'd riven by a 48V supply, but that might be wrong.

The 48V power supply might have its 48V GND connected to the mains PE, but if it is an old server supply propably not, only the 12V are done so. The 48V are typically totally floating.

Apart from that, yes the video is extremely instructive. It is very easy to blow a scope when connecting uninsulated to the mains or without a doff probe. A normal transformer have quite a lot of capacitive coupling, something to be aware of when measuring fex high side switching nodes in half/full bridges even over isolation.

Safest is of course to put the GUT (Gadget Under Test) at isolation, if you isolate the scope averything, including scope GND is floating and can ruin your day...
Or use a diff probe. In the end they might cost about the same.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on August 23, 2019, 06:56:08 PM
Just to belabor the point about scoping the coil voltage:
I now remember that I was using my little DSO-112 battery powered scope early on when I had been taking readings.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on September 16, 2019, 01:50:19 AM
The making of induction heater work coils the way I do it and why-
  That's the focus of my newest ZVS 1000 to 2500 watt  12 to 48 volt induction heater video.
It is here:
/>
I'd be glad to entertain any comments, questions or criticisms.
  As they say somewhere:  "There's more than one way to skin a cat".
It's just the anyone who doesn't do it my way is wrong. <G>

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on September 16, 2019, 04:18:26 AM
The LM2596HV chip is almost certainly counterfeit from a Chinese source.  I'd purchased some similar little switching regulator boards, and the "LM2596HV" parts on them are definitely not legitimate.  Haven't tested voltage, but their regulation loop becomes unstable at higher current (by 2A at least) and the switching frequency is not within LM2596HV spec. range (lower).  Many FETs I've bought on EBay are also counterfeit, especially Alpha-Omega Semiconductor parts.  Sometimes they meet DC parameters, but are high on capacitance.  I'm now using normal electronics distributors for most parts.

If anyone is interested in building their own Royer oscillators, I've designed a couple improved (in my opinion) circuits.  In particular, these avoid the trade-off of power dissipation of the gate-drive resistors vs. gate drive strength that exists in the versions being discussed here.

My designs still share the issue of drawing a spike of current during power-up.  That shows up in Spice simulation and in real life.  A beefy slow supply isn't always a good solution either.  The power current peak builds up in the inductors feeding the resonant circuit.  When oscillation does start, that current momentarily makes the oscillation voltage much higher than during normal operation, which can fry the FETs.  I've added TVS diodes to most of my builds, although they aren't shown in schematics below.

My favorite circuit, shown here in a version intended for running off full-wave-rectified 120VAC (wall-plug power here in USA)"
 [ Invalid Attachment ]

And here is a very slight tweak to the above.  Power-coupling inductors L1 and L2 become a center-tapped transformer, and L4 is added as the power source inductor.  Power-line common-mode-chokes work well for the L1/L2 transformer.  The new L4 now has twice the frequency, so can be smaller.  This is my preferred implementation.  (Except for situations where the working coil can be center-tapped, in which case L1 and L2 can be eliminated.):
 [ Invalid Attachment ]

Finally, here is my previous Royer oscillator design - a smaller tweak to the circuits in the commercial units.  It just adds two PFETs to disable the gate-drive resistors once the gate is pulled high, allowing lower-ohm resistors (for faster gate rise-times) without excess power dissipation.  This version happens to be for 19V (laptop power supply).  It works at higher voltage too.  A friend in Germany built one for rectified 220V using 1200V SiC FETs.
 [ Invalid Attachment ]

I've played a little with induction heating, but used these more for driving home-wound inverter transformers to get a few kV (8kV peak for the one I'm playing with at the moment for DIY plasma balls - using the circuit on 19.5V from a Dell laptop computer supply.



Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on September 16, 2019, 06:59:12 AM
Good stuff, Pete.

Pro tip, literal: the way we did coils with interesting geometry, was to use square or rectangular tubing.  This can bend very sharply without kinking or tearing, and zero-radius corners can be mitered (cut out a wedge in three sides, leaving a thin web along the back; solder or preferably braze the seam closed).

The hard part is finding the stuff.  McMaster does sell it, in short lengths (and McMaster typical prices).  Never saw where we got the stuff from, but it was by 25'+ rolls, probably not cheap either way.

Swaged fittings, and using ~telescopic sizes for couplings and adapters, is an excellent method, of course.

I highly recommend brazed joints.  These are super easy in copper: you can use a self-fluxing (phosphorus-bearing) rod like Harris Dynaflow, no paste flux needed.  A plumbing torch will do for small tubing like this, or a somewhat bigger torch for bigger stuff of course.  The filler is considerably stronger than the base metal, no worries about strength; actually if anything, that's a liability as the joints are inflexible and break rather than bend -- plan your joints accordingly.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on September 29, 2019, 09:24:11 PM
Okay, I know this took forever from when I first mentioned it and till I put it all together last night :)

The 1800W IH is now boxed up with all power supplies, water cooling, foot switch, current meter and insulated work coil.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on October 01, 2019, 02:08:23 PM
Test with using regular mineral wool for house insulation, to insulated the work coil in order to achieve higher steel temperatures.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on October 01, 2019, 05:49:05 PM
Hello Mads.
  You did a very nice job of packaging your 1800 Watt ZVS induction heater.
I like your idea of the big DC relay.  That should make it easy to add PID temperature control some day.

Several other induction heater experimenters have complained about ZVS heaters, saying that they do NOT start up reliably.  I have never had this problem.  Why do you think they say this?

I, too have been trying to get steel hotter than the curie point.  The only time it has worked for me is when I insulate between the work and the inside of the work coil, and cover the work, too:

You can see the high temperature part of this video at about 20 minutes in.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on October 01, 2019, 11:10:44 PM
Here's my first induction heating video, in, just about forever! :D


Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on October 03, 2019, 09:51:48 AM
Hello Mads.
  You did a very nice job of packaging your 1800 Watt ZVS induction heater.
I like your idea of the big DC relay.  That should make it easy to add PID temperature control some day.

Several other induction heater experimenters have complained about ZVS heaters, saying that they do NOT start up reliably.  I have never had this problem.  Why do you think they say this?

I, too have been trying to get steel hotter than the curie point.  The only time it has worked for me is when I insulate between the work and the inside of the work coil, and cover the work, too:

You can see the high temperature part of this video at about 20 minutes in.

Pete Stanaitis
---------------

I got compact, but that also comes with some compromises, I have to run it for a long while to know if I can keep the temperature down or it will run away :)

I think the sole reason for unstable induction heaters is due to people using too small power supplies, most likely iron core transformers that are way too small, it gets overloaded, pulls the voltage down or saturates the core and boom goes the MOSFETs.

For a PI control, I would not use a mechanical relay, but a solid state, a mechanical DC relay at this current would get worn out too fast.

What kind of insulation material did you use in that video?

Here is another try, just with horizontal instead to contain some more heat, accidentally melted some plastic on the enclosure from that :)

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on October 03, 2019, 04:12:16 PM
Quote
For a PI control, I would not use a mechanical relay, but a solid state, a mechanical DC relay at this current would get worn out too fast.

What kind of insulation material did you use in that video?
------------------------------------------------------------------------------
I agree with using a solid state relay.  I have been looking at this one:
https://www.amazon.com/100A-Solid-State-Relay-SSR/dp/B00KO46YCU
but it is only rated for 32 volts.
I think they get pretty expensive when you get to 50 volts or so.

Insulation:
I use 8 pound density  one inch thick Kaowool that is rated for about 2300 degrees F service.
It is the material the we use in propane gas forges.  In that test I was simply tearing up small bits of it and stuffing them in beside the work.
If you search "kaowool" on ebay, you will find many sizes and thicknesses.
I see today that it is available in thicknesses down to about 1/4 of an inch.  I wish I had realized that when I bought another kind of "ceramic" insulation a few months ago. 
It was this stuff:
https://www.ebay.com/itm/INSWOOL-2300-PAPER-Flexible-Refractory-Paper-1-8-x-50-partial-rolls-also/131815404577?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649
  I bought it because it is thin.  It insulates quite well, but this material has some sort of chemical binder that burns away leaving the ceramic material behind.  With the binder gone, the insulation has very little strength left and it falls apart if mishandled.

Pete Stanaitis
---------------

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: RocketScienceSmurf on October 04, 2019, 10:23:15 AM
When it comes to switching on and off I plan to try IXYS IXFN520N075T2 since I have a few of them I got a year or so ago. Any thoughts on using this kind of power mosfet module?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on October 05, 2019, 05:19:30 AM
Wow, that's a beefy FET!  It should work well, as would the SSR mentioned.

With either one, the key requirement is a good clamp diode across the Royer oscillator (induction heater) power input, and bypass capacitor on the DC supply at the input to the FET/SSR.  The Royer oscillator has a pair of inductors at the power input.  (That's great for power switching.  Any circuit with a large bypass capacitor at its power input is difficult to switch quickly, as the capacitor needs charging.)  With the inductor input, at switch-off transition, the voltage will spike past the switch's rating unless a clamp diode is added.  The clamp diode and power input (before switch) bypass capacitor and switch should all be placed in close proximity to each other with low-inductance wiring.  Distance from that set of three parts to the power supply and to the Royer oscillator power input is not so important.  Just those three components need tight grouping.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on October 05, 2019, 05:33:55 AM
Quote
When it comes to switching on and off

What will you be switching on and off?
What supply voltage?
what frequency?
what current?

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on October 07, 2019, 08:24:21 AM


I think the sole reason for unstable induction heaters is due to people using too small power supplies, most likely iron core transformers that are way too small, it gets overloaded, pulls the voltage down or saturates the core and boom goes the MOSFETs.

 
I am not following, how do you saturate a transformer by overloading it , ie taking to much current?
I totaly agree that if to much power is drawn, the rectified voltage on the caps drops and this leads to a very unstable operation of a Royerlike circuit, not unlikely that the high/low Mosfet latchup and crossconduct.

What is a reasonable power supply then?

@kamelryttaren- for a induction heater I would definiately look into the large gate capacitance on that. Those are probably best for a 12/24V switching converter.
But for a power supply on/off modulation to the converter it should work even if 75Vds is quite close to 50ish volts.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on October 07, 2019, 08:46:10 AM


I think the sole reason for unstable induction heaters is due to people using too small power supplies, most likely iron core transformers that are way too small, it gets overloaded, pulls the voltage down or saturates the core and boom goes the MOSFETs.

 
I am not following, how do you saturate a transformer by overloading it , ie taking to much current?
I totaly agree that if to much power is drawn, the rectified voltage on the caps drops and this leads to a very unstable operation of a Royerlike circuit, not unlikely that the high/low Mosfet latchup and crossconduct.

What is a reasonable power supply then?

You are right, overloading the transformer secondary side has nothing to do with core saturation, I wrote some gibberish there.

Since the Royer is super power hungry and f.ex. a specified 2kW unit can also draw exactly that, with changing size work coil and work piece, the current can suddenly be much higher than you expected. I would say that just to be sure, 25-50% current head room is preferable.

However, you need more current head room the lower voltage you are operating at. A 12VDC supplied homemade Royer with IRFP250N would get unstable below 10VDC, so it is far more prone to failure than a 36VDC suppplied Royer circuit from China.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on October 07, 2019, 04:38:53 PM
Quote
However, you need more current head room the lower voltage you are operating at. A 12VDC supplied homemade Royer with IRFP250N would get unstable below 10VDC, so it is far more prone to failure than a 36VDC suppplied Royer circuit from China.

Yes, I'd say that is exactly correct.

About 2/3 of the way down this page you can see 4 scope traces of gates turning on with differing input voltages from 12 to 48.
https://spaco.org/Blacksmithing/ZVSInductionHeater/1000WattZVSInductionHeaterNotes.htm

As far as the "core saturation" thing goes:
  I am no expert on transformers, but I have rewound a couple of microwave oven transformers and I have taken some data on their "regulation" or whatever you want to call it, under varying current  draws.
/>  In this video, I show two charts that demonstrate  transformer output voltage reduction as current increases.

Pete Stanaitis
--------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on October 07, 2019, 05:09:36 PM
Thanks, I got it. Sounds very plausible. And always very informative, thanks for the link.

So around 3kW rock stable server supply @51VDC should suffice for a 2kW Royer, I presume? And if not, I double that, haha :)
Assuming I keep it sufficiently away from the coil? No problems with induced voltages? AFIK they are quite shielded given you earth the chassis.

MOTs are due to cheapness almost always close to saturation (small core cross area and few primary turns) running with high magnetising currents.
I would only use them longer periods with a variac or similar.
They run hot in idle and hotter during load. I doubt many of them can take a full 230VAC max voltage according to spec without going into total saturation, therefore fuses...

Our kitchenette at the office smells of hot xformer insulation at around 1 pm every lunch...

They also have quite a high leakage, providing some current limit capability and therefore not so good regulation, ie they are not "stiff"
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on October 07, 2019, 08:46:35 PM
Yeah, you solve the regulation by knocking out the shunts.  (Which also can saturate, which happens under a heavy load.  Apparently they only need a soft (current-limiting) output over a certain range of load condition -- which is fine for a magnetron that drops some known voltage when forward biased, and if anything, the fault current under a short circuit would help by blowing the fuse faster.)

You can also solve the saturation by adding extra turns in the space where the shunts were; 10-20% more turns should do.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: RocketScienceSmurf on October 08, 2019, 08:37:53 AM
@kamelryttaren- for a induction heater I would definiately look into the large gate capacitance on that. Those are probably best for a 12/24V switching converter.
But for a power supply on/off modulation to the converter it should work even if 75Vds is quite close to 50ish volts.

I don't fully understand what you mean. I haven't paid too much attention to the gate capacitance yet but my power supply is close to 50V so I believe I am safe in regards to the Drain to Source maximum voltage. The gate signal will most likely be handled by a proper mosfet gate driver like MCP1407. The IXFN520N075T2 would be used to turn the induction heater on and off. I have other mosfets in the induction heater circuit. From my previous measurements I feel like time it takes for the trench fet IXYS mosfet module to fully conduct is probably shorter than the time it takes for the induction heater to start oscillating so I think that losses in the mosfet module will be reasonable.

When it comes to the power supply I am confident mine can supply 150A without sagging.

[sidenote] What is the "Kelvin Source ( Gate Return ) Terminal" mentioned in the picture?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on October 08, 2019, 09:06:54 PM
I thought that the large FET would be not so good to use in the Royer oscillator as a switching element, but as a on/off switch instead. Just as you intend.
I totally agree that a mosfet with a rise time in 10s of nanos will fully conduct long before an oscillator at a couple of tens kHz ore slightly more is upp and running at full amplitude. Just as long as you smack the gate with a good drive signal.

I am interested in your induction heater, tell us more.
Låter intressant!.

Beefy supplies are always useful.
Ä
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on October 08, 2019, 09:15:28 PM
It's probably better to hit it slowly (gate rising over 100s us?) in case there's a lot of inrush current.

Better still to put in a current measurement circuit and control the gate accordingly, but that's maybe not so big a priority with well behaved power supplies.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on October 08, 2019, 10:48:46 PM
Nah, fast turnon. Let the power supply bleed :)

I think you want to raise voltage so fast before oscillator starts operating and it might show any kinds of unstable behaviour.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: RocketScienceSmurf on October 09, 2019, 09:56:00 AM
Nah, fast turnon. Let the power supply bleed :)

I think you want to raise voltage so fast before oscillator starts operating and it might show any kinds of unstable behaviour.

My power supply almost scares me a little. The three phase 6kVA transformer is massive and the five parallel Kemet C4DEFPQ6380A8TK film-capacitors can supply an insane amount of peak current and I suspect that an unintentional short circuit would result in all sorts of smoke, flash and fire. My power supply is definitely NOT the weakst point of my setup.

I haven't done any work on my induction heater for about a year :( I bought some water cooling blocks for the mosfets but I haven't got around to putting it all together yet.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on October 10, 2019, 08:58:36 PM
Just a Mosfet thought, relating to the recent posts about IGBT/Mosfet gate drive:
  I can't see ANY reason not to turn on these devices ASAP.  By that I mean that these devices, as far as I know, are SWITCHES and are NOT meant to be run in linear mode.  A main reason that they can handle such high currents is their low RDS(on).  So, if they are NOT fully on, very high amounts of power can easily be produced.
  Don't "baby" the gate.  Get it up to "full on' ASAP.  As previously mentioned, if the power supply can't hack it, get a better one. 

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on October 11, 2019, 02:07:09 AM
I mean you can go ahead and ignore the hesitance of a professional engineer but sure, damn the torpedoes right?

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on October 11, 2019, 04:45:55 AM
Unless you are concerned about radiated EMI, I don't see any down-side to fast turn-on.  I'm still not clear if this FET is being used to switch the power input to a Royer oscillator, or is being used (a pair) to implement the Royer oscillator itself.  Either way, fast turn-on should be fine.  As a power switch, it's feeding the power-input inductors, so current rise will be ramped by the inductance and supply voltage, so not an instantaneous current load to the supply.  If within the oscillator, it's ZVS, so fast should be good there too.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: RocketScienceSmurf on October 11, 2019, 08:51:11 AM
It's probably better to hit it slowly (gate rising over 100s us?) in case there's a lot of inrush current.

Better still to put in a current measurement circuit and control the gate accordingly, but that's maybe not so big a priority with well behaved power supplies.

Tim

You are probably right but I don't have the "know how" to adjust switching time and to measure current at the speed probably necessary.

To better explain the intended use for the IXYS module I will use it between my power supply and the induction circuit. Mosfets inside the IH are NOT IXYS modules!

My power supply (as showed in another forum thread) is a 6kVA three phase transformer connected to 3+3 45L10/45LR10 rectifier diodes rated at 150A. After that I have a large choke from the ESAB welder the transformer came from and after that 5 pcs C4DEFPQ6380A8TK capacitors. The plan is to connect the IXYS module as close as possible to the ground rail of the capacitor bank of the power supply.

The water blocks are for the mosfets in the induction heater since these get quite hot and I will most likely water cool the gate resistors and the tank capacitors.

The main reason I need a robust switch to turn it on and a power supply that can supply a stupid amount of current instantly is that I must start it with the object intended to be heated already inside the work coil.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on October 11, 2019, 11:33:19 PM
Unless you are concerned about radiated EMI, I don't see any down-side to fast turn-on.  I'm still not clear if this FET is being used to switch the power input to a Royer oscillator, or is being used (a pair) to implement the Royer oscillator itself.  Either way, fast turn-on should be fine.  As a power switch, it's feeding the power-input inductors, so current rise will be ramped by the inductance and supply voltage, so not an instantaneous current load to the supply.  If within the oscillator, it's ZVS, so fast should be good there too.

If there are no capacitors then turn-on doesn't matter much, and a turn-on time of microseconds is fine -- this will help kick the oscillator into motion without generating much EMI.

Although it will take some time for the oscillator to start up, and I wonder if it should be ramped after all.  Cooking the oscillator at low amplitude and high voltage is going to put a ton of heat into the transistors...

Switching speed is adjusted by series gate resistance.  Rising and falling edges can be controlled by putting a diode in series with the one, and a resistor in parallel with the diode.  One direction sees the lone resistor, the other sees both in series.  Or use a diode each for independent control.

If the load is inductive (if there's no capacitors on there, it just goes right into the bias resistors and inductors), then rapid turn-off is likely to damage the switch itself.  At minimum a catch diode is suggested.

The biggest problem is fast turn-off.  Removing gate drive power shuts off the oscillator, and the inductors discharge into the transistor drains, which subsequently explode.  A TVS can be connected from GND to each drain, say a 1.5KE120 or SMCJ120A, which should absorb most of that energy before the transistor does.

Better is to have gate supply from a separate circuit so it stays always-on while the drain power is switched.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on October 12, 2019, 06:16:12 AM
Thank you for the clarification about FET use.  I do see that clearly in your earlier posts now, but was confused at some point.

BTW, don't see anyone answering your "Kelvin connection" question, so:  One source lead is for the high-current connection (source-to-drain current).  The other is for gate-drive return.  That way the inductive and/or resistive voltage drop of the high-current connection isn't added to the gate drive signal.  The two source connections are identical in this part (not typical for other packages).  However, you should use only one for the high-current path, and the other only for gate-drive return.  (Connect gate drive to gate and that second source terminal.)

As long as there's a catch diode from the drain to the incoming supply V+, and a good capacitor from V+ to V-, I'd go ahead with fast switching.  (Inductance of the diode/cap/FET loop needs to be LOW.)  If you need to slow down the switching, the simple way is to add a resistor in series with the gate.  Another small diode and resistor are often added in parallel with the main gate resistor to make turn-off faster than turn on.  If you need a more uniform voltage ramp on the output, adding additional miller capacitance (gate-drain capacitance) can help.  I've personally had trouble with RF oscillation when adding external miller capacitance, I think due to gate and source lead inductances, but it can work.  Any additional miller capacitance is a second step after inserting gate resistance.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: DICKEYBIRD on October 20, 2019, 10:34:22 PM
Okay, I know this took forever from when I first mentioned it and till I put it all together last night :) The 1800W IH is now boxed up with all power supplies, water cooling, foot switch, current meter and insulated work coil.
Hi Mads, I enjoyed your video with the details about your putting the IH & support bits into an enclosure.  I noticed that you put a separate switch for each power supply & was curious why you did that.  I have the same modded PSU's (3) as you and have the mains power for all 3 wired through one switch.  I haven't finished the enclosure yet or actually powered up the the IH but I have powered-on the PSU's in series & then applied a test load of approx 200W for 5 minutes or so.  They seemed to be working OK.  Should I worry about not switching them on separately?

Thanks very much for all the good info you guys take the time to post here. :D
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on October 21, 2019, 01:24:00 AM
A positive thought for you guys who  are using multiple 12 volt power supplies:
 People are often asking me how to control the power that the  work coil transfers to the work of the ZVS heaters that I use....
If you are using a 48 volt power supply, as I do, then the two main ways to accomplish that is to either:
1. move the work closer to or farther away from the work coil until the current is at the level needed.
or-
2. Change work coil design to match the needs.
But, in many cases, the operator needs to adjust the power one direction or another as the work heats up.
An example of this would be when heating a piece of iron or steel-   Initially, if the work piece must be inserted into the work coil before the power is turned on, the current sucked up by the workpiece may exceed safe limits for the circuit, but once the part reaches the curie point, you need MORE POWER.
In this case, if you were to arrange your power supply switching so that you could start out with, let's say 24 volts and then switch to 36 volts or to 48 volts as needed, you'd have at least 3 power levels available.

Just a thought.   I am not saying that the switches would be simple, but it certainly isn't impossible.
And------- note that I did NOT suggest starting out with 12 volts.
I hope you all know why by now.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on November 24, 2019, 04:46:34 PM
Effect of frequency on melting copper???

Recently, a guy asked me how to change his ZVS induction heater's frequency so he could keep the current up as the copper melted in a graphite crucible.  He has observed that the current drops at the moment the copper melts.  I think many of us have seen that effect.  My guess is that the molten copper shorts out the  carbide crucible or somehow reduces eddy currents in the graphite.  He must think that changing the frequency will get the current back up to where it was before the copper melted.  I don't think that's a viable option with these ZVS heaters.

Anyway:
A confession:
   I lack a practical understanding of the exact degree to which frequency affects the heating of copper and brass.
  After reading everything I can find on the internet, it seems generally accepted that higher frequencies are better for non-ferrous metals.
  That's nice, but it also appears to me that the people who are heating copper parts commercially almost always use induction heaters that have a lot more power than do the 1000 watt to about 2500 watt heaters that are the subject of this thread.
  Then there's the (apparently) very popular "15KW" induction heater of which this is one example:
(search this on Ebay)-
"220V 15KW 30-100 KHz High Frequency Induction Heater Furnace"
(Be warned, however, that this heater really only puts out about 7.5KW AND, you need a pretty good water cooler to use it!).

There are plenty of videos where the operator melts copper or at least heats some parts up for soldering or brazing.
  But we never, as far as I know, get to see what frequency the thing is running at when this takes place.


All that said, if appears to me that you need more than about 5KW to melt a kilogram of copper or more in a graphite crucible anywhere in the range of about 25 kiloHertz to about 100 kiloHertz.  Once you have a lot of power available, then you can start messing around with frequency, if your circuitry allows it.

Conclusion:
I don't have a good answer for that guy.  We can change the frequency of these ZVS heaters by changing the L (size and shape of the work coil) and C (adding or subtracting capacitors) in the circuit, but that's not what he thinks he  wants.  I told him that insulation and MORE insulation is his friend.

Your comments, please.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on November 24, 2019, 11:55:18 PM
Guessing the frequency rises at the same time.  In other words, more magnetic field is being reflected than absorbed, when it melts.

The higher frequency means the work coil is a higher impedance, so draws less power, roughly speaking.  This could be compensated by switching in more capacitors.  Or running with more capacitors, but beware of controlling supply voltage to avoid burning up the transistors from too much current.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on November 25, 2019, 02:37:09 AM
I could picture frequency rising if the copper shape changes significantly during melting, spreading out to a shape that blocks more of the magnetic field within the work coil.  If the frequency goes up, then the impedance goes down, not up.  The capacitors stay the same, so higher frequency comes from lower inductance, which is lower impedance.

Copper resistance roughly doubles as it melts.  That's the most likely explanation for power dropping.
https://www.sciencedirect.com/science/article/pii/S0022369717304341

At the start, before the copper heats significantly, the electrical conductivity of the copper being heated and the work coil are roughly the same.  That makes direct induction heating quite inefficient.  I don't know much about the graphite crucibles, what their electrical conductivity may be.  Conductivity may be too low for much heating from the induction field.  If the graphite is conductive enough to generate much heat, that will reduce as the copper melts and its shape fits the crucible inner wall.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on November 25, 2019, 05:28:44 AM
Did I say up, I meant down... ::)

In any case, it's a matching thing, always is.  If the load resistance is rising, you simply need fewer turns or more volts to match it, that's all.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on December 02, 2019, 04:10:14 AM
Some interesting recent comments, but still don't have a good handle on best frequency for melting copper in a graphite crucible.
   It certainly isn't all that simple, is it?

I have found a few "Induction Heating Calculators" around on the internet.  I think they probably do a fairly good job of estimating a few variables when it comes to iron and lower alloys of steel, but they seem to miss the target when getting into non-ferrous metals.
  For instance, here's one of the "calculators":
https://www.plustherm.com/power-calculation.html

It took me some time to figure out how it works.  (Note that you put in the whole number for the "Inductor Efficiency", not the percent as a decimal.  That is- if your material is  Aluminum, enter "40" not "0.40").

In the "Results" area, I am very dissapointed that it usually outputs an extremely low frequency in the "Minimum required frequncy" box.  Why doesn't it know what frequency it actually used for that particular calculation?
  And--- I am pretty certain that the results from their "minimum" frequecny and the ideal frequency would produce differing readings.

At first, I left the "Working Frequency" blank, so the calculator could choose, but even when I did enter a number, it didn't seem to help much.

After fooling around with the thing for a while, I realized that we aren't actually melting copper anyway with these 1800 watt to 2500 watt ZVS induction heaters, we are heating a graphite crucible!  Well, there's no graphite in the Material choices box, so now what?  I made a few guesses in the Inductor Efficiency box, estimating the graphite to be somewhere between "60" and "85", but no conclusive results.

It seems that some of the parameters in the  "Details" area not closely related to the input parameters at all.

What do you guys think?

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on December 02, 2019, 10:51:33 AM
Minimum frequency is just informational; note it's driven by the minimum size of material being heated, which is quite a bit smaller than the crucible if the crucible itself isn't doing the heating.

There isn't really an ideal frequency because any frequency works.  Nonmagnetic metals aren't more or less lossy with respect to frequency, they just have a resistance.  (Steel is notably more lossy below a few MHz, because that's around the magnetic relaxation rate for mild steel.)

This calculator does not produce any coil data, so it doesn't do anything with resistivity.  It's assuming you know how to design a coil and match it to a power supply.  Not very helpful I'm afraid.

It also doesn't ask for insulation thickness and quality, it's only using material for heat capacity; and in the solid phase at that (i.e., not including latent heat of fusion!).

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on December 24, 2019, 11:01:26 PM
Cobalt in your future?
Many readers of this thread have asked about methods to melt copper faster.
  We typically use ceramic insulated graphite crucibles, which work pretty well, but
1. don't push the 2500 ZVS induction heaters to their power limit, even when on 48 volts
2. their graphite is comsumed during the process, which limits one crucible to about 10 "heats" before it gets so thin that it is dangerous to use.

So I was thinking that a steel crucible might melt copper a lot faster than do the graphite crucibles due to their magnetic/hysterisis characteristics.  Problem is that they lose their rapid heating abilities well below the melting point of copper.

I just realized that Cobalt has a curie point of 2067 degrees F, well above the curie point of iron  (1418 F, or thereabouts).  I wonder if there's a cobalt crucible in my future, but I won't lose any sleep thinking about it, since that metal costs about USD$75.00 per pound.   And, there probably aren't a lot of Cobalt crucibles on sale at Ebay.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: klugesmith on December 25, 2019, 01:39:02 AM
I had a summer job in 1973, at a place which made lithium niobate for optoelectronic applications, & related single-crystal materials.
They grew boules from scratch, using graphite resistance-heated elements.  Imagine 0 to 10 volts at up to 1000 amps, 60 Hz three-phase, from conventional power transformers very close to the station.

Anyway, some of the  crystal materials were initially prepared from high-purity powders,
melted by induction heating in crucibles of pure platinum.
Crucibles got dented and battered from removing residual solid between batches.
Eventually they were weighed very carefully & traded in for new ones from the precious metal supplier.

Looks like Pt crucibles are still a thing, 46 years later.
http://www.platinumcrucible.com/
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on December 25, 2019, 06:41:29 PM
PT crucibles is da thing when used for dry ashing, ie heating to white hot to render all C N O H into gas, leaving only non volatiles left. These can be dissolved in fex nitric acid and analysed fex with ICP /AAS. Safer than using perchloric acid and nitric acid etc directly on the cellulose filters used for collecting water samples...
Fex heavy metal content in the blow down water on a nuclear reactor. Especially content of Fe-56,Co-60,Ni-58 and other activation products as a measurement of corrosion on the actual reactor itself.

Amazing that they let a summer worker do that kind of work, but they wear us down milli Sivert for milli Sivert :)

Sorry for Off Topic...
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: John123 on December 25, 2019, 08:08:52 PM
Anyone tried running a flyback LOPT on one of these? Bet it runs in the audible range with all those tank capacitors.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: klugesmith on December 25, 2019, 08:39:46 PM
Pete's thread makes me question whether to buy an induction heater driver, or finish making one from scratch.  The latter project, stagnant for a few years, is just a collection of capacitors, documented ferrite toroid cores, and some copper for tank electricity and coolant conduction. Oh, and some Cerrobend (low melting-point metal) in case it helps when making tight coils of copper tubing.

Back to the tangent about Pt crucibles. RTT's account led to failed search for "fex" in that context.
Found FEX as a line of turbo expanders used in chemical plants.  Their blades might not run as hot as those in jet engines, but have to contend with process fluids that would be very corrosive even in an open crucible at room temperature.

Ref. this industrial service newsletter, with a drawing in Christmas colors. :-)
Axial flow with scroll cases on both ends! Maybe to share parts with other machines.
https://rotatingmachinery.com/pdf_files/newsletter-april-june-2014.pdf
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 02, 2020, 01:38:14 AM
A little more  information for people who want to make specialized work coils:
I just updated my work coil spreadsheet.
It is here:
http://www.spaco.org/Blacksmithing/ZVSInductionHeater/Work%20Coil%20Data.xlsx

I added an area down toward the bottom that I call "guessitmates.  This is where I enter the data that I do have, usually  physical coil dimensions and measured (not necessarily actual) inductance.    As I have said many times before, if the coil's inductance is higher than about 1.0 microhenry, then the ZVS driver will run at a low enough frequency to avoid overheating of a properly cooled system.


Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on January 11, 2020, 12:28:35 PM
Thank you Pete, for continuesly sharing your calculations and findings to optimize and get the most out of these units :)

I bought some other versions and sizes of the "Chinese induction heaters" and just finished looking at the 150W unit.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 12, 2020, 03:18:17 AM
Thank you for your kind words, Mads.
   Nice demo of the little 150 watt unit.
Several members of our blacksmith club enjoy making things from horseshoe nails. This unit is just right for that.
If you power it from a 12 volt car battery (or appropriate rechargeable battery pack), you have a very portable heat source.

Salt spoons and finger rings are but two examples.
Works well for making tiny swords from double headed (duplex) nails, too.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on February 17, 2020, 06:00:59 AM
Water Cooling Notes.  Here is the reply I gave to a guy recently relating to water cooling the work coil:
----------------------
I have a couple of questions if you don't mind.
where did you get your radiator from? I am using a 12x 5 cpu cooling radiator and it can't keep up with the required cooling.
----------------------------------------------------------------------------------------------------------------------------
I bought several radiators like the ones you see in my videos at a local online  auction house several years ago. They were all used ones. That style of radiator has one continuous tube that runs from input to output.  There’s no “tank” at the top, so it doesn’t really matter which end is the input. Although I have not tried this myself, I think the heater core from an automobile passenger heater would work well, using the fan/blower that comes with it.  The radiator on my 2500 watt unit holds a bit over 2 quarts of water, by the way.
---------------------------------------------------------------------------------------------------------------------------------------------------------------------

There are several air pockets in the lines that may be the issue. Any suggestions on how to bleed the air out of the lines?
-----------------------------------------------------------------------------------------------------------------------
   Re: air pockets-  It may not be apparent on my setups but:
  Note that I have an expansion  chamber (later on, I used a plastic pop bottle, upside down with the bottom cut out) on both heaters.  I either take water from the outlet just below the expansion port or from the other end of the radiator.  Makes no difference, as long as the water level is higher than my piping to the workcoil , and as long as any bubbles that form are allowed out the expansion vent so they are  not trapped within the system. This expansion opening is also very important since the water expands significantly  when heated, and its also a sort of safety feature in case the water were to boil.
  One other REALLY important point: The pump MUST feed water into the BOTTOM of the coil.  This is the only way to be certain that the work coil is FILLED with water.  I once made the mistake of doing that backwards and produced many bubbles of steam before I realized what I had done.
------------------------------------------------------------------------------------------------------------------------------------------------------------

I have the same pump as you, but I am using 3/8 "clear tubing. I also can't get the pump to really circulate the water like it should be in my opinion.
---------------------------------------------------------------------------------------------------------------------------------------------------------------------
  Try to keep the pump and the radiator at about the same level.   Some guys put their water supply (usually a pail of some sort) on the floor, while the induction heater and pump are up on a bench.  The pressure output drops rapidly as the height increases, leading, of course, to lower flow.  Them if your radiator is very restrictive at all, flow gets even worse.

Link that will take you to most of my videos and webpages on the subject:
https://spaco.org/Blacksmithing/ZVSInductionHeater/1000WattZVSInductionHeaterNotes.htm

Pete Stanaitis
---------------

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hammertone on February 29, 2020, 11:09:36 AM
I stumbled upon a couple of these 2100W, 48V server supplies, and thought they would be well suited for these powerfull induction heaters, but I have failed miserably in locating the pinout, so also failed to get them to turn on.
I hope someone on this list can help me out.
The power supplies are from the SUN M4000 Sparc Enterprise servers, manufactured back in 2007.
The model number is AWF-2DC-2100W
Part number is ECD15020005/02
There are similar PSU's which put out  12V, they are well documented for the use as battery chargers, and do not share the same pinout as this supply.





Thanks in advance for any insight offered.

Cheers, Finn Hammer
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on February 29, 2020, 12:26:19 PM
I stumbled upon a couple of these 2100W, 48V server supplies, and thought they would be well suited for these powerfull induction heaters, but I have failed miserably in locating the pinout, so also failed to get them to turn on.
I hope someone on this list can help me out.
The power supplies are from the SUN M4000 Sparc Enterprise servers, manufactured back in 2007.
The model number is AWF-2DC-2100W
Part number is ECD15020005/02
There are similar PSU's which put out  12V, they are well documented for the use as battery chargers, and do not share the same pinout as this supply.





Thanks in advance for any insight offered.

Cheers, Finn Hammer

Maybe some google translate can help you through the russian: https://forum.bits.media/index.php?/topic/10447-%D1%80%D0%B0%D1%81%D0%BF%D0%B8%D0%BD%D0%BE%D0%B2%D0%BA%D0%B0-%D0%B8-%D0%BF%D1%80%D0%BE%D1%87%D0%B8%D0%B5-%D1%82%D0%BE%D0%BD%D0%BA%D0%BE%D1%81%D1%82%D0%B8-%D0%BD%D0%B0%D1%81%D1%82%D1%80%D0%BE%D0%B9%D0%BA%D0%B8-%D1%81%D0%B5%D1%80%D0%B2%D0%B5%D1%80%D0%BD%D1%8B%D1%85-%D0%B1%D0%BF/page/111/&tab=comments#comment-1734682

Found it on google image search for "delta awf-2dc-2100w psu pinout"
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hammertone on February 29, 2020, 02:34:57 PM
Thank you very much, Mads.

Just what was needed, the psu immediately woke up with a scream.
There has been some complaints about the noise level of the fans, now I agree...


Cheers, Finn hammer
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 26, 2020, 11:50:01 PM
Melting Aluminum with the ZVS induction heaters:
   I guy recently asked if he could melt a half  Kg. of aluminum.   I haven't tried melting aluminum at all. 
I had assumed the aluminum would soak up energy about as well as does copper; that  is almost none.
There are a few videos out there but I saw only one where the guy was measuring current.  And, indeed, I saw no appreciable change in current when he added strips of aluminum to a graphite crucible.    The only caveat is that he was using a Litz wire coil on a Chinese 1000 watt unit and was only getting about 8 amps at about 40 volts .  I'd assume that Litz wire probably gave him a much higher frequency.  And, his crucible was quite a way from the inside of the coil.
  Maybe I need to try it out just to verify/determine the relationship between the two metals. 
Any interest in that?

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 27, 2020, 12:21:46 AM
Depending on the frequency and thickness of the graphite crucible, the magnetic field may be mostly blocked from the aluminum, all dissipated within the graphite.  In that case, properties of the inside material don't matter.

Even if the crucible is providing all the heating, I believe it will take much more than 1kW to melt 500g of aluminum.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on March 28, 2020, 10:08:16 PM
Also as the metal heats up, its resistance rises significantly and it can absorb more heat directly.  Getting started is slow but maintaining a heel is easier.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 22, 2020, 12:39:36 AM
I am using some of my  at-home time to clean up my two ZVS induction heater breadboard setups.

For the 1000 watt unit, I still had been getting 12 VDC for the water pump and the pcboard cooling fan from an automobile battery. Other than that, the whole system runs from the (USA) 120 volt mains.  And I had to plug in 3 other things to get it going:
-Radiator cooling fan
-48 volt power supply
-5 volt wallwart for the frequency meter

  So I am in the process of adding a walwart to replace  that 12 volts DC and a dedicated power strip for the rest.  This way, all I have to do is to plug in that one power strip to bring the whole thing to life.
   This whole system is mounted on 1 inch thick board that is about a foot wide.  That board is sitting upon a cart that is about 2 feet square.  It takes up a lot of valuable shop space. So, once I am done, I can put the whole system up on a shelf until it is next needed.   The cart can be stored someplace else until I get another wild idea.

2500 Watt ZVS induction heater:
  The big issue here is that, since the power supply requires 240 VAC, I have to locate the machine close to an outlet that puts it IN THE WAY between my engine lathe and my power hammer.
So, I am adding a new 240 volt outlet on the other side of that shop wall, so I can move the heater into another, slightly less cluttered room.  This has turned into a big deal since I have to work inside the wall, struggling with 10 gauge wire to add the new outlet.  Also, since it's not a good idea to go hardware store shopping these days, I am getting parts online.
This system is on its own wheels, so it will continue to take up floor space, but at least it won't be in the way of the lathe, powerhammer and grinder. But I will also add a wallwart to this one to remove the need to take water pump power from the pcboard, giving me a couple of extra amps for output power.

I think I may do a couple of aluminum melting experiments, just to see if my hypothesis that aluminum acts a lot like copper is true or not.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 05, 2020, 03:07:28 AM
Notes on the negative effects of increasing operating frequency of the ZVS induction heaters:
---------------------------------------------------------------------------------------------
Many of you are probably tired of hearing me talk about being careful to keep the Mosfets from spending much time in the linear mode.
But here I go again:
  We can easily see that the idle current varies when we change work coil designs.

From my own work coil data spreadsheet, I recently noticed that idle current and frequency seem to vary together.  When I plotted 9 pairs of idle current versus frequency observations,  I could see the relationship easily. I got a correlation of about 0.8.  Of course "correlation" does not necessarily equal "causation", but it seems reasonable to me.  It see this relationship as another indicator that higher frequencies create more lost power within the ZVS circuit components themselves.  This, in turn, leads to more stress on, particularly, the Mosfets and the tank capacitors.

But why do we care, as long as things seem to work okay?
 Well, here are a couple of reasons:

1. Assuming that you or your system has a maximum safe current limit, the higher the idle current, the less current there is available for heating the work.
In my case, the maximum sustained current that I will allow on my 1000 watt unit is about 22 amps TOTAL.  And, for my 2500 watt unit, it's 50 amps, TOTAL.
I hope that is clear to everyone.

But--- I think this item is even more important:
2.  The Basic ZVS power circuit that most of these Chinese ZVS induction heaters use works most reliably at a frequency below about 100 kHz.   And, the lower, the better, in my view!!!

So, when you design a new work coil, you need to know whether it will overstress the system or not.
You can quickly check it out by  powering the system up and noting the idle current. That current is primarily heating up the Mosfets and the Capacitors.  Of course it also heats the work coil, but as long as we are water cooling it, we won't worry about that.
   If the idle current is higher than about 8 amps, I think you will be stressing the system more than it was designed to be stressed.
See Chart:
 [ Invalid Attachment ]

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: myearwood on May 07, 2020, 08:02:24 PM
Good day learned ones.
I have been watching videos and reading.
I recently purchased a 2100W zvs induction unit from China. It arrived wired for China. I asked for 120vac, but they ignored that. Is it possible to get an extension cord for my oven socket and use 240v 50amps. I would need clear wiring for a stove socket to this Eaton MPR-48.

I am new here, but not new to technical forums.

Thanks in advance

Mike
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 07, 2020, 11:53:26 PM
Post a link to the product that you bought.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: myearwood on May 08, 2020, 12:10:53 AM
Hello Pete. You have very informative videos. I'm glad of your time.

www.wish.com/share/8263ljngsh
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 08, 2020, 03:27:50 AM
Sorry, but I don't know what wish.com is and I do not wish to register there.

I need a link to the advertisement for the place you bought it.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: myearwood on May 08, 2020, 05:48:00 AM
Not a problem. I can post screenshots tomorrow. Wish is an online store. Much is sourced from China.
Mike
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: ritaismyconscience on May 08, 2020, 07:12:08 PM
It might work even if you plug it into 110V. There seems to be a switch mode power supply underneath and a lot of them work with voltages from 100-240V.

There should be a white label on it, you can see what voltage the power supply is rated for
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: myearwood on May 09, 2020, 02:55:38 AM
Hi all. I know it is an Eaton EPR48-3G. Eaton says it "derates" at 110. The attached data sheet does not say that.

I really appreciate your help.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 09, 2020, 03:49:25 AM
This datasheet does show input voltage derating:
http://www.amppower.de/datas/APR48-3G.pdf
See right hand graph on page 2.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: myearwood on May 09, 2020, 06:16:40 AM
Hi Pete. Thanks. The question at hand then is will 1100watts be enough to melt metal? Just plug this in to 120vac with 15amp fuse? It does turn on. I heard you talking about needing to wait for power buildup. Also needing a 12vac for the fans and pump. Can I plug it into the stove outlet with appropriate plug?

Thanks again.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 09, 2020, 04:35:04 PM
Answers, one at a time:
-Will 1100 watts be enough to melt metal?
    "Yes" to small amounts of copper, aluminum, brass,  silver and gold, etc. using an insulated graphite crucible in the work coil.
     Maybe up to about 200 grams or so.  You can see from my copper melting videos that I have melted as much as about 500 grams with the 2500 watt unit.
    "No" to steel or other ferrous metals.  Once they pass the "curie point" in temperature, (about 1500°F or 815°C) the heating rate drops substantially.

-Also needing a 12vac---
  Aren't the fans and pump built in to and powered by, the device?    For my own units, the fans and pumps are 12 volts DC not AC
Here is a unit that looks like the one you bought:
https://pt.banggood.com/220V-2100W-Mini-Induction-Heating-Machine-Heater-Air-Water-Double-Cooling-DIY-Device-Science-Model-Kit-p-1410359.html?rmmds=buy&cur_warehouse=CN
If it is like yours, I think I see a water pump and I see wires running from the fans to a circuit board below them.

-Can I plug it into the stove outlet with the appropriate plug?
   I assume that your (kitchen?) "stove outlet" is a 220 single phase outlet to answer this question.  If this is true, you could melt about twice as much copper.
   Depending on the stove and the age of the installation, it may have either a 3 wire or 4 wire plug/receptacle.  You will be handling a lot of power and ,
   of course, lethal voltages, so you need to know what you are doing from a safety and fire prevention standpoint. 
   So my answer is "Yes", but I suggest that you  get some help to get the wiring correct.  You don't get a second chance if you get it wrong.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: myearwood on May 09, 2020, 08:54:46 PM
Advantages of asking a master. You are correct on your assumptions good sir. I saw discussion of this daughter board being unwise to feed the fans and pumps off the main supply, which may not be needed.
I certainly followed your concerns about water. What of a breaker to power the coil. Is that recommended?

I'd have to replace the china supplied plug.

I will get pro assistance for oven.

Thanks very much again.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 09, 2020, 09:32:30 PM
Re:
Quote
What of a breaker to power the coil. Is that recommended?
I am pretty certain that most modern switched mode power supplies, such as the one there, are smart enough to only connect their output to the load after they are ready to deliver full power.
  So, a "breaker" or switch between the power supply outlet is not crucial for normal operation.
But I do like to have a way to shut off power to the while still allowing the water pump and fans to run while the induction heater board  cools off..
However, if the pump and fans ARE powered by the DC power supply, (through a small buck converter) you will need to make sure that their buck converter input is connected directly to the power supply's output, not to the induction heater board.
  I can't tell from the pictures whether there is a buck converter off of the 48 volt supply for the pump and fans or whether there is a small separate 220 volt to 12 volts (or whatever that pump takes) power supply that powers them.

Since I do not have one of these exact units in front of me, I have to do some guessing here.
---- I used to be part of a group that had to write completely unambiguous test questions.
You may have NO idea how hard it is to do that.
It can be just a hard to write completely  unambiguous answers to the questions of others sometimes.
So, if what I say does not make sense, feel free to question my answers or to ask for clarification.

I do not consider myself to be a "master" at anything.  Maybe just a page or two ahead of you in the textbook,
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 28, 2020, 10:33:50 PM
Put Parts into Workcoil before turning DC power on?

Several people have asked if they can put the workpiece into the work coil BEFORE applying power. This video demonstrates me doing exactly that with samples the draw 3 different amounts of current with no ill effects.  I am using the 2500 Watt Chinese ZVS Induction Heater, running at 48 volts for this test, but I am pretty sure the same relationships will exist with other sizes.  Even IF I NEEDED to put the work into the coil first, I'd want to carefully run a test where I put the parts in AFTER the DC is on, just to make sure I wasn't exceeding safe current levels.

/>
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: arctichominid on June 03, 2020, 02:10:18 PM
Hello everyone,

New here. But figured I should share what I'm working on.

I submitted a project to the Hackaday 2020 prize that essentially is about shredding aluminum cans, and using a
ZVS induction heater to melt the shredded can material in a crucible that will have a stepper motor raise and lower
the crucible as well as dump the crucible contents into a heated baking tray.

https://hackaday.io/project/171954-aluminum-recycling-micro-factory

Here's the sub-reddit where I've begun collecting useful links related to electric forges, casting aluminum and casting copper and aluminum together
to make aluminum bronze.

https://old.reddit.com/r/ArcticMicroFactory/

I have loads of questions for everyone here if you're available to chat.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on July 26, 2020, 06:53:13 PM
Some folks have commented that, although there's a lot of ZVS induction heater stuff on my spaco.org website (and on youtube), it's hard for some to find the exact information they want.
  To make it a bit easier to navigate that data, I recently put a flowchart at the head of both of my main ZVS induction heater pages.  It is designed to direct the user to the pages/youtube videos that will be most helpful.

Here's where the flowcharts reside:
https://spaco.org/Blacksmithing/ZVSInductionHeater/1000WattZVSInductionHeaterNotes.htm

https://spaco.org/Blacksmithing/ZVSInductionHeater/ZVS1800Watt/1800-2500WattZVSInductionHeaterNotes.htm

Sorry that the flowchart is not interactive.  That's for another day.

 [ Invalid Attachment ]

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Epiconix on September 10, 2020, 06:18:51 PM
Hello all,

I have 2 brand new (Chinese) ZVS boards and a 1000W power supply (48V).

No matter what i'm doing, the boards draw only around 3V
the 2000W board have 2 fans connected to it, and I can see that they are working for a tenth of a second every 1-2 seconds (video attached).
I tried with another power supply (12V 3A) and i got the same result.

What am i doing wrong? what/where is the issue?



ZVS 1000W Link:
https://www.aliexpress.com/item/4000337672504.html?spm=a2g0s.9042311.0.0.27424c4dICyq8l
ZVS 2000W Link:
https://www.aliexpress.com/item/4000822524105.html?spm=a2g0s.9042311.0.0.27424c4dICyq8l
1000WPowerSupply Link:
https://www.aliexpress.com/item/33042313383.html?spm=a2g0s.9042311.0.0.27424c4dX1JfDZ
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on September 10, 2020, 06:21:28 PM
Hi and welcome to HVF

I think your power supply is too small. The overcurrent protection shuts it down and it tries to restart every 1-2 second.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Epiconix on September 10, 2020, 06:40:37 PM
Hi and welcome to HVF

I think your power supply is too small. The overcurrent protection shuts it down and it tries to restart every 1-2 second.

Thanks, FYI I needed to go out and buy a beer when I saw your ZVS unboxing video, heard you sipping that beer, couldn't help it.

I get the same result with the ZVS 1000W, you still think the power supply is the issue?
can you recommend on another Power supply ?


Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on September 10, 2020, 08:12:46 PM
Hi and welcome to HVF

I think your power supply is too small. The overcurrent protection shuts it down and it tries to restart every 1-2 second.

Thanks, FYI I needed to go out and buy a beer when I saw your ZVS unboxing video, heard you sipping that beer, couldn't help it.

I get the same result with the ZVS 1000W, you still think the power supply is the issue?
can you recommend on another Power supply ?

You seem to know my youtube channel :) I am a big fan of using old server power supplies, they are high quality, regulates fast to changing loads and usually have very high output currents in small packages. With a SMPS it is a good idea to have some 25-50% head room for the initial inrush current, when you try to power it on.

A ZVS driver like these is extremely power hungry and the higher voltage, the worse it gets :) So a 1000W power supply at 36V might be good enough for running the 1000W at 800W, but the 1000W 48V version might make the ZVS inverter try to pull 1200W.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on September 10, 2020, 11:50:18 PM
The link for power supplies that you sent shows a whole range of models.
  I hope you really DID get the 48 volt 1000 watt unit.  If so, what AC input voltage are you using?
If, by chance, you are NOT using a 220/240 volt AC input, the power supply probably will not make the full 1000 watts.
You might try testing the supply itself with an appropriate load to see what it is really capable of.  (the heaters from old electric clothes driers can be very handy for this.
I agree with Mads that the server power supplies are a great power source for these induction heaters.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on September 22, 2020, 01:48:11 AM
Melting some Cast Iron with the 2500 Watt ZVS Induction Heater:
I just put up 2 videos on the subject:
-The first one:  Making some minor changes to the 2500 watt unit and a few other preparations:
/>
-The second one:  Actually doing the experiment to see if we CAN melt ferrous materials:
/>
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: klugesmith on September 22, 2020, 05:42:28 PM
Nicely produced, Pete.  Thank you.  Is your ingot mold made of graphite?

Not to go too far off track, but who knows where to find that youtube video of aluminum-pouring gone wrong?
We see two teenage boys in white lab smocks and partial PPE, opening a small front loading kiln on a workbench. 
One takes out a crucible of molten Al, about the size of yours, and pours it into what appears to be an iron mold.
A couple seconds into the pour there's a sharp pop.
Camera continues to roll as the kids deal with small pools of Al starting fires on the bench,
at least one significant droplet that landed in pocket of someones's smock, etc.
No significant injuries or permanent scars from that incident.

We can only hope that youtube has a net positive effect on safety of new generations of nerds.
Pete's productions are good.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on September 22, 2020, 08:15:40 PM
Quote
Nicely produced, Pete.  Thank you.  Is your ingot mold made of graphite?
Thank you for your kind words, klugesmith.  Yes, it is made of graphite.  I spend 'way too much time explaining its fabrication on the preparation video.
Regarding the aluminum going wrong:
  I didn't say it in the video, but before I did the test, I baked the sandbox and crucible for an hour in a 250°F kitchen oven, to make certain there wouldn't be ANY moisture.
  i/we have had a couple of incidents in the past that brought that issue into crystal clear view!!!
For instance:   a bit of grease left in a bearing shell reacts exactly the same way that water does.   Seeing molten babbit squirting 10 feet into the air does leave a lasting impression.  Dry, Dry Dry!!!

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: extremeodd on September 23, 2020, 12:35:16 AM
Oh man this thread is a wealth of info on these various Chinese zvs induction heaters, was well worth the time it took to go through it in it's entirety. First off I just want to thank you Pete for your meticulous documentation of seemingly every last detail when doing or changing anything with these circuits. I've been to your site quite a few times while trying to learn what I can about how these work, why these work, why these like to randomly die, and how to make them work better. By far you've been my most helpful resource.

I currently have like 4-5 of the "5-12v" modules that are all the same design with slight variances in smd component sizing, one that uses a center-tapped coil (single toroid, meant for driving the included flyback transformer) that claims 12-36v input and has ifrp250n mosfets, and finally, one of the ones that is sometimes listed as "1000w" but still runs the bog-standard 2 caps/2 chokes setup with ifrp260 mosfets. Oh, almost forgot the 3.7v-4.2v unit, mostly because I've yet to find anything it has any application for due to its extremely weak performance https://www.ebay.com/itm/ZVS-Induction-Heating-Machine-Module-Low-Voltage-3-11V-4-21V-High-Frequency/292287860722 (https://www.ebay.com/itm/ZVS-Induction-Heating-Machine-Module-Low-Voltage-3-11V-4-21V-High-Frequency/292287860722).

I'm using these units mostly for heating up a "Dynavap VapCap" which is a 'flame powered' herbal vaporization device. Basically it's a stainless steel one hitter that has a cap that when heated to the correct temperature makes a loud click via a bimetallic disk, this indicates that it's ready for a drag. Normally the user would heat the end with a butane torch while constantly rotating but this leads to inconsistent performance and hotspots which can cause combustion instead of vaporization. That's where the cheap 5-12v induction heaters really shine, with a solid 12v psu it'll heat the dyna to temp in less than 10 seconds usually.

Not being one to leave well enough alone I've been trying to optimize the setup to work well on lower voltages as heating times dramatically increase once the voltage is under 12 and I'm wanting to make some portable setups that more pocket-friendly than 4 18650s allow. Most of my focus has been on flux control/concentration, stainless is a fairly poor metal for magnitism and I figured focusing any stray flux could be very beneficial. With that said, if finding solid info on the how/why of the chinese zvs induction heaters is tedious, finding anything about flux control in induction heating is just painful.

Theres a company that produces a putty of sorts that you cover the outside of the with and it concentrates the flux in the center of the coil (or on the inside of the coil if it's doing bore heating, on one side of a flat coil, etc) The companies site can be found here https://fluxtrol.com/magnetic-flux-control-in-induction-installations . Sadly I haven't found much in the way of disccusion on the topic anywhere on the internet and it doesn't appear the average person can buy it directly from the company.

The other "technique" I came across with some actual discussion, albeit it's posts from a single guy, using a ferrite U core with a corner broken off to create an intensely focused spot of heating. Basically, the U core goes through the work coil and the work piece is placed in the core's gap. He was using it to solder copper joints super cleanly with no visible color changing/torch damage. It appeared to work impressively well and I've had some limited success trying to use it for my purposes. If you want to try it yourself just take a ferrite core and clamp it with the one leg inside the work coil and a piece of metal (I use razor blades) pinched in the other leg. With 12v and about a 4ish amp load it'll almost instantly make the area that is in contact with the ferrite red hot, this could have applications for blade heat treatment imo.

Otherwise, my time with these devices has been mostly spent trying to fully grasp their operation and what causes them to randomly die. I've had 3 of the 5-12v units fail with no obvious signs of why or what failed and I managed to kill the centertapped work coil unit by using it to load test a 12v 30a psu I modified to be adjustable to ~20v. It simply stopped drawing current when I got up to about 12a@18v, I'm assuming one or both of the ifrp250n mosfets blew. Looked into replacing them and it's honestly not worth the trouble considering I can get another for $18 that also includes a DST flyback transformer.

Figured I'd try to contribute what little I've gleaned that I didn't see come up in the thread yet.

PS: Here's a quick tip for when you copy ebay/amazon/whatever URLs to share: You only need to copy up to the "?", everything from the "?" on are user data parameters that only serve to make the link excruciatingly long. (Also they are used for tracking where the person came from, amazon uses it for commission on affiliate links) For amazon the cutoff spot is "ref=pd", you only need the stuff before that for the link to work.
For instance:
https://www.amazon.com/SainSmart-Voltage-Switching-Generator-Ignition/dp/B00ZTTVX4O/ref=pd_bxgy_img_2/140-7389242-9697844?_encoding=UTF8&pd_rd_i=B00ZTTVX4O&pd_rd_r=e7161629-c58e-48bd-ba5a-87f0877b1b5a&pd_rd_w=QDj6G&pd_rd_wg=NjoIX&pf_rd_p=ce6c479b-ef53-49a6-845b-bbbf35c28dd3&pf_rd_r=JSFD7Y9DMKQYSXXED3MC&psc=1&refRID=JSFD7Y9DMKQYSXXED3MC

Turns into

https://www.amazon.com/SainSmart-Voltage-Switching-Generator-Ignition/dp/B00ZTTVX4O/

Or
https://www.ebay.com/itm/Digital-LC100-A-LCD-High-Precision-Inductance-Capacitance-L-C-Meter-Tester/123977420869?_trkparms=aid%3D111001%26algo%3DREC.SEED%26ao%3D1%26asc%3D225086%26meid%3D8c6a5bce9c874c0f83bed2e2a2b03f4f%26pid%3D100675%26rk%3D3%26rkt%3D15%26mehot%3Dnone%26sd%3D292287860722%26itm%3D123977420869%26pmt%3D0%26noa%3D1%26pg%3D2380057&_trksid=p2380057.c100675.m4236&_trkparms=pageci%3Aac9eeaf5-fd23-11ea-87c6-2eff4665e3f4%7Cparentrq%3Ab7f388d81740aadcf4cf25a4fff40952%7Ciid%3A1

Turns into

https://www.ebay.com/itm/Digital-LC100-A-LCD-High-Precision-Inductance-Capacitance-L-C-Meter-Tester/123977420869
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on October 31, 2020, 03:43:55 PM
Controlling the power to the work with a DC to DC Solid State Relay:
  Several people have asked about controlling the power to the work for these ZVS Induction Heaters.   Some ask about using frequency change to do it, others would control the input voltage to the board.    Sometimes, particularly when heating ferrous materials, the work demands wide variations of power as it initially heats up and then passes the curie point.
  Most of us don't have DC power supplies in the 48 volt, 50 amp range that are voltage-controllable from about 24 to 48 volts, so that's not an option.
So I am presently working to handle this power control by doing two things:
1. Inserting a DC to DC solid state relay into the DC line to the ZVS heater so I can use a thermocouple to sense temperature, and then control power with a PID controller
2. Using a simple meachanical "Jack" to adjust the position of the work within the work coil to keep the ZVS heater board's power consumption within safe limits.
  It the two videos that I link here, I show my progress to date.   Just recently I received my PID controller, but it was the wrong model, so I am using a simple on-off switch to control the DC to DC SSR as I watch the temperature of the work piece.

Part 1,  A cautious test of a DC to DC SSR to control my 2500 Watt ZVS Induction Heater:
/>
Part 2,  Higher power test of the SSR and using my workpiece "Jack" to control power requirements:
/>
Once I get the correct PID controller, I will install it and make an update video to demonstrate what it can do.
I will also make a short video showing the workings of a "fake" 100 amp DC to DC SSR.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on November 11, 2020, 04:29:56 PM
I finally got my new Chinese knock-off PID controller to drive the 40 amp SSR.  In this video, I heat up some copper, just enough to draw a total of about 36 to 38 amps from the 2500 watt ZVS induction heater. ----and it works well!!!:
/>
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 03, 2021, 07:01:15 AM
Here's a bit of a history lesson about the range of ZVS Induction Heaters and accessories for sale on the internet.  It is intended to help makers in choosing WHAT to buy:
/>
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on January 03, 2021, 09:00:09 PM
Hi Pete, good to hear from you again and coincidently on the same day where I just for fun searched for induction heaters on ebay and saw a version that you briefly mentioned.

 [ Invalid Attachment ]

Have you tried this unit yet?

I am wondering if the display is just for voltage/current measurements and it nice to see that they use a huge power film capacitor now, but the small PCB tracks is too little mass for effectively cooling it :(
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 04, 2021, 05:53:58 PM
I don't know much more about that model than you do, Mads.
But here's another one, of the same general type, that has a somewhat better description:
https://www.ebay.com/itm/ZVS-3000W-High-Voltage-Induction-Heater-Module-Flyback-Driver-Heating-Board/284075222797?
It says that the display does show some status information, but I don't know what else it may show.

It is baffling to me why the Chinese continue to do such a poor job of "Selling" the Features, Advantages and Benefits of their products.
For instance, I think it would be very beneficial for them to show a picture of the display in action, and then point out the value of each piece of information.

But I DO have to give these guys credit for the level of detail about operation that they document.
I also give credit to the tech guys who do the design.  I think they are really listening to the us guys who are their potential customers.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on January 04, 2021, 06:46:37 PM
Hello again, Mads.
Regarding the Display on the ZVS unit that you asked about:

I think the display shows:
Amps
Volts
Watts
Power On/Off
Maybe Mosfet temp
If a short occurs, which Mosfet(s?) have died.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on January 21, 2021, 03:00:50 AM
I know video is off-topic melting foundry and does not even use a ZVS melter, but the result is what I am still seeking using this same 2500W ZVS induction heater - jump to 5:05 in video to see the clean casting pour.

Link to graph shows copper's resistivity linear increase as temp rises all the way to liquid melting state.  A higher electrical resistance in liquid-state bodes well for clean melts - less stirring/gas pick-up, etc. for both copper and copper alloys. ----->>  https://en.wikipedia.org/wiki/Electrical_resistivity_and_conductivity#/media/File:Resistivity_Cu-Ag-Au.svg

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on January 25, 2021, 08:37:01 AM
I know video is off-topic melting foundry and does not even use a ZVS melter, but the result is what I am still seeking using this same 2500W ZVS induction heater - jump to 5:05 in video to see final clean casting pour.

Link to graph shows copper's resistivity linear increase as temp rises all the way to liquid melting state.  A higher electrical resistance in liquid-state bodes well for clean melts - less stirring/gas pick-up, etc. for both copper and copper alloys. ----->>  https://en.wikipedia.org/wiki/Electrical_resistivity_and_conductivity#/media/File:Resistivity_Cu-Ag-Au.svg

You need a crucible to get high enough temperatures. Properly also need a good amount of insulation around it and some serious water cooling to keep the work coil alive. With just 2500W you will properly see that you need to spend much longer time to get to the right temperature than with a more powerful heater.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on February 08, 2021, 08:36:59 PM
A rusky discovers the joy, trial and tribulations of melting with induction.

"I melt non-ferrous metals. Hot video. An induction heater melts copper."
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: klugesmith on February 09, 2021, 06:24:13 PM
Slightly off topic, I got a propane-fired Cast Master crucible furnace set for Christmas.  Insulation is refractory fiber blankets, and the whole thing is less powerful than one I made for backyard foundry more than 30 years ago.  No air blower, no BTUH or kW rating.

Need to whine about misuse of English technical words in many online advertisements, how-to-do-it videos, etc.

1. Too many accounts say smelting, when they mean melting metal.
It's not smelting if there is no chemical reduction of rock (ore) to metal at high temperature.

2. Too many youtube accounts refer to a crucible furnace as a foundry.  Nope, a foundry is a factory or workplace for producing metal castings.  In my experience, making good molds took more materials, equipment, and practice than melting and pouring metal.
Title: Turbo Charger Add-On Board for ZVS 2.5KW Induction Furnace
Post by: hightemp1 on February 12, 2021, 02:55:37 AM
Slightly off topic, I got a propane-fired Cast Master crucible furnace set for Christmas.  Insulation is refractory fiber blankets, and the whole thing is less powerful than one I made for backyard foundry more than 30 years ago.  No air blower, no BTUH or kW rating.

Need to whine about misuse of English technical words in many online advertisements, how-to-do-it videos, etc.

1. Too many accounts say smelting, when they mean melting metal.
It's not smelting if there is no chemical reduction of rock (ore) to metal at high temperature.

2. Too many youtube accounts refer to a crucible furnace as a foundry.  Nope, a foundry is a factory or workplace for producing metal castings.  In my experience, making good molds took more materials, equipment, and practice than melting and pouring metal.

Agree completely.  Chinglish -ya think someoneone would clue them in.  I ditched my 5 gallon super heavy furnace a few years ago and now really regret it.  Possibly lightness of cheap ones could be good though - less energy, although more maintenance?  Could easily soup up the burners by adding more air/gas?



To Pete & Company:
On the subject of souping things up... I think we should add a "tubo charger" (add-on mini board w fan) to the ZVS 2.5kW induction board so it can take full advantage of the 3KW power supplies everyone is using.  What says ya all?

I was reading one of Pete's older posts on adding caps and making longer coils with more turns to achieve lower frequencies (lower freq. gives deeper, faster heat penetration -not to mention safer component operation (correct me if I misread this)).  Seems industry induction furnaces are also using lower frequencies in the 5-10 khtz region, while the supplied coils that China sends us are about 5X, or 40kilohertz.  My background prevents me from whipping up a modified daughterboard schemetic.    Any thoughts from you engineers on upping up our 2500W boards to get more juice out of em, and to take full advantage of the 3kW power supplies?

No clue how to do it, but ideas so far include:
1. adding a similar sized third fan to add-on board so as to increase components/power by 50% to 3.7kW. Simply just copying/enlarging current design so this add-on board would be 1/2 the size of current board.  CRAZY 50% IDEA - Circuit board would probably explode into a million pieces and you with it  So option A is now a realistic five percent option or just pushing board without any modifications at all.
2. A full 20% boast to 3kW by modifiying current board, and future daugterboard so that freqencies are closer to industry's 10kilohertz standards .   A less expensive modification, but maybe more efficient & safer. So a turbo board/fan would be maybe 1/4th the size of the mainboard, and the mainboard would have some minor tweeking, somehow incorporated into the add-on board. 8)  Maybe this MOD could just be more CAPS and whatever necessary changes needed for more Caps?

#3 alternative - I think Pete proved that lower frequencies come at a cost = higher idle current).  If so, then maybe more than a 20%boast, say a 33% boast would be better. Though this may entail making larger traces, effect larger traces would have -- maybe a need for larger components, etc. so this may be impratical and unecomicial.  Basically, clueless here on whole turbo concept. (Correction: LOWER FREQ=LOWER IDLE CURRENT).   Still, hopefully some practical MOD for maxing out these 2.5kW boards with 3kw PS exists?

#4 alternative - 10% boast or 2.75kW - since engineers over design by this amount.  Just tweaking a few things here and there- ideas?

All brainstorming ideas welcome...

SURVEY - What Power Increase do You Vote For & How To Best Get It:

A. 50% 5% No Mods, just push current design
B. 20% Minor overhaul with turbo board
C. 33% Major Overhaul with larger "turbo board"
D. 10% Minor tweeks with smallish add-on board


UPDATE: LOOKS LIKE CHINA BEAT US TO IT - A 2500kW BOARD WITH SLIGHT MODIFICATIONS CLEARLY OPERATING AT NEARLY 3kW !  However Chinglish description warns against operating at maximum for prolonged time -Danger. So maybe we can run our 2.5kw boards at 3kw for short times and maybe 2750 all day?

https://www.ebay.com/itm/3KW-ZVS-Induction-Heating-Heater-Board-3000W-Module-with-DC-Power-Supply-Bundle/274555462249?
Looks like same board with possibly a very small daughter board for extra 12v connections. We can do better?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on February 12, 2021, 03:35:25 PM
Quote
I think Pete proved that lower frequencies come at a cost = higher idle current).

I think it's just the opposite.   Lower frequencies get you LOWER idle current.

See:
https://spaco.org/Blacksmithing/ZVSInductionHeater/Work%20Coil%20Data.xlsx

By the way, "they" are now producing 4000 watt and 5000 watt versions of this basic device.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on February 13, 2021, 06:51:01 AM
Quote
I think Pete proved that lower frequencies come at a cost = higher idle current).

I think it's just the opposite.   Lower frequencies get you LOWER idle current.


See:
https://spaco.org/Blacksmithing/ZVSInductionHeater/Work%20Coil%20Data.xlsx

By the way, "they" are now producing 4000 watt and 5000 watt versions of this basic device.

Pete Stanaitis
---------------

Thanks for correction, Pete.   Cool, so that means we can get more power into work loads just by lowering frequency?  And lower frequencies also give us deeper, faster heating for heavier loads, like 1.5 lbs of bronze in graphite crucible.  Your thoughts on a "turbo charger" add-on board???

Another question on why amps and watts crash from about 40a and 2.2Kw to about 25 amps and 1.1kW respectively, when melting copper.  Just went back to ohms law.  So if voltage is constant from our power supplies, and amps are dropping then resitance somewhere has to be increasing.   I also found out that copper's resistivity increases linearly 20 times from room temp to melting temp with a noticeable increase at melt temp.  Can copper's twenty fold resistivity increase be the reason power crashes so much?  Some have said thinning crucibles or melted copper shorting things out, are why watts/amps plummit?  Anyone?  If copper's increased resistivity is not the answer, where is the increased resistance to blame coming from? 

Update: Resistance and Resistivity are not the same!!  So, coppers 20 fold resistivity increase is not the same as its resitance increase. Anyone kow what percent copper's resistance increases between room temp and melt temp???


Assumption:  Total power at start is 2kw, Volts 50 & 40 amps -- and at end Watts are 1kW, volts are 50 and amps are 20
 
& we know these formulas from grade school:  W=V*I  & R=V/R
 
So solving for resistance at beginning of melt and end of melt, I get (1.25ohms at start & 2.5ohms at end of melt)

I am thinking increase heat somewhere is causing the extra resistance.  Maybe the graphite's resitance is increasing and or copper's is to?

Interestingly, I just measured .25 ohms in supplied coil and .85 ohms in the crucible we are using though the crucible ohms seamed to jump around alot -could be my meter.

Math is the only absolute truth - resistance (ohms) has got to be nearly doubling!

I doubt the coils resistance is going up much so the graphite's must be going up.  But from what I've read, yes graphite resistance goes up, but does not double, maybe goes up only 40% max.  Not sure what else could be causing remaining ohmage increase.  Of course, the reason amps are dropping and ohms are rising could have nothing to do with graphite's resitance, but maybe something to do with how the ZVS circuit behaves.  But right now I think graphite's increased resistance is at least one of the culprits for power decline.  Possibly, Pete's theory about copper's resistance increasing with heat is correct and accounts for the remaining increase in amps. And the theory about a thinner crucible due to burning would also decrease power.  So maybe all three are contributing??? 

1.graphite resitance goes up with temp maybe 40%
2.Copper load resistance goes up maybe 50% with a noticable jump at melt temps.
3.graphite loss by burning is reducing power by 10%

This is all just my initial theory and may change tomorrow or in five minutes, so take it with gain of salt.  Why is induction heating so complicated?

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on February 18, 2021, 01:43:50 AM
Induction Heating Manual: https://ia803006.us.archive.org/18/items/HighFrequencyInductionHeatingFrankCurtis1944McgrawHill/High%20frequency%20induction%20heating%2C%20Frank%20Curtis%2C%201944%20Mcgraw%20-%20hill.pdf
(see page 26 for schematic for a 15kw heater from 1944)


More theory & practice on blacksmithing---https://www.youtube.com/watch?v=a7W9DSpnm5E
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on February 19, 2021, 05:36:49 AM
Pete's prior experiments with the 1000W ZVS heater showed, I think(?), that one could dramatically cut frequencies.  If I recall correctly, by altering the coil (more turns, wider diameter and slightly wider spacing between coils), and also by doubling capacitors, Pete was able to cut frequencies nearly 3X from approx 120 to 40KHz.  Seams like a good start to altering this board - from shallower metal surface heating for heat treating applications, to deeper heating for furnace melting of non-ferous metals.  Currently, a 1kg graphite crucible inserted into a stock coil is running at about 40KHz.  By doubling the coil's stock turns and doubling capacitor bank, hertz should be reduced to maybe 20KHz.  20Khz is much closer to where industy runs induction furnaces for melting non-ferrous metals (approx 5kHz). 

1. I counted 12 caps on the board.  Does anyone know off-hand what these caps are rated at and where to source exact or similar working substitutes? 
2. Pete used 8 gauge wire to bridge caps from main board to a 2nd board.  Not sure but I think 8 gauge should handle 50 amps on 2.5kW board so will start with that.
3. Will add 4 more turns to a coil and increase coil diameter enough so a longer silghtly wider #1.5 crucible can be used for larger melt capacity.

By doubling capacitor bank, and increasing coil turns/width I should get frequencies closer to my target.  Don't know for sure, but these two changes should not make the unit any less stable, and even possibly make it run cooler - at least the capacitors? 

As far as tweeks go regarding beefing up the board to accept more power, I am afraid that I would not know where to start so will just start with this frequency MOD.  Suggestions, comments more than welcome.


DISCLAIMER:  DON'T FOLLOW ME CAUSE I HAVE NO CLUE ABOUT ELECTRONICS or ZVS circuits !!

UPDATE: FREQUENCIES AT OR BELOW 20kHZ MAKE ANNOYING SOUNDS AND CAN WRECK HAVOC WITH YOUR OTHER SENISITIVE ELECTRONICS. MAYBE, NOT SURE, PERMITS ARE REQUIRED TO OPERATE AT THESE LOWER FREQUENCIES.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on February 19, 2021, 07:05:09 PM

The capacitors:
BM brand.
0.33 mfd  630VAC, ((1200VDC)
Type MKP

Regarding the uncommented post about the video about "Induction Heating 101":
You will be paying at least USD$1000 and probably more to get one of those running by the time you get a water coolant system running.
If you shop around, you can find some of them for as (apparently) cheap as USd$600, but then they want USD$300 plus for shipping!  And that's without a water cooler.

I am not saying that you shouldn't do it.  I know several guys who HAVE gone that way and they are generally pretty satisfied.
One HUGE caveat though---  Those things are always billed as "15KW" machines, although their maximum input power is only about 7.5KW.
But of course, as they say in China, "if you can cheat, cheat".  I am surprised that no one has taken them to task on that issue.
All that said, however, they do a good job, and generally seem to be pretty robust.

In my opinion, here's THE best guy to watch if you want to learn more about that unit:
(EJ of the Anvil):
https://www.youtube.com/channel/UCRRCSJ1ZNe3Sa-PmfsJsPzA

Pete Stanaitis
---------------



Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on February 20, 2021, 04:58:43 PM

The capacitors:
BM brand.
0.33 mfd  630VAC, ((1200VDC)
Type MKP

Regarding the uncommented post about the video about "Induction Heating 101":
You will be paying at least USD$1000 and probably more to get one of those running by the time you get a water coolant system running.
If you shop around, you can find some of them for as (apparently) cheap as USd$600, but then they want USD$300 plus for shipping!  And that's without a water cooler.

I am not saying that you shouldn't do it.  I know several guys who HAVE gone that way and they are generally pretty satisfied.
One HUGE caveat though---  Those things are always billed as "15KW" machines, although their maximum input power is only about 7.5KW.
But of course, as they say in China, "if you can cheat, cheat".  I am surprised that no one has taken them to task on that issue.
All that said, however, they do a good job, and generally seem to be pretty robust.

In my opinion, here's THE best guy to watch if you want to learn more about that unit:
(EJ of the Anvil):
https://www.youtube.com/channel/UCRRCSJ1ZNe3Sa-PmfsJsPzA

Pete Stanaitis
---------------

Thanks for cap specs: found a pack of 10 for about $10.  Ok to add all 10 to the "turbo board"? Good place to add caps on turbo board may be parallel with water tubes since the twelve caps on main board end there?

Very good video Pete, but currently I am less interested in how to use heaters and more in how they work.  Here's one from the boomer on theory and DIY zvs boards.  The program he uses for troubleshooting circuits looks interesting, at least for those who know how to use it. 

"THE BOOMER" waxing-out on induction zvs control circuit.
/>
Got me thinking about DIYing a bigger board.  Our 2.5Kw boards are pretty simple with common parts.  Other than the 6 mosfets and 12 special caps(about $25 total retail), I could probably have cobbled together everything else and slapped something silly together like everybody else.  Also wondering if this basic zvs design is scaleable up to 10kW -anyone here using 7.5kW or above?  China is out with a new 5kW unit and I suspect 6kw shortly.  Maybe at 10kW there is a better design since DC power supplies for ZVS get more expensive?

 Topic: What happens when L and C are not matched?  (Read 1221 times)
https://highvoltageforum.net/index.php?topic=638.msg4193#msg4193

Check out this massive crucible  8)

https://books.google.com/books?id=aCgDAAAAMBAJ&pg=PA24&dq=popular+science+induction+furnace&hl=en&sa=X&ved=2ahUKEwjq7Z-0wvfuAhWqd98KHek-AcMQ6AEwBHoECAIQAg#v=onepage&q=popular%20science%20induction%20furnace&f=false



Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on February 20, 2021, 07:42:56 PM
There has been people building large scaled up ZVS induction heaters before. Its now 10 years since Rogerinohio did it and I would guess that the Chinese manufacturers just copied his ideas on how to scale it: https://4hv.org/e107_plugins/forum/forum_viewtopic.php?id=129196
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on February 27, 2021, 04:50:48 AM
Not sure if anyone has posted pics of board yet so here are some pictures of the 2.5kw version I received with the two fans removed:

A magnifying glass (not pictured) showed that solder on one of the power diodes did not make it through to the back side of board?

Any tips on making a beefier turbo board? How bout 10 more caps, 2 more mosfets with matching diodes, resistors, etc. Making deeper traces somehow and overload protection for mosfets?


UPDATE: Mads, thanks for that link.  Very good stuff though appears site is no longer active.  I wound up on your website and if you don't mind I'd like to post your advice and tips on ZVS induction heater design:

"Considerations
The MOSFETs used need a voltage rating about 4 times higher than the supply voltage and a on-resistance below 150 mΩ. In ZVS operation the switches see a voltage that is π times input voltage, so 4 times rating of input voltage leaves some head room for playing it safe.

If supply voltage gets over 40 VDC, consider using resistors between 470R-800R for the gates. Supply voltage needs to be minimum 12 VDC, lower than 470R gate resistors can be used in that case, if supply voltage dips under 10 VDC, there is a risk of MOSFETs failing from overheating by working only in the linear region or short circuit if one of them stops switching.

Supply voltage should not exceed 60 VDC, as this is very close to 200 VDC across the MOSFET. The internal construction of MOSFETs with a higher voltage rating makes them unsuitable for use in a self oscillating circuit like this Royer oscillator.

A MMC is made from 27 capacitors to avoid excessive heating in a single capacitor. The capacitors will still heat as massive current flows between the tank and work coil. To get a good result, a large tank capacitance is needed, if a capacitance lower than 4 uF is used, results might be disappointing. It is strongly advised to use a capacitor with made from polypropylene (MKP) or similar that can handle large RMS currents, it might even be necessary to water cool the capacitor too. A MMC as the one I use here can only withstand short run times and will even then heat up.

The value of the inductors are advised to be between 45 to 200 uH and depending on core material the number of turns varies a lot, use a LCR meter to check the values.

Water cooling of the work coil is a must! Even at just small runs with moderate power input as the ones I have conducted, the work coil would take damage from heat."

(6)IRPF260N mosfets
(12)0.33 mfd  630VAC, ((1200VDC) Type MKP
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on February 28, 2021, 04:33:46 PM
Quote
Any tips on making a beefier turbo board? How bout 10 more caps, 2 more mosfets with matching diodes, resistors, etc. Making deeper traces somehow and overload protection for mosfets?

The best "tip" I have is this:
Simply look around on the internet for sellers of the 4000 and 5000 watt units to see what they have done.
Places like Ebay, Amazon, Aliexpress and Banggood.
To increase power, they usually simply add pairs of Mosfets and duplicate the driver circuits, adding more capacitance and thicker traces as they go.  Some manufacturers/sellers have also added protection circuitry, too.
But, of course, you have to find bigger power supplies to match all that output.  5000 watts at 50 volts is 100 amps, right?

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 02, 2021, 01:26:36 AM

But, of course, you have to find bigger power supplies to match all that output.  5000 watts at 50 volts is 100 amps, right?

Pete Stanaitis
---------------

Regarding bigger power supplies.  I think two of the common 3kW servers in series gives 100 volts @ 60 amps, not 100 amps, (parallel conncetion may be much safer) though you probably could wire PS in parallel.  Update: I think two 3kw PS in series @ 100 volts posses a serious electrocution risk since DC would be floated.  Not sure, but regulations i think require proper grounding for anything over 50 volts.

Yes, another obstacle, but more volts may be less of a problem than more amps with ZVS design.

Point is, I am not looking to get 5kW out of present board.  The TURBO add-on board option is maybe only a 10-30% boast so we can take full advantage of our 3kW power supplies?  A 2.5kw to 3kw over-clock gives 25% more power (500 more watts, or 10 extra amps @ 50 volts) to our work loads. 

Yes, for sizes of 4kW to 20kW heater/furnaces one would not modify this board, but would either start from scratch or buy expensive new. 

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 02, 2021, 04:37:50 AM
Quote
I think two of the common 3kW servers in series gives 100 volts @ 60 amps, not 100 amps,
100 volts is NOT an option for these things.  The supplies would need  to be in configured in  parallel. 

I'd work on coil design to maximize power TRANSFER.
Maybe even consider a cobalt crucuble, since cobalt's curie point is MUCH higher than that of iron bearing steels.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 02, 2021, 05:21:16 AM
Quote
I think two of the common 3kW servers in series gives 100 volts @ 60 amps, not 100 amps,
100 volts is NOT an option for these things.  The supplies would need  to be in configured in  parallel. 

I'd work on coil design to maximize power TRANSFER.
Maybe even consider a cobalt crucuble, since cobalt's curie point is MUCH higher than that of iron bearing steels.

Pete Stanaitis
---------------

Mads did three 750W power supplies in series for 2.25kW?  I read ZVS circuits work better with more volts and have trouble handling more current?  I think MADS said Mosfets peak out somewhere in these higher kW ranges, but china appears to be using them at least for the 4-5kW units.  Ok, so they are using dual 3kW PSs at 50 volts, as you say, then traces must be huge and more expensive mosfets, caps - either way components get pricier I suppose.
Update: I think two 3kw PS in series @ 100 volts posses a serious electrocution risk since DC would be floated.  Requlations require proper grounding for anything over 50 volts.Floating DC ground at any voltage posses risk.

Thinking coil only MOD might only give 5% max boast, if that.  I know thinner means more power, have you tried the 2 or 3kg skinny crucibles designed for those induction desktop gold/silver melters?  If I remember correctly, BadPeter could melt over 3 pounds with the 1800W unit though it took about an hour.

Cobalt currie point definately makes it an interesting unknown wildcard, but lots of power is going to have to go into a dense heavy metal, and resitance is not as good as graphite.  Sure its been done before but can find no literature and can find no crucibles.   Melting point of cobalt is actually about 25 degrees less than iron so maybe works with copper alloys esp. brass using a barrier coating.  Cobalt is so dam expensive but if you get 5 times more melts then ?   Crucible expense is a problem. 

But to your point - can cobalt give us more power?  There is a Russian You-Tuber who also thinks it is worth investigaing, and maybe has the resources to do test.  I guess time will tell us if Cobalt can give a power lift and/or compete with graphite.

Update: Pete, I think Dave says below that when increasing power for a given coil shape, voltage and current should increase proportionately.  With proposed turbo board, maybe most of increased power would be attributed to amperage increase only since voltage is fixed on PS.  So maybe coil shape should idealy change?  Also, as previously mentioned, I'd like to somehow get frequency closer to 20kHz or even 10kHz.  Adding 10 extra similar caps and a longer, wider coil should get me fairly close?

CHECK OUT THIS DESIGN:
/>
Picked up some good tips, no; but, you guys all need to give him some pointers, I think.

Disclaimer: I am an electronics knob.  Do not follow what I do.


Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 02, 2021, 06:22:52 AM
Such ZVS oscillators can run with more voltage or more current.  My little Jacob's ladder runs at 170V (but under 2kW).  IGBTs work well at higher voltages.  I've seen one running on rectified 220V line for an induction cooking appliance prototype (using SiC FETs).  For normal induction heating, the work coils need lower voltage at higher current.  For a given work coil, current and voltage need to increase together.  That is generally below line voltage, even here at 120Vrms.  (And isolation from line is needed for safety.)

Using two additional low-threshold-voltage FETs for gate drive (instead of pull-up resistors and diodes) can dramatically reduce gate drive power while simultaneously speeding up gate drive:
https://highvoltageforum.net/index.php?topic=1439.msg11002#msg11002
I've made several ZVS oscillators with variations of the above circuit.  So far none at 5kW, as I don't have any use for such, nor a sufficiently-large DC supply.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 02, 2021, 07:05:01 AM
Such ZVS oscillators can run with more voltage or more current.  My little Jacob's ladder runs at 170V (but under 2kW).  IGBTs work well at higher voltages.  I've seen one running on rectified 220V line for an induction cooking appliance prototype (using SiC FETs).  For normal induction heating, the work coils need lower voltage at higher current.  For a given work coil, current and voltage need to increase together.  That is generally below line voltage, even here at 120Vrms.  (And isolation from line is needed for safety.)

Using two additional low-threshold-voltage FETs for gate drive (instead of pull-up resistors and diodes) can dramatically reduce gate drive power while simultaneously speeding up gate drive:
https://highvoltageforum.net/index.php?topic=1439.msg11002#msg11002
I've made several ZVS oscillators with variations of the above circuit.  So far none at 5kW, as I don't have any use for such, nor a sufficiently-large DC supply.

davekni, thanks so much for your hands-on reply.  Would really appreciate your suggestions for my turbo add-on board idea for these 2.5kw ZVS induction heaters(basically somehow increasing power to the stock 2.5kW circuit board from China so output is closer to the power supply's 3kW capability).  Increasing power by 20% puts 10 more amps or 500 extra watts into work load.  Not a huge deal of course, but for my application it would be really sweet.

What's your opinion of the 3kW induction furnace in the video from schematic above?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 02, 2021, 10:18:05 PM
To clarify, voltage and current increase together for a given coil, given frequency, and given impedance of object being heated.

If I understand correctly, you have two goals:
1) lower frequency
2) increase current at a fixed supply voltage

Lowering frequency for a given coil is done by adding capacitors.  The resonant current (current oscillating between capacitors and coil) will increase.  Power supply current (input power at fixed voltage) might go up or down depending on impedance of the object being heated.  I don't know enough about what objects are being heated and how they respond to frequency.

More turns within the work coil will also lower frequency, and lower resonant current.  It will likely lower power at fixed voltage, as volts/turn is lower.

Adding capacitors will certainly help with lowering frequency, and might increase power draw.  If power draw doesn't increase, then add yet more capacitors and reduce work coil inductance (fewer turns).  That will increase power draw.

Now the critical piece - how to keep the board from overheating and failing at 3kW:
First, you will likely need larger inductors.  Lower frequencies require larger inductances.  Higher power (at fixed voltage) requires higher inductor current capability.  Both cause inductor physical size to grow.  I don't know specifications for what is there already, so don't know what margin may exist.  (BTW, total size can be a bit smaller with a single inductor feeding the center-tap of a ferrite transformer, transformer ends to the FET drains.  Simpler to stay with two inductors and increase size.)

I also don't know specifications for the FETs being used.  A significant portion of power dissipation may be switching loss with the simple pull-up resistor gate drive.  If so, reducing frequency may be sufficient to get 20% more current from the existing FETs.  Switching losses can be reduced with stiffer (lower value) gate pull-up resistors.  To avoid absurd power in the pull-up resistors, a pair of PFETs can be added in series to switch them on only when needed:


The above uses only one pair of power FETs.  However, it should work with many in parallel.  A single pair of PFETs should be able to switch out all of the pull-up resistors, allowing much lower pull-up resistance.

BTW, if you use a separate 12V supply for node "V12" instead of a pull-up resistor to the main supply, turning on that 12V gate supply first before the main 3kW supply solves the start-up issues that these circuits sometimes have.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 03, 2021, 04:00:14 AM
I couldn't find specs for: I320N20.
What is it?
What is its RDS on?

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 03, 2021, 05:53:33 AM
The full part number is IPP320N20N3 G, 0.032ohms 200V TO220.  I'm not recommending this specific part, although it does switch quite fast with very low miller charge.  Instead I was showing the circuit topology.  If I understand correctly, your goal is to modify an existing 2.5kW ZVS induction heater to 3kW.  My thought is to add two PFETs and lower-ohm gate resistors and a 12V supply to your exiting board.  For that you will need somewhat larger PFETs than I used, as my circuit had only a single pair of IPP320N20N3 G parts as the power devices.  That specific circuit is used as a 200W driver using a 19V laptop supply with only a bit of copper foil for heatsinking and no fan.  You will need PFETs that can conduct for short periods the current of all gate resistors in parallel for one side.  If you use 10-ohm gate resistors pulling up to 12V, that will be roughly 1A per resistor (after counting for diode and NFET voltage drops).  If there are say 6 FETs per side, then the PFETs need to handle short times (~1us) of 6A, say STP10P6F6 or FQP7P06 or IRF9Z24.

In other words, I'm suggesting a circuit topology, not those specific parts and values.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 04, 2021, 07:58:06 PM
The full part number is IPP320N20N3 G, 0.032ohms 200V TO220.  I'm not recommending this specific part, although it does switch quite fast with very low miller charge.  Instead I was showing the circuit topology.  If I understand correctly, your goal is to modify an existing 2.5kW ZVS induction heater to 3kW.  My thought is to add two PFETs and lower-ohm gate resistors and a 12V supply to your exiting board.  For that you will need somewhat larger PFETs than I used, as my circuit had only a single pair of IPP320N20N3 G parts as the power devices.  That specific circuit is used as a 200W driver using a 19V laptop supply with only a bit of copper foil for heatsinking and no fan.  You will need PFETs that can conduct for short periods the current of all gate resistors in parallel for one side.  If you use 10-ohm gate resistors pulling up to 12V, that will be roughly 1A per resistor (after counting for diode and NFET voltage drops).  If there are say 6 FETs per side, then the PFETs need to handle short times (~1us) of 6A, say STP10P6F6 or FQP7P06 or IRF9Z24.

In other words, I'm suggesting a circuit topology, not those specific parts and values.

Top notch advice -thx so much.   I am least qualified in this forum to suggest mods to my proposed add-on Turbo Board, hither known as TB.  To clarify application, basically trying melt more bronze in a taller 1.5kg graphite crucible for larger size castings (2lb range). And, also wish to add aluminum melting capability with taller/wider based 150ml. crucible.  Looking for frequencies in the 10-20kHz range -present china config with 1kg graphite crucible is 40 kHz. Here are ideas for TB, good or bad as they may be: 

1.Board currently has 12 caps.  Would a good start be adding 10 same spec'd caps in parallel directly attached to current ones on the underside of board? At the same time widening traces of all the caps, both inductors, and one trace to each mosfet?
2.Both inductors measured about 37uH in circuit.  Go bigger OK, maybe start by doubling them to, say 75uH?
3.Present circuit board has (6) IRPF260N mosfets (pic added).  Maybe start by adding two of the same spec'd mosfets - thinking no room on circuit board for them? (see pictures of circuit board in post #323 further up this page-17).  If no room, then both new mosfets would go on the proposed TB, along with matching diodes (there are no diode markings on china board -can someone please suggest a part number?), and matching resistors (1 power resistor (see pic-5W 470ohm) and 1 regular resistor(would this be the resistor you suggest decreasing. If so, I assume replacing them all and what resistance might be best?).
4.Incorporating a 300W 12 volt DC adapter into TB.  Didn't have time to figure that one out. but would appreciate advice on attaching/wiring it in TB.
5.20% more amps is goal, but bigger coils unfortunately give less power to load.  To keep from increasing inductance on a new taller coil design for bronze, use same amount of material in the coil but widen spacing between coils.  For Al the newly designed copper coil's inductance will have to increase to accomadate both wider and longer coil.  Lower Al melting temps should hopefully allow for this. Both new coils could be packed with high temp castable ceramic insulation, and swapped out with flared brass hardware.

As you know by now my electonics knowledge is sparce.  Above ideas are both specific in places and vague in others but at least is a start to TB.  Don't trash em too badly.  If you would please assist by filling in gaps, offering superior plans, both general and specific to proposed TB for Chinese 2.5kW ZVS induction heater.  Finally, since my schematic reading abilities suck, simple wiring explanations of components to TB would be sweet.  Thanks kindly in advance.  :)

PS if your niech is blacksmithing, I think same MODS specific to circuit boards would apply.  Basically we are just trying to juice an extra 10 amps or 500W out out this stock chinese 2500W zvs induction heater, and at the same time take full advantage of the commonly used 3000W server power supplies.


Disclaimer: I am an electronics knob.  Do not follow what I do.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 05, 2021, 12:22:33 AM
There are many better MOSFET options compared to IRFP260.  That part must be cheaply available in China (or some Chinese knock-off version).  3kW looks possible from 6 of these parts IF they are kept cold.  That appears to be the strategy of the claimed-3kW versions on EBay.

If you can change MOSFETs, a quick Digikey search shows FDA70N20 as a good option - higher gate threshold voltage, lower resistance, about 1/3rd of the gate charge.  No need to increase gate-drive power with better FETs.  These list as $3.17 in singles or 10 for $26.63 on Digikey.  May be cheaper other places.  Or, for a bit more money, IXTQ76N25T is a 250V option for a bit more voltage margin.  There are many others that are better than IRFP260.  The IPP320N20N3 part I used is also better, but in the wrong package for a retrofit here.  There are likely many IGBTs that would also work here too, but there also might be unpleasant surprises with such a more substantial change.  (Gate zener diodes would need to increase to 15V for IGBTs too.)

In short, if you are willing to change FET parts, there is no need for additional parts.

Fans blowing down on the board would be great - both to keep FETs cool but also to keep caps and ECB traces cool.

Additional caps on the back is fine.  With fans blowing on the board, adding trace thickness may not be necessary.  If back-side caps get too warm, blow air at the board from both sides.  By fans blowing on the board, I mean something like the 3kW EBay models show, but duplicated for the back-side caps.

That leaves the only hard problem: inductors.  Power inductor design is somewhat complex.  Current rating is even more important than inductance.  For these powdered iron cores, saturation current isn't a hard value, but rather a continuing drop of inductance as current increases.  You need higher current for increased power and higher inductance for lower frequency operation.  Think of something big, like this 3" core on EBay:
https://www.ebay.com/itm/Large-Toroid-Core-3-Inch-Diameter/202733471216?_trkparms=aid%3D1110006%26algo%3DHOMESPLICE.SIM%26ao%3D1%26asc%3D20201210111314%26meid%3D63b757a18a3244079717fbb1554b26b4%26pid%3D101195%26rk%3D2%26rkt%3D12%26mehot%3Dpf%26sd%3D274175297983%26itm%3D202733471216%26pmt%3D1%26noa%3D0%26pg%3D2047675%26algv%3DSimplAMLv9PairwiseUnbiasedWeb&_trksid=p2047675.c101195.m1851
Wind with perhaps 2 strands of 14AWG wire.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 05, 2021, 03:08:09 AM

In short, if you are willing to change FET parts, there is no need for additional parts.


Please forgive my ignorant questions -now it's clear to all why I flunked physics.  Replacing just the six current FETs with the more efficient FDA70N20 FETs will increase power to work piece in current ZVS stock configuration by roughly how many watts?

Example: Currently when Peter places his #1 crucible in his matching coil it draws about 40 amps @ 50 volts.  Just by upgrading to superior FDA70N20s FETs, the same coil/crucible will now draw more amps, say 48 amps @ ZVS's fixed 50 voltage?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 05, 2021, 04:13:43 AM
Sorry, I wasn't clear.  Changing FETs avoids the need for additional FETs (and associated diodes and resistors).  Changing FETs will not by itself change power significantly.  (Perhaps 2% increase due to lower voltage drop and faster switching.)  Changing FETs will keep the circuit from burning up when caps and/or work coil are changed to increase power to 3kW.

So you still need to add caps to reduce frequency, which may increase power.  If not enough increase, then add yet more caps and decrease work coil inductance.  (Or, change to 250V FETs and increase input voltage from 50V to 60V.)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 05, 2021, 05:08:14 AM
Just call me a knob -thanks for not laughing out loud.:-[  Assuming work coil inductance stays about the same, can you ball park about how many more capacitors would be needed for a 10% boast?  Currently, I think there are six .33uf in parallel or 12 total caps.

Many of these 3kW server PSs do have hidden POTs for cracking up voltage.  So that may be the quickest way.  48v to 53 would be a nice 10% boast. 

Just noticed that my ZVS board model has an isolated fan line not tied into circuitry.  Some have so fan speed automatically varies accordingly to load -easiest way to MOD for auto fan speed? If too complicated, a boast converter could crank amps on fan to maybe 15 volts?

I don't have any coil that size in my parts bin but have some smaller ones.  In the video above by Schematix, he gangs two smaller inductors together to effectively get same larger inductance.  I don't have a scope to test circuit, but do have an inductance meter to measure - you said maybe 3 turns.  Current inductance on each existing Inductor is 37uH, roughly about what uHenry should I be shooting for on each pair of two smaller ganged inductors?   

Indeed, I am a little slow.  Thanks so much for your patience and time, and for sharing your knowledge.  :)



Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 05, 2021, 05:43:13 AM
You don't need to do anything with the two inductors that are connected between the power supply and the tank circuit.  Their purpose is to minimize coupling back to the power supply.  They don't contribute to the tank's operation in any meaningful way as far as I can tell.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 05, 2021, 07:41:34 PM
A 10% voltage boost should give a 21% power boost.  Voltage is 1.1x.  Current will follow, so be 1.1x.  1.1 * 1.1 = 1.21x (power is current * voltage).

If you stay at 40kHz the existing inductors might suffice.  They appear to work for the 3kW units sold on EBay, which look like this 2.5kW unit with large fans added.  If you add caps to reduce frequency, you will almost certainly need larger inductors.

A bit of explanation on the inductors - please ignore if it doesn't make sense:  When current increases, iron-core inductors saturate, so inductance drops.  At high current, inductance can drop to a tiny fraction of what you measure with a meter.  The low inductance causes high current ripple.  High ripple current can cause two problems.  First, it may cause your DC supply to shut down due to over-current spikes.  Even if the supply doesn't shut down, high ripple current is heating the supply's output capacitors, which may lead to eventual failure of the supply.  (The supply's intended server use doesn't draw significant ripple current from the supply.)  The other issue which may show up first is added FET current.  The high ripple current comes from the supply and is conducted through the FETs.  Even with the better FETs I suggested, if ripple current gets high enough, the FETs forward voltage drop will be too high to keep the opposite FETs off.  The result will be fried FETs.

Unfortunately, there is no simple way to design inductors.  Measuring with your meter gives inductance at low (almost zero) current.  Saturation can make the inductance drop by an order of magnitude or more at high current.  High-current inductors are usually custom-made for specific applications.  I don't know of any good source for pre-made 50A inductors.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on March 05, 2021, 08:48:19 PM
How about buffering the server with a beefy set of caps that takes the brunt of the ripple? In a post LC filter style.

You would probably need inductors(L) between the supply and these cap banks (C) or otherwise it might either be difficult to start the supply due to the charging current trips overcurrent protection, or the reduced ESR might cause instability problems due to regulation loop going haywire.


Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 06, 2021, 03:41:03 AM
A 10% voltage boost should give a 21% power boost.  Voltage is 1.1x.  Current will follow, so be 1.1x.  1.1 * 1.1 = 1.21x (power is current * voltage).

If you stay at 40kHz the existing inductors might suffice.  They appear to work for the 3kW units sold on EBay, which look like this 2.5kW unit with large fans added.  If you add caps to reduce frequency, you will almost certainly need larger inductors.

A bit of explanation on the inductors - please ignore if it doesn't make sense:  When current increases, iron-core inductors saturate, so inductance drops.  At high current, inductance can drop to a tiny fraction of what you measure with a meter.  The low inductance causes high current ripple.  High ripple current can cause two problems.  First, it may cause your DC supply to shut down due to over-current spikes.  Even if the supply doesn't shut down, high ripple current is heating the supply's output capacitors, which may lead to eventual failure of the supply.  (The supply's intended server use doesn't draw significant ripple current from the supply.)  The other issue which may show up first is added FET current.  The high ripple current comes from the supply and is conducted through the FETs.  Even with the better FETs I suggested, if ripple current gets high enough, the FETs forward voltage drop will be too high to keep the opposite FETs off.  The result will be fried FETs.

Unfortunately, there is no simple way to design inductors.  Measuring with your meter gives inductance at low (almost zero) current.  Saturation can make the inductance drop by an order of magnitude or more at high current.  High-current inductors are usually custom-made for specific applications.  I don't know of any good source for pre-made 50A inductors.

Thats' good stuff!!  Pete, do you know if your PS's voltage can be tweeked - seems like 10% is a safe overclock, if you have the POTs? 

David, wondering if you watched the 3kw heater build on page 17 by Schematix-"the dreamer" ?  He has a riduculously huge work coil for a 3kw heater and more caps than we do.  When he cranks up the volts to 50 he experiences exactly what you are referencing I think (at 15 minutes, 15 seconds into the video  - ferrite coils were getting very hot.  Most importantly though is his solution for expensive bigger ferrite coils (doubling or paralleling two smaller sized coils).  Seams like the way to go for bigger coils, no? His design is a little different but the ganged coils did the trick of both cooling em down and smoothing out wave forms on his scope.   

Just dawned on me that longer and wider work coils decreases inductance, not the other way (doh!) I keep mixing em up - but bigger is great for getting closer to target frequency.   

So far here are the proposed mods:  Crank up volts 10%, doubling up fan-cooled caps to underside of board, ganging pairs of ferrite inductors to increase inductance, plus one longer and one bigger work coil.  Now, at last, 500 more watts will be available to work coils with more conducive frequencies for melting metal, and the full power of the PS will be utilized! Thank you Peter & David!!!!

Addendum: Adding two more mosfets is still on the table but a little more complicated, at least for me.  I get how replacing all existing FETs with better quality ones allows for more stability (good for long metal melts) and marginaly more efficient system, but not sure what adding two more mosfets does?

Another dumb question.  Pete, I noticed you have a DC switch to power on ZVS after PS is powered on (now china copied you and includes them).  Looks like you also now have a SSR relay.  Wondering if I could ommit buying the DC switch and just use the relay to power on ZVS?  I like the relay idea as a future temperature controller, but wondering if it can also replace the DC switch.
https://www.ebay.com/itm/173467987699?chn=ps&mkevt=1&mkcid=28&var=472165317799

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 06, 2021, 05:59:49 AM
Yes, I saw the video.  BTW, the inductor cores are powdered metal (iron alloy), not ferrite.  Using two identical cores with the same number of winding turns will double inductance, but leave current rating the same.  Using two cores with 71% of the turns (sqrt(0.5)) will give the same inductance as original with 1.41x current capability (sqrt(2)).  90% of the original turn count might be about right for your plans - 11% more current capability and 1.62x original inductance.  He is running 38kHz per the scope capture.  If you plan to lower frequency below ~30kHz, then you may need yet more inductance.

For a given work coil turn count, wider increases inductance, longer decreases inductance.  If you add turns to make it longer, then longer increases inductance due to the increased turn count.

Adding two more FETs is similar to replacing the existing ones, but requires additional 5W 470ohm resistors and diodes, and still doesn't switch as fast (more losses).  The replacements I suggested are closer to adding 4 more FETs.

Some SSRs can't handle inductive loads well.  You may need to add a diode across the ZVS DC power input to prevent voltage spiking at turn-off.  Of course, that diode needs to be in the direction to not conduct normally (cathode to + supply, anode to - supply of ZVS board).
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on March 06, 2021, 08:48:52 PM
Tip: You can rip out the filter inductor of old ATX power supplies for the 5V and 12V part, and possibly the PFC correction circuit. They are mostly powdered iron or material with similar properties (low AC ripple but high DC current)  and sized maybe somewhat in the right direction (1,5 inch and larger) so parallelling cores to allow for higher current might work if you have some large ones and perhaps not the largest IH.

Or if you are lucky and have some humungously large ferrite cores you can always gap them with paper to get the right inductance but they still saturate HARD at much lower flux densities than powdererd iron cores so you have got to know your core, and hard saturation would at best trip overcurrent in your power supply, at worst fry the lot.

Using a relay for on/off PWM regulation of the IH might pose some problems, as they have to be rated for a huge current. But with a low PWM frequency (after all time scale is in the seconds/minutes) wear and tear might be keept reasonable but still I expect relays to fail frequently.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 11, 2021, 07:07:09 PM
Yes, I saw the video.  BTW, the inductor cores are powdered metal (iron alloy), not ferrite.  Using two identical cores with the same number of winding turns will double inductance, but leave current rating the same.  Using two cores with 71% of the turns (sqrt(0.5)) will give the same inductance as original with 1.41x current capability (sqrt(2)).  90% of the original turn count might be about right for your plans - 11% more current capability and 1.62x original inductance.  He is running 38kHz per the scope capture.  If you plan to lower frequency below ~30kHz, then you may need yet more inductance.

For a given work coil turn count, wider increases inductance, longer decreases inductance.  If you add turns to make it longer, then longer increases inductance due to the increased turn count.

Adding two more FETs is similar to replacing the existing ones, but requires additional 5W 470ohm resistors and diodes, and still doesn't switch as fast (more losses).  The replacements I suggested are closer to adding 4 more FETs.

Some SSRs can't handle inductive loads well.  You may need to add a diode across the ZVS DC power input to prevent voltage spiking at turn-off.  Of course, that diode needs to be in the direction to not conduct normally (cathode to + supply, anode to - supply of ZVS board).
  Most excellent sirs - thanks again!

Drawing close to 60 amps could be an issue especially with longer runtimes so modding larger traces may be advised.  Going by Schematix's video the traces that need to be enlarged are the ferrite inductors, the caps and the middle mosfet pins to each cap.  Seems like the wire gauge used in Schematix video was overkill if fan cooling is used as well.  Assumption - board is rated for 52 amps @ 48 volts, if fan and water cooled.   Plan is to add 500W to work-load by tweeking/increasing POT's power supply by 5 volts, then a 5 amp boast will naturally follow with a ZVS circuit.  Wondering if 16 gauge stranded copper wire will be sufficient to handle extra 5 amps/volts and extra 500 Watts of power?? Many of the traces appear to be hidden either under the large caps or invisibly sandwiched between top/bottom of board?  Still, all components have through-hole pins to back of board so it should be a simple matter to build up some solder on bottom of board to the caps, inductors and center pin of each mosfet, then solder a connectiing wire to all those pins, no?  While expanding traces, this may be a good time to reinforce any other suspect joints, like my previously mentioned power diode's lack of solder, or maybe just lightly beefing em all up as a matter of course. Also, since the capacitor doubling plan is to use existing cap through holes, consideration should be given to expanding holes or seeing if they both will fit in existing holes. As Dave explains in next posts, enlarging circuit board holes is a bad idea.

Another thought just occured to me - the two planned larger crucibles may naturally draw more amps @ 48 volts.  Possibly, one of the crucibles would draw 10 more amps by itself so increasing PS voltage would be unnecessary, or at least not by 10%.  If larger crucible/coil combos naturally draw more amps then larger traces become even more important, I would think, since in this case the 500W increase could come from an amperage only boast of 10 amps - volts remain untouched at 48. So maybe a thicker 14 gauge wire added to traces would be better here? 

Lastly, wondering what effect changing the work coil diamenter from quarter inch to 3/8 inch would have on the power level going to the work-load?  Or put another way, if you have a 1/4" and a 3/8" coil basically the same diameter, length, coil spacing, etc. is the 3/8" one more efficient and if so, by how much?

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 12, 2021, 06:15:43 AM
The largest current is from caps to work coil.  The other paths (to FET drain pins and to inductors) conduct less current, although still significant.
Skin depth for copper at 30kHz is 0.38mm.  Copper foil will be more effective than large-diameter wire.

If modifying an existing board, don't try to enlarge holes.  Holes are plated with copper inside.  Enlarging them will break the copper connection between top and bottom layers.  For adding caps, solder to the lead stubs of the existing caps.  Not super rigid mechanically, but will work fine electrically.

Larger work-coil tubing diameter should increase efficiency a bit.  So will reducing space between the work coil and crucible, although that makes insulation thinner so increase crucible heat loss.

Yes, hitting exactly 3000 watts will be tricky.  Any slight change in coil geometry or crucible placement or even crucible temperature will change the power draw a bit.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 12, 2021, 07:31:02 AM
The largest current is from caps to work coil.  The other paths (to FET drain pins and to inductors) conduct less current, although still significant.
Skin depth for copper at 30kHz is 0.38mm.  Copper foil will be more effective than large-diameter wire.

If modifying an existing board, don't try to enlarge holes.  Holes are plated with copper inside.  Enlarging them will break the copper connection between top and bottom layers.  For adding caps, solder to the lead stubs of the existing caps.  Not super rigid mechanically, but will work fine electrically.

Larger work-coil tubing diameter should increase efficiency a bit.  So will reducing space between the work coil and crucible, although that makes insulation thinner so increase crucible heat loss.

Yes, hitting exactly 3000 watts will be tricky.  Any slight change in coil geometry or crucible placement or even crucible temperature will change the power draw a bit.

Wow, great info on thru holes, may also explain why I've had problems getting solder to stick on some hosed circuit boards I have repaired.  To solve rigidty issue I noticed the short bottom stand-offs have threads for accepting screws so I will find out the thread size and buy half a dozen long screws, then maybe attach to 1/8" plywood or acrylic sheet cut to same size as circuit board. Rigid connection between new plywood board and existing circuit board will not only protect caps but will also give a solid foundation for mounting fans to cool them.
Since going to 3/8 coil only gives minimal power increase, I will use up my large roll of 1/4" before moving onto 3/8".  In keeping with a minimalist "use what you got" mentality, rather than using copper foil, thinking about pounding 14 gauge wire flatter, annealing, repeat, etc., then drilling very small holes in DIY "foil" to attach to stubs and new caps, inductors, and existing mosfets.

Update: would this 63amp 125volt DC circuit breaker make an OK on/off DC switch?
https://www.ebay.com/itm/1P-63A-DC-125V-Circuit-breaker-MCB-C-curve/324176560751?hash=item4b7a6dc26f:g:qmMAAOSwLrleyr5V
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 12, 2021, 05:28:04 PM
I have been using a similar breaker as an on-off switch for my 2500 watt system for a few years now, and it works okay so far. However, I don't think I'd depend on them for overcurrent protection.  I have almost never turned that breaker OFF unless the system is down to its idle current state.   If you look at some of the kits or "combo's" that several sellers of ZVS heaters offer, you will see that they usually do include one of these for that exact purpose.

You can spend a lot of time trying to figure out  how those breakers really work for DC arc quenching.  Then, when you think you know enough, you still have to consider how much "Chinese de-rating" you need to put into your calculations.   At less than USD$3.00, what do you really think is going to be inside the shell?
After all, they can "fake" just about anything, and usually do.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 12, 2021, 07:47:20 PM
I concur with Pete on the breaker use.  Probably fine as an on-off switch with some limited life.  I bought a similar cheap 250V DC breaker.  It had no internal magnet to force the arc into any quenching chamber.  It did not quench an arc, at least not in the couple seconds I allowed it to continue arcing.  I expect 50V will be less of a problem.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 13, 2021, 07:00:26 AM
The largest current is from caps to work coil.  The other paths (to FET drain pins and to inductors) conduct less current, although still significant.
Skin depth for copper at 30kHz is 0.38mm.  Copper foil will be more effective than large-diameter wire.

Well, it is not exactly copper foil but I think thickness matches the capacitor's protruding stubs nicely.  Stubs are protruding about 1/20" of an inch and "foil" is also 1/20" of an inch thick.  Took me half hour of hammering, annealing, and straighting. All you blacksmiths could have done it in a couple of minutes.  Used regular construction 8 gauge soft wire, annealed it once - 8 gauge is border line as to even having to anneal at all to get to .05" or 1.25mm.  Will mark holes to drill in copper by pierceing stubs to a piece of copy paper -holes wil be large enough to accodomate extra caps.  Still need to hammer another similar cap strip, six shorter cap to mosfet strips(12 gauge), and 2 cap to inductors to coil post strips(8 gauge).  There are no traces in back so hopefully I will not melt circuit board when filling holes with 60/40 solder? -the circuit board's traces must be hidden under the caps on top side of board.  This trace MOD should now allow the unit's traces to easily hande 3000 Watts and 60 amps, up from china's stock 2,5KW.  Thank you David for the "modified foil" tip - flatter traces will be much neater than wire, and solder joints should be significantly studier than they would be with wire. ;D

Update: There are six mosfets and currectly 12 caps, possibly 2 series of 6 in parallel -no clue really?  In the Schematix video when he enlarges traces, he attaches the middle pin of each mosfet to the nearset cap.  I assume that I can do the same with all six of my mosfets?

In hindsight using 8 gauge wire seams like over-kill for just trying to get an extra 10 amps.  Can one have traces that are too large, or put another way do larger traces take power away from the work load?  For all I know 14 gauge is more than adequate.

UPDATE: AliExpress just came out with a circuit board only:
https://www.aliexpress.com/item/4000736471927.html?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 13, 2021, 06:18:50 PM
No downside to added copper here, other than the work to add it.  The circuit board should handle soldering temperature fine as long as the board isn't under simultaneous mechanical stress.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 14, 2021, 02:52:57 AM
No downside to added copper here, other than the work to add it.  The circuit board should handle soldering temperature fine as long as the board isn't under simultaneous mechanical stress.

Like Pete says china over specs and underpowers everything so it is comforting to know that pumping up traces should do no harm - thanks again. :)

Another question for you Pete - I know you have stated that the maximum sustained amperage you allow on your 2.5kw system is 50 amps.  So of course I am curious with these proposed MODs if you would feel confortable pushing your rig to 3kw? Please Dave or anyone else feel free to comment as well. Dave, I know you are not fond of the existing FETs(IRPF260N) specs but for 3kw(50volts @ 60 amps or 55v@55a maybe they are OK?

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 14, 2021, 04:57:54 PM
Maybe my self-imposed current limit of 50 amps at 48 volts is a bit arbitrary, but it's all too easy to stick something a bit too far into the work coil and see a short current surge well above that.   So---- I wouldn't be comfortable with a 60 amp limit.
Recently I have been making some "flattened" solenoid coils in an attempt  (not successful yet) to get small pieces of tool steel well beyond the curie point for forging.  During one test, I stuck a larger piece of bar into the coil and got a huge current surge that shorted the 60 amp DC to DC SSR that was in series with the load.  The Mosfets on the  pcboard didn't blow however.
  I guess you get to decide.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 14, 2021, 06:17:21 PM
IRPF260N will work fine when cool.  As they heat up, on-state-resistance goes up and gate threshold voltage goes down (as with all FETs).  At high current and high temperature, the on-state FET can't pull it's drain voltage low enough to keep the opposite FET off.  When current is high and temperature reaches some threshold, the off-state FETs conduct enough current to generate more heat and yet higher temperature.  Result is thermal runaway.  (Better FETs could handle more current and higher temperature before reaching a runaway condition.)

In other words, it will work fine with aggressive cooling.  If FETs get too warm, thermal runaway and fried FETs will result.  If the FETs fry, then you have a good excuse to replace them with better ones.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 15, 2021, 10:33:04 AM
Thanks guys!  Certainly won't be pushing FETs anytime soon, but future target will be 2.75kW (53volts @52 amps) for now.
Here is mofset datasheet for stock IRPF260N -makes no sense to me but posting for those who it does make sense:
https://www.vishay.com/docs/91215/91215.pdf

And here is one possible suggested mosfet upgrade #FDA70N20:
https://www.mouser.ie/datasheet/2/308/FDA70N20-D-1806350.pdf

If I do blow em up and upgrade to the above mosfets, then I am wondering if any of the mosfet's matching two resistor(s) values or two diode(s)specs also changed with changed mosfets? Or put another way, can above mosfets simply be swapped out without changing any other components?

Or alternatively, as mentioned in previous post, a cheaper mosfet upgrade would be to add two identical mosfets to the existing six mosfets, 8 total.  If two new mosfets are added do you foresee any problems in wiring them off the current board and on to my proposed plywood or plastic board that would be housing/supporting cooling fans underneath the existing circuit board?  Or more to the point, would Dave or someone please be so kind to explain to a knob like myself how the 2 extra mosfets and new resistor(s) and new diodes(s) would all be connected?

Pete, thanks for leaking your tool steel project, sounds like a great idea.  Do you think 3kW will get you there? :)   If not, Dave suggested possibly using even better mosfets that have 250v ratings that can easily handle 60 volts?  Don't know if these better FETs could handle 50-60amps though, or how other components would comply or handle over 3kW, etc.? --- Dave, do I dare ask if a 3.5Kw (60v 58a) MOD is physically possible and economically feasible with this unit? 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 15, 2021, 05:58:13 PM
Existing gate resistors and diodes should work fine with better FETs.  I don't have any specifications for the diodes.  If they work for IRPF260N, they should work for the better FETs too, as they have lower gate charge and lower on-resistance.  Just changing FETs should be fine.

Adding FETs is more tricky.  Wiring inductance can lead to parasitic high-frequency (RF) oscillations.  If you can extend the copper planes of the existing board, then it should work.  I think using better FETs with the existing board is easier and more likely to be successful - why I suggested that initially.

60V 60A should be possible with the 250V FETs I'd suggested IF the inductors are upgraded too.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 15, 2021, 06:47:03 PM
Existing gate resistors and diodes should work fine with better FETs.  I don't have any specifications for the diodes.  If they work for IRPF260N, they should work for the better FETs too, as they have lower gate charge and lower on-resistance.  Just changing FETs should be fine.

Adding FETs is more tricky.  Wiring inductance can lead to parasitic high-frequency (RF) oscillations.  If you can extend the copper planes of the existing board, then it should work.  I think using better FETs with the existing board is easier and more likely to be successful - why I suggested that initially.

60V 60A should be possible with the 250V FETs I'd suggested IF the inductors are upgraded too.

Wow, who knew such a cool MOD could potentially be so easy. So rather than the ZVS board being limiting factor, the power supplies may be limiting us from going above 3kW.  However, there seams to be several different used servo type 3kW PSs out there, and lucky for us, many have documentation on tweeking up volts by 10% and since ZVS loves more volts amps will follow so maybe a 20% bump or 3.6Kw PS is possible with the right model of PS?  Getting a KW more power just sounds so much better than getting .5kw more - especially if reliable. :) 

Dave, I just went back and looked at the 250v FET you suggested and it actually is only a 200v model.  When you get a chance please suggest a good 250v FET. Thanks again and so sorry to be such a nuisance.  Promise not to post for another week! :o
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 15, 2021, 08:59:07 PM
Sorry, typo in my earlier reply #334, now corrected.  I'd listed the same part number twice.  Real 250V option is IXTQ76N25T (or IXTH76N25T).  There are other 250V options too.  IRFP4332PBF is fine, higher gate charge, but still less than IRFP260.  There are more options too if those aren't readily available.

No concern about posting - especially when pointing out a typo that needs fixing.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 16, 2021, 09:59:24 PM
Broke my promise David only to say how admirable I think it is that you, Pete and so many others here take the time to help knobs like me out especially considering how valuable your time is.  :)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 20, 2021, 03:35:23 AM
Most of the used server-type 3KW power supplies out there are very good ways to power our +2.5kw IHs.  Alternatively, if you have several MOTs lying around like me you could make one.  For a target 3.5kw version it could get complicated though.   To get a 3.5kw rated transformer PS, I am thinking four MOTS may be required.  Reason being, 30 turns of proper gauge wire is about the maximum number of turns you can get for each transfomer, so to get 60V you'd wire two in series - so far so good?  Power-wise, assuming each MOT is equally rated at 1200 Watts, two MOTs in series would only give 2.4kw.  So maybe two more transformers added in parallel would be required, giving 4.8kw?  Pete and company please find the flaws in my thinking.  Another question is -- 4.8kw would require 220v outlet.  All my MOTs are rated at 120v - can a 120v MOT even be used with 220V?

UPDATE:  None of my 120v circuits have 30amp fuses at the panel.  Maybe add a 30amp 120v circuit and do three MOTs in series wired with 20 turns each with heavier gauge wire and still get 3kw?

If you'd like more info on how to properly rewire a microwave transformer check out Pete's YouTube video.
/>
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 20, 2021, 03:20:50 PM
I wouldn't be in favor of ever putting a 30 amp breaker on a 120 volt line  except in the unlikely event that the circuit had 10 gauge or heavier wire.
How about putting the primaries of two MOT's in series across 220 volts? 
Maybe this goes without saying, but please keep "one hand in your pocket" as they say,  when working with mains voltages.  Plenty of opportunities to get dead.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 21, 2021, 03:05:14 AM
You can use the primaries in series for 240V or 220V.  Best when the two series transformers are identical (same initial model number and exact same output turn count).  If you aren't sure of the turn counts being identical, then use series primaries only with parallel secondaries.  Turns ratio still needs to be close.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on March 21, 2021, 07:20:23 PM
I am not sure about US MOTs with 120V primary, but our here in EU are very cheaply wound and therefore they are well into saturation with a magnetising current in the amps range causing high idle losses. Judging from size I would say they are rated at about 3-400W at 100% duty.

My advice would be to use atleast three in series for 220V supply.

Also measure the primary resistance. Calculate the actual primary current that would lead to 5% losses (0,05*U primary *I primary = I primary^2 *Rprimary) and then you get a ball park figure about the actual 100% duty rating.

Also, as they are designed for high leakage with a magnetic shunt they have a lot of voltage drop when loaded. This helps stabilise output voltage though when you add enough capacitance as the leakage inducatnce and rectifiying caps form a LC filter. But the voltage will be load dependant.

MOTs are fairly easy to rewind though due to a large winding window, and with the windings separated not concentric but side by side in a split bobin. I have never successfully opened an EI transformer core to rewind it and successfully reassembled it with the same efficiency ( idle current, low noise etc) but maybe it is a matter of just getting enough experience.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 22, 2021, 04:21:14 AM
I am not sure about US MOTs with 120V primary, but our here in EU are very cheaply wound and therefore they are well into saturation with a magnetising current in the amps range causing high idle losses. Judging from size I would say they are rated at about 3-400W at 100% duty.

My advice would be to use atleast three in series for 220V supply.

Also measure the primary resistance. Calculate the actual primary current that would lead to 5% losses (0,05*U primary *I primary = I primary^2 *Rprimary) and then you get a ball park figure about the actual 100% duty rating.

Also, as they are designed for high leakage with a magnetic shunt they have a lot of voltage drop when loaded. This helps stabilise output voltage though when you add enough capacitance as the leakage inducatnce and rectifiying caps form a LC filter. But the voltage will be load dependant.

MOTs are fairly easy to rewind though due to a large winding window, and with the windings separated not concentric but side by side in a split bobin. I have never successfully opened an EI transformer core to rewind it and successfully reassembled it with the same efficiency ( idle current, low noise etc) but maybe it is a matter of just getting enough experience.

You guys know your stuff.  I could not find a single video of two or more 120v MOTs being used in series on 220V here in US, let alone three or more.   So I think knowing that 120v mots can be used with 220 is great (CORRECTION--matching 120v MOTs  to 120v wiring is necessary else 120v MOT's insulation may fry with 220v - two 120v MOTs can be used with 220v if their  primaries are wired in series so each MOT gets 120V). 
And, 30 amp 120v circuits are very rare in the states. Only use I could find, is people who plug their travel trailers up to their homes since many motor homes require 120v @ 30a.  So if one has to make another circuit it would seem logical to use the more common, more useful 220v circuit. 

Shocking that EU mots are rated at less than 500 watts.  The larger ones here in the states claim 1200+ watts, but an electrician who worked on microwaves once told me that most all these larger MOTs are all basically the same and only rated at 1kw.  He did not say what duty cycle that is - for me, I have never ran a micowave for more than 30 minutes. Come to think of it, my 1200W rated unit does possibly cycle power on and off -that sucks.   For 3kw or larger PS, I agree that three or more in series would be required.  Assuming 1kw per MOT is correct, then with losses from rewinding, it may take 4 to get  ++3kw, at an unknown duty cycle?  For melting larger quanties of metal in bigger crucibles a 50% or better duty cycle would be best (does 50/50 mean on half an hour then off half an hour)?
 
Again assuming that most of these larger MOTSs are all pretty much the same, then maybe overloading one will not be a huge deal.  I found one person who has re-wound several MOTs for welders (two Mots in series on 120V)- he says that they all eventually fail due to one MOT taking too much of the load.  He just accepts having to do another every few years.  When they do blow, does it take all of them or just one, and maybe some type of simple overload circuit; or, even simpler for me, a fuse(s) could be added in the circuit somehow to prevent blow ups?

Update: Randomly checked five mots in mothballs.  Only 2 were identical & looked medium sized.  3 larger ones varied in weight, windings metals & primary ohm readings.  So findling matching larger sized MOTs may be more difficult so getting very close voltage matches with matching iron cores seems to be challenging.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on March 22, 2021, 08:09:27 AM
Well, I have a 1000W EI transformer core (actually it is an old autotransformer 220/127V from the time we had 127V in Sweden) and it is substantially larger than any MOT I have seen on video, US /EU/JP/UK/wherever

I would say twice the size. All MO are made in China. All MOTs are made in China. I really doubt they would spend a single won extra on copper or magnetic steel than necessary, considering you blast your MOT at full power not more than a few minutes and the heat loss is absorbed by the thermal capacity.
When you operate your oven at longer periods it is with reduced power, it is run with low frequencye PWM (like 1 sec on/ 1 sec off)

At my office we have a couple of MO in the pentry, and during lunch hour they blast away constantly as each individual trying to heat the lunch box and soon the smell of food is mixed with that good ol´overheated Xformer core insulation smell. They are frequently replaced due to breakdowns  but finally the house maintainence crew have found some good brand tht last longer than usual. And now in Corona times everyone is working from home so...

My point is that an EU transformer ACTUAL rating is close to 500W, but not officially so. There is no rating on these. The nameplate says something like the magnetrone power output and perhaps something about minimum current drawn from wall socket.

You have to investigate this by perhaps doing the primary resistance evaluation. I would expect an US MOT to have some margin due to the less demanding requirments of lower primary voltage and higher frequnecy or, if the manufacturer have reaaaally done it dirt cheap it has the same flux density as an EU MOT. And hence they capaity is larger.
 
A power supply for an induction heater needs to be able to deliver full power for atleast an hour.
A home made welding xformer have a very low duty, unless you are an experienced welder who can just lay down fillet after fillet successfully
But eventually you stick one pin to much or something like that and wham the magic smoke is gone and you test your circuit breakers...
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 22, 2021, 09:37:15 AM
Well, I have a 1000W EI transformer core (actually it is an old autotransformer 220/127V from the time we had 127V in Sweden) and it is substantially larger than any MOT I have seen on video, US /EU/JP/UK/wherever

I would say twice the size. All MO are made in China. All MOTs are made in China. I really doubt they would spend a single won extra on copper or magnetic steel than necessary, considering you blast your MOT at full power not more than a few minutes and the heat loss is absorbed by the thermal capacity.
When you operate your oven at longer periods it is with reduced power, it is run with low frequencye PWM (like 1 sec on/ 1 sec off)

At my office we have a couple of MO in the pentry, and during lunch hour they blast away constantly as each individual trying to heat the lunch box and soon the smell of food is mixed with that good ol´overheated Xformer core insulation smell. They are frequently replaced due to breakdowns  but finally the house maintainence crew have found some good brand tht last longer than usual. And now in Corona times everyone is working from home so...

My point is that an EU transformer ACTUAL rating is close to 500W, but not officially so. There is no rating on these. The nameplate says something like the magnetrone power output and perhaps something about minimum current drawn from wall socket.

You have to investigate this by perhaps doing the primary resistance evaluation. I would expect an US MOT to have some margin due to the less demanding requirments of lower primary voltage and higher frequnecy or, if the manufacturer have reaaaally done it dirt cheap it has the same flux density as an EU MOT. And hence they capaity is larger.
 
A power supply for an induction heater needs to be able to deliver full power for atleast an hour.
A home made welding xformer have a very low duty, unless you are an experienced welder who can just lay down fillet after fillet successfully
But eventually you stick one pin to much or something like that and wham the magic smoke is gone and you test your circuit breakers...

Ok, so you are saying that even though the MO may be pulling over 1kw intermitantly, the duty cycle for an induction furnace should be 100%, so one would need six or more MOTs in series for correct duty cycle?  Not sure I would need 100% duty cycle though?  Due to cost of graphite, I would not want a melt to last longer than 1/2 an hour. so maybe I can get by with only a 55% duty cycle or 4 MOTs in series?  Just call me slowhand.  Aggressive air cooling would help.  Though another potential problem is cost - pulling 60 amps requires maybe 6 guage wire and wiring 4 MOTs with that would not be cheap?  Suddenly, the cheap MOT idea may not be so cheap anymore - help??

Other minor tweeks I've read are:
1.leave the shunts in or out - see it done both ways, or even using wood shunts - which is best?
https://theeducationalblog.quora.com/Repurposing-rewinding-a-microwave-oven-transformer
2.lengthen the primary wire for increased stability - trade-off power reduced?
3 a. magnetic wire for more power -though more prone to damage while wiring
3 b. thicker insulated welding type wire for less power but easier to work with.
4. E type MOTs can be cut open for easier bobbin installs, otherwise rewind old fashion way
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Mads Barnkob on March 22, 2021, 10:09:06 AM
I built a handheld 150 Watt induction heater, but it failed and caught on fire!

I thought I would be able to get "some" run-time on 6x 18650 batteries, but apparently not :)

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 22, 2021, 10:22:55 AM
Great idea.  V.2.0 could make a fortune.  8)
See through acrylic is always facinatiing.
Maybe a trigger hold/release button.
Some flashy LEDs and/or Laser Rays
Ray Gun sound effects, and/or
Nuclear Reactor Core sound effects.
"The Heater" or the "Mini Nut Buster"
Ok, I quit.  ::)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 22, 2021, 03:31:56 PM
I am a bit concerned about all this talk about "3 or 4 MOT's in series across 220 volts" or whatever.
Are we all on the same page here?

When I suggested that a person could put two 120 volt MOT's in series, I meant that you would put the PRIMARIES in series.  This supplies 120 volts to each transformer, so each transformer can still deliver full power, whatever you guys think that is.
  If you saw the spreadsheet in my video:
 
/> you saw that the "regulation" wasn't all that bad, at least in my opinion.  By the way, that was the SMALLEST of 3 sizes of MOT's that I have in my shop.

But, if you are considering 3 or 4  120 volt MOT PRIMARIES in series with the 220 volt line, you are simply reducing each individual MOT's input voltage well below 120 volts and thereby reducing the available output power.  But, even worse, with a lower-than-designed input voltage, you would need even more secondary turns to get the voltage up to where you need it.
    I guess I'd spend more time scrounging up a couple of big fans than I would spend worrying about over heating some MOT's that you get for free.
At that price, you could even have two banks of MOT's and allow them to cool alternately if you really NEED heat for longer periods of time.

Pete Stanaitis
---------------

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 22, 2021, 05:51:17 PM
Yes, sorry everyone, all this rewiring is really Greek to me so basically ignore anything that I blurt out - please forgive my ignorance.   :-[

To me it is just not clear how to best re-wire same rated MOTs to get a +3kw PS with, say a 60 volt 60 amp rating, and a duty cycle allowing 1/2 hour runs for melting metal.  :-\

My pea brain assimulates comments and thinks  3 or 4 Mots with series primaries & cooled, might work but really no clue.

Update: another suggestion - runing two as Pete recommends and having a third wired so it can be added in series, when Power falls at melting temps; or just runing 6 in a some even more confusing 3x3 parallel/series combo?

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 22, 2021, 05:57:03 PM
Quote
I would say twice the size. All MO are made in China. All MOTs are made in China. I really doubt they would spend a single won extra on copper or magnetic steel than necessary, considering you blast your MOT at full power not more than a few minutes and the heat loss is absorbed by the thermal capacity.

They spend extremely little on copper.  All the MOTs that I have are wound with aluminum wire.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 22, 2021, 09:49:58 PM
Quote
I would say twice the size. All MO are made in China. All MOTs are made in China. I really doubt they would spend a single won extra on copper or magnetic steel than necessary, considering you blast your MOT at full power not more than a few minutes and the heat loss is absorbed by the thermal capacity.

They spend extremely little on copper.  All the MOTs that I have are wound with aluminum wire.

I have unwound two mot transformers - both had aluminum secondaries and copper primaries.  The MOs that I find have had:  burnt out magnetrons, burnt out switches, burnt stirrer motors, or are were just tossed because inside paint pealed or MO was just too greasy.   Do not believe I have ever found one with burnt transformer windings.

UPDATE: Just checked 5 of my MOTS & two have copper primaries and four have aluminum primaries.  Depending on size copper MOTS weigh 1-2lbs. more and have smaller gauge wire 15-16 vs 12-14 for alum.  Did not check secondary metal.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 22, 2021, 10:31:47 PM
Quote
They spend extremely little on copper.  All the MOTs that I have are wound with aluminum wire

The older MOT's seem to have copper primaries. 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on March 22, 2021, 10:56:26 PM
Quote
The older MOT's seem to have copper primaries.

Interesting.  The only reason for split material that comes to mind is that they hadn't yet figured out how to make reliable connections to aluminum.  The secondary has enough voltage to break through any thin oxide layer that forms withing the crimp connection, while the primary may not.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 23, 2021, 05:16:31 AM
Taking a break from the MOT 3.5KW(60v/60a) power supply design.  Everyone seems to have different ideas, and a best design is a just a clusterfuch to me. 

On ZVS design, found this good simple intro video for nobs like myself.
/>
PS. Pete, his video uploads your modified schematic from your website of the 1kw ZVS heater - immitation is most sincere form of flattery.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 24, 2021, 06:10:24 AM
Dan Hartman used an array of 8 large MOTs connected to a 240V 50amp circuit; and, illustrates his recommended way of hooking up anywhere from 2 to 12 transformers for welder use. He advises against rewiring with magnetic wire for higher amp welders due to diffculty of wiring 6-10 guage, also said sourcing 6-8g was difficult or impossible. (regular wire is fine for high amp/short burst stuff like spot welders but not high duty cycle apps cause you basically lose half your Watts -best to use magnetic wire so you get full power out of MOT).  As Pete has pointed out, Power Supplies for ZVS have different requirements than welders.  Higher voltages are better for ZVS and huge amperages used by welders not so important.  My unique application here requires a PS cable of 60V @ 60A with a continuous run time of 30 minute.  I know this DIY stuff can just be seat of the pants guess work but I am still questioning  - What would your MOT array look like to meet above requirments??

I know some of you have opined on this already so if you prefer ignore.  I just have a mental block on figuring out the best wiring configuration for a more robust 60v@60a Mot PS. Maybe, assuming 20 turns of 8 gauge regular stranded building insulated wire on each MOT, then would 4 MOTs in some kind of series/parallel combo possibly work?  Also just noticed that for welders Dan connects the primaries in parallel and the secondaries in series.  That seems to be just the opposite what we are advocating here for a Mot ZVS power supply?

UPDATE: In video link above the guy has a whole video series on rewinding MOTS.  For Continuous Use he recommends like Pete to not use thick insulated wire but magnetic wire for getting higher duty cycle.  Good tip, again like Pete said, is to wrap two or more lesser gauge wire in parallel.  Problem with ordinary household wire is over 50% of its area is damn insulation.
/>




Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: MRMILSTAR on March 24, 2021, 04:03:03 PM
For 120 volt input, I've often wondered about the possibility of wiring the primaries in series and the secondaries in series for two 120 volt MOTs. The result would be essentially like a NST with a center-tapped secondary. The input voltage and output voltage of the combined MOTs would be the same as a single MOT. Each MOT would have 60 volts on the primary. The possible advantage would be that the cores would not be in saturation and they would act more closely like a proper transformer.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 25, 2021, 06:34:11 PM
For 120 volt input, I've often wondered about the possibility of wiring the primaries in series and the secondaries in series for two 120 volt MOTs. The result would be essentially like a NST with a center-tapped secondary. The input voltage and output voltage of the combined MOTs would be the same as a single MOT. Each MOT would have 60 volts on the primary. The possible advantage would be that the cores would not be in saturation and they would act more closely like a proper transformer.

Sounds great!   If I understand correctly series wiring of MOTs with 120 volts seams to remove some of the innate MOT inefficiences due to MOT design (less copper/alum. in primary windings).

Assumption:  {I have 4 same size MOTs rated @ 1KW and can get 25 turns of 8 gauge wire on each MOT & get 1 Watt Volt per Turn}

If I wire each pair in series, then wire the series pair in parallel would I then get 50V output using 220V mains outlet.
And if I use the same wiring on the output, ignoring inefficiencies for the moment, what would be the AMPERAGE output?
Or alternatively, if the output is all wired in series, what would be the amperage output?
And, of course, is it practical to wire in above configurations?  If not, what would be the best way to wire 4 MOTs to get approx. a 3KW (55v@55amps) power supply? ???
 :(
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 26, 2021, 04:47:43 PM
Quote
Assumption:  {I have 4 same size MOTs rated @ 1KW and can get 25 turns of 8 gauge wire on each MOT & get 1 Watt per Turn}

I suspect you mean one VOLT per turn, not one watt.
Do you KNOW this or are you estimating?     In my own tests, I have been getting about 0.7 volts per turn.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 26, 2021, 11:46:52 PM
Quote
Assumption:  {I have 4 same size MOTs rated @ 1KW and can get 25 turns of 8 gauge wire on each MOT & get 1 Watt per Turn}

I suspect you mean one VOLT per turn, not one watt.
Do you KNOW this or are you estimating?     In my own tests, I have been getting about 0.7 volts per turn.

Just an estimate though based on other youtube videos the larger MOTs usaully get about 1V per turn.  Thanks for correcting - I need some more of that.  Still struggling with how best to wire 4 same size larger MOTs to get somewhere close to 3KW @ approximately 50 volts on a 220V outlet?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: klugesmith on March 27, 2021, 04:33:12 AM
Just an estimate though based on other youtube videos the larger MOTs usaully get about 1V per turn.  Thanks for correcting - I need some more of that.  Still struggling with how best to wire 4 same size larger MOTs to get somewhere close to 3KW @ approximately 50 volts on a 220V outlet.
Hope we are not cluttering Pete's thread with this discussion.

Obviously, no combination of series or parallel connections can magically increase the useful power per transformer core.   For max power per core, each primary needs to get the voltage it was designed for, and to run more than 15 minutes you need forced-air cooling.  (or oil immersion and active liquid cooling)

I agree that 1V RMS per turn needs a pretty big MOT (large core area passing through the coils).  You need even more core area if you are in 50-Hz land, or if you want to avoid the near-saturated operating point that's acceptable in fan-cooled MOT's but not in ordinary power transformers. https://www.electrical4u.com/core-of-transformer-and-design-of-transformer-core/

If you want to combine the power of four identical MOT's, and they have 220 V primary windings, the primaries all go in parallel on your 220V power source.  If the MOT's have 120 V primaries, two primaries in series can go on 220V power, but if it's split phase 240 as in North America, I'd recommend connecting both to the neutral wire at mid voltage - what is gained by letting the load circuit determine the voltage division?

For your 50V output goal, it seems like connecting four 12.5 V secondaries in series is best.  Then the current is automatically the same in all transformers.   Fewer turns to wind on each core.   You can easily operate at 3/4 or 1/2 or 1/4 voltage by switching out some of the transformers, and the switching can be done on primary side if "unused" primaries are short-circuited by the switch.

For efficiency it's important to use as much copper as you can fit. Divide core window area by number of turns & choose your wire accordingly.  If it were my project, I would try stranded copper building wire.  If the insulation occupies too much area, combine uninsulated stranded wire into a bundle of the right size and add a thin insulating sleeve (heat shrink or nylon braid?) over that. The resulting bundle can assume non-round cross section for better packing into the window.  Or do a multifilar winding with smaller wire, but the parallel secondary windings on same core need to be very closely matched.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 28, 2021, 07:06:39 AM
Quote
Hope we are not cluttering Pete's thread with this discussion.
I hope so too, am so slow - so sorry to waste everyone's time.  Will try to be as breif as I can.  Your specific example is good for uneducated like me.  For us in the states with 120V MOTs, could you please be even more specific on how to wire the 4 primaries? I understand and think all your other ideas are spot on - thank you!

If I understand, each primary leg of 220V line gets 120V, and with 2 MOTs in series on each leg, from our previous example, output/leg would be 25V@ 80amps, assuming 1kw/mot?? If so, then 4 series connected secondaries would get 50V @ 80A.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 28, 2021, 04:03:36 PM
Quote
and the switching can be done on primary side if "unused" primaries are short-circuited by the switch.

?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: klugesmith on March 28, 2021, 06:42:35 PM
No time to draw schematic.  The model I had in mind is four MOT's, with 12.5 volt 60 amp secondary windings permanently connected in series.    Primary windings independently switched to mains voltage. 
To operate at 3/4 of nominal secondary voltage, we can turn off one transformer, using a 2-way switch so its primary is shorted when not connected to power source.  Otherwise that transformer would greatly impede the output current, while developing 3x nominal voltage across its primary.

It would be conceptually simpler, maybe practically simpler, if the switching is done in secondary circuit with sufficiently big switches.  And what I talked about would not work if two 120-V primaries are connected in series across 220 volt power -- then you can not short one primary, it would put whole 220 V across the other primary.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on March 30, 2021, 06:33:54 AM
If the MOT's have 120 V primaries, two primaries in series can go on 220V power, but if it's split phase 240 as in North America, I'd recommend connecting both to the neutral wire at mid voltage - what is gained by letting the load circuit determine the voltage division?


 Thank you klugesmith - so each 120V leg of a 220V AC line would have 2 mots in series? How then would each of those 120V pairs be tied together?
Regardless of MOTs rating(120V or 220V), would output be the same - 50V@ 60A?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on April 02, 2021, 05:14:09 PM
Odd that the only actual 4-array MOT low-voltage applications I could find were done with 220V mots.  Anyhow, I think I can get 27 winds of 8 gauge (four 12 gauge in parallel) through each mot.  So now have a question on the secondary wiring as well.  If I get 48 volts on each series wound primary leg can I wire the secondary in parallel to get more amps?  Anyone care to tackle any or all my many questions?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 02, 2021, 05:15:58 PM
Quote
So now have a question on the secondary wiring as well.  If I get 48 volts on each primary leg can I wire the secondary in parallel to get more amps?
   I don't understand this question.   Are you suggesting to apply only 48 VAC to the primaries?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on April 02, 2021, 05:24:15 PM
I was thinking that primaries could be in some kind of series/parallel combination to maintain 120V to each mot and full power to all of them?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on April 02, 2021, 06:12:24 PM
I was thinking that primaries could be in some kind of series/parallel combination to maintain 120V to each mot and full power to all of them?

Apparently not.  :'(  How about linking two seperate 220V lines together somehow? 

If no way, then this makes 120V mots much less useful than 220v mots since hooking maximun of 2 Mots to 220V seams to be the maximum?


If so then, how about using 4 mots on 120V mains in some sort o series/parallel combo with a dedictated 30 amp circuit?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: klugesmith on April 02, 2021, 06:52:13 PM
Pictures might help here.

There is no power to be gained from putting less than nominal voltage on any MOT primary.  Half voltage on primary would not allow double current in primary, because of primary wire gauge and resistance.
There's no power to be gained from putting more than nominal voltage on any MOT primary, because core saturation will get you.

If you have 120V MOT's, the primary options appear to be

* 120V 30A circuit, all primaries in parallel.

* 2 MOT's in parallel on one 120V circuit, and 2 MOT's in parallel on a different 120V circuit.

* 240V center-tapped circuit, with 2 MOT's on each side (all four primaries having one end connected to neutral).

* 240V circuit without center tap (uncommon in North America): MOT's are in pairs with primaries in series.  For any pair with primaries in series, the secondary windings must be matched, and secondaries permanently connected in series or in parallel. That makes the 240V divide evenly between the two primaries.  You could wire small 120V indicator lamps across each primary, to show that the voltage is split properly.

* You might get satisfaction with three MOT's, primaries in series on 240V circuit for 80 volts each.  Then you are far below saturation, and could enjoy much less no-load current and core loss.  Limitation could be copper loss, at (say) 15 amps primary current for about 3600 watts total power.  The three secondaries need to be matched, and wired in series or parallel.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on April 02, 2021, 09:56:27 PM

* 240V center-tapped circuit, with 2 MOT's on each side (all four primaries having one end connected to neutral).


Pictures are what my thick skull needs and I finally found one but can not post due to copy-rights.  Maybe why using 120v mots with 220v is so uncommon is because the MOTs voltages need to be almost perfectly matched, not to mention this complex series/parallel array.  Unfortunately, my 220V does not have a neutral, just two hots and a ground, I think.(my 220v outlet does have a common neutral/ground wire.  So I will try to explain -- all four mots are connected in parallel, two on each hot leg with a series connection across the hot lines.  So I think voltages & amps would be approx. 50V & 60A on primary side?

If ok, now wondering how to wire the secondary to keep same VA power thinking is just tying all beginning primaries together and all ending primaries together?

UPDATE: My outlet is older style with a common neutral.

UPDATE: Here is a great example of the design I was looking for:
/>  & here is a schematic of an 8 mot welder:
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 02, 2021, 11:31:11 PM
hightemp1, what country do you live in?

In countries that have 220 volt service to the home with NO NEUTRAL, just a protective ground, Microwave Ovens for home use will have transformers with 220 primaries.
I don't  think I have ever seen  a 220 volt Microwave Oven in the USA.  At least not for home use.
So---- if you are finding references to 220 volt MOT primaries, you must be seeing videos or webpages produced in Europe, etc..
   I don't know what the "standard" household outlet uses for breaker rating, but, if it's high enough, you can just keep paralleling 220 volt transformers across the line.  Or, if you can, you could wire in extra circuits until you get to the overall wattage that you need.
However, since you have talked a lot about 120 volt MOT's, that leaves ME confused, since you say; "Unfortunately, my 220V does not have a neutral, just two hots and a ground, I think. "  It's the "I think" that worries me.
You might be better off simply "biting the bullet" and locating a commercially available power supply.
Heck, maybe it would be easier (and safer) to round up 5 good sized 12 volt car batteries and put them is series to get 65 to 60 volts.
You know, if you do go the MOT way, you still have to rectify the output and do something to smooth it so the induction heater isn't calling for current close to a zero crossover.
  Some of the folks on this forum would certainly be up to that challenge, but it's not as easy as it sounds.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on April 02, 2021, 11:52:12 PM
Here in the good old USA.
Yea, almost every MOT array of more than two on the internet is with foreign 220V rated MOTs.  Seams to be more complex and dangerous using 120v mots on 220V.
I think, just an expression - I know for a fact my 220V outlet has no neutral :-)  WRONG
Commercial PS is definately the way to go.  This is just something I always wanted to do, and do not recommend even attempting unless you are an engineer, or have lots of time and mots on hand and can have it verified by electrical engineers.
5 good batteries would be hard to find or terribly expensive, but always a reliable alternative.
Rectification is not just simple bridge rectifier properly rated with smoothing cap?
Also, using Dan's welding design with variable SCR power control, gives more options with power input/output. Adding more MOTs is doable too.
Did Dan Hartman's 120V MOT primary array on 220V schematic on page 51 make sense to you.....?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 03, 2021, 05:35:40 AM
Quote
I know for a fact my 220V outlet has no neutral :-)
Then there is something wrong with your wiring. OR-----
The receptacle you are looking at is wired according to an old or outdated standard.
If you check the voltage between either hot leg and "ground", I think you WILL see about 120 volts.

If you plan to proceed with this experiment, I strongly suggest that you somehow start with a properly wired outlet.
  I have to admit that, back in 1985, I wired a single purpose circuit where the motor required only 220, in the way you mentioned.
But when I rewired it recently, knowing that I'd need both 220 and 120. I ran  a new, separate neutral so the green wire is simply the protective ground.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on April 03, 2021, 10:21:02 AM
Found this quote from internut so must be true.
220 does NOT need a neutral. Some newer appliances require it for parts of the system that run on 110 volts. Almost all installations more than a few years old do not have a neutral
Please read next two posts for the real deal...

I HAVE NO ELECTRICAL BACKGROUND SO PLEASE IGNORE WHATEVER I MAY BE THINKING !!!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 03, 2021, 05:33:07 PM
Quote
“ 220 does NOT need a neutral. Some newer appliances require it for parts of the system that run on 110 volts. Almost all installations more than a few years old do not have a neutral


Sorry, but this is misleading and, in some places, downright wrong .

The following only relates to the USA----
Yes, an appliance that doesn't use any 120 technically does not need a neutral.
The "some newer appliances--- " sentence IS wrong.
Should read " ANY "220" appliance that uses 120 volts requires a neutral."  (220 volt Clothes dryers have been using 120 volts to run the timer and motor for decades.
Regarding the " Almost all installations----" sentence:
How can he even say that?  It simply assumes that 220 appliances NEVER use 120.
Check the NEC code book and any state and local code books  if you don't trust me.
There IS a bit of ambiguity on this subject, but in general, it's only certain existing installations where the neutral is not required.

Note: my comments are simply made in the spirit of keeping readers safe and alive.  I don't want to get into a "pi----ing" contest.
This will be my last post on this 220 wiring topic.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: klugesmith on April 04, 2021, 02:19:16 AM
This is in support of Petes' statement.  In old 220V outlets with 3 contacts, in NEMA land, the 3rd one is "neutral", not "ground".   Unlike green wire ground, it's intended to conduct current during normal operation.   ( In a recent twist, light switch circuits in new construction must have neutral as well as ground at the switch.  Smart switches used to be allowed to use green wire to return the half mA of control current. In even older buildings, only hot and switched-hot wires run to the switch box.)

In my house from 1950's, electric clothes dryer has a 3-prong plug that fits a matching receptacle.  NEMA 10-30, long deprecated but still perfectly legal.

The center pin is connected to white wire from the panel. It's the Neutral, technically the Grounded Conductor.   Referred to as Center Tap in my previous post here.   Used for 120V timer/controller, and also serves to ground the dryer's metal frame and skin for safety.  There is no separate pin for green wire, technically Equipment Grounding Conductor.

I could easily and legally install a receptacle for 4-prong plug, NEMA 14-30.   Equipment grounding connection would be made through the steel conduit in which the other wires run.  I measured its resistance back to panel, and it's substantially lower than that of the white wire.

For 240V hobby experiments, I made a 4-wire 10-AWG extension cord.  The plug end has 3-prong dryer plug + green wire pigtail with a ring lug, attached to a stud on grounded box and secured with a wingnut. 

By the way, in case this gives anyone ideas, the "grounded" and "grounding" resistances were measured from the far end of extension cord kluge.  With a low voltage transformer I sent 20 amps through white and green wires, round trip to grounded bus bar at the panel.  The far-end junction voltage was monitored through one of the hot wires, while the circuit was switched off.

Not really the right thread for this, but the work was part of an exercise to get 277 volt power at home.  Voltage out of the wall was typically about 251 V, easily stepped up beyond 277 with a dual variac.  In experiments with a 315 VA control transformer configured to Boost that, I got 321 V with 8.75 A one day (2800 W).  Load being three 120V lamps in series: two 1000's and a 1500.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on April 06, 2021, 09:42:33 PM
My sincere apologies to you Pete, and entire forum. Thank you for emphasizing safety first.  Thanks to this thread, I believe I now have the knowledge to proceed very cautiously building a 3kw MOT power supply. Any further questions would be in another thread or forum.  As always, thank you Pete, klugesmith and company for sharing your professional knowledge. Lots of great advice, excepting mine, gleaned from the masters.

PS -- I was of course dead wrong.  My 220v outlet is an old style Nema 10-50 having a neutral that also served as ground, assuming it was properly installed 60 years ago by whoever built the house.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on April 07, 2021, 09:18:58 AM
I say only 400V 3N~ @50Hz :)
 
I did actually used two phase + neutral with  2 identical 10A variacs to create a 0-400V variable AC source for testing a equipment ( an electrical heater at 4kW) . But gettin your head around 3phases takes some time.

Back on track :

I was thinking about the induction heater , is there a good way to asess the heat loss in the resonant caps ? Or is it just to bop until they pop?
I have a couple of handfulls of Wima FKP, and datasheets have diagrams with maximum voltage applied for difference capacitances and frequency.
My experience is that these datasheets are very conservative.

That should be translatable to rms current based on capacitive impendance at specific frequency? (Urms /Xc = Irms)

And then parallell enough for a good total capacitance and sufficient current handling?  Provided a proper layout with good current sharing of course.

Or am I missing something here?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on April 08, 2021, 11:10:51 AM
I reply myself.

Had a look at the datasheet and it has been updated this february and now not only diagrams of maximum voltage applied vs frequency but also diagram of maximum current vs frequency is listed, for different voltage ratings and capacitances .

The constraining factor for the given data is  hot spot temperature rise internally of the capacitor <15 degrees (Centigrade/Kelvin).
typically a 0,47 uF FKP1 cap can take about 8amps of current maximum under this condition, but it variates a bit with the voltage rating.
Higher voltage rating means larger size in general due to thicker dielectric and therefore higher current rating.

Fex for a 68nF cap:
630VDC   = 2,5 A (27,5mm pin spacing)
1000VDC = 3A (37,5 mm pin spacing)
2000VDC = 6A (37,5 mm pin spacing)


The reason for my interest is that the caps represent a much higher cost than the transistors. A 0,1uF 630VDC FKP1 cap costs as much as a transistor, but I also realise that there are many more ways for the transistors to perish (ie release of magic smoke) due to erraneous behaviour of a self-oscillating induction heater.

Hot spot temperature rise is proportional to power loss so by keeping surface temperaure down , fex with forced convective cooling it is possible to really push the caps to the limit, aswell as the semiconductor switches.

The copper tubing is less prone to temperature degradation but resistive heating will lead to higher resistance and more loss, and eventually power limiting, depending a bit on the actual design. Contacting the tubing is probably the limit here. Or what is the general opinion?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 08, 2021, 05:33:10 PM
Quote
The copper tubing is less prone to temperature degradation but resistive heating will lead to higher resistance and more loss, and eventually power limiting, depending a bit on the actual design. Contacting the tubing is probably the limit here. Or what is the general opinion?

In my opinion, you should always be water cooling the  copper tubing, so "resistive heating" should not become an issue, since the tubing never even gets close to 100°C (in my experience).

Regarding capacitors:
The zvs induction heaters that I have all use this capacitor:

BM brand.
0.33 mfd  630VAC, ((1200VDC)
Type MKP

The 1000 watt units have 6 of them in parallel
The 1800 watt units have 9 of them in parallel
The 2500 watt units have 12 of them in parallel
So far, I have never had one of them fail.

I always force air-cool the induction heater boards, pushing the air downwards towards the circuit components.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on April 10, 2021, 03:07:15 PM
Incidentally, one could even operate with no resonant capacitors at all, just brute force inverter capacity.  This maximizes transistor use, putting all reactive power through them -- but if they're cheap, maybe that's not such a big deal.  The catch is, you still need to handle that current somewhere, somehow; in this case it comes back to the DC link / inverter bypass caps, which need to handle all that ripple current.  The only consolation is that the ripple voltage is small, so much lossier, bulkier types can be used -- DC link film, "filmlytic", the kind with the bolt studs, and even polyester (MKS) and electrolytic are suitable, as long as you have enough of them.  I think it's a poor tradeoff overall, that you'll end up spending more here (in cost or size) than the resonant version.

The one upside is that frequency can be freely controlled, which makes for interesting possibilities in special applications (like variable depth case hardening).  Of course, this requires industrial scale machines to be practical (100s kW, to heat up the surface fast enough while the core of the part remains cool).

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 10, 2021, 10:47:31 PM
If you do decide to use the same capacitor that is on the units that I have, they cost about USD$1.00 each.  Here's one example:
Ebay Item  174496593032

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: rikkitikkitavi on April 11, 2021, 09:25:43 PM
Those I recocgnise, they are typically used in a half bridge simple induction cooking stove plate. Due to that topology they are subjected to high voltage stresses. 

Considering that only a few , 2-4 or so is used in each cooking plate with a output of 1+kW parallelling a bunch explains why they can take a lot of power.
Normally it is not thte caps that fail in those cooking tops, but the IGBTs.
Have become hero of the neighbourhood  as I fix expensive  4 plate stoves for a handful of dollars , compared  to  what the dealers charge for a replacement module (no warranty give n though as other parts might have taken a beating, but I always rely on the circuit breakers for that...)

I have seen these black boxes, a  grey box and some round white caps with axial leads (probably more inductive )

Unfortunately no datasheet with specs easiliy avialable but your  table is good data to be used as an indication

Thanks for the input!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on April 19, 2021, 09:35:16 PM
New work coil related video----
 Finally trying to actually forge and then harden something using the 2500 watt ZVS induction heater as the sole heat source.
-I went through a LENGTHY process of (mostly) empirical coil "design" that may be of interest to some:

/>
  You WILL need to have a fresh cup of coffee or your favorite beer to watch this one.

Pete Stanaitis
---------------

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on April 24, 2021, 05:54:59 AM
Cold quart of milk is my perferred poison. 

That is a great way to heat those strikers. 
Maybe even cover with a nice warm kaowool blanket.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: hightemp1 on April 24, 2021, 09:19:58 AM
This blacksmith architect toasts three ZVS 2500W heaters before learning about AC initial voltage surges from transformers. 
At 6:35 in video - reminds me of superman bending steel - very cool  8) 
/>Now, the powerful small coil may be ramping up the amps above 2.5kw?  Judging by time it took bend metal rod -maybe running at or near 3kw - doubt if over 3kw, else PS's over-current of approx. 60 amps would have kicked in?
Pete and company enlighten him on the finer points of induction heating. Maybe a FAQ page at the start of this thread would be a good idea, esp. since I have diluted much of the meat so badly.

Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 08, 2021, 08:14:40 PM
I finally watched the video suggested by hightemp1 and I did make a comment.  I also emailed him directly with a link to some of my pages, etc..

TOPIC of this post:
Attempting to protect my DC to DC SSR's from unintentional overcurrent induction heater events
------------------------------------------------------------------------------------------------------------
I thought I might be able to put a fast acting automotive or electronics type of fuse in series with the DC power to protect the SSR.  So far I have learned that simple cartridge and blade type fuses aren't all that simple after all.
Terms like "standard", "fast acting", "ultra fast acting", and "slo-blow", etc. get in the way.
My first test was to insert a 40 automotive type blade fuse in DC circuit.  It didn't blow at 50 amps, at least not in the several minutes that I tested it.
After reading some more, it APPEARS that many fuses of these types don't blow QUICKLY until, maybe twice their rating.
I just tested a 15 amp fuse that blew at 30 amps and
a 20 amp fuse that blew at about 45 amps.
Who knows if that's even fast enough to save a Mosfet (or whatever) inside a DC-DC SSR!
I will do some more testing, but I doubt that I will be able to arrive at a fuse size or type that will be able to withstand 50 amps for at least 20 minutes a time and still be able to blow at, let's say, 55 amps, reliably.  And, even then, will it blow fast enough to save the SSR?
Plan "B"  will be to read the voltage drop across the current shunt that powers my analog ammeter with an Arduino  and simply shut off the SSR in case of a current surge beyond a programmed value.  I'd require a manual reset of the Arduino after such an event.
"Beats sitting in a bar", says I.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: T3sl4co1l on June 10, 2021, 05:14:41 AM
Or rather, they are exactly as simple as they are, nothing more -- a coupon of metal that heats up under load, and maybe melts at some point.  The breaking current varies with time delay (and construction, hence the various types) and can easily be double the rating.

Fuses only protect the wiring, and function best when large fault currents are available to clear them -- a lead-acid battery or a mains circuit, for example.  A switching power supply with current limiting, might happily sit there simmering into a short circuit, until it blows up (maybe a good reason to choose "hiccup" type overcurrent limiting supplies -- the hiccup should be pretty safe over the long term).

Fuses are especially poorly suited to protecting semiconductors.  A "semiconductor" fuse is so named because it is just capable of protecting the most robust kinds of semiconductors -- diodes and SCRs.  All other types (MOSFETs, IGBTs, etc.) will have long since failed and turned to plasma, by the time the fuse opens.  (Which can still be of some value, for example reducing the hazard of shrapnel and arc flash emitted from an industrial IGBT module.)

For that, one needs a current limiting or protected switch -- using semiconductors to protect semiconductors, with some clever circuitry to protect itself in the process.

I've made a few such units myself, though nothing rated for quite this current.  One is unidirectional 40V 8A and extremely fast (so fast it's difficult to use!), another is unidirectional 40V 10-12A and fast (limiting current within 20us and faulting in 0.3 to 7ms), and the last is bidirectional +/-30V 20A and slow (permitting a fault condition for up to 150ms; it has a special design to deal with the heat dissipation).

I have made a prototype scaled-up version good for +/-30V 100A, but it's not behaving quite how I expected, and I don't currently have plans to make a proper PCB version.  If there's enough interest, I could develop it further.  (e.g., how many people would be willing to pool funding for that; or what kind of unit cost would you be willing to pay for such a device.)

BTW since these are electronic fuses, they also have on-off buttons and can be interfaced to controls, so they can replace the SSR too.

As for the proposed method, that can work -- don't even need an Arduino, just a logic gate or two and comparator.  The trick is in the details, turning on and off.

Tim
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on August 06, 2021, 11:34:11 PM
Subject:  Protecting DC-DC SSR's from operator error when inserting work piece too far.

I am now working on  an Arduino  controlled device to sense current from the DC power supply and shutting off the power if I see a potential overcurrent (which I write into the Arduino code.

Progress report #1,  can I get enough voltage across the existing 75 mv shunt to do what I need to do?
/>Answer: No, but---


Progress report #2.  What if I add a single rail opamp?
/>Answer: I think so, but not quite there yet.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on September 13, 2021, 01:10:07 AM
DC-DC SSR Protection Device finally working correctly!!!
  Well, it has been a long time in getting there, but I think I finally have a working device.

See:
/>
For those of you who might be looking at my solution through a microscope <G>,  I could still shorten the Arduino sketch significantly by eliminating all LCD code  if I need the system to act even faster.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on September 15, 2021, 12:10:23 AM
Here' a "follow-on video to the one I published yesterday.
I am simply catching up and cleaning up after all the time I spent making my SSR protection device work correctly.
In this one I tackle 3 subjects:
- A block diagram of my heavily expanded 2500 ZVS heater "system", with pictures to match the diagram,
-A simple spreadsheet that details current draw for a number of non ferrous materials.
-and, a little bit about repairing one of my recently failed DC to DC SSR:
/>
You guys should feel lucky that I won't bother you with ANY of several videos that I made DURING the process; those that were so boring and mistake-ridden that I'd be embarrassed if the ever saw the light of day.

Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on March 13, 2022, 07:16:44 PM
Mosfet Gate Drive Concerns for ZVS Induction Heaters and other applications:
---------------------------
Mosfet gate driver concerns-
or-
Achilles heel for ZVS gate drive circuits
or
Running Mosfets with gate voltages too far below maximum

My Conclusion:
We must carefully design gate drive circuits to consider linear mode operation, even when running in "switched mode"!

Discussion:
Over the past several years, I have been experimenting with small ZVS induction heaters (1000 watts to 2500 watt models).  Early on, I had more than my share of circuit failures mostly due to overheating of the Mosfets.  ( I do NOT have these problems any more)
More recently, I have been using/attempting to use DC to DC SSR's to switch power to induction heaters.   It appears to me that all too many of these devices have poorly designed gate drive circuits.  I have even made a couple bad choices myself when attempting to repair a couple of them.
Much of this is documented on my relevant webpages and in my youtube videos.
  In addition to the above, I see, from time to time, other folks building  switching mode Mosfet circuits without, in my opinion, proper attention the detail regarding the gate drive voltage profiles, particularly when the Mosfets are momentarily running in linear mode.
Sometimes, I even have trouble understanding the gate drive voltage specs for a particular Mosfet from its datasheet.

Recommendation:
So--- I offer this link as a useful tool in understanding the care that must be taken in gate drive design and implementation: 

See:
https://www.infineon.com/dgdl/Infineon-ApplicationNote_Linear_Mode_Operation_Safe_Operation_Diagram_MOSFETs-AN-v01_00-EN.pdf?fileId=db3a30433e30e4bf013e3646e9381200

Pages 4, 5 and 6 are the most relevant
.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 19, 2022, 06:16:42 AM
Today I summarized my thoughts on what these 1000 watt to about 2500 or 3000 watt ZVS Induction Heaters can do well and can't do so well.
The video is here:
/>
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on July 31, 2022, 06:36:38 PM
Power Transfer OUTSIDE the work coil:
  In this short video, I am addressing a viewer's specific question about power transfer OUTSIDE of the work coil.
His question was rather general in nature, so I simply ran a couple of tests.  I could do a lot more testing if I had more information about his exact use:

/>
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on August 08, 2022, 09:27:10 PM
I just did a few more tests of workpieces that are of  larger  Diameters than the diameter of the  work coil, using my 1000 watt ZVS induction heater.
This is Part 2:

/>
Pete Stanaitis
---------------
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: Solhi on December 07, 2022, 05:49:44 AM
Has anybody in to his thread ever measured the magnetic field/flux of the 3.5 KW induction cooker under load? If yes, wat is the result?

Edit: if no, can somebody try to take a measurement? Maybe one with and without load
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: tarak on May 07, 2023, 05:54:18 PM
hi is any one still following  and posting here
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on May 07, 2023, 06:25:28 PM
Thank you for asking!!!
I am the original poster for this thread.
   It still gets between 50 and 100 views per day.
I consider it to be a resource to those just getting into induction heating with the 1000 to 2500 watt or so induction heaters sold all over the world.
I know it takes a while to locate specifics by scrolling through all the information here.
So, Here are a couple of links links that take you directly to my many  youtube videos and webpages on the subject;
https://www.youtube.com/playlist?list=PLVxWen9M87dBhvInCQ-3pELWdxCM3XH4I
https://spaco.org/Blacksmithing/ZVSInductionHeater/1000WattZVSInductionHeaterNotes.htm
I am certain that I have posted them previously here, but this is a quicker way to get started.
I don't post testing and communications here where the folks who write to me asking for consultation on projects that have proprietary concerns.
But I do have to say that, over the last few years I have seen these small induction heaters used on some pretty interesting  "out of the box thinking" projects. 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: thewindowguy on June 21, 2023, 08:46:45 PM
Hi dudes and dudettes!

New member and first post. I am trying to build a induction heater setup. I have read this entire thread and seen all of Pete and Mads very informative youtube clips on zvs induction heaters.

This post will be a test of my english and ability to explain what I am trying to say in a good enough way.

My goal with the IH is heat steel to around 450-500'c. I work with restoring old windows. In my work I also have to restore window fittings. First tumbling in acid to remove paint and rust. After that heating and quenching in oil. The current process is slow. So I think and hope that heating with IH would be a more efficient way.

I have ordered the 2500w kit from Amazon. (https://www.amazon.se/Induktionsv%C3%A4rmare-G%C3%B6r-det-sj%C3%A4lv-h%C3%B6gfrekvent-Induktionsuppv%C3%A4rmningsmaskinmodul-V%C3%A4rmekretskortsmodul-Flyback-drivrutin/dp/B0C62CG6L1/) sorry it is in swedish, but I think the pictures explains the parts.

First question:
One thing I have come across a lot is this warning: "No matter in any case, no-load is allowed (no-load refers to the output port without any load). Once the power is turned on, the board will emit high-frequency noise, which will cause the MOS tube to be completely damaged and cannot be repaired."

I interpreter that as "Machine should not be even idle without metal in the coil". However, I have seen many clips when they talk about the energy consumption of 4-8 amps in standby mode. But nowhere have a seen any discussion on the topic.

How serious of a problem is it? Is it depending on idle time?

The reason of interest is that the parts I wanna heat are quite small and probably quite quick to heat. If I have to turn everything of between heating, quenching and hanging the part to dry and ten start it up again, I will loose some efficiency.

Second question:
For the same reason as first question I would like to use a footpedal to engage heating. Or maybe have I misunderstood it and there is nog engage exept turning on the machine?

 Thanks in advance
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on June 21, 2023, 09:57:09 PM
Quote
"No matter in any case, no-load is allowed (no-load refers to the output port without any load).----
They mean that you must not apply power to the ZVS circuit if you do not have a proper work coil connected.
As you have already observed, there is NO PROBLEM with having the unit powered up with no work piece in the work coil.

Quote
Second question:
For the same reason as first question I would like to use a footpedal to engage heating. Or maybe have I misunderstood it and there is nog engage exept turning on the machine?
The "footpedal" can be used to trigger an appropriately sized mechanical relay or an appropriately sized DC Solid State Relay.
I assume that you will already have your work piece in the work coil. I have tested this case and I see no problem with it.
---As long as you already know that the current that will be drawn by that work piece will not exceed the capabilities of the relay or the ZVS board.
In this case, of course, you will need to have your DC power supply "ON" and only use the relay to send the DC  power to the ZVS board.

If any of this is unclear, please feel free to ask for more information.  !!! Your English is very good!!!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: davekni on June 22, 2023, 05:47:54 AM
Quote
I interpreter that as "Machine should not be even idle without metal in the coil". However, I have seen many clips when they talk about the energy consumption of 4-8 amps in standby mode. But nowhere have a seen any discussion on the topic.
I'm guessing they mean with no coil attached.  Coil without load inside it should be fine.  Unless that causes startup transient issues, which are common, but I haven't heard being related to load inside coil.

Quote
Second question:
For the same reason as first question I would like to use a footpedal to engage heating. Or maybe have I misunderstood it and there is nog engage exept turning on the machine?
If you have a high-current foot pedal switch, it could be in series with 48Vdc supply connection to induction circuit.  Or high-current relay connected to foot pedal as Pete mentioned.  If in series with line voltage to 48Vdc supply, may cause issues due to slow rise of 48Vdc.  There are fixes for this, such as a separate low-power supply feeding gate pull-up resistors that is always on.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: thewindowguy on July 06, 2023, 04:07:39 PM
Thanks for your answers guys! 👌

I just got the delivery a couple of days ago. So the last few days I have been tweaking, soldering and building. Today I just tried it out. It works, I can not say if it work as it should since I never tried one before.

Your instructions on Spaco was really useful. Now a first setup is done and I will modify it more and make it cleaner in the future. I will also add the de-attachable fittings you have so I can change coils.

I was a pain to make Mads modifications so be able to mount the coil on the under side of the circuit. The screws to the mount where soldered with heaps of solder. My regular weller station where not up to the task of heating enough to melt all solder at once. I sucked it away and later on used my stained glass solder station and it made all the difference.

At one spot I feel like it is really small separation from the coil mount to the next path. As you can see in the attach picture that ends with 104.

My amp meter says the voltage is 53,5v which makes me thinking that either the server driver is putting out to much or the amp meter is not calibrated. I have not checked it out with multimeter yet. But is will probably show me which of them is right.

The power supply fore the fans was a bit weak to drive fans, pump and 3 temp meters. It becomes a bit hot and I have to find a new that's up for the task.

Otherwise it work. It makes square tube red hot in a few seconds (actually 10+ or so). So I am quite happy with the setup. I will try it for the real tasks in a week or so.

Thanks again guys!    🌹
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on July 06, 2023, 04:24:38 PM
Thank you for the feedback.
  Looks like you are off to a really good start!
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: thewindowguy on July 13, 2023, 01:28:40 PM
I could not, even if my life depend on it, find fittings for anything thing smaller than 10mm.

It seams like the only business that uses 6mm (1/4 inch) are hydraulics. And most of it is not corrosion protected. I ended up with aluminum fittings used for racing.

After returning home it dawn on me, will it work with electricity. I know that aluminum is conductive, but I have no idea how the finish, mixing with copper and dc will effect the setup.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on July 13, 2023, 05:30:45 PM

Refrigeration and air conditioning companies seem to use these fittings.
I convert my systems to 1/4" OD copper tubing and fittings for that exact reason.  (I am in the USA)
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: tarak on October 27, 2023, 07:29:26 AM
hi again  just checking in, these heaters bought from china seem to have a lot of problems, I had problems and also found the china chip for the fans was wrong I did replace it with the higher voltage one still had problems, so I started testing with some heavy DC batteries starting with 24vDC. however could not get the unit to oscillate.
the main problem from the the unit was delivered in England and was dropped over an 8ft gate onto concrete,  I told them what happened, and like one of you other members had the same off 5pound so after some real hard words with amazon and the 'so called maker' I got told by amazon they would refund my money all of it, the 'maker' told me to keep the unit. as a tech I inspected it but there was cracked board. These boards are double layers no easy to trace and so no circuit. bottom line we stripped it saved what we could and we started to put a circuit together from what could be seen. still had a problem with it no oscillating.  the I had to return to Australia.
I suggested we look at this for anther way to use it for the shed heating, using the water radiators that they use there. I decided to stop relying of the 'self' oscillating and drive it with an SG3525. so thats where I am at the moment, me in Australia and my brother in England, we needed the shed warmed up as we were working on a new project for home power generation. Its dam cold in the winter there so the 'shed heater' project was started. there was on guy  actually use the 'radiator' clip on unit so I thought why not. winter coming up again over there so I have to try and get it to work. lets see if any one else is interested 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on October 27, 2023, 04:36:56 PM
Sorry to hear about all of your troubles.
From what you have written, it appears that you did not get a replacement board from the maker.  So you attempted to build the circuit from scratch using that mosfet driver.   The actual circuit that those heaters usually use is  shown about 1/3 of the way down this page,if you are interested:
https://spaco.org/Blacksmithing/ZVSInductionHeater/1000WattZVSInductionHeaterNotes.htm


I don't understand why you would want to go to all the trouble of using an induction heater to heat a shed (by heating water?) or to heat any space at all.   A watt is a watt.  If electricity is your only power source, why not simply use a resistive element or an off-the-shelf electric space heater instead?
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: tarak on November 24, 2023, 04:04:48 AM
hi,   ' I don't understand why you would want to go to all the trouble of using an induction heater to heat a shed (by heating water?) or to heat any space at all.   A watt is a watt.  If electricity is your only power source, why not simply use a resistive element or an off-the-shelf electric space heater instead?'
 to help you understand, most houses in England are heated by a closed circuit hot water radiators. my brother had a spare one, so we looked at using that to heat the shed as a test unit.  the water 'boilers' are what is an instant flow though gas heated system. two inputs, one for the heating and one for hot water. England want to force people to go to a heat pump, shit house, or electric. There is a replacement that you can get that uses induction heat circuit. so we though why not build one for the shed. The problem as I have said the unit was damaged, so I took it apart saved what I could, the circuit show above in the part, was ok but the 'mosfets there were four, not two as shown. so I did for a start use just the two, but this royer circuit relies on one switching on first. This was not happening, so both on at a time, but I had a ammeter in the output, so on with on, with finger on the off switch, so if the current went quickly to much we turned off. hence the using a two output to switch with a dead time. If you try to use the inverter circuits most use the 12v to drive a transformer, push pull to get a high voltage output. the problem with most driver circuits, they do not have a 12v output to drive the gates, usually only five volts. I also have gone over to use IGBT's. the SG3525 it one but again only 5v out. I did did another switching circuit, this drove the IGBT's this was for another job, the problem is with having a driver to drive the IGBT's there become a problem switching at high speed, the time lag from one to another all gets out of sync.  so just use electric heaters?? have you seen the price of the gas and power there? induction is cheaper to run. I think. So the testing goes on.

 
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: petespaco on November 24, 2023, 06:40:13 AM
Quote
induction is cheaper to run. I think.
No.  I tell you again:  A watt is a watt.  You will gain nothing for all your trouble.  In fact, if you intend to use the device to heat water, you will waste power in your power supply and in circuit losses that wiil NOT heat your water.
 
If your goal is to heat the water in an existing system, you can use inline electric water heaters like this one:
https://www.fleetfarm.com/detail/kat-s-circulating-tank-heater/0000000047511?Ntt=005101472&gad_source=1
I am certain that there are MANY similar heaters like this available all around the world, and they won't require any ancillary equipment or electronics.
Title: Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
Post by: klugesmith on November 24, 2023, 11:57:26 PM
Came to agree with Pete: a watt is a watt.   
For heating a shed, plain old resistance heater will be as efficient as any induction heater.

Heat pumps were mentioned, and they _can_ heat water or room air more efficiently than electrical resistance.
Hence promotion of heat pumps by governments, even over heaters that burn natural gas.

[off topic] Gas is much too cheap, because of free-market policies that don't adequately account for depletion of national reserves and costs of climate change.  100 years from now, it may seem bewildering that reserves built up over 100s of millions of years were consumed in a century or two, mostly to be burned for heat!
SimplePortal 2.3.6 © 2008-2014, SimplePortal