Author Topic: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"  (Read 16950 times)

Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile
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.  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?

« Last Edit: May 26, 2019, 07:45:45 PM by hightemp1 »

Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile

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.

« Last Edit: May 30, 2019, 02:35:21 PM by hightemp1 »

Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile
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?





« Last Edit: May 30, 2019, 02:41:40 PM by hightemp1 »

Offline petespaco

  • High Voltage Technician
  • ***
  • Posts: 127
  • Karma: +8/-0
    • View Profile
    • French Creek Valley Activities
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.



Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile
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?
« Last Edit: May 27, 2019, 12:11:51 AM by hightemp1 »

Offline T3sl4co1l

  • High Voltage Experimenter
  • **
  • Posts: 79
  • Karma: +8/-0
    • View Profile
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

Offline petespaco

  • High Voltage Technician
  • ***
  • Posts: 127
  • Karma: +8/-0
    • View Profile
    • French Creek Valley Activities
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.

Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile
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.
« Last Edit: May 30, 2019, 04:34:05 PM by hightemp1 »

Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile

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?
« Last Edit: May 30, 2019, 07:36:58 PM by hightemp1 »

Offline T3sl4co1l

  • High Voltage Experimenter
  • **
  • Posts: 79
  • Karma: +8/-0
    • View Profile
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

Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile
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.


« Last Edit: May 30, 2019, 07:40:09 PM by hightemp1 »

Offline Mads Barnkob

  • Administrator
  • High Voltage Expert
  • *****
  • Posts: 1142
  • Karma: +17/-0
  • Denmark
    • View Profile
    • Kaizer Power Electronics
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 :(
http://www.kaizerpowerelectronics.dk - Tesla coils, high voltage, pulse power, audio and general electronics

Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile
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.  :(
« Last Edit: May 30, 2019, 08:31:24 PM by hightemp1 »

Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile
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.

« Last Edit: May 30, 2019, 11:15:36 PM by hightemp1 »

Offline petespaco

  • High Voltage Technician
  • ***
  • Posts: 127
  • Karma: +8/-0
    • View Profile
    • French Creek Valley Activities
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.

Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile
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.
« Last Edit: May 31, 2019, 02:22:57 PM by hightemp1 »

Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile
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
« Last Edit: May 31, 2019, 03:39:14 PM by hightemp1 »

Offline petespaco

  • High Voltage Technician
  • ***
  • Posts: 127
  • Karma: +8/-0
    • View Profile
    • French Creek Valley Activities
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.

Offline hightemp1

  • High Voltage Experimenter
  • **
  • Posts: 56
  • Karma: +0/-0
    • View Profile
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.   
« Last Edit: June 02, 2019, 12:13:50 AM by hightemp1 »

Offline badpeter

  • High Voltage Enthusiast
  • *
  • Posts: 27
  • Karma: +1/-0
    • View Profile
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

High Voltage Forum


 


* Recent Topics and Posts

post Re: making a useable induction forge.
[Transformer (iron core)]
T3sl4co1l
Today at 05:07:22 PM
post making a useable induction forge.
[Transformer (iron core)]
rustedone
Today at 04:20:15 PM
post Re: 220ac to 440dc 5a boost convertor
[Voltage Multipliers]
rustedone
Today at 04:03:58 PM
post Analog Panel Meter Shunt Calculation and Custom Scale Design
[Beginners]
Mads Barnkob
Today at 07:56:20 AM
post Re: Mysterious Chinese HV capacitors
[Capacitor banks]
MRMILSTAR
Today at 05:40:55 AM
post Re: Mysterious Chinese HV capacitors
[Capacitor banks]
davekni
Today at 03:33:52 AM
post WTS [US]: pole transformer, SRSG, PTs, NSTs, capacitors, variacs, [MORE]
[Sell / Buy / Trade]
acobaugh
October 15, 2019, 09:36:55 PM
post Mysterious Chinese HV capacitors
[Capacitor banks]
MRMILSTAR
October 15, 2019, 09:12:09 PM
post Re: DRSSTC with litz-wire primary and 40 x TO247 IGBT H-Bridge
[Dual Resonant Solid State Tesla coils]
Mads Barnkob
October 15, 2019, 12:38:24 PM
post Re: How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
davekni
October 14, 2019, 10:02:07 PM
post Re: DRSSTC with litz-wire primary and 40 x TO247 IGBT H-Bridge
[Dual Resonant Solid State Tesla coils]
Weston
October 14, 2019, 08:46:54 PM
post Re: How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
MRMILSTAR
October 14, 2019, 08:28:19 PM
post Re: How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
davekni
October 14, 2019, 07:27:22 PM
post Re: DRSSTC with litz-wire primary and 40 x TO247 IGBT H-Bridge
[Dual Resonant Solid State Tesla coils]
davekni
October 14, 2019, 07:01:34 PM
post Re: How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
MRMILSTAR
October 14, 2019, 04:23:56 PM
post Re: How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
klugesmith
October 14, 2019, 07:27:49 AM
post Re: DRSSTC with litz-wire primary and 40 x TO247 IGBT H-Bridge
[Dual Resonant Solid State Tesla coils]
T3sl4co1l
October 14, 2019, 07:10:39 AM
post Re: How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
MRMILSTAR
October 14, 2019, 05:43:57 AM
post Re: so far video
[Spark gap Tesla coils]
davekni
October 14, 2019, 05:40:27 AM
post Re: Drsstc 4 (new direction)
[Dual Resonant Solid State Tesla coils]
davekni
October 14, 2019, 05:23:48 AM
post Re: Drsstc 4 (new direction)
[Dual Resonant Solid State Tesla coils]
oneKone
October 14, 2019, 04:37:49 AM
post Re: How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
davekni
October 14, 2019, 03:47:03 AM
post Re: How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
klugesmith
October 13, 2019, 11:05:25 PM
post Re: How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
davekni
October 13, 2019, 09:29:25 PM
post Re: How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
MRMILSTAR
October 13, 2019, 04:43:47 PM
post Re: My first DRSSTC on bricks
[Dual Resonant Solid State Tesla coils]
davekni
October 13, 2019, 07:04:20 AM
post Re: How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
davekni
October 13, 2019, 07:00:57 AM
post Re: DRSSTC with litz-wire primary and 40 x TO247 IGBT H-Bridge
[Dual Resonant Solid State Tesla coils]
davekni
October 13, 2019, 06:36:13 AM
post Re: DRSSTC with litz-wire primary and 40 x TO247 IGBT H-Bridge
[Dual Resonant Solid State Tesla coils]
Weston
October 13, 2019, 12:41:07 AM
post Re: My first DRSSTC on bricks
[Dual Resonant Solid State Tesla coils]
Hydron
October 13, 2019, 12:17:39 AM
post Re: My first DRSSTC on bricks
[Dual Resonant Solid State Tesla coils]
Laci
October 12, 2019, 11:21:39 PM
post How to measure voltage output of AC flyback transformer
[Transformer (ferrite core)]
MRMILSTAR
October 12, 2019, 08:37:36 PM
post Re: DRSSTC with litz-wire primary and 40 x TO247 IGBT H-Bridge
[Dual Resonant Solid State Tesla coils]
davekni
October 12, 2019, 06:57:55 PM
post Re: so far video
[Spark gap Tesla coils]
thedoc298
October 12, 2019, 05:28:01 PM
post Re: 220ac to 440dc 5a boost convertor
[Voltage Multipliers]
rustedone
October 12, 2019, 04:58:28 PM
post Re: My first DRSSTC on bricks
[Dual Resonant Solid State Tesla coils]
Netzpfuscher
October 12, 2019, 03:53:12 PM
post Re: My first DRSSTC on bricks
[Dual Resonant Solid State Tesla coils]
Laci
October 12, 2019, 02:12:59 PM
post Re: 220ac to 440dc 5a boost convertor
[Voltage Multipliers]
rustedone
October 12, 2019, 10:43:48 AM
post Re: DRSSTC with litz-wire primary and 40 x TO247 IGBT H-Bridge
[Dual Resonant Solid State Tesla coils]
ScottE
October 12, 2019, 07:38:47 AM
post Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
[Electronic circuits]
davekni
October 12, 2019, 06:16:12 AM
post Re: DRSSTC with litz-wire primary and 40 x TO247 IGBT H-Bridge
[Dual Resonant Solid State Tesla coils]
davekni
October 12, 2019, 05:52:51 AM
post Re: so far video
[Spark gap Tesla coils]
MRMILSTAR
October 12, 2019, 05:13:28 AM
post Re: Problems with my first DRSSTC
[Dual Resonant Solid State Tesla coils]
davekni
October 12, 2019, 04:41:19 AM
post Re: so far video
[Spark gap Tesla coils]
thedoc298
October 12, 2019, 02:58:19 AM
post Re: so far video
[Spark gap Tesla coils]
thedoc298
October 12, 2019, 02:56:17 AM
post Re: so far video
[Spark gap Tesla coils]
MRMILSTAR
October 12, 2019, 02:38:43 AM
post so far video
[Spark gap Tesla coils]
thedoc298
October 12, 2019, 12:26:44 AM
post Re: Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"
[Electronic circuits]
T3sl4co1l
October 11, 2019, 11:33:19 PM
post Re: My first DRSSTC on bricks
[Dual Resonant Solid State Tesla coils]
Netzpfuscher
October 11, 2019, 11:17:17 PM
post Re: DRSSTC with litz-wire primary and 40 x TO247 IGBT H-Bridge
[Dual Resonant Solid State Tesla coils]
Hydron
October 11, 2019, 10:06:32 PM

Sitemap 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30