Help for people buying the "12-48 Volt 1800/2500 Watt ZVS induction Heater"

[betalab99]

Btt:
In which circumstances are the mosfets blowing up?
Just with the empty coil or during load?



If the mosfet burns out, the zener diode are typical going to burn also.
I would recommend checking these first or to replace the one on at the blowing mosfet.
These zeners are essential to limit the voltage at the gate.

Mosfet blows almost instantly - no load.

I can have the PSU connected without switching coil on and it shows up the correct voltage on the LCD display.  Soon as I press the button to turn on coil the relay clicks then mosfet dies.

Will change diodes and check transistors.

Will probably be cheaper buying a different unit then keep buying mosfets.  Can get a 2000W version for £20 delivered - https://www.aliexpress.com/item/1005006593756901.html

[betalab99]

I'm sure this would be an easy fix for someone as there's so few components but I'm defeated.  Diodes and transistors test ok - other mosfets test ok - it's just that same position that keeps blowing.  That's me out of mosfets so it's getting chucked back in the 'to do' box.  I'll leave this pic here in case someone has any ideas over the next few years    p.s.  pic is without the LCD for a better view

p.p.s  Just remembered that it sparks somewhere in the relay but I don't think it was at the contacts...

[betalab99]

Or - it might be easier to just bin this one and buy something else.  Unfortunately that's not really how my OCD operates

Depending on your financial situation, now that you have found a good power supply, it might be worthwhile to simply order a 2500 watt model such as this one:
https://www.aliexpress.us/item/3256803245349461.html?
I have had very good luck with them.

Over the years, dozens if not hundreds of others have had problems similar to yours with this general class of ZVS induction heater.  That's why i put that troubleshooting guide out on the web.

I've bought one similar to the one you linked but it says max 50amp - "Maximum DC working current: 50A"

The PSU I have says 62.5 amps at 48v.  I guess I need a 50amp resettable fuse in case I run it too high?

[petespaco]

My main way to handle the overcurrent issue is to install an analog ammeter (the type with an external shunt) in series with the power supply output.  Then be careful to watch the meter to make sure that the current  doesn't exceed 45 amps or so when inserting work into the coil.
A mechanical fuse generally won't be fast enough to avoid problems, in my experience.

  I have gone farther than this myself by building a circuit using a DC to DC solid state relay to interrupt  the DC power if the current exceeds about 45 amps.

Here I show the analog ammeter setup:

/>This will take you a long way toward making a useful setup. 


Here I show my SSR solution:

/>
Lastly, take some time to view related videos from my ZVS induction heater playlist:
https://www.youtube.com/playlist?list=PLVxWen9M87dBhvInCQ-3pELWdxCM3XH4I

[betalab99]

Thanks for taking the time to reply with that info.

The board turned up in bits - single layer of bubble wrap all the way from China does that!  I'll get one of these going eventually 

[betalab99]

At last!!  I received a relatively intact 2000W board 

All I need now is a smaller coil for heating bolt heads.  Does anyone know of a calculator/spreadsheet for calculating coil size.  Had a look on here but couldn't find anything.

I did find this one that looks promising - https://electronbunker.ca/eb/InductanceCalcML.html but think I will need the capacitance of the board.

Looking at some youtube videos I'm guessing it will be a short double coil.  Is a case of just making one and testing it?

[petespaco]

Go to this  Youtube playlist and look for the videos that deal with coil design:
https://www.youtube.com/playlist?list=PLVxWen9M87dBhvInCQ-3pELWdxCM3XH4I

If you make the work coil's inductance too low, you run the danger of excessively high frequencies which can cause the Mosfets to spend too much time in the linear mode, overheating and eventually destroying them them  (as you will see from some of the videos on that playlist..

I don't know exactly which board you have procured.  It may have as few as one or two large capacitors for the tank circuit or as many as a dozen smaller ones, (usually all in parallel, so just add them up).  You will need to look at them and see what value they are.  Even though there will be small sources of capacitance from other parts of the  circuit,   the summing of those capacitors will be close enough to determine, within about 25%, the frequency of the tank circuit.

I hope some of the experts here also offer opinions.

[betalab99]

I'm slowly getting through those videos - it's a lot of viewing!   

This is the one I got - https://www.aliexpress.com/item/1005006593756901.html

Works great but the coil is too big.

[petespaco]

Hello again, betalab99.
  Yes, there's a lot to go through.
But, as I have been thinking about your plans, I realized that, although I do talk a LOT about coil design for these particular ZVS induction heaters,  I don't make it easy to understand the importance of work coil inductance on net power output and circuit reliability.
So I have  produced a paper on this exact subject.  It's still a work in progress and I am still not sure exactly what I will do with it or how I will integrate it with the other ZVS stuff.
But, just for you, Here it is:

WorkCoil Inductance limits Coil design options

Workcoil inductance must be held high enough to prevent circuit over heating and to maximize available net current to heat the work
Circuit overheating leads to component failure; particularly the mosfets and their gate drive components,
Note that lower inductance means higher idle current.
  This is important because total current - idle current = net current available to heat the work!  We have a total current available of about 20 amps for the 1000 watt heater or about 50 amps for the 2500 watt version.
 As an example; If the idle current of the 1000 watt heater is increased from the stock coil's 4 amps at 48 volts to 8 amps by making a coil with too few turns, the net power is reduced as follows 4 X 48 = 192 watts For the 2500 watt heater, we have a loss to net current available of:  8 X 48 = 384 watts.
 
Workcoil inductance must be held high enough to prevent circuit over heating and to maximize available net current to heat the work.  Circuit overheating leads to component failure; particularly the mosfets and their gate drive components.
Note that lower inductance means higher idle current.  This is important because total current - idle current = net current available to heat the work!  We have a total current available of about 20 amps for the 1000 watt heater or about 50 amps for the 2500 watt version.  As an example; If the idle current of the 1000 watt heater is increased from the stock coil's 4 amps at 48 volts to 8 amps by making a coil with too few turns, the net power is reduced as follows:  4 X 48 = 192 watts.  So, 1000 - 192 =   808 watts to heat the work.
 And, for the 2500 watt unit,  we have; 2500 -384 = 2115 watts to heat the work.
 

[betalab99]

Found this very interesting PDF on coil design if anyones interested 

https://sg-induction.com/wp-content/uploads/2021/03/Coil-Design-Induttore.pdf

Also - whats the lowest operating frequency these reliably operate at as I need to heat steel bolt heads.  Am I correct in thinking lower is better?

[petespaco]

Is low frequency better for heating steel?

"Also - whats the lowest operating frequency these reliably operate at as I need to heat steel bolt heads.  Am I correct in thinking lower is better?"

You haven't told us what size of bolt you want to heat or what temperature you want to heat it to or what you want to do with it once you get it hot.
I make a point of this because, for steel, with this style of heater, the Curie point is going to be your worst enemy of getting things really hot.  Assuming that your coil design allows plenty of current flow when you first power up with the bolt head in place, all goes well until the part gets to about 1500°F. Then above that temperature, the power begins to drop and may settle at only about half of what it had been and the temperature levels off at only a couple hundred degrees hotter than that 1500.

Continuing on, as is often the case with these things, there's no simple answer to your question.
You can find references that say that high frequencies are better for non ferrous materials and that lower frequencies are better for iron bearing materials.
But you can also find info sources that tell you that high frequencies give you deeper penetration.  And so it goes.  Personally, I prefer the lower end of the spectrum, around 40 kHz, for heating steel with my 2500 watt units.

With the type of heater that you have, you don't get to simply choose any frequency that you want.  As I mentioned a few posts ago, you are limited to by the circuit itself and by your coil design.  Honestly, I haven't gone below about 38 kHz with any of my experiments, so I don't know  if lower than that would be better.
Another issue is circuit "Q", "Quality"  It appears that you don't transfer as much energy to the work if the value of C and L get too far out of balance.  In your case, if your capacitor bank is about 2.3 mfd, you wouldn't want you workcoil's induction to vary too far from 2.3 micro Henrys.
I was going to end this post by saying: "summing up---", but I am just making a few points that show, as far as I am concerned, that there is no easy answer.

 Again, I hope some of the experts will chime in here on this circuit Q issue.