DIY Induction Heater Builds and References

[Piteer]

I used an esp32 that I programmed so I could set the target temperature over bluetooth. Then I connected a 36v 15a mosfet to that so I could power on and off the induction heater according to the current temperature. Worked very well until I blew a power supply because I made my coil of less wraps to increase the power.

I'd like to make something like this but reliable, it's proving to be quite difficult to make a reliable ZVS circuit. Some other guy on another forum says it will never work reliably unless we move to very advanced circuits, I'm not entirely convinced it can't be done but I'll probably be paying an engineer to make it more robust so I can use it in a product I'm working on.

I made 3 Induction heater Prototype with IR Temp control, the first is tested for 6 months and still running, now i testing No 3 and so far it work. I used Arduino and a Optocoupler for turn on the Gate 12V supply for the Chinese heater. The HalfPint got too hot so i changed to the original IH. Still have some Program fail but it work !


https://ibb.co/Jwn9pCC7

20250121_224202 hosted at ImgBB

Image 20250121_224202 hosted in ImgBB

ibb.co

 

[Gomaruana]

I made 3 Induction heater Prototype with IR Temp control, the first is tested for 6 months and still running, now i testing No 3 and so far it work. I used Arduino and a Optocoupler for turn on the Gate 12V supply for the Chinese heater. The HalfPint got too hot so i changed to the original IH. Still have some Program fail but it work !


https://ibb.co/Jwn9pCC7

20250121_224202 hosted at ImgBB

Image 20250121_224202 hosted in ImgBB

ibb.co

Very cool! Reminds me of my build, I should look if I still have some images from it. I'm was using it to heat up a full sized ballvape, so that's why it broke down in the end. I did use one for about half a year or something but that prototype had a bit less metal mass to heat.

 

[waungus]

Having a bit of trouble.. I'm following the guide on page 13, but I think I have a different version of the board. It still has a third contact pad and I can't see a ground trace between the mosfets. Is there another guide for this version of the board? I'd appreciate any guidance. here's an imgur link with a picture of my board

https://imgur.com/a/8suJONL

 

[waungus]

Having a bit of trouble.. I'm following the guide on page 13, but I think I have a different version of the board. It still has a third contact pad and I can't see a ground trace between the mosfets. Is there another guide for this version of the board? I'd appreciate any guidance. here's an imgur link with a picture of my board

https://imgur.com/a/8suJONL

I went ahead with some steps and think I might understand it a bit more now... is this diagram I drew up correct?

What purpose does the magnetic wire loop on the other side serve? I'm confused because to me it seems like that's there if you don't want to add a switch, but if you're not adding a switch whats the point of cutting the trace? Is it there if you cut the board and the trace on the bottom is unaccessible?

Also with this board, where would I connect the negative?

https://imgur.com/a/4a5Ur6p

 

[Mr_Hodag]

Is their a build/how-to guide for a temperature controlled induction heater?

 

[waungus]

I went ahead with some steps and think I might understand it a bit more now... is this diagram I drew up correct?

What purpose does the magnetic wire loop on the other side serve? I'm confused because to me it seems like that's there if you don't want to add a switch, but if you're not adding a switch whats the point of cutting the trace? Is it there if you cut the board and the trace on the bottom is unaccessible?

Also with this board, where would I connect the negative?

https://imgur.com/a/4a5Ur6p

ended up re-reading the guide and scouring the rest of the thread - figured everything out. no need to reply to either of my previous posts if anyone planned on doing so

 

[Dns18_05]

I combed through the DIY induction heater threads on here the last few days and there's so much interesting information. I already built 2 IHs for my Dynavaps but I recently got a Tempest and I wanted to modify my first one for use with the Tempest and eventually build a more compact one based on the HalfPint design.
@TommyDee I saw that you were also active in the Tempest thread, are you using the IHs you built with your Tempest or only with Dynavaps so far?

 

[TommyDee]

Hey @Dns18_05 I haven't fired up the Tempest yet. I'll probably be using the Wand to start. I need to see where it lands on the power draw compared to my caps.

I haven't played with ball vapes but the idea you can heat them more aggressively seems likely. I'm still open to the idea of using much more powerful FETs.

 

[TommyDee]

Nice work @waungus Indeed, that low voltage switch is to enable to enable the gate circuit. Without that, the circuit will not oscillate. The FETs remain open even with power applied. By default, both the power circuit and the control circuit (gate circuit) are enabled simultaneously.

I think I made the loops to make it more stable to solder the coil to.

 

[Dns18_05]

Hey @Dns18_05 I haven't fired up the Tempest yet. I'll probably be using the Wand to start. I need to see where it lands on the power draw compared to my caps.

I haven't played with ball vapes but the idea you can heat them more aggressively seems likely. I'm still open to the idea of using much more powerful FETs.

Ah I thought maybe you had a Tempest 1 already but that makes sense

So my main issue is that the visual indicator of the Tempest should not be heated. I'm not sure if that also means it should not be inserted into a induction heater's coil. From what I read, only the sleeve around the ball chamber contains iron so it can be heated with induction
I just took apart my first induction heater and bent the coil so I only get 6 wraps of the coil around the ball chamber and that already gave me great results.
But I want to try to build a really portable heater based on your HalfPint design and if I only do 6 wraps, I'd have a lot of excess coil wire. As far as I know, you shouldn't change the length of coil wire, because it would increase current draw? Did you ever change it during your experiments?

https://imgur.com/a/qAz11Vd

 

[TommyDee]

I did the over-wraps to add the extra length. But yes, you are right, narrowing the coil narrows the field. It is most intense in the center. The little indicator is back a ways inside the metal, meaning it is somewhat shielded. I think the hint is indeed to keep coils short.

You could consider going 2S and shortening the coil wire to around 22".

 

[Dns18_05]

I did the over-wraps to add the extra length. But yes, you are right, narrowing the coil narrows the field. It is most intense in the center. The little indicator is back a ways inside the metal, meaning it is somewhat shielded. I think the hint is indeed to keep coils short.

You could consider going 2S and shortening the coil wire to around 22".

By over-wraps you mean that you wound the excess coil wire around the other windings, so the coil has 2 layers? Does that have any advantage compared to leaving it uncoiled?

So if I understand correctly, shortening the coil to 22" would increase the current requirement but the lover voltage of 2S batteries would get the wattage down to appropriate levels again?

 

[TommyDee]

Yes, just wrap the extra wire on the outside of the coil. This solves the problem of having two live induction 'legs' that maintain a field between them. Basically managing the wire 'responsibly'. Normally, you want those two legs to be parallel and close together.

And yes, the shorter wire and lower voltage can equal the same amount of power. Basically shifting voltage to current and making the same amount of power.

 

[Dns18_05]

Yes, just wrap the extra wire on the outside of the coil. This solves the problem of having two live induction 'legs' that maintain a field between them. Basically managing the wire 'responsibly'. Normally, you want those two legs to be parallel and close together.

And yes, the shorter wire and lower voltage can equal the same amount of power. Basically shifting voltage to current and making the same amount of power.

So I decided to get two 21700 cells and I'm currently waiting for parts. On my previous build, I had an led button that lights up when the module is active. But on this build, I want to cut the trace and connect that using a button as you did. What kind of voltage is present there? Is it just the regular supply voltage, so I could drive the led from there?

 

[TommyDee]

So I decided to get two 21700 cells and I'm currently waiting for parts. On my previous build, I had an led button that lights up when the module is active. But on this build, I want to cut the trace and connect that using a button as you did. What kind of voltage is present there? Is it just the regular supply voltage, so I could drive the led from there?

Correct, this trace is the mains power which only goes to the center of the two coils. The trace leads down to power the LED and drive voltage for the gates, which are capacitive in nature. Therefore, very little current through the switch. You need to account for the additional power requirement required for peripheral loads if you have it going through the switch.

 

[Dns18_05]

Correct, this trace is the mains power which only goes to the center of the two coils. The trace leads down to power the LED and drive voltage for the gates, which are capacitive in nature. Therefore, very little current through the switch. You need to account for the additional power requirement required for peripheral loads if you have it going through the switch.

Thanks for clarifying, I think the led in my button won't increase the current requirement that much
Do you know what usually breaks when a module burns out? I have two broken ones and tried to replace the mosfets but it still doesn't work. The diodes are fine but I'm not sure about the zener diodes
I also thought about building the circuit myself with better mosfets but I think I'll stick with the module for now

 

[TommyDee]

The only thing I've ever lost was a FET. Never had a replacement fail though.

These things are DC-AC converters. The coil is a sort of regulator. Without the coil, there is no voltage limit which is what burns out the FETs.

Better FETs is the way to go. I don't have the appropriate electrical engineering knowledge to know how to choose the best equivalent but I do know you want more voltage tolerance. You also want something capable of handling the power in the 80-100W range.

 

[Dns18_05]

The only thing I've ever lost was a FET. Never had a replacement fail though.

These things are DC-AC converters. The coil is a sort of regulator. Without the coil, there is no voltage limit which is what burns out the FETs.

Better FETs is the way to go. I don't have the appropriate electrical engineering knowledge to know how to choose the best equivalent but I do know you want more voltage tolerance. You also want something capable of handling the power in the 80-100W range.

I remember a list of better but still compatible FETs somewhere (I think in another induction heater thread here) but I have to find it again.
I had some problems at the beginning with running into the current limit of the PSU or running it without a switch and that destroyed two modules. It burned holes into the FETs but replacing them didn't fix it so something else must've been destroyed. But getting these components individually isn't worth it anyway if you can get the module for <4€

 

[TommyDee]

I got a couple of different FETs to test. Installed one pair and that worked. I didn't do any analysis to see how it compared though.

 

[Dns18_05]

In the first post they mention a few different FETs but I think all of them are through-hole so mounting them on the module's pcb would be unconvenient

 

[TommyDee]

You want the TO252 formfactor. You also want the highest voltage rating. I hit it up at 100v. The power we draw is always under 10 amps so this is almost moot.