Induction Heater IGBT help

Discussion in 'The Projects Forum' started by yappers, Apr 27, 2012.

Thread Status:
Not open for further replies.
  1. yappers

    Thread Starter New Member

    Apr 13, 2010
    16
    0
    I'm trying to build neon john's induction heater (schematic is attached) but I'm trying to mod it in order to heat up large flat small metal sheets and since I'm using 120 VAC, I won't be drawing more then 15A. However, my problem is that my transistors, the IRFP460 are blowing up on me when I send in my 170 VDC. My schematic is similar to the one neon john has, but with some slight differences shown listed below

    1. R1 in the schematic is altered in such a way that it will give me a max of 50% duty ratio
    2. the resistor on RT of the TL494 is a 10k and is non adjustable (fixed frequency)
    3. C4, C5 and C9 are all film capacitors from epcos.
    4. C10 is roughly .66 uF and is gotten from putting three film capacitors in parallel
    5. L2 is made of two different coils, both 12 gauge. One is rated for .024mH and the other is rated for .0268mH. It is similar to a bifilar coil but arranged differently. See attachment for the coil. The red in the picture means it is connected at those points

      If anyone can give me some advice, I appreciate it



     
  2. wayneh

    Expert

    Sep 9, 2010
    12,100
    3,034
    What frequency are you running?

    Those MOSFETs are borderline for your application. The quoted amperage of 18.4A is not much above your expected current. My personal rule of thumb is to go at least 4X, spec compared to continuous current expected. The spec is for ideal conditions you don't likely have, like a giant heat sink.
     
  3. yappers

    Thread Starter New Member

    Apr 13, 2010
    16
    0
    According to oscilloscope, my square wave (50% duty ratio) from my TL494 is about 250kHz oh and I do have a giant heat sink that is attached to both my IGBT/MOSFET It is 4.230" Wide x 3" long for each one. Also, what do you mean by "quoted amperage"?
     
  4. wayneh

    Expert

    Sep 9, 2010
    12,100
    3,034
    I mean that's what the datasheet says, 18.4A max continuous current at 25°C. Notice how that rating drops to 11.6A at 100°C. That tells me that your 15A is very near the upper spec for this MOSFET.

    Add to that your "fast" switching, with a rise and fall time on each cycle. Any time not fully on or off makes heat.

    Add to that the voltage drop. The current spec was quoted at ∆V=10v, I think. You're at 170V and, in short, well outside the safe operating area.
     
  5. yappers

    Thread Starter New Member

    Apr 13, 2010
    16
    0
    Oh OK, now I see what you're trying to say.

    So from the looks of your comment, I should look to either getting better IGBT/MOSFET or reduce the switching frequency.
     
  6. Experimentonomen

    Member

    Feb 16, 2011
    331
    46
    I've measured over 60A peak inverter current in my 2kW induction heater project.
     
  7. yappers

    Thread Starter New Member

    Apr 13, 2010
    16
    0
    I'm curious. What kind of MOSFET/IGBT are you using?
     
  8. wayneh

    Expert

    Sep 9, 2010
    12,100
    3,034
    Yup. You could add parallel MOSFETs but I think you'd get more bang for your buck by just choosing more robust MOSFETs. I use IRF540N for general purpose stuff but I'm not so sure about using them at your frequency.

    I could be wrong, but I think your inductive output is proportional to the frequency? So reducing the frequency would have a negative impact on the heating.
     
  9. yappers

    Thread Starter New Member

    Apr 13, 2010
    16
    0
    I'll go look for some more robust FETs. But your comment have me curious. Since I'm going to be using them for high frequency, is there a particular trait I should be looking on the FETs datasheet that I should be paying attention to when switching at a high frequency?
     
  10. crutschow

    Expert

    Mar 14, 2008
    13,000
    3,229
    The switching speed of most MOSFETs is usually determined by how fast you can charge and discharge the gate capacitance so the speed of the transistor itself is not a limiting factor. For example the IRF540N data sheet states a switching rise and fall times of 35ns, with a low impedance gate drive, which is fast enough for most power switching requirements.
     
  11. Experimentonomen

    Member

    Feb 16, 2011
    331
    46
    I used some unknown mosfet/igbt modules from an old 48V 73A 3.5kW three phase input smps.
     
  12. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Curious, why do you call your IRFP460's IGBT/MOSFETs?
     
    shortbus likes this.
  13. yappers

    Thread Starter New Member

    Apr 13, 2010
    16
    0
    To be honest, I tried other types of transistors as well which resulted into me using both IGBT and Power MOSFETs. It happened for way too long that I started mixing them up. Sad, I know.
     
  14. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Oh, OK.. I thought I was missing something.
     
  15. yappers

    Thread Starter New Member

    Apr 13, 2010
    16
    0
    OK. So the frequency was the problem. When I reduce the frequency to 25kHz, the system works just fine and when I increase the frequency, the FETs exploded and killed my MOSFET driver as well.

    However, at 25kHz, my metal object aren't really heating up to the point where it is hot to touch.
     
  16. Experimentonomen

    Member

    Feb 16, 2011
    331
    46
  17. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Your schematic seems to indicate that the output circuit is supposed to be resonant. Am I wrong about this? If I'm not then changing the drive frequency without a corresponding change to the tank components would seem counter productive.
     
  18. yappers

    Thread Starter New Member

    Apr 13, 2010
    16
    0
    Wow, you really melted that screw pretty well. But can you be a bit more specific on telling me what I'm doing wrong?



    You're not wrong. But it is obvious to me that I didn't design my resonant tank very well within the spec of the operating frequency. My speculation after looking at neon john's website again seem to indicate that the difference between our resonant tank is way too different to even be compared. But one difference I notice is that he used a lot more capacitors for his resonant tank. Right now, I'm trying to figure out the reason why this is needed and see if I can redesign my resonant tank.
     
  19. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    You've mentioned neon john's website more than once but you've not posted a link to the inspiration of your design.
     
  20. yappers

    Thread Starter New Member

    Apr 13, 2010
    16
    0
    To put it simply, there's no link because no such thing exist. I have an idea that I'm willing to try. That's it.
     
Loading...
Thread Status:
Not open for further replies.