Help With A Induction Heater

Discussion in 'General Electronics Chat' started by MARIO MIRANDA, Jul 4, 2007.

  1. MARIO MIRANDA

    Thread Starter New Member

    Jul 3, 2007
    3
    0
    Hi, I'm having trouble with a simple half bridge induction heater.
    The oscilattor is a TL494, working in a range of(30-280KHZ),the drives are two mosfet drivers:TC4421 and TC4422,inverter and no inverter.
    In the half bridge, i use two power mosfets IRFP460.
    The oscilattor and drivers are ok,delivering 15V signal to the power mosfets gates through 27ohms resistors.
    The power mosfets are for maximum ratings: 500V(VDS) 20A(ID) and is is mounted over a good aluminium dissipators.
    The problem is: The IRFP460 mosfets are going to get burst, when the power is turned on.
    The low voltage power supply feeds 15Vdc and the high voltage power supply feeds 170Vdc to the D(drain) from mains 120Vac.
    The first test was with 170Vdc in the D(drain)….it got burned
    Then I did a second test, using a variac, feeding 50Vac to both low and high power supplys, resulting in a 72Vdc in the D(drain).
    In this second test, I resolved to measure the drain current(ID) and the Vds(drain to source voltage) in one of the two power mosfets.
    Thus, in this time, I got only one of the two power mosfets burned after a few seconds, I believe that the ammeter lead wire connections, did a unbalance on the two parts, resulting in it burns.
    I am using a work 1.5uH work coil in parallel with a 376nf capacitor bank(resonance frequency in 212KHZ).
    A 470nf DC blocking capacitor in series with a 45uH matching coupling inductor.
    All this tests were made in a 30KHZ frequency , I did have no time to reach the resonance because the power mosfets got burned before that.
    So, Somebody can give me a hand on it?
    Regards
     
  2. cumesoftware

    Senior Member

    Apr 27, 2007
    1,330
    10
    MOSFETs and inductors do not mix, and I think the problem is that since a reverse surge voltage is created when the current reverses, which may damage the MOSFETs. Can you use a bipolar transistor instead? And/or a protection diode? I think that is part of the solution.

    Also, I suggest a circuit diagram. I cannot advance a concrete solution. I'm only speculating here.
     
  3. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    A schematic will help clarify your existing circuit hookup and make it easier for you to direct our attention to the device or devices that are not behaving as you believe they should.

    hgmjr
     
  4. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
    Are you certain that each MOSFET is turning all the way of before the other MOSFET begins conduction? I blew up a few IRF840s before learning how important "dead time" is.

    And yes, a schematic would be a big help in troubleshooting.
     
  5. MARIO MIRANDA

    Thread Starter New Member

    Jul 3, 2007
    3
    0
    Gentleman,
    I am sending three schematics.
    The diference between them is the final tank only.
    I did simulate ists circuits in Multisim, but without the originals tl494 and tc4421/3322 drivers because dont have its models; instead i used a 15V square wave signal generator feeding a 2 mosfets in a half bridge configuration, just like in the circuits attached.
    The interesting think is that i got some measures: Vds-maximum 130V and
    Id-maximum 1A,resulting a 130W maximum power, thus the IRFP 460 can handle 280W maximum power , then i dont understand why the fets did not handle, anyway i apreciate some comments about it.
    thanks
     
  6. MARIO MIRANDA

    Thread Starter New Member

    Jul 3, 2007
    3
    0
    Schematics-original, One And Two
     
  7. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
    You can improve turn-off time of your MOSFETs by including diodes in parallel with your gate resistors. Put the anode toward the gate, so the diode bypasses the resistor during turn-off.

    See the following for more info: http://focus.ti.com/lit/ml/slup169/slup169.pdf
     
  8. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    With the push-pull driver, don't you actually have 30V p-p into your gates? This will give you +/- 15V gate drive on each FET. The -15V seems excessive, but it may prevent excessive shoot-through currents.
    The IRFP460 can handle 280W (ST's datasheet says 220W) with an infinite heatsink, at 25C. Do you have heatsinks? Have you calculated the power dissipation?
     
  9. PA0DGR

    New Member

    Nov 21, 2008
    2
    0
    Collegae,
    At first ,excuse for my shocking english ( I am a dutchman ans are better in technics than in strange languages)
    The question: I have build a inductionheauter ( target is to melt silver - my other hobby is gold and silversmithing-) I use the same diagram as Mario Miranda ( 07-05-2007) but I have problems with tho both drivers ( TC 4421 and TC 4422) .
    Input a square wave. Rsults, after some seconds two very hot and smoking ic's and so end of the exercise.
    What goes wrong ??
    thnks PA0DGR
     
  10. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    Mario posted three schematics. You should post yours.
     
  11. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I don't know why a 1:1:1 transformer is being used to drive the gates instead of a half-H bridge driver. There's no way to provide "dead time" to prevent shoot-thru, which is virtually guaranteed to happen with the current configuration.

    I don't know of any N-ch power MOSFETs that have a shorter turn-off time than turn-on time. They may exist, but I am not aware of them.
     
  12. PA0DGR

    New Member

    Nov 21, 2008
    2
    0
    Attached the schedule of the induction heater. Please give me your reactions. Untill now I didn't succeed to make it work!
    I'll be grateful to anyone that can give me advice to make it work. Please help.
    PA0DGR
     
  13. Hesham2009

    New Member

    May 6, 2009
    2
    0
    I think it isnot easy thing to do
     
  14. bertus

    Administrator

    Apr 5, 2008
    15,638
    2,343
    Hello,

    Do you realize the thread is over a year old?
    The TS will be gone by now.

    Greetings,
    Bertus
     
Loading...