MOSFET over heating

Discussion in 'General Electronics Chat' started by R!f@@, Nov 1, 2011.

  1. R!f@@

    Thread Starter AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    The attached pdf is not the actual circuit. I have tried & tried but cannot find the schema for this typical model.
    A 42" LCD power supply.

    The attached pdf is the only thing I could find about the PWM chip used in this LCD. It is not the data but close enough.

    I am having a hard time with this supply. It ain't dead and no load voltage is as it should be. Friggin' thing just won't regulate. Even a slight load at a 24V out drops to 15.

    I changed all the caps due to bulging.
    The MOSFET's are too hot to touch and with small heat sink I doubt it would live long if I can get it to regulate.

    The pdf is something I found a few minutes ago, been trying to find the data for the PWM for weeks..

    Can any one explain me this.
    Say the pdf circuit is working properly under no load. what would be the course for the MOSFET to heat up to a level that is too hot to touch it.
    In my experience a transistor is never too hot to touch with in seconds when they are mounted on a sink.

    But say these guys heats up in seconds.
    What would be the likely cause ?
     
  2. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    FETs heat up due to excessive power dissipation. There are exactly two kinds of power dissipation in FETS:

    1) Conduction (ON state) power

    2) Switching losses if used in a switching converter

    Since you are generating 28kV, it has to be a switcher of some kind so you have both kinds of power dissipation. The power supply may have failed or may be getting overloaded by the circuitry it is driving.
     
  3. hert

    New Member

    Oct 24, 2011
    3
    0
    if you have a frequency meter, check either you are driving the FETs at a higher frequency than the normal specification. again the capacitor c2007 may be bad. you can change the two of them and replace them. does this help you?
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Slow rise/fall times on the MOSFET gates would do that, and a bad cap in the vicinity of the PWM IC or gate driver(s) could be the cause of that.

    You're not giving us much to work with.
     
    R!f@@ likes this.
  5. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    If you are generating 24V, and not 28kV, you need to post your schematic.
     
  6. R!f@@

    Thread Starter AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    I know Sgt....got one heck of a workload..
    But I think you gave me what I need for now. I'll show the works soon.

    @ Ron.
    U too did not get what I asked. No wonder too other members missed it.
    As I figured Sgt did.

    See that is what makes him unique.
     
  7. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Well, "hert" had a very good point as well. If the timing cap opened up and the frequency of the oscillator went into the stratosphere, the MOSFET(s) would spend most of their time in the partially conducting state - which links right back to what Bountyhunter was referring to; it's just another way to make that happen.

    It would be easy to overlook the timing cap; they're usually pretty small. If an aluminum electrolytic cap near it spilled electrolyte, it may have caused corrosion in the area of the timing cap.

    A number of years ago, some Asian manufacturers were using electrolyte that was much too concentrated, but that wound up backfiring because the package would get eaten through. No, I don't remember which, and I wouldn't post names if I DID remember.
     
  8. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    That was my point.
     
  9. R!f@@

    Thread Starter AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    In the actual supply the caps are replaced, beside the SMD's. This is where I am stumped.

    I got a scope now. I would give it a try and post the wave forms and pics together.
    But first me needs an isolation thingy. Man those things are never around when u need em.
     
  10. sheldons

    Well-Known Member

    Oct 26, 2011
    616
    101
    i service these things for a living and the problem you have -no regulation-could be due to any number of things- if you post the exact model nr of yr set and the id on the smps board ill look the schematic up and see whats going on
    if you are not aware of the proper procedures for the test and repair of an smps you can cause quite a lot of damage and extra work for yourself
     
  11. R!f@@

    Thread Starter AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    Sheldon, u better look at my threads

    I am well aware of SMPS and it's repair procedures.

    But if u can find me a schema it will be a lot of help.

    I will pics soon to night.

    My main problem is not a major issue probably. It's the little problems that takes a lot of time.
     
  12. tom66

    Senior Member

    May 9, 2009
    2,613
    214
    If it's a half-bridge architecture, check the two poly caps.

    You could also have cross conduction.
     
  13. R!f@@

    Thread Starter AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    Here's what I am talking about

    [​IMG]

    [​IMG]

    The TEA1601 is used for 12V and 24V. Both have regulation problem and the 4 mosfets are heating up.

    Here is a high res pic

    Model is Philips LCD 42PFL7432/98

    {ed}
    One thing bugs me is that at raw DC point, the filter cap 300VDC line, it measures a ripple of 0.4VAC rms. This is odd since I replaced the Filter caps due to high ripple.
     
    Last edited: Nov 2, 2011
  14. debe

    Well-Known Member

    Sep 21, 2010
    947
    184
    Its unlikely both 12v & 24v power supplys are faulty. Whats the power feed to both of them like? Is it not holding up or not clean?
     
  15. R!f@@

    Thread Starter AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    Exactly what I am thinking....I edited my post..see.
     
  16. R!f@@

    Thread Starter AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    Funny thing happened before.

    When I was measuring DC voltage at the primary side. Two MOSFET's blew. They blew when I try to measure the gate voltage of one of the MOSFET.

    And NO..I did not short circuit anything.

    I replaced the MOSFET's, still the problem remains and I believe I found one 10K (I think) SMD resistor at the same MOSFET between G & S.
     
  17. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,018
    1,538
    A resistor is often used between G and S to pull down the gate charge. To make sure the gate is or stays off. But you probably knew that.
     
  18. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    The 300V caps will always have some ripple on them if you are drawing any current.

    If you have both 12V and 24V linear regs not working, likely either:

    something is drawing excessive current

    An overvoltage caused them to fail and they are drawing current to ground
     
  19. R!f@@

    Thread Starter AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    No load, nothing nada.....just open circuit voltage is OK, even a light load cause the voltage to drop, when removed, voltage comes back slooooooooooooooowly.
     
  20. hert

    New Member

    Oct 24, 2011
    3
    0
    again you also have to check if your transformer is in good condition. you know a short in the trans. can cause that especially at the secondary side of the trans.
     
Loading...