mosfet failure 150watt inverter

Discussion in 'The Projects Forum' started by charanbandi, Jun 29, 2012.

  1. charanbandi

    Thread Starter New Member

    May 17, 2012
    18
    0
    I am working on inverter 150w using pwm Ic3524 to switch IRFZ44 mosfets.I am not getting the output.

    I have tested mosfet individually by connecting it to a battery i.e across drain and source using a 200ohm resistor without giving any gate signal i.e leaving the gate terminal and i am getting about 1amp current flowing across the source and battery negative terminal i.e Id is 1amp.
    But according to the datasheet of mosfet zero gate voltage drain current is 10µA max.
    Why i am getting like this??If that is the failure of that mosfet i have tested 8 mosfets and all are showing the same current??
    Is there something wrong with my testing??
    Thank you ..
     
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You should have the gate connected to the source terminal to ensure that Vgs=0v.

    If Vgs exceeds ±20v, even for an instant, the MOSFET will be destroyed.

    I don't know why you are using a CA3524, as it's designed to drive BJT's (transistors) instead of MOSFETs. The 22k resistors are the path for discharging the gates when it's time for the MOSFETs to turn off, and that will take a fair amount of time. During that time, the MOSFETs will be in partial conducting (resistive) mode, causing rapid heating. This could be the cause of MOSFET failures.
     
  3. charanbandi

    Thread Starter New Member

    May 17, 2012
    18
    0
    even if i havent connected the gate to the source.i shouldnt be getting any current (Id) why i am getting 0.8 to 1 amp??
    and
    how to test a mosfet whether it is working or not??!!
     
  4. crutschow

    Expert

    Mar 14, 2008
    12,988
    3,226
    Not true. The gate looks like a capacitor and, if it is left floating, then it can be at any voltage due to stray charges which will cause it to conduct current.

    To test a MOSFET you can apply a few volts to the drain through an ammeter with the source grounded. Then slowly increase the voltage of the gate. At the transistor's threshold voltage (as stated in the data sheet) the transistor will start to conduct current.
     
  5. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    You must understand about using power FETs as switching devices: they have a lot of gate capacitance. To get them to switch quickly, which is ESSENTIAL in any switching converter, you must drive them from a low impedance driver which can really crank current. You are driving through a 1k resistor.

    We did use the 3524 chips a lot in our power converters (back in the 80's, that was the best available) but we always built discrete drivers to crank the FETs or bipolar power transistors. Driving off the 3524 didn't work well.
     
    charanbandi likes this.
  6. charanbandi

    Thread Starter New Member

    May 17, 2012
    18
    0
    what should i do? to improve the switching charcteristics of mosfets?/?
    what are low impedence drivers????
    I couldnt understand??
     
    Last edited: Jun 29, 2012
  7. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    An example would be to use external transistors like 2N3904/2N3906 as pull up/down gate drivers. The drivers in the 3524 can't source and sink enough current.

    http://www.diodes.com/_files/design_note_pdfs/zetex/dn80.pdf
     
    charanbandi likes this.
  8. charanbandi

    Thread Starter New Member

    May 17, 2012
    18
    0
    thank you sir! I think it will help me alot
     
  9. charanbandi

    Thread Starter New Member

    May 17, 2012
    18
    0
    now the problem with ic 3524 11th pin is giving 7.5v as gate to source voltage but 14th pin giving only 1.8v as Vgs.mosfet is conducting from 3.76v..what should i do?if i change the value of pulling resistors what will happen??
     
  10. praondevou

    AAC Fanatic!

    Jul 9, 2011
    2,936
    488
    Where do you measure this voltage? At pin 11 and 14 or at the gate terminals?
    Verify the resistor values 1k, 22k.

    Disconnect the mosfet gates (and short the gate with the source terminal) and test again.

    How did you measure the pin 11, pin 14 voltage ( oscilloscope, I hope) ?

    What is connected to pin 1?
     
  11. charanbandi

    Thread Starter New Member

    May 17, 2012
    18
    0
    i have measured the voltage at gate terminals only

    yes! with oscilloscope only.what if i measured with multimeter??(I mean is there any difference??other than it will measure the rms voltage??)

    pin1 is unconnected first and then i have shorted to pin9 ,same result i m getting...!!

    I dont understand disconnect the mosfet gates we should measure the voltage across gate pin and source only na!If we shorted them we.ll get 0 Vgs!!!??/
     
  12. praondevou

    AAC Fanatic!

    Jul 9, 2011
    2,936
    488
    Did you check the resistor values? Are they really 1k and 22k ? (btw 22k is kind of to high as was mentioned earlier - depending on switching frequency and load current this will cause excessive heating of the MOSFET)

    I meant: Disconnect the MOSFET gates from the resistors and short them directly to their source terminal. Then measure the output voltage at pin 11 and 14 with a scope. Are they equal in pulse width and voltage level?
     
  13. charanbandi

    Thread Starter New Member

    May 17, 2012
    18
    0
    thanks!The problem got fixed now.I have changed the ic,now i m getting 8.2volts at 14th pin and 5.45v at 11th pin with feedback unconnected i.e pin1 left unconnected..

    If i have given the 12volts battery supply to center tapped transformer,I am getting Id as 0.85amps and i have used 25ohm rheostat in series with drain but I couldnt get any output..I have used 10watt bulb connected across the output.what else i need to take care of???
    I havent used any filter across the output
     
  14. charanbandi

    Thread Starter New Member

    May 17, 2012
    18
    0
    what if the voltage and pulse levels are not equal at both pins??
    I have used 25ohm rheostat to make Id equal in both mosfets.
     
Loading...