lower side mosfet keeps blowing

Discussion in 'General Electronics Chat' started by ak52, May 3, 2015.

  1. ak52

    Thread Starter Member

    Oct 15, 2014
    145
    4
    Hey guys,
    I am working on a 3KW solar charger and i am facing some problems in the mosfet triggering section.I have attached a snap of the gate driving section ,the IR2110 high and low side driver is used.
    When ever i start pulsing the low side mosfet keeps blowing.Can some one please specify any flaws in the schematic if there are any..

    For testing purpose when i connect a resistive load at the battery terminals ,everything works perfectly,but when i connect actual lead acid batteries ,the lower side mosfet blows :(

    This is my current configuration.
    IC U5 is not soldered,Jumpers J5,J6 and J3 are set,Inductor i used is 0.07MH,150 Amp.
    C17 is not soldered.


    Thanks in advance,
    AK
     
  2. ScottWang

    Moderator

    Aug 23, 2012
    4,850
    767
    Your uploaded file was unsuccess.
     
  3. ak52

    Thread Starter Member

    Oct 15, 2014
    145
    4
    Hello Scott,
    I uploaded it again.
     
  4. ScottWang

    Moderator

    Aug 23, 2012
    4,850
    767
    I'm not sure where did you get that idea to charge the battery, normally this IC was used for the inductive load as motor, because the motor can be work for two directions, but when the lower mosfet is turn on, the current of battery will flows though D to S of mosfet as shorted and destroyed.
     
  5. ak52

    Thread Starter Member

    Oct 15, 2014
    145
    4
    I'm afraid i don't understand,are you suggesting that a high side driver is enough for this operation?
     
  6. ScottWang

    Moderator

    Aug 23, 2012
    4,850
    767
    Why you want to using this circuit to charge the battery, where did you get the circuit?
     
  7. ak52

    Thread Starter Member

    Oct 15, 2014
    145
    4
    Hi ,
    I designed it myself,taking reference an old inverter where the same H bridge configuration is used ,with the same gate driver.

    Circuit point of view i am stuck with the current circuit,i cannot change it at the moment,can you suggest me any alternatives ..
    AK
     
  8. ScottWang

    Moderator

    Aug 23, 2012
    4,850
    767
    The ic ir2110 was used for motor, why you want to using oscillation in charge circuit?
     
  9. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,768
    1,102
    When Q3 turns on it shorts the battery. Why do you have Q3 there?
     
  10. uwed

    Member

    Mar 16, 2015
    64
    17
    Such half bridge configurations usually have an inductor (could be the motor-internal one in case of a drive) for current control. Otherwise you directly connect voltage sources of different potential (solar panel, batttery) and/or ground them, the unlimited current will destroy the MOSFETs, see comments above.
     
  11. ak52

    Thread Starter Member

    Oct 15, 2014
    145
    4
    Hi guys,
    Thanks for all the replies ,the thing is i cannot change the circuit(The PCB has already been made) ,although it is wrong(referring to Q3),imy only option is to use the IR2110 as a high side driver only.
    I don't use Q3, as in i de-solder it, and remove resistor R22.i.e i use the IR2110 as a high side gate driver only.
    Please refer the schematic i posted in the first post for understanding my point.
    Would it be ok if the above point is done?

    AK
     
  12. uwed

    Member

    Mar 16, 2015
    64
    17
    You would still directly connect the battery (= voltage source) to the solar panel (= voltage sorce). Current would only be limited by the panel's internal resistance and the internal battery resistance and the MOSFETs' on-resistance. It might work well but I don't know how (or if) lifetime of panel and battery are reduced. Maybe everything perfectly ok long-term, maybe you will have to replace e.g. the battery every couple of months. Does anybody of the reliability or battery specialists know more?

    You have drawn a red connection in your circuit drawing. Is this an inductor? If not, you could put one there without serious changes on the PCB layout, and could make a current control work. In that case you might have to add free-wheeling diodes because the MOSFET's internal one is usually poor.
     
  13. ak52

    Thread Starter Member

    Oct 15, 2014
    145
    4
    Yes that is an inductor.A rather low value one though ,0.07mH,150A.Just enough to turn my PWM signals to a rough triangular wave form.
    About the free wheeling snubber,would a simple 1N4007 do the trick?
     
  14. uwed

    Member

    Mar 16, 2015
    64
    17
    If it is an inductor and you already have triangular waveforms, a possible reason for blowing up Q3 is the poor quality of the internal diode (especially during deadtime). 1N4007 will not work (and would be in the wrong voltage range). If the MOSFET is 90A, the free wheeling diode should be in that range, too. Since you don't feed current from the battery into the panel, you don't need the MOSFET Q3 there, only the diode (simple buck-converter).
     
  15. ScottWang

    Moderator

    Aug 23, 2012
    4,850
    767
    What is the purpose for the inductor, when you added the inductor that it will make the situation getting more worse when the Q2 is turn off and the Q3 is turn on.
     
  16. uwed

    Member

    Mar 16, 2015
    64
    17
    The inductor in this buck-converter prevents a direct connection between two voltage sources at input and output (theoretically unlimited current) and allows current control with PWM.
     
  17. ak52

    Thread Starter Member

    Oct 15, 2014
    145
    4
    Generally pulse charging lead acid batteries is not suggested,you could say it smoothens the pulses,but more importantly used for ripple control. Instead of Q3 need to put a high range schottkey,a buck converter as uwed pointed earlier.

    About the flyback diode,my maximum charging current would be close to 30Amps,so any diode which would withstand 30Amps or more could do the trick?I have a few of those diodes such as "STPS60150C" which would do the trick?
     
  18. uwed

    Member

    Mar 16, 2015
    64
    17
    Generally, with a quick calculation (assuming a constant battery current):
    i_diode,rms = sqrt(1 - u_battery/u_panel) * I_battery,avg

    The STPS60150C looks good.
     
  19. ScottWang

    Moderator

    Aug 23, 2012
    4,850
    767
    If you wan to using the STPS60150C then you should in parallel with them as A1 connected to A2 to double the rating current.

    The turn on condition of Q2 should be like as this :
    Vds = 0V.
    Vd = Vs.
    Vd = V_solar
    Vg to Gnd equal to Vd +Vgs
     
  20. uwed

    Member

    Mar 16, 2015
    64
    17
    It would be a waste not to use the parallel diode ...
     
Loading...