ir2110 heating up

Discussion in 'The Projects Forum' started by assuc, Apr 23, 2014.

  1. assuc

    Thread Starter Member

    Mar 23, 2012
    77
    1
    Greetings,
    Hope you are all fine
    I have made a BUCK CONVERTER using IR2110 high side Mosfet drive, mosfet is irf540N. the problem is when initially it is powered the IC works for 3-4 secs but then gets heated and output is disturbed. Can anyone help telling me why the IC gets heated is there any problem in the given schematic

    UPDATE: Switching frequency is 30kHz and voltage across R3 appears to be 2.5volts
    buck.jpg
     
    Last edited: Apr 24, 2014
  2. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,282
    1,236
    Try removing R3 and C1.
     
  3. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,004
    1,525
  4. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,515
    2,369
    What switching frequency?
    Normally you would use a cap in the place of C1 and C2 depending on the switching freq.
    Max.
     
  5. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,282
    1,236
    Normally the driver just needs to drive the gate capacitance of the FET. It uses the charge stored in the bootstrap capacitor to do that. When you add the resistor it not only needs to drive the capacitance of the FET but also the gate to source resistance. That forces you to use a really big cap for the Cb or it will discharge between cycles. Normally it is kind of good practice to have some high resistance from gate to source so the FET doesn't get accidentally turned on, but not in this case. It's ok in the bottom FET (If you like) because it is driven directly from the 15 volts.
     
    shortbus likes this.
  6. assuc

    Thread Starter Member

    Mar 23, 2012
    77
    1
    switching frequency is 30kHz
     
  7. assuc

    Thread Starter Member

    Mar 23, 2012
    77
    1
    I will do it and let you know but will the voltage appearing on the Gate will be enough to drive the gate easily??
    Also C1 won't help to deliver charge at the gate of FET?
     
  8. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    R3 is really quite low in value; if you want to keep it in there, increase it to 10k at least.

    Why aren't you using the low side too? If you did, you would have a synchronous buck, which uses a MOSFET as an ideal diode. Of course, you need to turn the lower MOSFET off when the inductor current reaches zero (or close to it).
     
  9. assuc

    Thread Starter Member

    Mar 23, 2012
    77
    1

    okay thanks, i presume, that changing R3 will ensure that IC wont heat up .
    Also to use the technique of synchronous buck is my next step as this is the first time i have been using the gate driver IC to drive MOSFET previously i was using resistor network on the gate for switching FET which proved not to be of good way
     
  10. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,004
    1,525
    OK, don't mean to highjack but this is a case where it's easier to ask in the thread.

    Why would R3 drain the boot cap? I know 1k is too low of a value for R1, but, as the drain to source starts to conduct, wouldn't the gate voltage rise with it? And the BS cap would still just be supplying the needed turn on Vgs?

    And the BS cap, when mosfet isn't conducting is still at ground potential and charging, so how is this 'draining' the boot cap? The only difference is with R3 the gate is also at source/ground potential at the same time. preventing it from turning on.

    The only way I can see R3 deleting the BS cap would be if the gate driver was "slightly" on. But that can't happen due to the driver internal circuitry.
     
  11. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,282
    1,236
    Yea, It's not the end of the world, but the gate has to be 10 volts or so above the source for the FET to turn on good. So all the time it is on the 10 ma is drawn from the boost cap, whereas without it once the FET is on it's only leakage currents.
     
  12. assuc

    Thread Starter Member

    Mar 23, 2012
    77
    1
    I will do it and let you know but will the voltage appearing on the Gate will be enough to drive the gate easily??
    Also C1 won't deliver voltages at the gate of FET?

    @RONV
     
  13. RamaD

    Active Member

    Dec 4, 2009
    254
    33
    I hope you are not using 300V supply! IRF540N Vds is rated only for 100V!
     
  14. assuc

    Thread Starter Member

    Mar 23, 2012
    77
    1
    No its just a reference pic
     
  15. assuc

    Thread Starter Member

    Mar 23, 2012
    77
    1
    https://flic.kr/p/o2vtFU
    i have tried using the above configuration but same thing happens the ir2110 starts heating up after 3 seconds...please tell me what to do?
    Note: 15Vdc is from solar panel output.
     
  16. crutschow

    Expert

    Mar 14, 2008
    12,995
    3,229
    True. But when Q1 is on, the gate-source voltage is high with the current being supplied by the capacitor. If R1 is too low it will drain the cap during this turn-on time and the Vgs may go below the turn-on point causing high dissipation. So you want the R1C time-constant to be much longer than the longest on-time of Q1.
     
  17. crutschow

    Expert

    Mar 14, 2008
    12,995
    3,229
    Are you actually measuring 15V at the input to the circuit?

    I see no reason to remove C1. Try adding it back (but without R3).
     
  18. assuc

    Thread Starter Member

    Mar 23, 2012
    77
    1
    what do u mean by measuring? 15 vdc is the optimum voltage that is measured many times before applying it as Vdc

    Please also see i have updated the schematic in my above post.
     
  19. praondevou

    AAC Fanatic!

    Jul 9, 2011
    2,936
    488
    What diode is D1? As far as I remember Vs is not supposed to be lower than COM for more than ... I forgot that exact value. When D1 conducts, Vs IS lower than COM. How much? Did you measure? Did you try to use a low voltage drop schottky for that? Just a thought.

    EDIT:
    Well, it's 5V. Vs can be as much as 5V lower than COM.
    I would still measure if you don't exceed any max ratings (due to transients). Measure with a scope directly at the output of the IR2110. VS-COM, VGS.
    You could also post a picture of your setup.
     
    Last edited: Jun 19, 2014
  20. assuc

    Thread Starter Member

    Mar 23, 2012
    77
    1
    D1 is 1n4048, VGS IS 2-3Volts.
    vs-com is not measured.
    any further advices
     
Loading...