Gate driver for mosfet problem

Discussion in 'General Electronics Chat' started by raychar, Nov 17, 2011.

  1. raychar

    Thread Starter Member

    Nov 8, 2011
    82
    0
    Oh, the gate driver burnt again, How vulnerable this IR2117 is! For sure, I haven't made any wrong connections nor short-circuits. I made change on motor power supply voltage from 12V to 25V with all powers down first.

    Is the sequence of power up a matter? First turning on the power of control circuit, then the power of driver, then the motor/load power. I sometimes make reverse sequence. Does anyone have knowledge of this.

    Thanks
     
  2. jimkeith

    Active Member

    Oct 26, 2011
    539
    99
    Does the mosfet fail too?
    Does the diode fail too? this could be the smoking gun--it must very fast recovery and high voltage as it really takes a beating.
    What kind of transient voltages do you get when the FET turns off? snubbers present?

    This topology has been used for years, but any circuit 'enhancements' by the user are likely to introduce issues--I once consulted with a user and solved his high failure rate by reverting to the recommended circuit.

    I am not aware of PS sequencing issues, but cannot rule this out--this kind of problem could be experimentally demonstrated--harden the circuit working back by isolating various inputs etc.--might take a bag of parts...

    Per the spec sheet, 25V is excessive--recommended range is 10 to 20V
    http://www.irf.com/product-info/datasheets/data/ir2117.pdf

    Good luck
     
  3. jimkeith

    Active Member

    Oct 26, 2011
    539
    99
    Hope that you understand that this device isolates only the high side power device from the low side power device, and that the circuit driving it must be isolated from everything but the negative DC bus--circuit common may not be grounded--inputs and everything else must float--are electrically live--not easy to work with...

    Also, power sequencing and drive signals must be coordinated to prevent shoot-through where both the high and low sides are turned on simultaneously--guaranteed to fail!
     
  4. raychar

    Thread Starter Member

    Nov 8, 2011
    82
    0
    Hello,

    I attached the schematic, the diode now is changed from BYV27-150 to 1N4148, still the motor run slowly. Are there anything wrong in the circuit.

    I tried to add a 100ohm resistor from ground to the gate of mosfet but with no help.

    Thanks
     
  5. jimkeith

    Active Member

    Oct 26, 2011
    539
    99
    OK, I see your problem--throw away your gate driver--it must be used with a half-bridge that has two power devices--the driver for the upper device (what you are attempting to drive) must charge its boot-strap capacitor via first turning on the lower device (that you do not have). Also, there is no free-wheeling or commutation diode to handle the reactive energy of the motor when the FET turns off--without it, the motor becomes a source of destructive voltage. Check out the thumbnail for a simple circuit that should work for you--no driver chip required.
     
  6. raychar

    Thread Starter Member

    Nov 8, 2011
    82
    0
    Hello,

    The previous schematic was for my testing, the IR2117 is actually to be implemented in loading in the attached new schematic. Can I forget the original boost diode and cap and replaced by another voltage supply as shown. Are there any incorrect in this circuit.

    Thanks
     
  7. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,009
    1,530
    The low and high drivers are independent of each other in these half-bridge drivers. The low can be used by itself, but not the high. Depending on the mosfet, even on a low side switch, a driver is a good idea. To give enough current to charge the gate fast, to prevent heat in the mosfet.

    We also don't know what is being used for the timing circuit. If timing is from a 555,or a logic circuit, a driver is a real good idea.
     
  8. raychar

    Thread Starter Member

    Nov 8, 2011
    82
    0
    Hello,

    The circuit which including input logic are:
    555 -> comparator (to get pwm pulse) ->74LS07 buffet(to get higher voltage output)-->to gate driver -->mosfets

    Thanks
     
  9. praondevou

    AAC Fanatic!

    Jul 9, 2011
    2,936
    488
    Btw, the thread started actually HERE.
     
  10. raychar

    Thread Starter Member

    Nov 8, 2011
    82
    0
    a better clear image.
     
Loading...