Microcontroller + Gate Drive + Power Mosfet

Discussion in 'The Projects Forum' started by ashkarmalik, Nov 12, 2012.

  1. ashkarmalik

    Thread Starter Member

    Jul 1, 2012
    55
    1
    Atmega running at 15Khz + A 9A Mosfet agte Drive(13v)UCC37322P + N Channel Power Mosfet

    I connected the setup and controlled the pwm via Switches ,the voltage is varring as desired but when the power mosfet is been loaded with a power motor then the PWM started to jerk and the mcu reboots again and again i.e when power drawn from the mosfet rises the mcu reboots and the power is stable all the time MCU Terminals.
    What is the problem that is making the MCU to reset when the power is been climbing.
    If I can Huck up a buffer in between the the gate and MCU then would that be fine.
     
  2. JDT

    Well-Known Member

    Feb 12, 2009
    658
    85
    You need to completely isolate the micro-controller from the motor circuit. Use a high-speed opto-isolator between the MCU and the MOSFET driver.

    Power your MCU from a separate power supply.

    Some interference suppression on the motor will also help.
     
  3. ashkarmalik

    Thread Starter Member

    Jul 1, 2012
    55
    1
    Sir,the main problem is only due to EMI radiation from the motor or is there any other problem that is causing the MCU to reboot
     
  4. tshuck

    Well-Known Member

    Oct 18, 2012
    3,531
    675
    Well, you'll need to post a schematic so that we can see what possible errors are floating about....

    ....sounds likeeither your source is not rated properly, or your resets are floating...
     
  5. jwilk13

    Member

    Jun 15, 2011
    228
    12
    Sounds like this.
     
  6. ashkarmalik

    Thread Starter Member

    Jul 1, 2012
    55
    1
    No problem with the floating Reset pin,they are connected up rightly :)
     
  7. DickCappels

    Moderator

    Aug 21, 2008
    2,647
    632
    You wrote:
    "If I can Huck up a buffer in between the the gate and MCU then would that be fine."

    My thought exactly; if you aren't driving the gate sufficiently, then switch will be more like a variable resistor, and both turn-on and turn-off can loose more power than usual. Put a MOSFET driver in the circuit.

    By the way, what is the power supply voltage of the ATMEGA?

    The EMI problem for the motor, as already stated, might be part of the problem.
     
  8. ashkarmalik

    Thread Starter Member

    Jul 1, 2012
    55
    1
    Atmega Vcc=5v via LM7805 regulator
     
  9. ashkarmalik

    Thread Starter Member

    Jul 1, 2012
    55
    1
    Atmega is been powered by a LM7805 regulator
    And am using the TI's gate drive
     
  10. DickCappels

    Moderator

    Aug 21, 2008
    2,647
    632
    Is "Ti's gate drive" a MOSFET gate drive chip from Texas Instruments, or do I misunderstand?

    Then it sounds like the next step is to clean up the motor drive, and by the way, make sure that the ATMEGA is well decoupled from the motor power supply. Also avoid ground loops and (this is important) put some emi filtering on the ATMEGA output that drives the gate driver (Output pin => 100 ohm resistor => [capacitor to ground + small RF choke from the ungrounded end of the capacitor and the input to the driver chip]).

    If the problem persists, please post the schematic.

    And one other thing: is the 15 kHz the ATMEGA's clock frequency? If so, I wonder whether you PWM is fast enough for your low pass filter (or the other way around) and also wonder about the cleanliness of your clock source.
     
  11. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    See sigline on bypass capacitors.
     
    ashkarmalik likes this.
Loading...