• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 83 subscribers
  • Views 1763 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

3-level SVPWM for inverter

erka
Prodigy 70 points

I had been using 3-level SVPWM implementing in C2000 F28335.

We use V/F or Sensorless Vector Control for 3-level NPC inverter (3.3kV AC). 

Before using Vector control we identify stator resistance by dc injection method. At that time our inverter's IGBT burned. You know HV-IGBT is very expensive.

So we checked the PWM output by Typhoon HIL602 product. The result of U phase shown like (0,1,0,0) or (0,0,1,0) very shortly (about 8.5~12us) some time (not usually) . Actually our deadband time is 6us. 

How to solve this(0,1,0,0) or (0,0,1,0) problem?

  • 0
    TI__Mastermind 27240 points

    Hello,

    We don't have an example project for this PWM and we have no idea what your implementation code is.

    I would recommend that until you verify the pwm gen is correct, use low voltage and limit the power supply current to clamp at some safe lower level so that you not only protect the IGBT, but also the motor.

    rgds,

    ramesh

  • 0 in reply to Ramesh Ramamoorthy
    Prodigy 70 points

    Thank you Ramesh

    I also testing in 400V &1200V inverter. But there are no problem on 400V &1200V inverter.

    Actually how to check by Lecroy scope for triggering this (0,1,0,0) or (0,0,1,0) signal?

  • 0 in reply to erka
    TI__Mastermind 27240 points

    It is getting too particular and is difficult to give a straight answer. If you don't see problems with 400V and 1200V inverter, then I will have to suspect the turn off rate of the high voltage igbt. You may consider increasing the dead band. As for the scope is concerned, you need to figure out.

    In any case, I would verify the scope waveform without inverter and make sure the pulse pattern is all right before feeding in dc bus.

  • 0 in reply to erka
    Intellectual 655 points
    If you have access and the possibilities to modify the hardware, this would be my approach:

    Implement a fast safety mechanism on your DC link bus that disconnects the inverter from the DC bus in case of a short circuit, e.g. as in [1] or similar on IEEE. Alternatively, measure the DC link current or voltage and use the EventTrig function on your DSP to trip a fuse/ turn off the IGBTs (in case the issue lies in the PWM generation rather than false triggering due to EMC problems).

    Then measure the event of the safety circuit together with your gating signals of interest. You can use the capture function on the scope that triggers when the pre-defined event happens. This could give you the possibility to better track down the issue.

    Also, since you're using IGBTs, I recommend advanced gate driver circuits that have a short circuit protection for the IGBT implemented, such as the ACPL-333J-000E. This should further reduce the risk to blow your IGBTs.

    Lastly and most importantly:
    High voltages together with big DC link capacitors are a dangerous combination. You might want to consult a professional to help you in the lab before putting your fingers into the system. Until then, I highly emphasize Ramesh's comment to stick to lower voltages.

    Cheers,


    [1] Wolf, C. “Electronic Fuse for Prototype and Production Test.” Pcim'99. Europe. Official Proceedings of the Thirty-ninth International Power Conversion Conference (1999): 423–6, 423–426
  • 0 in reply to Ramesh Ramamoorthy
    Prodigy 70 points

    Thanks for answering

    Actually we use Concepts gate driver(1SP0335V2M1) for IGBT( in case of 3.3kV AC)

    I also guessing our dead time (6us) is maybe not enough for HV-IGBT. I will increase dead band later.

    Our inverters dead time is 400V -> 2us, 1200V -> 4us.

    But this signal (0,1,0,0) or (0,0,1,0) why occur in 3.3kV, so i do not understand.

    Before supply PWM for 3.3KV  inverter, I always check the PWM, ouput voltage, current in 400V inverter for driving motor full load.