1. Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
    Dismiss Notice

Generating uni-polar PWM signal using IR2113

Discussion in 'Homework Help' started by Ng Jing Xi, Nov 13, 2013.

  1. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,445
    Likes:
    937
    Location:
    Canada, of course!
    I think you are WRONG.
    The high-side IGBT must be driven so that its source charges the capacitor to a voltage ABOVE the supply voltage so that the gate voltage pulse is about 10V higher than the source voltage.
    It is a charge-pump.

    The power supply is a mess and I don't know why it is not fixed or replaced.
    I have a suspicion that the L-in must not be inverted because when it goes high you want the low-side IGBT to turn on and pull the output down.
     
  2. Ng Jing Xi

    Ng Jing Xi Member

    Joined:
    Oct 1, 2013
    Messages:
    97
    Likes:
    0
    I tried changing the power supply and the same issue occurs. Could it be that the boostrap capacitor is not of the correct sizing?

    I will post up the waveform of the Ho and Lo later
     
  3. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,445
    Likes:
    937
    Location:
    Canada, of course!
    Your circuit inverts the Lo signal that might be wrong, then the IGBTs might be both turned on at the same time and are shorting the power supply.
    The datasheet should explain the calculation for the value of the bootstrap capacitor. If its value is too low then it will not cause your problems and odd waveforms.
     
  4. dave

    Dave New Member

    Joined:
    Jan 12, 1997
    Messages:
    -
    Likes:
    0


     
  5. JoeJester

    JoeJester Active Member

    Joined:
    Nov 6, 2009
    Messages:
    699
    Likes:
    105
    Location:
    Grand Prairie, TX

    Ok,

    You have VS connected to +Vout. VS is the VB return. You need to draw whatever your going to load with this thing to give us a complete drawing.

    If you use a dual trace scope, the HO should, except with a slight delay, follow the HIN signal. Same applies to the Low side.
     

    Attached Files:

  6. Ng Jing Xi

    Ng Jing Xi Member

    Joined:
    Oct 1, 2013
    Messages:
    97
    Likes:
    0
    These are the waveforms that I manage to capture. The channel 1 shows the TP(IGBT1HIGH), while channel 2 shows TP(IGBT1LOW),
     

    Attached Files:

  7. Ng Jing Xi

    Ng Jing Xi Member

    Joined:
    Oct 1, 2013
    Messages:
    97
    Likes:
    0
    The rough idea is this drawing below. The Vout will be connected to a filter circuit to get a 50Hz sine wave
     

    Attached Files:

  8. alec_t

    alec_t Well-Known Member Most Helpful Member

    Joined:
    Jul 10, 2011
    Messages:
    9,255
    Likes:
    1,218
    Location:
    Cardiff, Wales
    Could you please shrink your image file size down before posting? For people on capped data tariffs or with slow internet connections large files are inconvenient.
     
  9. JoeJester

    JoeJester Active Member

    Joined:
    Nov 6, 2009
    Messages:
    699
    Likes:
    105
    Location:
    Grand Prairie, TX
    Ok, your going from a low level 50 Hz sine wave to a 12 V PWM, to a larger level PWM, and back to a larger level 50 Hz.

    Now, your diagram doesn't indicate the part numbers for Q1 through Q4. I assume there is a junction where Q1/Q2 connection crosses the +V out line.

    I also see +VDC and -VDC ... what are those values?
     
  10. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,445
    Likes:
    937
    Location:
    Canada, of course!
    The digital 'scope makes a big mess of the PWM waveforms.
    I expected to see a 50Hz sinewave there but I see only 800Hz very rough mountains.
    The PWM rectangular waves are not even rectangular, instead they are a mess.
    I wonder why the HIGH has half the amplitude of the LOW and has a lot of DC on it?
    I wonder if the power supply voltages are still jumping up and down?
     
    • Like Like x 1
  11. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,445
    Likes:
    937
    Location:
    Canada, of course!
    The size of the file is HUGE because it has a dark grey background instead of white. It seems to be a photo taken in moonlight.
     
  12. JoeJester

    JoeJester Active Member

    Joined:
    Nov 6, 2009
    Messages:
    699
    Likes:
    105
    Location:
    Grand Prairie, TX
    At 250 uS/cm with a 68 kHz Triangle wave, I'm not surprised the PWM signal looks terrible. We are looking at 0.015th of the modulating frequency and 17 cycles of the triangle wave per cm.
     
  13. shortbus=

    shortbus= Well-Known Member

    Joined:
    Mar 23, 2008
    Messages:
    1,214
    Likes:
    75
    Location:
    youngstown, oh
    Thats what I'm trying to say. Just don't have the literary skills to get it a crossed.

    If the high side is off at the same time the low side is off, the boot strap won't charge. He is controlling the Hin and Lin on the same gate driver with the same signal.
     
  14. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,445
    Likes:
    937
    Location:
    Canada, of course!
    Lin is Hin that is inverted. Then maybe the output is trying to go high and low at the same time which messes up everything, especially the power supply voltages.
     
  15. Ng Jing Xi

    Ng Jing Xi Member

    Joined:
    Oct 1, 2013
    Messages:
    97
    Likes:
    0
    The inverter should produce a 12V unfiltered waveform.
    For the IGBT, i am using TO-220FP, STGP10NC60KD.

    Sorry, What do you mean by "where Q1/Q2 connection crosses the +V out line."

    For the +VDC and -VDC, i am using a +12V from the DC-DC converter. But This DC-DC converter is also powering up the IC chips in the circuit. Is it wrong?
     
  16. Ng Jing Xi

    Ng Jing Xi Member

    Joined:
    Oct 1, 2013
    Messages:
    97
    Likes:
    0
    Actually my DC supply is okay but whenever I connect it to the IGBT, it just get messed up. I think it's because of the driver circuit that is causing this effect.
    When I connect the -VDC of the IGBT to the GND of the overall circuit, the power supply goes haywire.
     
  17. Ng Jing Xi

    Ng Jing Xi Member

    Joined:
    Oct 1, 2013
    Messages:
    97
    Likes:
    0
    I am sorry for the large file. I got no idea why the file is so big. I took the picture under the white light. Hence, I got no idea why is there a dark grey background.
     
  18. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,445
    Likes:
    937
    Location:
    Canada, of course!
    Can you post a photo of the two-channels 'scope showing the high frequency gate signal at the HIGH IGBT and the high frequency gate signal at the LOW IGBT at the same time to see if both are turned on at the same time?
     
  19. Ng Jing Xi

    Ng Jing Xi Member

    Joined:
    Oct 1, 2013
    Messages:
    97
    Likes:
    0
    For your better overview of the waveforms, I increased the sine wave to 5kHz so that you can view the PWM signal more accurately.
    Channel 1: IGBT 1 Gate Signal
    Channel 2: IGBT 2 Gate Signal
     

    Attached Files:

  20. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,445
    Likes:
    937
    Location:
    Canada, of course!
    I was correct from the beginning. The IGBT's are fighting each other. Both are turned on at the same time with one pulling high and the other pulling low causing very high power supply current but no signal at the output. No wonder the power supply voltages are messed up.
    Also it looks like something is reducing the level of the HIGH gate drive signal.

    The IR2113 must already be inverting one input but you are also inverting the LOW input.
    If you do not invert the LOW input it will be fixed except it will have no dead time between switching the output high and low.
     
  21. Ng Jing Xi

    Ng Jing Xi Member

    Joined:
    Oct 1, 2013
    Messages:
    97
    Likes:
    0
    But is the dead time necessary for the H-bridge inverter.
    Let me try that out and I'll get back to you.
     

Share This Page