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.

TPA3255: the waveform of output pins OUTA and OUTB is abnormal

Part Number: TPA3255

Some times ago I tested the TPA3255 amplifier.TPA3255 works in PBTL mode,supply voltage is 50.15V,signal frequency is 7.5kHz,  load impedance is 2 ohms pure resistance,TPA3255 output voltage is 31.8V RMS,so output power up to 505W.  And then i tested the waveforms of the TPA3255 power output pins OUTA and OUTB. The details are shown in the figure below. Blue and green are the waveforms at the OUTA and OUTB pins,Yellow is the waveform at the load end。As far as I understand, the waveforms at OUTA and OUTB should be completely inverted,but the actual waveform is not like this. As shown in the picture, OUTA and OUTB will simultaneously go high at a certain moment。Then, will the MOSFETs of the upper and lower arm inside the TPA3255 be turned on at the same time?Will this damage the internal MOSFETs? Hope your help.

Thansk

  • Hi user4101711,

    It looks like the only times you are seeing this occurrence is near the peaks of the signal (when the PWMs of the FETs are max and minimum duty cycle). It could be that what you are seeing is the pulse injector working. 

    Do you have the device connected in pre-filter PBTL or post-filter PBTL? Is input C and D grounded? 

    Regards,

    Robert Clifton

  • Hi Robert Clifton,

    Can you explain the specific meaning of "pulse injector working"?

    And i have a 2 ohms pure resistance connected in post-filter PBTL. All data is measured on the TPA3255EVM,so input C and D are grounded as shown below:

    What needs to be explained is that i tested the waveforms of the TPA3255 power output pins OUTA and OUTD  Instead of pins OUTA and OUTB. The test point is shown below:

  • Hi Jerry,

    If you want more details about the pulse injector I'd recommend taking a look at section 9.4.1.2 in the datasheet. But to put it simply, since the device has bootstrap architecture, you aren't able to drive a 100% duty cycle on the FETs since the bootstrap capacitors need to have a charge time, thus a special circuit is implemented that injects a pulse if the FET remains in an "on state" for a certain amount of time. 

    I forgot to ask if you also made sure that M1 and M2 are set appropriately for PBTL mode? 

    Regards,

    Robert Clifton

  • Hi Robert Clifton,

    Thanks for your reply. Does it mean that the TPA3255 works normally when pulse injector happening,it won't damage the chip? 

    Another question is that  is the OC threshold 34A for PBTL mode?Can i set the Rocp value to less than 22 ohms for a higher OC Threshold?

    Finally I made sure that M1 and M2 are set appropriately for PBTL mode.

    Regards,

    Jerry






  • Hi Jerry,

    It still seems to be a bit premature for pulse injection to be taking place since it's happening early. Are you using the EVM or your own board? If it's your own board, can you show me the schematic?

    The OC threshold is 34A for PBTL mode however the OC only looks at each full-bridge to determine if the threshold of 17A has been passed. So you don't need to adjust the OC threshold.

    Regards,

    Robert Clifton