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.

TPA3251: PFFB problem and OC protection issue

Part Number: TPA3251
Other Parts Discussed in Thread: PCM5242, ,

Hi, 

   When customer use PCM5242+TPA3251 device to devolpment  audio amplifier, we find some problem:

1. when use below PFFB circuit , it can work well, but the R_Z ( R3.3ohm/1W)   are burn up easily.  

     Another, when system shut dow,  it exsit POP noise.  pls. give some advise . 

2. When use the PCM5242+TPA3251 and refer the new PFFB circuit as below , TPA3251 can't work, can you help to check ? 

3. When TPA3251 set to BLT mode and for two channel woofer (4ohm), the PVDD is 32V. TPA3251 is easily to OCP . The ROC_ADJ set to 22Kohm. We can't find the reason. 

  • Hello,
    We are looking into this, Justin will get back to you soon later.
  • Hi Aggil,

    Sorry for the delay.

    1. Yes this was a potential issue with the old PFFB design, too much power at high frequencies was going through the resistor, which is why we recommended a 1W resistor. However since then we have updated the design to avoid this issue, allowing the use of a lest costly resistor.

    The shutdown pop is a bit large if nothing is done about it. One solution that may work depending on your surround circuit is to reduce the DC blocking cap on the input. This is because that DC cap is discharging through the feedback causing the pop after shutdown. If lowering the DC blocking cap does not bring the pop to acceptable levels, then I would advise adding a circuit that will enable a path for the DC can to discharge at shutdown.

    2. What circuitry is between the Rin and the PCM5242? Have you probed any of the feedback nodes or input at startup to get a sense of what is happening?

    3. Does this mean the TPA3251 is outputting at first and not immediately shutting down?

    Justin
  • Hi Justin

     Thanks. 

     We have found the answer about question 2 and 3, but there is another question need to be solved. 

    When we use new PFFB circuit, the shutdown POP   happened too. Can you give a way to solve it  ? 

  • Hi Aggil,

    The shutdown pop is a bit large if nothing is done about it.

    One solution that may work, depending on your surrounding circuit, is to reduce the DC blocking cap value on the input. This is because that DC cap is discharging through the feedback causing the pop after shutdown, and reducing the capacitor value will reduce the discharge amount and therefore the pop.

    If lowering the DC blocking cap does not bring the pop to acceptable levels, then I would advise adding a circuit that will enable a path for the DC cap to discharge at shutdown. This could be a FET with a resistor enabled when reset is set low, allowing the cap to discharge over this bleed resistor and the load, or you could cut off the load completely with a FET circuit.

    Justin
  • Hi Justin
    we will try to use 4 channel SPDT device to realize the DC cap discharge and feedback circuit cut off. thanks
  • Hi Justin
    We can't solve the OCP issue. when the audio signal is low, the AMP is ok, but when the audio signal is increase, it happen into the OCP easily. Would you pls. give some advise ?
  • Hi Aggil,

    What amplitude input causes the amplifier to hit OCP? When you say OCP is that the CLIP_OTW pin going low? Is it a solid low or a pulsed low? Does the amp shutdown or just clip?

    Justin
  • Hi Justin

      The input signal is as below ,pls. refer.  When protection hanppen, the CLIP_OTW pin is also going high, but the FAULT pin going low, and TPA3251 shutdown one of the channel  or two channel, it can't self clearing,  It must be restarted.. Attachment is the SCH and the PCB LAYOUT, and a vedio include CLIP_OTW, FAULT  pin waveform, would you pls. give some guidance to solve this problem ?

    M200MKIII+WIFI.pdf

    M200MKIII+WIFI.PCB

    CH1. 2 DIFF  INPUT

    CH3 FAULT

     

    CH2: CLIP_OTW

    CH3: FAULT

  • Hi Aggil,

    Thank you for all the information. I was able to play the video and look at the schematic, but the PCB file would not open. I got the following error:

    "ERROR(SPMHDB-238): The design is corrupted. It may have been copied from a different architecture using ASCII mode; copy using binary mode."

    From the video it looks like the TPA3251 might be hitting its current limit and continuing to operate through the first few bursts, and on the last one when the fault happens, the CB3C counter limit is hit and the channel shuts down. this would be the FAULT Low, OTW High result you see in that condition. This can shut down either an individual channel or both channels.

    If you are still having issues with the 3.3 Ohm resistor you can try making the series capacitor to this resistor a bit smaller. This will reduce the power dissipated in the audio band by a bit but be careful not to make it too small and loose stability. Unfortunately its likely that you can still hit more than 1W through the resistor in peak cases, but it would reduce the occurrence. Any new values must be tested for stability in worst case use cases for your system. This is one of the main reasons we updated the design of the recommended PFFB circuit.

    Justin
  • Hi Justin
    You can open the pcb latout file using ALTIUM DESIGNER or PROTELL 99SE or another tools, the file is ok.
    We connnet the same input audio signal to the TPA3251EVM, we test the CLIP_OTW waveform, it is as same as our PCB, But the EVM don't happen to protection.So we want kown how to optimize the circuit ?
    Also, it is the same result when we add the PFFB circuit or not .
  • Hi Aggil,

    Here is the information I have currently:

    PVDD: 32V
    Load: 4 Ohm
    OC ADJ: 22k (16.3A CBC)

    So it should not be able to hit overcurrent.

    Since you said one channel shuts down sometimes and not both, could you check the voltage on the VBG cap when operating? I have seen this with a customer in the past that the return current of the VBG cap was forced to cross the output return current causing the VBG voltage to raise. This causes one channel to shut off at high output.

    Looking at the layout there seems to be a decent path for the VBG to get back to the GND pins in the center of the device. Although I would like to verify this by measuring the voltage when operating and increasing the input. If this is the problem you sill see the VBG voltage rise as the input/output increases.

    Justin