• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 471 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TPA3221: RESET not working properly, outputs stay off.

Gordon Cook
Prodigy 110 points
Part Number: TPA3221

This is regarding the reset of the TPA3221 power amplifier. The schematic is the same as the evaluation board and it is being used in the stereo BTL mode, single ended inputs. The gain is set to 30 dB master mode. Clock is set to 600 KHz, internal with a 50K resistor. 

PVDD is 27 volts, there is an external DVDD, AVDD, and GVDD +5 volt supply.

The problem is that setting the RESET from low to high does NOT always turn on the output stages. If the reset pin goes from high to low then back to high within approximately 200 to 600 mSec, then the outputs will turn on. If the reset pin is low for greater than about 800 mSec, then set high, the outputs remain off and held to ground. 

The  TPA3221 IC had been replaced once. 

The reset line is being driven from an ADAU1701 GPIO pin to 3.3V, but using the 5V to set the RESET high has the same behavior.

The +5 V supply is stable and not sagging. The PVDD supply is also stable. 

It doesn't matter if there is a load on the outputs or not. 

The FAULT line goes high to low for about 10 mS when the outputs turn on. It goes low for about 8mS when the outputs do not turn on.

I would appreciate any insight into this problem.

  • 0
    TI__Expert 6295 points

    hello,

      RESET from low to high does NOT always turn on the output stages---when it is not startup, can you help monitor AVDD and GVDD as well?

    we will do furher analysis.

    Regards

    Linda

  • 0 in reply to Xiaolin Qin
    Prodigy 110 points

    Thank you for your reply.

    VDD, AVDD, and GVDD are all stable at +5VDC +/- 50mV from an external supply.

    According to the TPA3221 data sheet, Section 9.4.1, bringing RESET from low to high after all supplies are stable will start the power up sequence. 

    I am using the RESET to put the power amplifier into a low power state to avoid switching PVDD on and off. This is a product with an auto turn on and off function and when it is off the power consumption has to be very low.

    Are you saying that using RESET in this way is not reliable? Is there a better way to accomplish a low power shut down?

    Regards,

    Gordon

  • 0 in reply to Gordon Cook
    TI__Mastermind 20750 points

    Hi Gordon,

    I've seen reset issues similar to what you have described when the customer used a 10uF DC blocking capacitor on the inputs instead of the recommended 1uF. 

    Was there a reason you chose to control the device in a low power state using the reset pin rather than the CMUTE pin?

    Regards,

    Robert Clifton

  • 0 in reply to Robert Clifton56
    Prodigy 110 points

    Hello,

    The product is a subwoofer with an auto on/off function. It detects the audio signal and turns on, and when no audio signal it turns off.

    It has to meet EU spec for power draw when "OFF" of less than 0.5W from the AC power source.

    In order to meet the "OFF" power draw specification  the PA IC must be put into a low power state, so I am using RESET rather than switch the PVDD supply on and off. The CMUTE function does not reduce the power enough.

    With 30 dB gain setting, the input caps are 10uF to pass the low frequencies..

    I have a fix, which is to cycle the RESET line high for 100 mS then low for  ~100 mS then back to high. This reliably turns on the output.

    Unfortunately this also delays the turn on somewhat, but that will have to be acceptable.

    Regards,

    Gordon Cook

  • 0 in reply to Gordon Cook
    Prodigy 110 points

    Hello,

    I set the input coupling capacitors to 2.2uF and that cures the RESET turn on problem, as you suggested. I can compensate the frequency response to correct for the increase in high pass frequency cut off.

    Thank you,

    Gordon Cook