• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 145 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.

TAS5508C: Delay in output channels 4 and 8 from the PWM section?

Moritz Götz
Prodigy 10 points
Part Number: TAS5508C
Other Parts Discussed in Thread: TAS5508, TAS5558, TAS5548, TAS5352A

We use the TAS5508C and we measured a delay between channels. We used differnt Chips which all showed the same result.

We could measure a delay of 2,7us between Channel 1 and 4 and Channel 1 and 8. Channels 1,2,3,5,6,7 had no delay between them.

The interchannel delay in the tas5508 was set to 0. All channels were configured as default.

We tried different configrations of the signal processing in the tas5508, e.g. send the same input signal to the output via the output mixer, and we suspect the delay coming from the pwm section

For the TAS5508A there was a one sample delay (described in the Errata for the Chip) for Channel 3,4,7 and 8 in the PWM section, which is still there in the TAS5508C (the biquads for the channels are configured to implement one sample delay as default ).

Do you know of this delay and can you confirm this behaviour?

  • 0
    Prodigy 10 points

    Btw. the used Samplerate is 48kHz.

  • 0 in reply to Moritz Götz
    TI__Intellectual 1490 points

    Hi,

    TAS5508C has the errata corrections for all TAS5508x and TAS5518x. 

    So the Errata is not suitable for TAS5508C.

    TAS5508C is so old that we don't have comprehensive data to support it. Sorry for that.

    If the delay is 2.7us (not 2us and 7us) between channel 1 and 4 and channel 1 and 8?

    And if the delay is 2.7us and PWM fre is 384khz/352.8khz, maybe it is a delay of one upsampling period.

    If you are sure that channel 1 has delay, you can use BQ or  interchannel delay to compensate.

    And we recommend you use some new amps just like TAS5548/TAS5558 to replace TAS5508C.

     

  • 0 in reply to Donny Peng
    Prodigy 10 points

    Hey, thanks for the reply.

    The delayed channels are 4 and 8 and the delay is around 2.7us.

    The biquad is not suitable, the smallest delay possible at 48kHz is around 20us. And the interchannel delay is also not suitable, because its not high enough, around 1us. We already tried that.

    In the end we compensate the delay by implementing subsample delays with fir filters before the TAS5508.

    We just wanted to check if you can confirm the behaviour of the chip.

  • 0 in reply to Moritz Götz
    TI__Mastermind 34680 points

    Hi 

      Donny is on a sick leave, he'll check if there's still EVM at hand to check this behavior. 

      Could you please help confirm how do you test this delay? Is there some waveform could show us? Thanks. 

  • 0 in reply to Shadow He
    Prodigy 10 points

    There is a screenshoot of a measurement:

    All 8 Channels get the same signal, a 10kHz Sinus. In our hardware the TAS5508C pwm outputs go to 4 TAS5352A Amplifier. All the 8 output channels are equally designed. Instead of a loudspeaker connected to the differential outputs the are 3,9Ohm Resistors connected. The measurement was done on the outputs pins of our hardware.

    The measurement was done on different hardware boards to be sure that it´s not the effect of tolerances of the used parts.

    The TAS5508 has the default configuration except that the interchannel delay for all channels is 0.

    CH1 is connected to Channel 1

    CH2 is connected to Channel 4

    CH3 is connected to Channel 5

    CH4 is connected to Channel 8

  • 0 in reply to Moritz Götz
    TI__Mastermind 34680 points

    Hi 

      Thank you.

      Sadly we still didn't find any document left to prove this point. We'd wait Donny back to office and check if possible doing some test about it.