• State TI Thinks Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 736 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.

TPA2008D2: Problems with amp being unstable and shutting down

Peter Brunnengraeber
Prodigy 70 points
Part Number: TPA2008D2
Other Parts Discussed in Thread: PCM5122,

Hello all,

I'm trying to build an audio output stage for an IoT device that is based around a Raspberry Pi.  The built-in audio of the Pi is very poor and susceptible to cpu noise, so I have created a design utilizing the TPA2008D2 amplifier and a PCM5122 DAC.

I've been struggling for the last several weeks with this project because of issues with the amplifier.   It seems to be whenever there is even the slightest noise in the source audio signal, or even when I turn up the gain on the amp, it will shut down requiring I remove and re-apply power to restart it.  I've been on chat with the support team, but they have referred me here.   I will try to put as much detail as I can.

Here are the snippets of the schematics (pardon the messiness):

Note that D5 has been removed and U2 is bypassed.

The 1nf caps are documented on the datasheet page 14/15 for the application note of longer wires from the amp to the speaker. I have 13cm from the PCB to the headphone jack and from the jack to the internal 8ohm speakers. The ferrite bead is on the cable coming off of the PCB.

The 470uF caps are currently bypassed per my discussion with the customer support team, but was based on what I see recommended in the SLOU011 datasheet (pg. 28 / 3-12 ; C2 & C3) for the evaluation platform for this amp.

I've attempted a separate board populated only with the amp and it's components.  Current draw was only showing 18ma draw at 5.07v.  The scope measurement is at the audio output jack.  I see the switching and this looks good to me.

I did attempt to put some additional 0.1uf caps between the audio signals (test both with phone and DAC) with no effect on keeping the amp running. I also tested adding a 50k ohm resistor in line in addition to the cap with similar results. I was able to get the amp to play back audio from the DAC by adding 0.1uF caps between the audio input to the amp and ground, plus a 2.2nF cap between the output of the DAC and the input to the amp. There is some distortion noise in the background and the amp still was shutting down occasionally. I also couldn't turn up the gain much past half way or it too would cause the amp to shutdown.

Might there be something with isolation? I noticed also that the DAC output while the amp was off was averaging at 280mV Pk-Pk (see left), whereas when the amp was running I was seeing 1.73 Pk-Pk and pulses at the switching points (see right).

So, I'm at a loss for what to do next... Is there something that I missed on the evaluation platform?

Thank you kindly.  -Peter.

  • 0
    TI__Guru* 84915 points

    Hi, Peter,

    Welcome to E2E and thank you for your interest in our products!

    Do you have details about the power pins of the TPA2008D2 before and after sending an audio signal? Are they stable or do you also see a noise issue there?

    Best regards,
    Luis Fernando Rodríguez S.

  • 0 in reply to Luis Fernando Rodriguez S.
    Prodigy 70 points

    Hello Luis,

    The supply is measuring 5.07v at the power pins and the draw with the amp running (no audio) is 18ma.  I hadn't looked for noise at the power pins, so I will do that as well as check the consumption while the amp is running and report back.

    Also I noticed I had forgotten to mention...  The scope screens in my first post: yellow trace were at the test point between DAC output and amp input on the left channel positive, green trace was on the left audio output positive.

    Thanks again!  -Peter.

  • 0 in reply to Peter Brunnengraeber
    Prodigy 70 points

    Hello Luis,

    Ok, so you were right to have me look at the power pin with the scope. Some very strange results here...

    I ran my audio test routine which is spoken word first only from the left, then only from the right, then both channels, and then out of phase on both channels.

    The voltage measured on the scope stays stable at 5.05V the entire time.  The yellow trace in the images below are the 5+ rail and the green is still the left audio output.  First the scope when no audio is playing:

    Once the audio is playing:

    There is indeed noise on the +5 rail when the audio is playing.  I turned off the trigger to try and capture this... there is what seems to be the audio waveform as distortion on the vcc pin.  When this noise gets too high, that is when the amp shuts down.  Playing audio out of just one channel or out of phase stereo both cause this on the +5 rail.  Normal stereo audio does cause some, but very little of this distortion.

    Even stranger, the current draw when audio is playing out of only one channel bounces around between 19ma and 25ma.  When audio is playing from both channels, the current draw is stable at 19ma.

    So this makes sense that when there is too much noise on the positive rail, that it is probably destabilizing the amp oscillation and causing the shutdown.  What do you suggest next?

    Thanks!  -Cheers, Peter.

  • 0 in reply to Luis Fernando Rodriguez S.
    Prodigy 70 points

    Luis Fernando Rodriguez S. said:

    Do you have details about the power pins of the TPA2008D2 before and after sending an audio signal? Are they stable or do you also see a noise issue there?

    Hello [user="Luis Fernando Rodriguez S."],

    Do you have any further suggestions on what I can try next?   I tried some additional decoupling caps closer to the chip, and I added some beefier electrolytic and film caps to the power pins without much success.

  • 0 in reply to Peter Brunnengraeber
    TI__Guru* 84915 points

    Hi, Peter,

    Thank you for the details you provided.

    One more test, could you use a separated power supply on the TPA2008D1? I mean, could you try with an independent supply separated from the circuit +5V rail?

    Best regards,
    Luis Fernando Rodríguez S.

  • 0 in reply to Luis Fernando Rodriguez S.
    Prodigy 70 points

    Hello Luis,

     Sorry I had not mentioned it, but yes the last post was with a separate bench supply just for the amp.

    Thanks!  -Peter.

  • 0 in reply to Peter Brunnengraeber
    TI__Guru* 84915 points

    Hello, Peter,

    Thank you for your response. During your test, do you power off the rest of devices in the board and just test the TPA2008D2 with the external supply? Do you still see this issue under these conditions?

    Best regards,
    Luis Fernando Rodríguez S.

  • 0 in reply to Luis Fernando Rodriguez S.
    Prodigy 70 points

    Hello Luis,

    Yes...  So I ran the amp just by itself and connected my mobile phone as a sound source.  The phone was running on battery, so it was isolated.  I generated test tones as the audio source.   Photo #7 with tones playing out of only the L channel I had 53ma draw.  Photo #8 is with tones playing stereo and I had a 18ma draw.  You can see the disturbance on the power rail (yellow trace).  The green trace is the left channel output.

    #7  

    #8

    I did need to keep the 0.1uF caps between the each channel of the audio source to ground and a 2.2nF cap between each channel of the audio source and the input to the amp.  Removing the 2.2nF cap caused the amp to shutdown as soon as there was an imbalance of audio between the L and R channels.  This cap is in addition to the original 0.1uF caps on the audio inputs from the reference design.

    Thanks!  -Cheers, Peter.

  • 0 in reply to Peter Brunnengraeber
    TI__Guru* 84915 points

    Hi, Peter,

    Thank you for the details you provided. One more question, are all these measurements made with an out-of-band filter? I mean, is all the noise you are getting in the audible band? As far as I understand, some of our audio devices have integrated modulators that add some high frequency content at the output. Could you try adding a measurement filter as mentioned in this App Note?

    https://www.ti.com/lit/an/slaa313a/slaa313a.pdf

    Best regards,
    Luis Fernando Rodríguez S.

  • 0 in reply to Luis Fernando Rodriguez S.
    TI__Guru* 84915 points

    Hi, Peter,

    Do you have any feedback on this?

    Thank you.

    Best regards,
    Luis Fernando Rodríguez S.

  • 0 in reply to Luis Fernando Rodriguez S.
    TI__Guru* 84915 points

    Hi, Peter,

    I'm closing this E2E thread for now. But please feel free to provide more information and details about this issue. We will be glad to help you.

    Best regards,
    Luis Fernando Rodríguez S.

  • 0 in reply to Luis Fernando Rodriguez S.
    Prodigy 70 points

    I am currently out of the office.  I will take a look next week when I return.

  • 0 in reply to Peter Brunnengraeber
    TI__Guru* 84915 points

    Hi, Peter,

    Please let us know if you have any update on this. We will be glad to provide additional support.

    Best regards,
    Luis Fernando Rodríguez S.

  • 0 in reply to Luis Fernando Rodriguez S.
    TI__Guru* 84915 points

    Hi, Peter,

    I need close this E2E thread for now since this E2E case has been open for a month. However, please feel free to post your results on a new e2e thread. We will be glad to continue the support.

    Thank you!

    Best regards,
    Luis Fernando Rodríguez S.