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.

TAS3204 noisy

Other Parts Discussed in Thread: TAS3204, TAS5611, TAS5612

Dear Suzan,

I have a noise problem using the TAS3204:

In order to get a hight output from the TAS5611 I have to amplify the signal coming from the TAS3204EVM by 4 giving me a unacceptable high noise level. 

The TAS3204EVM was a bit noisy ever since but I could kind of live with it. The Amplifier and the TAS5611 do have very little noise. 

I would really need help with that issue as the system (TAS3204 + TAS5611) should go to market in January with a annual quantity f 10000.

Any help of how to reduce TAS3203 noise please.

Thanks,

Thomas

  • Thomas,

    What is the charateristic of the noise? I assume that it's the analog gain you're giving, right? have you tried to apply the gain in front of TAS3204, whether it has the same amount of noise?

    by 4, meaning 12dB gain?

    Regards,

    Susan

  • Thanks for your reply Susan,

    when streaming audio from the input to the output I get this noise pattern: I have a random noise with about 50mV, wich is already to much. But there is a 400mV burst of 2.7MHz repeated every 115kHz. My power supply is a standard transformer with a linear regulator on 470uF.

    The noise increases with my process flow running.

    Yes I have to amplify by 12 dB to get the 100W in 4Ohm out of the TAS5612 at 1V input to the system.

    Maybe you send an FAE to sort this out. It is pretty urgent.

    Thanks for your help,

    Thomas

    DeToma GmbH

  • Thomas,

    I have brought up this isse to my boss, I will let you know when I hear anything back.

    Regards,

    Susan

  • Susan,

    it is getting urgent over here. I was told that we already have some thousend EUROs worth of TAS3204 here, produktion is to start in January and I still have not solved that noise problem.

    I need support !!!

    Thanks,

    Thomas Deuble

    DeToma GmbH

  • Hello Thomas,

    We took a look at your situation, seems the problem is not related to the TAS3204.  We believe it's a system hardware related.

    You mentioned you see a 2.7 MHz , so with Fs = 44.1 kHz the SCLK (or BCLK) is 64 X 44.1 k = 2.8 MHz.  It could be the issue you see.  If your sampling rate is 48 k then the SCLK is 3.07 MHz.  You may try something to test this theory?  If so, we suggest you dampen the I2S clocks and provide good decoupling on the TAS5611.

    Others could be related:

      - PCB ground: recommend not to have split grounds, should have one common ground.
      - DC-DC buck converter: Still one common ground but should have very good layout and decoupling.

    Anyway, we don't believe the issue is casued by your TAS3204 designs, should be something else hardware related.  Hope it helps.

    Regards,

    Li Zhang

     

  • Alright I did remove the 2.7 MHz and the noise is still there of course.

    Can I please have an FAE over here to take a look at my project.

    As I said we have to start production and I would really expect TI to give me some support on that!

    Thanks.

    Thomas Deuble

    DeToma GmbH

    Lauenbühlstr. 59

    88161 Lindenberg

    Germany

  • Hey, I have to get in on this topic because we have the same problem with the TAS3204 -  we're getting audible noise at the output, the design is based mostly around the EVM schematic as a reference, and we're not using the digital inputs.

    We have production level ready to go, and this issue is just cropping up. Interestingly we didn't seem to have this problem with the EVM and it makes me wonder if there has been a batch problem with the DSP?

  • Also, we've confirmed that the noise is much different in characteristic and level between different TAS3204 chips if we swap them around.

    It really does seem to be chip quality dependent.

  • David, nice to learn that we are not the only once.

    Our design is based on the eval board schematics as well. 

    I even incorporated active emphasis and de emphasis that makes the noise a little better but not acceptable.

    Please see the schematics attached.

    @TI: Really need need help on this now. I would be prepared to send the prototype as well.

    Thanks.

    Thomas Deuble

    DeToma Germany

  • Since this is not something I could help with, I've informed the management team to see how we could help.

    And also some people are still on vacation, please be a little patient with us.

    regards,

    Susan

  • David,

    Just a thought: since you mentioned that you didn't detect that kind of noise on the EVM and suspected that there's some issue with the chip quality,  Have you tried to take the TAS3204 chip on the EVM and put it on your board to see what happened?

    Regards,

    Susan

  • Heya Susan, 

    Thanks for the  update, it's appreciated.

    Actually yes, that's one of the things we considered trying today, but we're not keen to risk the development board, and also we're still investigating the problem ourselves, if it happens to be that it's our fault that the chips are producing noise then I don't want to break the good chip.

    A quick search on here at the same time revealed that we're not the only ones to be suddenly having this problem now suddenly with the current chips in our stock, hence jumping on. If we come to absolutely no conclusion, I will try to convince the CTO that it's a good idea to try the known chip on the DEV board, but those things are expensive, we would prefer not to destroy it,

    Cheers

    David.

  • Oh, and also for some reason I no longer have access to the TAS3204 EXTRANET - that's the second time in a few months. Are you able to re-instate it or do I have to contact someone else?

  • Sorry, David, I am not in charge of the extranet, it could be under maintenance. If you still couldn't access it tomorrow, please let us know, I could try to find the person who's in charge of this.

    Susan

  • David,

    You shouldn't "lose" your access if you already have one.  Anyway, you may try to contact the below mailing list, and fill in some required info:

    purepathstudio@list.ti.com

    Regards,

    Li Zhang

     

  • DeToma,

    We will have someone take a look at the schematics, although I don't anticipate we can find anything, you did mention you follow the EVM schsmatics, right?

    Regards,

    Li Zhang

     

  • David,

    I would say it's high unlikely.  TAS3204 has been in the market for long time, many customers used the devices for years and certainly we had many batches in the past. The product (and all other TAS3x devices) is NRND now, not because it has "issues", simply it's an "old" device.  One major reason is the development tool for the device, the PPS tool for TAS3x was developed some time ago, it doesn't support the latest Windows environment.

    Regards,

    Li Zhang