RE: TAS2563YBGEVM-DC: PPC3 Software can not connect TAS2563 EVM

Hi Mikko,

As you mentioned on the other thread, you should be using the TAS2563QFN EVM version of the app.

The warning message is expected for now and can be ignored.

Have you verified that EVM is set as your default playback device and that volume is not muted? You can use any software to play audio such as media player or web browser.
As part of the debugging, you can try by going into "Device Control" panel, selecting "ROM Mode" from the top-right corner and click "Apply".
Do you see the IRQ LED, on the left side of the board, light up at any point during your test?

Best regards,
-Ivan Salazar
Applications Engineer

Ivan Salazar said:

you should be using the TAS2563QFN EVM version of the app.

Hello Ivan,

Thank you for your reply. Do you mean this on context of an advice or did you check from my TI account that I was approved with the correct QFN version of EVM App? I recall reading that difference between QFN and DSBGA was, that for instance the amount of EQ biquads is different. But I was unsure which one was which.

In my tuning tool, after applying Smart Amp in 4 order mode, I can see 5 available biquads in EQ. DRC has two bands, dynamic EQ single band. Would these match with expectations for QFN -version?

Ivan Salazar said:

Have you verified that EVM is set as your default playback device and that volume is not muted? You can use any software to play audio such as media player or web browser.

Yes, all seems to be ok here. TI EVM board is shown as USB sound card, volume control works. 

Ivan Salazar said:

As part of the debugging, you can try by going into "Device Control" panel, selecting "ROM Mode" from the top-right corner and click "Apply".

No, this does not help, however thanks for pointing me here as now I think we are closing to a solution, as

Ivan Salazar said:

Do you see the IRQ LED, on the left side of the board, light up at any point during your test?

No, it does not. It does quickly flicker when I attach power connections to VBAT and GROUND, but does not stay on.

On my image, speaker is not connected by purpose. As if I read IRQ messages under "Device Control," there are several active:

Further, If I remove  PSU wire from VBAT+ connection and measure output impedance from Amplifier + / - connectors, I get sensible result of 10.3kOhms.

However, if EVM is switched on, tunings are applied and connection is on, same measurement gives me just 13 Ohms, despite speaker driver is not connected! With the speaker I see 0.2 Ohm resistance. This status from 10.3kOhms to 13/0.2 Ohms switches when I enter to "Tuning and Audio Processing" module. It happens with my current tunings or with New Default Mono -setting.

Some sort of current overload protection? 

So to re-iterate, steps to reproduce this issue:

1. Apply 3.7V DC or 5.0V DC to VBAT + GROUND
2. Connect USB cable, activate connection to PPC3
3. Choose "New Mono" Tuning"
4. Measure Output impedance from amplifier output. You should see 10.3k Ohm (without driver, with driver you should see driver DCR)
5. Enter "Tuning and Audio Processing" -module, confirm that settings are correctly transferred
6. Re-measure Amplifier Output impedance. Now I see 13 Ohms or 0 Ohms with driver
7. Enter Device Control, check IRQ error status. I see Device Power Down, Power up and PLL Lock activated. 

One more thing.

Looking at this thread here:

https://e2e.ti.com/support/audio-group/audio/f/audio-forum/1357095/tas2563-audio-tuning-module-differences-when-using-tas2563qfn-evm-and-ppc3-evm-mb-tas2563ybgevm-dc

OP has both TAS 2563 EVM and TAS 2563QFN EVM -modules activated on his PPC3

Arthur from TI describes that QFN module is the one with 3 band DRC. Like I did mention on my previous post, I can only see 2 bands in DRC.

To me it seems that I was given/did apply for DSBGA version of EVM module and try to use with QFN board.
What do you think?

-Mikko

Hi Mikko,

You're using the QFN version of TAS2563 EVM, so I mean you should also use the QFN version of the software. WCSP and QFN are only compatible in ROM mode, if you try to configure WCSP device with QFN software in tuning mode it won't work, the same goes for QFN device using WCSP software.

Checking on the system, you've been granted access to the WCSP software only, you should get access to the QFN software soon but let me know if you don't.

The IRQ flags you see enabled are expected and don't represent an issue in itself.

I don't expect the device to work if you access the "Tuning and Audio Processing" panel right now, as you're using an incorrect version of the software.
However, I would expect you to get audio output if following the procedure I described of using "Device Control" panel only set to ROM mode. Alternatively, you can go to "Test and Measurement" panel which does a similar ROM mode configuration. Note that the moment you go back to "Tuning and Audio Processing" the device will be configured incorrectly.

Best regards,
-Ivan Salazar
Applications Engineer

Actually, Issue returned after some idle time for the EVM board. I will keep on trying stuff in order to find that what launches it and how to get the sound back

Hi Mikko,

Can you try setting your VBAT to something like 4V or 5V? If that helps, you may also try with lower like 3.5V and check if issue is reproducible.

There are some buck/LDOs in the board that would require 3.9V at least, so the 3.7V you're using may be just in the edge of operation limits.

Best regards,
-Ivan Salazar
Applications Engineer

Hello Ivan, quick question about this. I am looking at EVM board schematics now, can you please point me out to these buck/LDO's you mention?

Still think this is somehow related to over current. Today, I suddenly noticed my device was MUCH louder than yesterday, with no tuning change in between - I just disconnected everything and carried them to audiolab.

I was able to revive the board by starting tuning all again from scratch and adjusting master volume to -20dBFS 

Great, thanks Ivan.

I was able to narrow my "no sound" issues to story, where it systematically did happen on a situations where we started reaching the max output power ow 6.1W AND there was plenty of peak current. Attached is the acoustic impedance of the system I am working with - there is a large mid-range area which was close to 4 Ohm nominal load. With Sine-sweeps, my issue never happened. But with Noise-based signals or music, frequencies easily stack up and transient peaks can be surprisingly high.

I was able to make those "no sound" issues vanish by carefully tuning the VBAT limiter, DRC and temperature -part of the smart bass. Just generally making sure I do not push the EVM too hard. 

Hi Mikko,

Thanks for sharing all the details. This is great for later reference of TI and the community.

Using the limiter is a good idea to prevent large peaks and DRC can help raising the average amplitude to compensate for the reduced peaks.

Best regards,
-Ivan Salazar
Applications Engineer