Failed to create ID on LE-Audio DK

[Gerry]

Hi Nordic team

According to this post I eliminated all calls to cs47l63, ina231 max14690 pmic in order to migrate the project to my custom hardware.

After that I get the following log on both the audio DK and on my custom hardware:

HR [00:00:00.001,129] <inf> audio_sync_timer: Audio sync timer initialized
HR [00:00:00.001,770] <dbg> nrf5340_audio_dk_nrf5340_cpuapp: core_config: Setting gpiote latency to low power
HR [00:00:00.001,831] <dbg> nrf5340_audio_dk_nrf5340_cpuapp: remoteproc_mgr_boot: Network MCU released.
*** Booting Zephyr OS build v3.1.99-ncs1-1  ***
HR [00:00:11.464,447] <inf> fw_info: 
	 nRF5340 Audio nRF5340 Audio DK cpuapp 			    
	 NCS Version: 2.1.2 			    
	 Cmake run : Mon Nov 28 10:32:47 2022
HR [00:00:11.464,477] <inf> fw_info: ------- DEBUG BUILD -------
HR [00:00:11.464,477] <inf> fw_info: 
	 HEADSET right device
HR [00:00:12.030,792] <inf> board_version: Compatible board/HW version found: 0.8.0
HR [00:00:13.649,932] <wrn> ble: Failed to create ID
HR [00:00:13.653,137] <inf> fs_nvs: 6 Sectors of 4096 bytes
HR [00:00:13.653,137] <inf> fs_nvs: alloc wra: 0, f98
HR [00:00:13.653,137] <inf> fs_nvs: data wra: 0, 5c
HR [00:00:16.367,767] <wrn> bt_hci_core: Controller to host flow control not supported
HR [00:00:16.370,819] <inf> bt_hci_core: No ID address. App must call settings_load()
HR [00:00:16.370,880] <inf> ble: MAC: 00:00:00:00:00:00 (public)
HR [00:00:16.371,459] <inf> ble: Controller version: 3303
HR [00:00:26.212,768] <inf> cis_headset: Regular advertising started

settings_load() is called in main - on_ble_core_ready()

"ble: Failed to create ID" confuses me. However, the service is visible on nRF Connect.

Streaming starts on the Audio DK

On my custom hardware strm_state in streamctr.c is always STATE_PAUSED

even when I call this custom function in streamctr.c

int stream_run(void)
{
	int ret;

	if (strm_state == STATE_PAUSED) {
		ret = le_audio_play();
		if (ret) {
			LOG_WRN("Failed to start playing");
		}
		return ret;
	}
	return 0;
}

What am I doing wrong here?

Kind regards

Gerry

[Vidar Berg]

Hello Gerry,

Are you able to share your project here or in a private support ticket so I can try to debug it here?

Kind regards,

Vidar

[Gerry]

Hello Vidar

Thanks for your reply.

I attached the project. The code runs on the audio DK (AUDIO_DK = true; in main.c)

test_klaud.zip

kind regards,

Gerry

[Gerry]

Thanks Vidar, I really appreciate your help.

The connected_cb in le_audio_headset.c is called and has no errors. Anyhow the gateway shows no active connection (blue LED)

Also in the log of the headset there are no further entries later, when the device should be connected.

kind regards

Gerry

[Vidar Berg]

Hi Gerry,

Gerry said:

Also in the log of the headset there are no further entries later, when the device should be connected.

For me, the logging continued after the connection was made. Below is the hex file I built from the project you sent. Could you please program this to your board and see if you still get the same result (i.e. no logs after connection)?

7103.zephyr.hex

Also, the error from bt_id_create() was fixed by this commit: https://github.com/nrfconnect/sdk-nrf/commit/edbf678dc8342d0e6d7f77d32d9f633acd219379 

Kind regards,

Vidar

[Gerry]

Hi Vidar

Thanks for your help.

On the debug uart I can also see cis_headset Connected now. I think I just missed the connected logentry on RTT.

The id_error disapeared with your fix, thanks for that.

The log now looks clean to me. when I press the Playbutton on the headset, the log shows: streamctrl: Playing stream. Nevertheless, strm_state in streamctrl.c always stays on STATE_PAUSED. Do you also see this behaviour?

kind regards

Gerry

[Vidar Berg]

Hi Gerry,

I am sorry for the delay. I have not been able to spend time on debugging your application to figure out why the streaming state does not get updated. Have you made any progress on this in the meantime? 

Kind regards,

Vidar

[Gerry]

Hi Vidar, 

I spent a lot of time debuging but still no stream on I2s and the state is not changing.

Would be great, if you find some time to find out, whats going wrong here.

Kind regards

Gerry

[Vidar Berg]

Hi Gerry, 

I built and programmed the test_klaud.zip project you shared earlier with the following modifications:

 - Removed the adc module

- Commented some GPIOs that were not defined in the nRF5340 Audio DK board files

After this I was able to stream audio from another DK programmed with Gateway FW. But I did not expect this to work as I thought the cs47l63 had been disabled? Have you started to integrate support for MAX98357A?

Kind regards,

Vidar

[Gerry]

Thanks for your help Vidar

I'm still not able to stream. Can you tell me, what GPIOs you have changed in the board files?

The cs47163 related code should be eliminated like shown in this post. Do I need to disable something in the board-files also since I dont have the peripherial chips?

I would expect to have a I2s signal at the inputs of the MAX98357A, so I would only have to enable the chip with P1.12 right? 

[Vidar Berg]

I tested on the Audio DK and assumed that it wouldn't work as I thought interfacing of the cs47l63 had been disabled the code. There were some GPIO functions calls I had to comment out because the project would not build otherwise, but I do not recall which file was in.

Here is the projected I tested with:

test_klaud_mod.zip

Gerry said:

I would expect to have a I2s signal at the inputs of the MAX98357A, so I would only have to enable the chip with P1.12 right? 

Can you try to probe the lines with a logic analyzer or scope to verify if data is being transferred?

[Gerry]

Hi Vidar

I probed the I2s lines, no signal, even with your modified code.

After investing 100+ hours in this example project and some days in the board files, and also with the help of nordic in several tickets, i was unfortunately not able to get the code to work without the audio DK peripherals. It seems more difficult than expected. Uncommenting the peripherial calls like mentioned here doesn't do the job. Unexpected behavior wherever i change the code. our streaming hardware has only one button, 5 leds and needs the I2s signal. I could never see an I2s signal on the outputs.

Unfortunately, we have to stop the project or at least pause it, because the effort gets too great. It would be very helpful if nordic had an example project of how the audio DK works without all the peripherals. I think most developers need this as a starting point to develop custom hardware.

Anyway thanks for your support Vidar

Best regards

Gerry

[Vidar Berg]

Hi Gerry,

I am sorry to hear about your experience with our LE audio solution. I agree that there are things we can do to improve the developer experience, especially when it comes to development with other HW. I will forward your feedback to the team who is working on this.

Best regards,

Vidar