QN9080 BLE RX problem after powerDown

[arpad_toth]

Hello,

We have a serious problem with the BLE stack, in QN9080 after going to powerDown0 mode then waking up the BLE scan doesn't report back any beacons. Transmission (TX broadcasting) still works after power down.

For reception it receives exactly the same callbacks in BLE driver when I issue Gap_StartScanning(report duplicates), the only difference is that it doesn't report any tags after waking up.

Have you tested this?

Is there an example which uses powerDown + BLE reception?

Env: MCUXpresso Software Development Kit (SDK) 2.2.0 + IAR

Cheers

[estephania_mart]

Hello,

What do you mean by the scanner is not reporting the beacons ?

Do you have an sniffer log so we can guarantee that the device is sending correctly as well as what is the device sending ?

If you do not send the device to low power, does this implementation work ? Are you sure that you are enabling everything that is needed once it comes back from low power?

Regards,

Estephania

[arpad_toth]

Hello,

BLE scanning is constantly running, all the callbacks are happening, but we dont get anything via gDeviceScanned_c.

More than 20 BLE beacons are transmitting constantly in the office, I checked with nRF.

Reception with the same code works perfectly before going into powerDown, but it stops receiving after wake up.

I took the wake up BLE restoration code from TX examples (I guess that's why TX is working even after wake up).

But I didnt find example with RX + powerDown. But why do I get callback that scanning successfully started?

Please see attachment for power down and restoration.

I updated the SDK to MCUexpresso_SDK_v221, same happens.

Only way we can restore BLE is to do a cold start.

[estephania_mart]

Hello,

Still, could you please provide a description of what you are trying to do? Are you just scanning for the beacons? Are you filtering ? Are you considering that the device will not only get the packets from your beacons, but also the rest of the devices sending BLE packets ?

Do you have enough memory for that ?

Could you please check and add  the following memory debug definitions ?

#define gUsePanic_c 1

#define MEM_TRACKING

#define MEM_STATISTICS

#define MEM_DEBUG

#define MEM_DEBUG_OUT_OF_MEMORY

In case you have a memory issue, you will have an assertion and/or it will enter the panic function.

How are you restarting the BLE communication ? In the task of the device which error are you getting ? Do you have a hardfault while debugging ? In which state does the device remains?

Regards,

Estephania

[arpad_toth]

Hello,

Without sleep reception works for hours, I don't think it is a memory problem. Also I don't get any panics. Runs happily, just reception is not working. I'm using passive scanning, no filtering. I cannot attach a debugger because it won't go to sleep.

It is would be really difficult to debug on my side.

Do you have an example with sleep+BLE reception?

[estephania_mart]

Hello,

Sorry but we do not have that example you are looking for. 

It is really hard to debug this way as we don't really have any information, without a sniffer log or the error that you might be getting specially with only one fraction of the code like that.

If the code works correctly when you use it in without low power , but it fails when adding that part you might have some intrusive code there or the sequence is missing some elements compare to the examples such as the HRS.

Normally, when using the low power you turn off some of the pins and you need to add the functionality again when recovering, you can check that part and you need to ensure that you power up everything as correctly.

Hope this helps

Regards,
Estephania