Cannot get HFXO to start in Murata MBN52832 Module (nRF52832)

[SP2020]

Hello!  We are having a problem getting the HFXO to start running in the Murata MBN52832 module (nRF52832-based) so that we can start advertising.  The 32MHz crystal is embedded in the module, so the crystal hardware should be OK.  We do the:

NRF_CLOCK->TASKS_HFCLKSTART = 1;

while (NRF_CLOCK->EVENTS_HFCLKSTARTED == 0)

but never get out of the while loop.  If we skip the Bluetooth setup the processor runs OK (just without the radio).  Any thoughts or suggestions?

[SP2020]

Out of our prototype run we've had about a 60% failure rate where the only thing on the entire board that isn't working is the HFXO/Bluetooth advertising.  I acknowledge the possibility that it is a hardware/manufacturing issue--it's just strange that this specific thing consistently doesn't work right.

[Kenneth]

I believe an manufacturing issue yes. I assume the shield must be de-soldered for further investigation and debugging. Have you been in contact with Murata on this? If you are able to de-solder the shield you should be able to read the chip markings for instance, then I can check our production data if you want.

Best regards,
Kenneth

[SP2020]

Sorry for the long delay in replying to this, but in case anyone else has a similar issue, we determined that our board manufacturer was exposing the modules to too high solder temperatures.  They were able to change their process and we have had almost no fallout since then.