BG95-M3 Needed to be power Cycled


I have a device in the field that lost connection. As it was in the field we could not observe the UART logs or what AT commands are failing.

But we had to remove the power from the system and then re apply it.

Has any one else experienced where a the a PCB need to be power cycleed as the BG-95 was not responding?


You mean lost connection on UART?
Have you set HW flow control? and are you using QUECTL EVB board?

Linkin WANG


It is hard to say if it was UART or if the device has gone into an unknown state.

I think it was the device and not just the UART. As we needed to remove power and re apply it, instead of a simple reset.

We are simply using UART via a transistors , with TX and RX pin.


yes, we are facing the exact same problem on the BG95-M2 / M3.

Some modules (approx. 5 in 300) sometimes enter an unknown state and it is not possible to boot them - neither via PON-TRIG nor PWRKEY. Disconnecting the power supply (and shorting capacitances) sometimes fixes the problem, but it occurs on these modules again after a while.

Maybe production outliers?



What is your power source? Is it battery ?

Thank you for your reply.
We haven’t encountered these issues before. If possible, could you please share your operation and schematic to rule out the possibility of these two factors?

And we think the problem is more likely caused by software dumps. Do you have test points on the USB_VBUS, USB_DM and USB_DP pins? Can you try to connect the USB interface and see if there are any dump logs for that port?

Well, I have no software dumps, and this is in the field.

I am suspecting it do with battery and voltage drop, below 3.3V. I might be wrong.

I will try to replicate this in our office to see if I can drop the voltage and not recover.