BG95-M3 stops responding to commands after a while

Hi, I’m having a problem with a BG95-M3 module. The system waits to get a GPS fix, requests the location with the QGPSLOC command, then disables the GPS and switches to NB-IoT network to send the data over MQTT.
This works fine, but after approximately 20 minutes the module stops responding to the commands send through the UART for about 5 minutes. Then, the module recovers itself and starts responding again.
I’m wondering if someone has experience something like this, or if this is a known bug.
The firmware version of the module is BG95M3LAR02A03_01.014.01.014.
This is an example of the commands that we are using

enviar_comando("ATE0")
enviar_comando('AT+QGPSCFG="outport",none')
enviar_comando("AT+QGPS=1")
enviar_comando('AT+QGPSCFG="priority",0')

for i in range(10):
    enviar_comando("AT+QGPSLOC?")
    sleep(10)

enviar_comando('AT+QGPSCFG="priority",1')
enviar_comando("AT+COPS=0")

After this, we send the data over MQTT and starts all over again.

Thanks!

pls run the command to disable psm and edrx, and try again

AT+CPSMS=0
AT+CEDRXS=0,5

Thanks for your reply. They are both disabled. Also, we are measuring the current consumption, and when it get in that locked state, the module is constantly using about 50mA - 60mA of current.
This usually happen when we are waiting to get a fix from the GPS. The module stops responding to the AT commands, and after some minutes it starts responding again.

You can first try to disable the GNSS function, and then re-perform the debugging.

I forgot to mention that after sending the data through MQTT, we are turning off the BG95 with an external microcontroller in order to save battery life.
In a new test we tried keeping the module running all the time and it worked fine.
Maybe shutting down the module and later turning it on has something to do with this problem?
Thanks.

Instead of turning off the module, now we are using PSM mode and it worked fine for a longer time. But eventually it stoped responding to commands again. I attach a graph showing the current consumption. The horizontal axis is in seconds and the vertical axis is in Amperes. You can see at the end the moment where the module locks up and stops responding to commands, and keeps drawing around 60mA of current.
Thanks!

The interval between the two high pulses seems to be 900 seconds. Is this a keepalive?

The system works as follows:

  • Enables GPS and waits for a fix.
  • When it gets the position, GPS is disabled and NB IoT is enabled.
  • The coordinates are sent through MQTT to a server.
  • The BG95 enters in PSM mode and the microcontroller also enters in low power mode.
  • After approx. 900 seconds, the MCU wakes up the BG95 and the process starts all over again.

The first two pulses that you see in the graph shows the system working as described in the previous list. But in the third measurement cycle, the BG95 locks up and stops responding to AT commands and keeps using about 60mA of current constantly. After a couple of minutes, the BG95 starts responding to AT commands again, but the problem is that the battery life is greatly reduced, because it keeps using 60mA whenever it locks up.