Lara-R2 behind the scene communication?

0 votes
We have a Lara-R2 connected to a NEO-8Q on a custom board. The communication works perfectly, the GNSS is also very good. I'm currently looking at the power consumption and found a "strange" behaviour.

The modem is hooked to the network and stays powered all the time. We use the DTR line to go in/out of power saving.

I log absolutely everything we do with the modem (in development/debug phase) and I discover some power peaks I cannot explain. Whatever we communicate or not, there is a 30-40 seconds where the modem uses about 100mA, go back to sleep ans 6 minutes later, it does it again. If I enable the GNSS or send data to our server, the 6 minutes gap doesn't change.

Is there an explication for this communication and is it possible to prevent it? It's not fantastic to have those peaks on a battery setup...
by nicolas asked Feb 13, 2018
0 votes
In collaboration with the uBlox team, we (they) found the problem.  It's actually a normal behaviour in roaming. The SIM tried to go back to it's home network on regular interval...
by nicolas answered Mar 9, 2018
Best answer
Thank you for posting a significant insight!!!
0 votes
I suspect that this is modem checking in with the network - it is done when modem is in power save mode.  Interval is controlled by the network and can't be changed.

Without checking with the network modem would not be able to stay registered or receive SMS or data from the network.
by grumpy answered Feb 13, 2018
If the peaks would be 30-40ms, I would agree with you, but 30-40 seconds, just for a keep-alive, it's a bit extreme. When we boot the board, power on the modem, configure it, send a message to the backoffice, receive the response, close everything, it doesn't even take 10 seconds. With all the initial connection overhead.
Oh, I read it as ms not seconds.
Then something is keeping modem on.  I would check +UPSV settings to make sure modem is going to sleep.
If you are using USB then you would need to suspend it.
we use +UPSV=3 with uart communication. I can see the modem going to sleep, the power drops drastically and we have tiny keep-alive spikes every 4-5 seconds. That part is good and I understand the network peeking principle. The spikes jumps to around 20-25mA, which is acceptable.

The problem is the 30-40 seconds of 100mA.  I'll continue to experiments with various parameters to see if I can put mi finger on the problem.
0 votes
This is far from being an answer, more of a follow-up.

I still didn't find what job is running at a 6 minutes interval.

I tried to enable a lots of URC, but didn't received any when the modem kicks off it mysterious job.

I wonder if there is a way to "monitor" all the modem activity, to find what the hell it's doing during those 30-40 seconds (the long seconds, not the quick milliseconds...) During the testing phase, I don't care if I receive gazilions URCs, all I want is to understand what's happening.
by nicolas answered Feb 14, 2018
0 votes
Some more follow-up...

I have the same behaviour on the EVK-R203 dev kit.

I emit a +UFACTORY=2,1 followed by a +CPWROFF and a power on to clean all the previously entered configuration. I still have the same thing, about a two minutes after the power on.

This is one of those time where I wish I could dig into the firmware's source to find out what the heck the modem is doing...
by nicolas answered Feb 14, 2018
website banner