Craig, Mark, Thomas, Dimitrie / Thomas just reported a very similar (same?) issue on a EU modem: https://www.openvehicles.com/node/2401 <https://www.openvehicles.com/node/2401#comment-6228> He managed to get a log from the issue, which shows the MUX startup to be broken and remain broken until power down. Specifically, MUX port 2 (just that one) gets stuck in some error state, which does not resolve by a soft power cycle. Thomas also found out even a short hard power cycle won't be sufficient to resolve this:
I found out, when i powered off and 1-2 sec. later on again. the modem got stuck. when i power it off and wait 1 minute and then power it on. it will start normal.
His modem firmware version is 35316B12SIM5360E (B12). @Thomas: there is a B13 available…
I've found later versions (up to B13) and release notes listed only here: https://techship.com/products/simcom-sim5360e-mpcie-sim/
…but you need an account to access them and must accept an NDA.
If you feel lucky you might try that. But it's also possible this is a hardware flaw in all SIM5360 versions. Regards, Michael Am 05.01.20 um 11:12 schrieb Michael Balzer:
Craig,
you'll need log level verbose to see all tx/rx going on (i.e. MUX commands), but debug should be sufficient for the power on sequence.
You can test setting the PWRKEY manually by "egpio output 0 1". If the simcom driver is now stuck in the power cycle loop and interferes, you should be able to stop that by "simcom setstate PoweredOff" or "power simcom off".
Regards, Michael
Am 05.01.20 um 05:16 schrieb Craig Leres:
The simcom on one of my modules wedged today and had some time to poke at it. When I turn on simcom debugging I see "AT" commands once a second:
D (1199981) simcom: tx scmd ch=0 len=4 : AT D (1200981) simcom: tx scmd ch=0 len=4 : AT
but no responses.
I wanted to see if I could talk to the simcom via the usb connector on the modem card so I sacrificed a micro usb cable and hacked the molded end off so I could plug it in without detaching the modem board from the main board (see attached).
Next I spent a few minutes with a spare module looking for a windows 10 driver. The sim5360 (rar) archive from techship.com didn't work but the windows 8 zip file from here did:
https://github.com/botletics/SIM7000-LTE-Shield/tree/master/SIM7000%20Docume...
I ended up with four com ports (at, nema, diagnostic, and audio) and was able to talk to the spare modem.
Then I went out to plug into the simcom in the car. The first thing I noticed was that the blue led on the modem board is not lit and (as expected) I got nothing from the at com port.
For completeness I tried rebooting the module but it help.
I now believe wedging the simcom involves running the car for a short amount of time. In this particular case I moved the car out of the garage to make room for some work on my other car. An hour later I moved it back. The run time was 1-2 minutes each time. A bit later when I tested Mark's new ios app I noticed that the car was not reporting a location.
I looked at the simcom docs a bit and it's too bad the simcom RESET pin isn't connected to a pio pin.
It'll be a couple of days before I'll need the gps to work in case there are ideas of what can be done via software to wake up the simcom. I might trying add a simcom command to set PWKEY to an arbitrary state to see if holding it low for more than one second does anything.
Craig
-- Michael Balzer * Helkenberger Weg 9 * D-58256 Ennepetal Fon 02333 / 833 5735 * Handy 0176 / 206 989 26