PR merged, well done Christian :-)

Just a minor issue: Mark wanted the cellular framework to not depend on the actual modem model or maker, he added a separate build switch CONFIG_OVMS_COMP_CELLULAR_SIMCOM to enable/disable the SIMCOM drivers.

With your new component layout, the cellular framework now needs to link the simcom_powering module, so the build will fail with SIMCOM disabled.

The hardware dependency cannot easily be resolved, as the framework needs to apply the specific power timing to power up the modem in the first place, before it can identify the modem (unless the user has configured a model).

Unless someone has a clever idea how to solve this, this can IMO still be addressed when we ever need to add a non-SIMCOM modem.

Regards,
Michael

Am 31.03.25 um 11:01 schrieb Info Zeitnitz via OvmsDev:

Hi Michael,
great! Thanks a lot for your effort and patience.
I first thought, that debugging the code without direct access to the hardware would be very difficult, but this really worked out fine.

So, I will issue a PR.

Best regards
Christian

Am 30.03.2025 um 17:31 schrieb Michael Balzer via OvmsDev:
Christian,

this looks very good now. I've mistreated my test modules in all ways I could think of including switching off the running modem via S1, the cellular system recovered nicely from all situations.

Regards,
Michael

Am 30.03.25 um 16:01 schrieb Info via OvmsDev:
Hi Michael,
I think I found the problem which caused the 5360 not to powercycle correctly. Actually the T_on time was in that instance set to 1126913089 ms, so this will take a while!
This was caused by a stupid mistake with an incorrect modulo calculation of the timing index .

I fixed the unused variables as well.

Best regards
Christian

Am 30.03.2025 um 14:26 schrieb Michael Balzer via OvmsDev:
PS: please fix the two unused variables now in gsmnmea.cpp:

/home/balzer/esp/Open-Vehicle-Monitoring-System-3/vehicle/OVMS.V3/components/ovms_cellular/src/gsmnmea.cpp: In member function 'void GsmNMEA::IncomingLine(std::__cxx11::string)':
/home/balzer/esp/Open-Vehicle-Monitoring-System-3/vehicle/OVMS.V3/components/ovms_cellular/src/gsmnmea.cpp:148:40: warning: variable 'vdop' set but not used [-Wunused-but-set-variable]
     float lat=0, lon=0, alt=0, hdop=0, vdop=0, pdop=0;
                                        ^
/home/balzer/esp/Open-Vehicle-Monitoring-System-3/vehicle/OVMS.V3/components/ovms_cellular/src/gsmnmea.cpp:148:48: warning: variable 'pdop' set but not used [-Wunused-but-set-variable]
     float lat=0, lon=0, alt=0, hdop=0, vdop=0, pdop=0;
                                                ^

Regards,
Michael

Am 30.03.25 um 14:21 schrieb Michael Balzer via OvmsDev:
Christian,

the user way to power off/on the modem is via "power cellular off/on", that's what I test/ed, and that works.

After having no issues now with the 7600, I did the tests with my old 5360 module.

That generally also worked, with one exception: on one of four crash reboots, the cellular system got stuck in PowerOffOn state, with state ticker at 0 and no more activity.

Logs attached.

Regards,
Michael

Am 30.03.25 um 13:08 schrieb Info via OvmsDev:
Michael,
one more thing: if you are performing another test, please check from the console, that the powering of the modem works as well.

cellular setstate PoweringOff
cellular setstate PoweringOn

It did not work when I started with the 7670 and I had to modify ovms_cellular (set m_powermode according to the state) to get it going.

Thanks
Christian

Am 30.03.2025 um 12:02 schrieb Info via OvmsDev:
OK, thanks Michael.
At least this was an easy fix.

Had a look at the NMEA code and I screwed up the istringstream object when checking for the CGNSSINFO string.
Fixed it and I hope this works now as well.

Best regards
Christian

Am 29.03.2025 um 18:42 schrieb Michael Balzer via OvmsDev:
Christian,

the GSM/PPP part now works smoothly with all standard situations tested so far.

But the GPS part doesn't work. It looks like the NMEA sentences are received in the MUX channel, but not forwarded to the NMEA module.

V (794230) gsm-mux: ProcessFrame(CHAN=3, ADDR=0d, CTRL=ff, FCS=fa, LEN=78)
V (794230) gsm-mux: ProcessFrame(CHAN=3, ADDR=0d, CTRL=ff, FCS=f4, LEN=82)

These should be the NMEA sentences, but there are no corresponding log entries from gsm-nmea.

New logs attached.

Regards,
Michael

Am 29.03.25 um 15:13 schrieb Info via OvmsDev:
Hi Michael,
thanks for testing.

The results look odd, so I went back to the schematics of OVMS modules and realized, that I was mistaken. The PWRKEY pin is actually inverted in the modules as well. This definetly explains the strange behaviour.

I changed now the logic in simcom_powering.h and the inverted PWRKEY level is now the default.

Could you please run another test?

Best regards
Christian

Am 29.03.2025 um 09:25 schrieb Michael Balzer via OvmsDev:
Christian,

well done on the 7670 support :-)

I've just done some first tests of your branch on the 7600, which turned up some issues. Logs attached:

Cold boot failed in 2 of 3 test boots, with the driver getting stuck in a loop (poweredon / identify / muxstart)

Regular shutdown fails to power down the modem in time -- for a deep sleep phase, this would mean the modem will remain powered on

After a crash, the modem driver gets somehow stuck in PowerOffOn state

It seems the test for reaction ("AT") sometimes gets skipped, even though the modem isn't ready yet.

On the failed cold boot log, you can see this at the end:

I (389178) cellular-modem-auto: Power Cycle
I (389178) Simcom: Power Cycle - T_off 2500 ms - T_on 200 ms
D (389178) events: Signal(system.modem.poweringon)
D (389188) events: Signal(egpio.output.0.low)
D (391678) events: Signal(egpio.output.0.high)
D (396678) events: Signal(egpio.output.0.low)
D (396878) events: Signal(egpio.output.0.high)
I (396878) cellular: State: Enter Identify state
D (396978) cellular: tx-cmd: AT+CGMM
D (396978) cellular: tx-cmd: AT+CGMM
D (396978) cellular: tx-cmd: AT+CGMM
D (396978) cellular: tx-cmd: AT+CGMM
D (396978) cellular: tx-cmd: AT+CGMM
D (396978) cellular: tx-cmd: AT+CGMM
D (396978) cellular: tx-cmd: AT+CGMM
I (397008) cellular: Identified cellular modem: SIM7600/Experimental support for SIMCOM SIM7600
D (397008) cellular: Remove old 'auto' modem driver
I (397008) cellular: Set modem driver to 'SIM7600'
I (397008) cellular: State: Enter PoweredOn state
D (397008) events: Signal(system.modem.installed)
D (397008) events: Signal(system.modem.poweredon)
D (406178) cellular: tx-cmd: AT+CPIN?;+CREG=1;+CGREG=1;+CEREG=1;+CTZU=1;+CTZR=1;+CLIP=1;+CMGF=1;+CNMI=1,2,0,0,0;+CSDH=1;+CMEE=2;+CSQ;+AUTOCSQ=1,1;E0;S0=0
D (408178) cellular: tx-cmd: AT+CGMR;+ICCID
D (416178) cellular: tx-cmd: AT+CMUX=0;+CATR=6
D (434808) cellular: mux-rx-line #0 (2/46): OK
D (434808) cellular: mux-rx-line #0 (3/39): RDY
D (434808) cellular: mux-rx-line #0 (12/23): +CPIN: READY
D (434808) cellular: mux-rx-line #0 (8/11): SMS DONE
D (434808) cellular: mux-rx-line #0 (7/0): PB DONE
I (434808) cellular: State: Enter MuxStart state
D (434808) events: Signal(system.modem.muxstart)

Regards,
Michael

Am 26.03.25 um 13:13 schrieb Info Zeitnitz via OvmsDev:

Dear all,
I finished to integrate the Simcom 7670 modem into the OVMS code.
See https://github.com/zbchristian/Open-Vehicle-Monitoring-System-3.git branch update_cellular.

Modem Powering
===============
I had problems to get the 7670 working with the current code. The modem was stuck in a power cycle loop.

I checked the Simcom documentation for the 4 models 5360, 7000 series, 7600, 7670 and the powering procedure via the PWRKEY pin is the same, but with different on/off timing. Depending on the hardware, the PWRKEY can be inverted as well. This is the case for my Lilygo TCall board, but I think not for the OVMS HW versions. At least version 3.1 connects the corresponding MAX7317 pin directly to the modem.

Assuming, that all hardware versions use the PWRKEY procedure for powering, I streamlined the code and moved the corresponding functions into separate files (simcom_powering.cpp and simcom_powering.h). This avoids the duplication of the code.

All this leads to substantial changes in ovms_cellular and ovms_cellular_modem_driver as well.

The code works now perfectly for the 7670, but testing is required for all other models.
If my assumptions about the OVMS HW versions are not correct, or the modems do not work as documented, some modifications might be needed.

GNSS Location
============
The 7670 no longer accepts the currently used AT commands to obtain the location as a GRMC sentence. So, I added code to the new 7670 class to request the location via the AT+CGNSSINFO command.
Since this has to be send explicitly again and again, I implemented a corresponding call in the StatusPoller of the 7670.
All the NMEA related code has been adapted to handle the corresponding CGNSSINFO response.

In conclusion, before doing a pull request, it would make sense that the code is tested with the other hardware and modem models.

So, tell me what you think and if the general strategy of the mods make sense to you.

Best regards
Christian

_______________________________________________
OvmsDev mailing list
OvmsDev@lists.openvehicles.com
http://lists.openvehicles.com/mailman/listinfo/ovmsdev
-- 
Michael Balzer * Am Rahmen 5 * D-58313 Herdecke
Fon 02330 9104094 * Handy 0176 20698926
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.openvehicles.com
http://lists.openvehicles.com/mailman/listinfo/ovmsdev
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.openvehicles.com
http://lists.openvehicles.com/mailman/listinfo/ovmsdev
-- 
Michael Balzer * Am Rahmen 5 * D-58313 Herdecke
Fon 02330 9104094 * Handy 0176 20698926
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.openvehicles.com
http://lists.openvehicles.com/mailman/listinfo/ovmsdev
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.openvehicles.com
http://lists.openvehicles.com/mailman/listinfo/ovmsdev
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.openvehicles.com
http://lists.openvehicles.com/mailman/listinfo/ovmsdev
-- 
Michael Balzer * Am Rahmen 5 * D-58313 Herdecke
Fon 02330 9104094 * Handy 0176 20698926
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.openvehicles.com
http://lists.openvehicles.com/mailman/listinfo/ovmsdev
-- 
Michael Balzer * Am Rahmen 5 * D-58313 Herdecke
Fon 02330 9104094 * Handy 0176 20698926
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.openvehicles.com
http://lists.openvehicles.com/mailman/listinfo/ovmsdev
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.openvehicles.com
http://lists.openvehicles.com/mailman/listinfo/ovmsdev
-- 
Michael Balzer * Am Rahmen 5 * D-58313 Herdecke
Fon 02330 9104094 * Handy 0176 20698926
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.openvehicles.com
http://lists.openvehicles.com/mailman/listinfo/ovmsdev
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.openvehicles.com
http://lists.openvehicles.com/mailman/listinfo/ovmsdev

-- 
Michael Balzer * Am Rahmen 5 * D-58313 Herdecke
Fon 02330 9104094 * Handy 0176 20698926