Ok, than next option would be simbase.com. They offer Sim Cards also for individuals. Prices seem a bit higher... Covererage is given also for Italy.
Anyone has experience with this service?

Thanks, flori

well, while the website iotcreators.com is still up theirs SIM cards no longer seem to be activated. From the iotcreators.com discord server:
a user asks: "... I'm wondering if iotcreators.com is still active ..." and
[STAFF] Afzal answers: "It is not. Sorry. I tried my best to keep it alive"

Thanks
Felix

@felmue said in ATOM-DTU-NB-IoT - modem does not respond anymore:

So for instance the M5Stack MQTT example will not work as the modem cannot connect directly to M5Stack MQTT server.

Good to know that. If you hadn't pointed that out, I would have fallen into the next trap.
So then second best option would be iotcreators? I don't find any information about the need to connect first to a proxy, and the SIMCom7020G is in the list of the certified modems. Should I go for it?

Thanks flori

Miotiq, in my opinion, is a good offer - I tested with one myself. And they list Italy as supported country.

However, when I tried a while ago, the Miotiq SIM did not allow the modem to connect directly to my server. The modem always has to use the Miotiq backend as a proxy. See here.

So for instance the M5Stack MQTT example will not work as the modem cannot connect directly to M5Stack MQTT server.

Thanks
Felix

I used your pass-through code example to send the AT+CSMINS? command wich gives this with inserted SIM card:

20:21:03.854 -> AT+CSMINS?
20:21:03.854 -> +CSMINS: 0,0
20:21:03.854 ->
20:21:03.854 -> OK

So it seems that the SIM card is not recognized... Maybe broken. Do you have any recommendation for Nb Iot SIM cards? Maybe https://www.miotiq.com/ for Italy?

Thanks flori

I get SimStatus: 1 when SIM card is installed and recognized. (I get SimStatus: 0 if I remove the SIM card.)

So yes, it appears that something is either wrong with the SIM card you are using or with the sim card slot in the DTU.

Thanks
Felix

Nice to know that there is no problem caused by this error.

Contacts facing upward, shape as indicated on the label, should be OK?
The SIM card does not use a PIN, nevertheless I tried this:

void nbConnect(void) {
  unsigned long start = millis();
  log("SimStatus: " + String(modem.getSimStatus()));
 if (GSM_PIN && modem.getSimStatus() != 3) 
    {
    modem.simUnlock("0000");
    }
  log("SimStatus: " + String(modem.getSimStatus()));
....

gives this:
19:20:52.856 -> SimStatus: 0
19:21:13.040 -> SimStatus: 0
19:21:13.040 -> Initializing modem...

So according to this, the SIM card is not ready, right? Could there be a mechanical problem with the SIM card itself?

flori

I get the same error, but it's not fatal. The error stems from the fact that in ATOM_DTU_NB.h a GPIO is defined, e.g.

#define ATOM_DTU_SIM7020_RESET -1

for the SIM7020 reset which isn't connected in the DTU.
If you change that to something like 13 the error goes away.

The program is probably stuck in Inititalzing modem as it tries to check for the SIM card and that it doesn't have a PIN defined.

Have you double-checked that the SIM card is placed in the correct orientation and that it doesn't use a PIN?

Thanks
Felix

Yes I agree, there must be something wrong with this module. I will investigate that later... First priority is to set up one DTU NB IOT module.

Unfortunately also with the working modem I get errors using the MQTT code example:

12:37:55.848 -> Initializing modem...
12:37:55.848 -> [ 62][E][esp32-hal-gpio.c:102] __pinMode(): Invalid pin selected
12:37:55.848 -> E (61) gpio: gpio_set_level(227): GPIO output gpio_num error
12:37:56.138 -> E (364) gpio: gpio_set_level(227): GPIO output gpio_num error

May they come from the reistallation of Arduino IDE?
Here are my compiling settings:

I found the error code here. For sure there must be something wrong with the serial communication.... Strange....

flori

thank you for the update. So it's confirmed something's wrong with the first SIM7020G. I still suspect the baudrate got changed somehow, especially since you are writing that the LEDs behave normally.

Thanks
Felix

hmm, since you already tried changing the modem baudrate I am at a loss what might have gone wrong.

Hi Felix,

thank you very much for your answer and the code example. I can upload the code, without errors. The Atom stops here:

19:19:49.162 -> ets Jun 8 2016 00:22:57
19:19:49.162 ->
19:19:49.162 -> rst:0x1 (POWERON_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)
19:19:49.162 -> configsip: 188777542, SPIWP:0xee
19:19:49.162 -> clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
19:19:49.162 -> mode:DIO, clock div:1
19:19:49.162 -> load:0x3fff0030,len:1344
19:19:49.162 -> load:0x40078000,len:13964
19:19:49.162 -> load:0x40080400,len:3600
19:19:49.162 -> entry 0x400805f0
19:19:51.594 -> Start SIM7020 passthrough test

Unfortunately I do not get answer thru the serial monitor... I've tried:
AT+CEER=?, AT+CGMI=?, AT+CFUN=1. No response... By the way the modem LED's show acitivity red is on, blue flashes with ~.5Hz....

After that I reinstalled Arduino IDE. Now there are only the M5Stack Boards and the Arduino Boards listed. As Libraries I installed only the M5Unified and the M5GFX. I'm clueless. What could I try next?

Update: I've tried another 7020G modem. That one answers....

*19:43:58.890 -> ets Jun 8 2016 00:22:57
19:43:58.890 ->
19:43:58.890 -> rst:0x1 (POWERON_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)
19:43:58.922 -> configsip: 188777542, SPIWP:0xee
19:43:58.922 -> clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
19:43:58.922 -> mode:DIO, clock div:1
19:43:58.922 -> load:0x3fff0030,len:1344
19:43:58.922 -> load:0x40078000,len:13964
19:43:58.922 -> load:0x40080400,len:3600
19:43:58.922 -> entry 0x400805f0
19:44:01.378 -> Start SIM7020 passthrough test
19:44:04.165 -> AT+CEER=?

19:44:04.197 -> +CEER: (0,1)
19:44:04.197 ->
19:44:04.197 -> OK
19:44:14.518 -> AT+CGMI=?*

Thanks, flori

hmm, since you already tried changing the modem baudrate I am at a loss what might have gone wrong.

Maybe my serial pass-through example can shed some light onto this issue?

Thanks
Felix

I'm working as well with the above mentioned hardware (7020G modem). For a while the program worked fine. Unfortunately since some time the modem does not respond anymore.

The example MQTT code given here does not work (no errors). I think I'm using the needed libraries, so I would not expect errors coming from there. The here already mentioned way to solve the problem by changing the baudrate was not successfull - still silence.

The Serial Monitor Output indicates that the initialisation is not possible:
21:23:10.998 -> waiting....1067s
21:23:26.485 -> waiting....1082s
21:23:41.922 -> waiting....1098s
21:23:57.420 -> waiting....1113s
21:24:12.866 -> waiting....1129s
21:24:28.348 -> waiting....1144s
21:24:43.793 -> waiting....1159s
21:24:59.276 -> waiting....1175s

so the M5Atom is still stuck here:

 while (!modem.init()) {
        log("waiting...." + String((millis() - start) / 1000) + "s");
    };

I really don't know how to solve this problem. Since I'm not an expert in programming I would be thankful if someone could help me out with a code to send simple AT-Commands via the Serial Input.

Thanks in advance
flori

yes. See here.

Thanks
Felix

please do not use the nb_dtu_reset_issue code - it was meant to debug an issue and can change the baudrate of the SIM7020 modem.

Thanks
Felix

In file included from N:.....\nb_dtu_reset_issue-main\src\main\main.ino:21:
c:\Users\fb\Documents\Arduino\libraries\TinyGSM\src/TinyGsmClient.h:118:2: error: #error "Please define GSM modem model"
#error "Please define GSM modem model"
^~~~~
exit status 1

Compilation error: exit status 1

My previous assumption that the reset-button kills the modem seems incorrect.

I've had another two devices and carefully did not use the reset-button at all but these modules also stopped working.

ok, I see. Yes, I think getting in touch with M5Stack support is probably best.

Thanks
Felix

Removing the Q5 completely changed nothing.

I ordered another two modules to check the signals but I start to think the issue is in the modem.

Maybe I come to a point where I should ask the M5 team for help. This is kinda annoying, I really like the design of the DTU.

I agree, I wouldn't expect 3.3 V peaks at the U1_RX side of the Q3. And yes, I guess the transistor could be broken, but it being broken in three devices - that is a bit hard to imagine.

I never have used the TinyGsmAutoBaud code myself. I normally run all my modems with just 115200 and so far that worked just fine.

Have you tried to measure what comes out of the modem, e.g. at transistor Q5? Maybe you can determine from that what baudrate the modem expects?

Thanks
Felix