0
\$\begingroup\$

I'm using the A6 GSM module to send data from my sensors board to a server via GPRS.

The problem is that most of times, usually when the module attaches to the network or connects to the server, noise appears on the UART A6 TX line and the controller reads garbage. After this noise appears, the next command sent to the module gets a CME error response. I suspect the noise it generated by the chip, maybe caused by the RF transmitter.

Sometines it works, but most of the times it fails. I've tried putting 100pF capacitors to GND on the UART lines and lowering the baud rate to 2400 baud, but with the same results.

If it has anything to do with this problem, I mention that I am applying 3.3V level on the UART line (the A6 supports 2.8V), since I did not have a voltage translator available when I made the board.

Has anyone encountered this problem? Is there a solution or should I use a different GSM module?

UPDATE

The GSM module is a ready made breakout board from Ebay. The module is connected to my microcontroller board by wires. My power supply is a R-78E5.0-1.0 DC-DC converter that provides 5V/1A. I've tried powering the module from a separate 2.5A power supply, but the results are the same. These are my schematic and setup.

Schematic and setup

The SIM card is connected via an extender. I've also tried without the extender, but with the same results. In the setup you can also see the logic analyzer connected to the UART lines and GND.

This is the output from the logic analyzer (the last waveform is module TX analogic):

Logic analyzer captures

The noise consists of bursts of impulses, from 2.2V to 3.7V, or longer pulses from 1.9V to 3V. I usually receive CME Error 50 (EXE_FAIL) and CME Error 58 (which is not described in the module documentation).

I tried connecting the antenna via an RG316 cable coax cable and the noise disappeared. With or without shielded wires to the microcontroller board, there was no noise. Does anybody know why?

Antenna connected by cable

\$\endgroup\$
  • \$\begingroup\$ (a) Do you have access to (and experience using) an oscilloscope? (b) Is this a ready-made A6 GSM module, or your own? It would help if you to edit your question and add a link to the web page for your specific module (if it's ready-made), and also add an in-focus, appropriately-cropped photo of your specific hardware setup. A schematic diagram of the module will also help, especially if it's your own design. (c) How have you confirmed that your power supply reaching the GSM module is adequate? (d) Getting the specific CME error code number may also be helpful. Do you have that? \$\endgroup\$ – SamGibson Jul 18 '18 at 14:05
  • \$\begingroup\$ I suspect it is a common mode RF noise problem that can be solved with a CM choke and shunt caps to earth ground. or simply STP UART cables with earth ground. \$\endgroup\$ – Sunnyskyguy EE75 Jul 18 '18 at 14:07
  • \$\begingroup\$ @SamGibson Please read the updated question. \$\endgroup\$ – Cristian M Jul 19 '18 at 8:02
  • \$\begingroup\$ @TonyEErocketscientist I will try with STP cable. I've updated the question with more details. \$\endgroup\$ – Cristian M Jul 19 '18 at 8:08
  • \$\begingroup\$ Near E field coupling of RF to wire mesh without ground plane for antenna to isolate logic is a good way to induce interference since it radiates radially from antenna. \$\endgroup\$ – Sunnyskyguy EE75 Jul 19 '18 at 11:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.