I'm working with SIM800L for remote-sensing communication via GPRS. The problem is, everytime the module is running initialization function (SIM800.begin(), different for every SIM800L library), it failed.

I have check whether the module is broken or not, by connecting RX & TX pin of the SIM800L module to the TX & RX pin of FTDI module, and send AT command from PC serial monitor. The module seems not broken, as it still can reply the AT commands.

Here is the piece of initialization code from the Adafruit FONA library: enter image description here

The program is failed to attempt to open comm with ATs, and keep looping on the AT sending. How can i fix this issue? Thank you.

[Extra: i use 2-cell 7.4V Li-Po battey, and it is regulated with XL4015 module to 4V, as the module work on 3.4-4.4V. Could this be the cause of the problem (too weak power supply)?]

  • \$\begingroup\$ How are you powering the module? Is is a bare module, and you have it attached to your own circuitry, or is it a complete unit with a housing and its own powersupply? There's a common problem when using bare modules that people often make. \$\endgroup\$ – JRE Dec 29 '18 at 13:30
  • \$\begingroup\$ i just add the powering explanation above hehe, sorry for not giving the detail on the first place @JRE \$\endgroup\$ – Marchio Kevin Dec 29 '18 at 13:31
  • \$\begingroup\$ The power supply for the module is separated with the power supply for the microcontrollers by the way \$\endgroup\$ – Marchio Kevin Dec 29 '18 at 13:32
  • \$\begingroup\$ Could you show your exact model, or link to a datasheet for it? \$\endgroup\$ – JRE Dec 29 '18 at 13:34
  • \$\begingroup\$ This is my refference: teachmemicro.com/send-data-sim800-gprs-thingspeak/… The only difference is i use regulator module for the Li-Po battery \$\endgroup\$ – Marchio Kevin Dec 29 '18 at 13:35

The most common problem with such GSM modules is that they draw current in surges.

The transmitter is the only thing that draws a lot of current, it only transmits sporadically. When the transmitter kicks in, the module will draw around 2A.

If the connections are poor, or the power supply weak, the voltage will drop when the transmitter goes on. If it drops enough, then the processor on the module will shutdown and then restart when the voltage goes back up.

The transmitter kicks in when you tell the module to connect to the network - that's part of the initialization.

The designers expect this kind of problem. That's why there's a 47uF capacitor on the module. It provides that short burst of current when the transmitter goes on.

If your connection is poor (high resistance through a breadboard or high impedance through long wires) then the 47 uF won't be enough.

  1. Use short wires for the power connections.

  2. Make sure your power supply can provide 2A without dropping voltage.

  3. Make sure the power connection is good (solder the connections, if possible.)

  4. Add a larger capacitor across the existing 47uF part. Use an electrolytic capacitor with short leads. Watch out for polarity - electrolytic capacitors tend to go "bang" if connected backwards.

Those should fix your problem.

  • \$\begingroup\$ I see. For the prototype, i will try to use 5V 2A adaptor. But for the real implementation, i'm going to use high capacity battery 12V 45Ah. Hope that it will be sufficient. But i've watched some SIM800L tutorial which said that the module will require current up to 2A only when bursting data package. So, for simple usage, like SMS, powering the module via USB port supply is enough, as it require less current. Is this true? \$\endgroup\$ – Marchio Kevin Dec 29 '18 at 14:49
  • \$\begingroup\$ You should not expect a difference between data and SMS. But you should be able to verify serial communication without attempting to register on the network. \$\endgroup\$ – Chris Stratton Dec 29 '18 at 16:35

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.