Sim800L stops responding after working 10-12 days, after that it needs power reset

Sim800l is not responding anymore after a few days of continuously working and requires restart/reset. It won’t recover by itself or respond to any commands unless a restart or power off is made. If I restart the sim800l it will work for another few days.

Same behavior on two different sim800l modules.

The module is powered by a 5v 2.5 A power supply in parallel to the raspberry pi. 115200 baud rate is set.

I have tried a DC to DC (2.5) step down buck converter to power the module with 4V 2.5A separate power supply with common GND for SIMcard, raspberry pi, and module.

I don’t think poor network or Transmission burst is the problem for now. It works for 12 days in the same network conditions and then stops working for a long time.

Is someone else experiencing the same behavior? Any suggestion or information would be helpful.

Hi, are you sure that buck converter can deliver 2A peaks and also be able to reliably step down to 4v from 5v? Most buck converters I know either need at least +1.25v more on the input than on the output, or they work with a smaller difference but only provide very limited current (usually less than an amp).

Another thing I would ask you is, how are you connecting the raspberry and the module together? using a logic level converter?

Yup i had faced some similar issues while i was working on a project, it was a very simple fix, that was to use a 1000uF 25V capacitor in parallel to the GSM powersupply.

Try that and let me know if that helps!

Even, if your data isn’t too crucial to sent every second, you got a rpi a full fledged computer running linux, just store the important data like a log, and use a Mosfet to power the SIM800 and then push the necessary data to the server.
That’s usually the efficient way ive figured out when i was working with the GSM Modules.

Parth Yatin Temkar

@viedana 1. Yes, I have used "LM2596 DC-DC Adjustable Step-Down Voltage Regulator Power Supply Module " from the information I get from datasheet I can tell that it gives me 2A normal operation and 3A as peak current. Voltage I have tested in my lab. I convert from 4.8 V to 4.1 V so that I generate less heat and voltage can be optimum for SIM800L module also.

  1. I connect GND, TX and RX from Rpi to Sim800L, +4.8V 2.5 A power supply to Buck converter and then +4.1V and GND to sim800l so that they share common GND. Like shown in picture.


Hi @ParthTemkar, Thank you for your reply. Do you mean it is happening because of DC power supply ripples?

Or what is the exact effect of the capacitor on power supply?

Is that can happen only because of power supply ?


Basically, that’s a simple fix, so that the current consumption your saying, specially the peak burst, that’s handled by the cap and it works more smoothly and efficiently.
and like i said, if this fix also doesn’t work for you (which should ideally work), then just keep your gsm on offline whenever its not doing any work.

Parth Yatin Temkar

Im sorry @hdhanian1, i just re-read your post and even the other reply of yours to @viedana,

Can you tell me what’s your main power source? which is powering the Raspberry Pi and the GSM through the Buck Converter?

Because what i can read from your comments i just saw, your feeding 5v 2.5Amp Supply to Rpi and gsm. if that’s the case, then i feel that is not efficient (So far from my experience), you should at-least keep 3Amp or preferably 5Amp supply. its like adequate for running everything very smoothly!

If possible even try this out.

Parth Yatin Temkar

Have you checked the output of the buck converter with an oscilloscope? According to the datasheet (on page 27) of the LM2596, you need AT LEAST 1.3v of difference between the input and the output, otherwise you are stressing the buck converter (not the other way around as you thought).

Also, as @ParthTemkar said, check that you have at least 2A extra only for the SIM800 (I think the Raspberry needs 2A-3A just for itself).

Also also, check the hardware design guidelines of the SIM800 (link: There, on page 32, it shows you a very simple circuit to interface the TX-RX pins of the SIM800 with a 3.3v MCU as the Raspberry; it is pretty simple but very important for stability (from experience I recommend using some Schottsky diodes).

@ParthTemkar Sure, I will try the capacitor (Not sure if it will work) and controlling Input power with GPIO and mosfet (i think will work but not really solving the problem. it is reset the Sim800L when it hangs). I will update the results here when I test it next time.

on your second point, its a nice idea too. As a back up currently I save the data locally when there is no Internet. And upload in free time. when we are not measuring. In principle it is the same.

@viedana Thank you for the information. Yes, you are right. I will make the difference at least 1.23V for the Buck converter. I followed the article for my connection. They say 3.3V logic level is fine. I thank you for showing the hardware info from the datasheet. I think I should trust on Datasheet infomation more from next time.

I try the given suggestion Since it is my personal DIY project and I need to test at least 20 days to make sure that the problem is solved. I will update the results in a Month or so.

Thank you so much for your help.

Ok, please let us know if it worked!
Good luck :+1:

Did you have any luck with your project?

1 Like