Arduino SIM800L Problem...

The datasheet at page 21 say that you can have peak up to 2A (I've found this http://www.seeedstudio.com/wiki/images/4/46/SIM800L_Hardware_Design_V1.00.pdf ), so are you sure that the power supply don't drop down too much during this peak?In theory if the voltage go below 3.4V it shut down automatically, page 25

I hope that this will help

I hooked up my multimeter and the voltage drops from 4.11 to 3.98 when I send a command. Maybe the peak is higher and not really easy to measure with a normal multimeter. I will try to hook up a scope to see, but I don't have one around right now.

I hooked up my multimeter and the voltage drops from 4.11 to 3.98 when I send a command. Maybe the peak is higher and not really easy to measure with a normal multimeter. I will try to hook up a scope to see, but I don't have one around right now.

Why not use arduino to try to reveal low peaks? You can use one of the analog input to measure the voltage of the other board (first you send an instruction to the sim800l and then arduino start to control the voltage), if it drop down a certain threshold it can turn on a led or something else, probably arduino's ADC is faster than a multimeter, or if you have an operational amplifier you can use it to activate an arduino's interrupt, I think that is the better choice to reveal peaks.

You can also try to use a battery as power supply but I'm not sure that it can be enough.

A multimeter isn't fast enough to catch most power glitches.  You'll need to use an oscilloscope. 

Alright, I should be able to get hold of one for testing, and I'll post the results here. Thank you.

Did you find something? Because I have the same problem...

No   I never got it to work, it was hard, but I had to give up on this.

Did you ever get this working?
Not sure why you are using such an odd supply voltage

The prescribed working voltage: 3.7~5V so you would seem to be in spec. However It's unlikely that you are accurately reading the minimum voltage 'spike'. Unless using a bulletproof bench supply, I'd be inclined to start by raising the supply to 5V and add an electrolytic close to the board to buffer current needs, especially since you state that your PS can supply 2A (known, or stated on the side?), yet the peak demand is also 2A. So a brownout triggered reset is quite likely.

A good quality bomb proof power supply is critical, especially on a test bench, and in some cases a bank of alkali cells is better than a flimsy PSU. And of course, even pricey PSU's can pass a large amount of ripple, so unless you are adding a lot of post filtering, 'wall warts' are not a great choice, especially when reaching their maximums.

Don't forget that you can use rectifier diodes to drop voltage, no need for regulators, zeners or power transistors etc. and no noise.

And of course as mentioned elsewhere a scope is invaluable for it's speed. I still use an old Philips CRT twin beam scope, but check out simple kits like DSO138 which work well for most needs (I love mine - given for xmas!). You can then put away your volt meter!

BTW, to another poster - ADC on an Arduino is a slow process, and for a narrow spike, timing would be critical to see it, assuming it lasted long enough. Nice cycling of available tools though.

No intention to teach egg sucking, just passing on experience.

Regards

Hello,

I could try adding that capacitor. I tested the small PSU, it can deliver 2A without much voltage drop. I bought that DSO138 like you said, specifically to test this a few months ago , but it does not show any spikes.

I've been waiting a while for this to arrive, so I'll let you know how I get on, assuming that my SIMs will work in it! I've not really hit the data sheets yet; I usually find that success depends on a few lines hidden away as an aside!