Do you have a link to the module you are using? The data I have is for the active device on the board. The board you have may have more circuitry that may mean these modifications are not necessary.
Weedpharma
Its a SainSmart SIM900:
I've looked for it on the internet and had lots of trouble finding support for it. Figured out just now its a Sim900 Core Board or STM32 board.
It also looks like this:
I bought it on amazon from the first link.
This information in the beginning would have saves some messing around.
The board has 5v input so doesn't need a diode. I would also try it without the extra capacitor.
Weedpharma
Ok
Its a very strange module/board or whatever. I've searched for tutorials online about it and they are pretty much non-existent. There are similar boards for it but not this same one.
Ok but then that is the way I have been trying it so far and its not working. So something else is going on. I started on this whole "power" trip because I read a few posts here and in other places where the users noted that the module seems to draw peak power bursts of 2A, which the USB cable was unable to provide or even an external 9v battery pack source which I have.
So Im wondering why it doesn't work.
Everything works fine up until the point after it connects to the APN, negotiates a connection, gets an IP and even makes the HTTPGet request, but then doesn't get anything back. The posts mentioned above insinuate that a lack of peak power (2A) could be causing the problem.
- How could I make sure if its a power problem or a code problem?
The code Im using is this one
The datasheet says that the transmit power "typically rises to 2A". You will need a serious power supply for that. Or a LM2576.
Thanks Isaac, thats what Ive been trying to figure out.
Where can I get a power supply like that or where (and what is) can I get an LM2576?
Ok, in the meantime, while I find a power supply for this, would the capacitor do the job?
I have at my disposal (came with my Arduino kit), some:
0.1microF (50V Ceramic disc)
22 microF (50V Ceramic disc)
100microF (35v-Radial)
weedpharma I believe recommended using a 10 or 1microF but I don't have those at hand. Can I try with the 22 one?
The capacitor is used to hold power and supply it for the surge. The size of the C needs to be able to supply the amount required in this use. As I don't know the actual current requirement and for how long it is required, I can only guess and try different values.
I would expect that the current is only required for a very short time.
Try the 22uF and see what happens.
Remember that the power supply needs to be able to hold 5v for the Arduino as well as recharge the C so should be able to supply at least 1A.
Weedpharma
Ok, thats another thing. For example:
- How many amps can the USB connection deliver to the arduino board?
- How many amps can the USB deliver to the GSM module?
- How many amps can a 9v battery pack deliver to the arduino board & gsm module?
I would like to know how I can find what those values are.
Thanks
- 0.5A, fuse protected
- 0.5A
- squarish 9V battery: 0.25 to 0.3A for maybe an hour, and will likely be very warm doing so.
6AA battery in series: 2.5A for maybe an hour, small amount for longer.
6C or 6D: much more.
Look up manufacturer spec on the battery you want to use. Oftentimes, it is right on the battery case.
For example, rechargeable NiMH (nickel metal hydride) can be found with 2700mAH rating
http://www.amazon.com/Sanyo-Rechargeable-Batteries-4-Pack-Discontinued-Manufacturer/dp/B000IV413S
Output for rechargables is generally 1.2V, vs 1.5V for alkaline.
These particular batteries seem to need charging before 1st use.
Ok cool! So I would be better off with a 6xAA battery pack?
Sounds doable.
Since I have the 9v battery pack sitting right here next to me, I have to ask,
In your expert opinion, would inserting the capacitor solve the problem (if in fact the issue is that the sim900 is wanting too much Amperage momentarily), event though I will go out and buy a 6xAA battery pack later today?
Thanks
It may do so. The C can supply current quicker than the battery. The battery can "slowly" charge the C that can then be supplied to the load quickly. It all depends on the amount of current required and how quickly it is required.
Weedpharma
Ok, turns out I found a 4xAA battery pack as well. I've got it loaded with batteries and ready to go. So ill give both of those a try right now:
-
4xAA battery pack
-
9V battery pack with C
Now as I understand, both of these battery packs need to be connected to the barrel jack but the GSM module needs to draw power from the Vin, right? Not the 5V anymore.
Can I still leave the USB cable plugged in to watch the serial monitor output?
The GSM needs 5v so DO NOT connect to Vin!
The C should be connected as close as possible to the 5v input to the GSM.
Weedpharma
What!?
Why not?! We have been talking about 2 possible solutions for this issue:
-
A more powerful power source (like a 6xAA battery pack).
-
Or a Capacitor
So you're saying I can't use option 1?
Where to get a LM2576: google shopping LM2576
Option 1: The SIM900 will fry if you connect it to something other than 5V.
Option 2: A big cap will be needed (1000 uF at least)
OK well I'm not sure what happened, maybe it was the imminent threat of being connected to a 6v battery pack, but I connected it to the USB to upload the sketch again and it worked 
I only tried it once so I'll try it a few more times tomorrow.
Marciokoko:
What!?Why not?! We have been talking about 2 possible solutions for this issue:
A more powerful power source (like a 6xAA battery pack).
Or a Capacitor
So you're saying I can't use option 1?
We were talking about powering from 5v, then you say you are going to connect to a higher voltage. My warning probably saved your GSM!
Weedpharma
My bad. I need more current not more V :-).