LowPowerLab Forum
Hardware support => RF - Range - Antennas - RFM69 library => Topic started by: eternaldrake on February 16, 2015, 10:15:22 PM
-
Hey there, I'm pretty new to Arduino and I am working on a project that's going to be using a network of sensors to monitor a parking lot, and I decided to go with Pro Minis (3.3V, 328p version) + RFM69CWs. I've read through the RFM69 library and a good bit of the threads on the forum here concerning interfacing the RFM69s with Arduinos (mostly UNOs). After wiring up two units I tried to do an initial test with my node running on battery and my gateway running via FTDI adapter and was confused as to why it wasn't working. Then I realized that using CR2032 coin cells with a max continuous output of 1 mA isn't exactly going to cut it haha :)
So going forward I had two immediate questions:
1. What batteries do you guys recommend for autonomous sensing? I read in another thread Felix recommended the 3.7+ LiPo batteries so I'm probably going to order a handful of those. I should be able to plug them into the RAW pin of the Pro Mini and get a 3.3V regulated output to power the RFM69. I can't find the maximum supply current from the 3.3V pin of the Pro Mini anywhere online but the I/O pins are rated at up to 40mA so I feel like this would work.
2. Are there any changes that need to be made to the RFM69 library to work with the 8MHz 328p?
My ultimate goal is once I get my sensors and gateways communicating, I want to implement a mesh network protocol for the gateways with the sensors fashioned in a star topology with their respective gateway, and sink the gateway network into a terminal with a raspberry pi to upload all this sensor data to a server. Sounds like fun doesn't it?
-
I'll just say forget coin cells.
LiPos are up to 4.2V when fully charged and drop towards 3V when you should consider them discharged. Your minis will likely drop voltage to some degree so you want to keep the input voltage above 3.3V + that threshold, check the specs from your source.
The lib should work just fine as is with your minis.
-
I'll just say forget coin cells.
LiPos are up to 4.2V when fully charged and drop towards 3V when you should consider them discharged. Your minis will likely drop voltage to some degree so you want to keep the input voltage above 3.3V + that threshold, check the specs from your source.
The lib should work just fine as is with your minis.
Alright, I'll go ahead and order a batch of LiPos then. Thanks for the quick response!
-
Tried testing again today running one Pro Mini as a Gateway via FTDI/USB and the other as a node powered by the 3.3V regulated output of an Uno I had lying around. The node blinked and seemed to transmit just fine, but the gateway which was listening didn't pick up any transmissions. I played with them for a while and tried adjusting the antennas (22 AWG wire of 7.8cm (915MHZ RFM69s)) and for about half a second the gateway picked up the nodes transmissions (3 transmissions received for each payload, which means ACKs werent getting back to the node) but I wasn't able to replicate it. I'm guessing I must have soldered something wrong somehow.
Sometimes when programming a node it will send only packet every 3-4 seconds, regardless of what delay i specify, which I found strange and am guessing is something soldered incorrectly with regards to the SPI lines.
Either way, I think I'm going to go ahead and order some Moteinos tonight so that I can get started ASAP on getting the structure of my payload and basic networking framework into place.
EDIT: Reading about antenna theory leads me to believe I might have messed something up there too. As I had them when testing, the antenna was just a piece of specific length wire soldered to the ANT pin, which I figured was all that mattered. The way I am testing the transceiver is they are standalone with a bunch of long wires soldered to each pin I am using, and these wires go into a breadboard where the Pro Mini is mounted. So in this setup, the antenna of the RFM wouldn't even have a ground plane except I guess the PCB of the transceiver itself, which doesn't seem like a setup that would work. wireless is hard