I have two RFM69HCW modules, each connected to Arduino UNO with a level shifter (CD4050) and a seperate 3v3 supply for the RFM module, decoupling capacitors in place. I use the RadioHead library with the default examples for RFM69. I'm pretty sure everything is wired up right, because the modules initialize over SPI without problem. However, I have hardly any range. When I touch the antenna's together I have a "wireless" connection, as soon as I move them apart they stop seeing each other.
I have attached the SMA connector with a short wire to ANT and GND of the RFM board. For good measure I added a small groundplane to the SMA connector, but it doesn't seem to help. The antenna was bought specifically for 433 MHz.
I'm pretty clueless how to go at this from here, because I have hardly any experience with antenna's. Suggestions are welcome.
Right, did the single straight antenna, still no connection. It pushed over 7 packets, that's it.
Can anyone here confirm they have a RFM69-module working with Arduino UNO, preferably with monopole antenna and soldered connections like I have. If so: what library are you using?
thanks for your reply. I can solder the SMA connector to the PCB, this will automatically ensure the antenna will be at right angles with the ground plane. I can solder the pin for the antenna to a wire and connect that to the RFM module. and use another wire to connect the ground plane to the ground of the RFM module. Or do all of these wires have to be etched traces on a PCB to make this even remotely work ? I included some links below that suggest that you should at least get /some/ range with seemingly easy solutions.
Will soldering the sma to the pcb work, with wires running to the module ?
I fear one of the answers to my own question regarding the use of wires to connect the sma connector is: microstrip. You can't use regular wires, and need either something coaxial or such, or build something that resembles this on a pcb.
However: this doesn't explain why some people still get great results with only a piece of wire attached to the module.
this doesn't explain why some people still get great results with only a piece of wire attached to the module.
This is because the ground plane on the module board acts as a ground plane for your antenna.
Looking at the photo it seems to me that you need to drill a hole in the piece of PCB material and attach the module with wires that are as short as humanly possible, if not shorter. That would mean mounting the module on the PCB and having the RF output pin close to the antenna. Wires of more than a few cm are too long. For best results that piece of PCB ( ground plane ) has to be as big square as the antenna is long.
In that first picture the results would have been even better if the wire was at right angles to the board.
I'll design a breakout board for the module, with the SMA connector soldered to the PCB, the center pin in a straight line to the ANT of the module, as close as possible. The bottom side of the PCB will act as ground plane entirely, on the top side everything I don't need will be ground as well.
Is it okay if I solder the outside pins of the SMA connector to the bottom part of the PCB ?
"has to be as big square as the antenna is long", so for my 17 cm antenna a 4 cm x 4 cm board would do ?
..Talking of ground plane:
I wonder how my simple antenna, made from 50cm of thin coax-cable, where the last 17cm is stripped off.
This antenna, hanging out of a soda-can falling in a parashut from 3000 feet up, sends data to groud-station without any problems at all.
I'm amazed, because most 433 MHz units are way smaller than 17x17 cm, and still have good range. I don't even think I have a pcb that large. Ofcourse, I can create a bigger square with aluminum foil for effect, see what it does. I do hope the soldering on a pcb will do the trick. Tomorrow will be etching day.
I use RadioHead library, the module is RFM69HCW. Seeing that other people make do with 17 cm of cable I keep wondering whether I do something else wrong.
Well just to be sure I added some code to read all registers from the module, and that works without problem. The radiohead library states it is intended for use with this module, so I guess changes for a software problem are very small.
Are you sure that your SPI transfer is ok?. I've got Arduino UNO too and I can't connect with RFM69HCW, because the signal in MOSI line is near of 5 V.
I have connected the MOSI, SCK, NSS, through a CD4050 level converter, MISO and DIO0 connect directly to UNO. Source is 3v3.
I'm pretty sure SPI is working all right, because I can read all registry values from the module and the results make sense, I compared them to the data sheet.
The antenna should be in the center of the ground plane not to one side for maximum efficiency.
It is meant to reflect the antenna and make the quarter length into a half wave dipole.
It had nothing to do with the antenna, though I will be trying several setups to see how they affect performance.
I use the RadioHead library. It is meant for all modules RFM69xxx. I'm using the HCW variant, as opposed to the older W or CW. I think the C designates the formfactor, the H stands for High Power (up to +20dBm). Now in order to use the H variant you have to set a specific register to use the high power boosters. The RadioHead library has not implemented this for the H-version. Without setting this it does not work.
My current range is 10 meters line of sight, but I still have to check all register settings and mount a decent antenna. Seeing that people almost squeeze a 1000 meters out of a 17cm wire I have to be able to do better. I have to look into the data rate I'm using now. I will get back on this.
I'll inform the guys at RadioHead about my findings.
