OH2 as a backend server

I used a Rubber Duck type on one end of the RFM69 devices and a single straight wire on the other end. For the rest its a RFM69 Rubber Duck antenna on the RFXCOM Transceiver and either a long wire on the AOKE relays or the built in PCB type on the Home Easy Stuff.
I’d say the AOKE’s seem more sensitive than the Home Easy Sockets, and with the RFM69’s its likely the length of transmission that causes the problem, which was a lot of packet loss which wasn’t that helpful when I was using it to monitor overloading of the Mains!!
I was an RF Engineer in a former life and listening to the channel with a scanner reveals the problem as an almost continuous stream of squarks together with the odd socket turning itself On without a command.
Ultimately 433MHz has a greater range than WiFi, which also means it has to compete with other stuff further away, especially after passing through the building fabric which attenuates both both frequencies.
The easiest way to see if its doable is to go old school and tune a scanner to 433MHz and see how busy it is, after discovering mine was unreliable I tuned in and discovered why

I used to work with a bunch of signals processing guys so I know enough to know I’m in way over my head when it comes to this stuff (take that Dunning-Kruger!).

I’m using straight wires all the way around. I tried curling them up so they would fit inside the enclosures (read that should work on some forum somewhere and my signal completely went away unless both devices were on the same table (the enclosures are wood and plastic), forget about across the room even. I also don’t really have a way to solder on and position another wire for the dipole so I have the antenna ground wired to the Arduino/RPi ground. Probably not the best design.

When I get to building a weather station and garden monitoring sensors I may revisit and see if I can improve the antenna situation a bit. Better grounding would probably be all I need, and maybe a Rubber Duck antenna would work out better for me on the Gateway.

For now what I’ve got going works. :slight_smile:

Yep, Antenna design can be interesting, and if I was designing even a basic antenna for range I’d just use a simple dipole of the correct length for 433MHz (around 17.3CM for each leg) assuming there isn’t any loading already on the board. Usually the Grounded bit is missing which if the science is to be believed will compromise the range somewhat, especially on the transmitting side, although maybe its relying on the power feed wires to do this job. There is a downside though as a better aerial would also pick up more interference from elsewhere so it would be down to how well the onboard receiving chips at each end deal with it.
Using straight wires assuming the transmitter and receiver antenna’s are in the same direction makes sense. Having them in different planes of being off axis will drop the signal considerably.
In the case of 433MHz around here, the problems are down to interference from other devices more than anything so assuming no acknowledgement is used the antenna on the transmitting device is more important than the receiver which may be just struggling with picking the local signal out from the rest of the noise.
I’d guess the devices with onboard printed antenna’s rely on receiving something very small and immediately amplifying it, such is the power of modern low noise electronics, and the low expectation of reliabilty.
The first thing I did with my AOKE relays is extend the antenna wire straight down and that is likely why they are more reliable than the sockets.