Transmission works with wires but not always with wireless

[dhouston]

It's not really 'problems' I have with the modules. They work fine for me, in bursts. I have a couple of project where one of my remotes acts sort of like a volume control and I have to continuously hold the transmit button down to get the receiver to change, takes about 15 seconds to run from one end to the other. About 3 or 4 seconds into holding the button, it craps out. If I release the button for less than a second and press it again, it'll work for another 3-4 seconds, then repeat. Again, no real problems, and I have a feeling that dhouston's hints earlier may solves this issue.

That's somewhat analogous to dimming a lamp using an X-10 palmpad to send RF to a module that translates it to powerline commands to control the lamp module.

X-10 uses the NEC protocol and the palmpad will send 32-bits continuously with each 32-bit burst preceded by a ~9ms/4.5ms lead-in and followed by a 40ms gap (so each code is ~105ms). It takes about 4-5 seconds to go from full on to full dim and it's very smooth.

I conduct range tests (and RF tune receivers) by clamping a button down on a palmpad. The receiver operates smoothly for the several minutes I leave the transmitter clamped.

IOW, I think you'll do better using the NEC protocol or some variation on it.

One other thing to consider (which I alluded to earlier) is that you usually have at least two timelines. One is for the RF bitstream and one is for the action to be taken by the receiver (e.g. adjust volume or brightness). Things will be smoother if the end action can finish during the gap between code bursts. That way the receiver is ready to capture the next burst(s) and act on it (them).

The NEC protocol (or variations of it) is the most popular (and oldest) protocol used for IR control of AV gear. If it has been so successful there, it makes sense to me to use it in similar low data rate RF applications.