I was flying in an industrial are the other day with my FPV setup and even when the picture was clear it would go all rainbow colored which was very distracting, additionally my range seemed to be shortened quite a bit. It looked something like the photo below which has been photoshopped since I wasn’t recording the downlink.
I came home and did some reading and research and discovered that this may be due to multipath interference, basically where the transmitted video signal reflects off of various objects and multiple copies of the signal come into the receiver, each one shifted slightly from the other. This causes problems with the timing of the colorburst and synchronization issues and makes it difficult to fly, which is why the video below is pretty restricted to where I fly, it kept going fuzzy or multicolored.
The most common solution to this problem is to use circularly polarized antennas. Circularly polarized antennas are better in situations where you have multipath interference. The most popular versions of that design are undoubtedly IBCrazy’s Skew Planar and Cloverleaf designs which he has posted over at RCGroups for anyone with some soldering skill to build. He also sells pre-built antennas on various sites which have the added benefit that they’ve been RF tested and tuned but are $30-$40 each. Using the information provided at the links below, many people have had good success without hundreds or thousands of dollars worth of testing equipment.
I built two antennas. First I built the Cloverleaf 3 lobe antenna to be mounted to the hexacopter. These are big antennas at almost 7 inches in diameter for 900Mhz, but for 2.4 or 5.8Ghz, they’re much much smaller. The 5.8 is smaller than a golf ball.
So far I’ve only been able to test the antennas for functionality and short range, in certain circumstances I’m able to tell that the reception actually is much better, but I haven’t been able to try it again in an environment with lots of structures to see if this resolves the multipath situation. I’ve not been able to test the video range either. The original antenna on the Multi was an IBCrazy Vee which had reasonable range and the base station had an omni-directional antenna, and when in diversity mode also had a second receiver with a flat patch style antenna.