Weird but easy way to get super long runs of data cable from OctoWS2811 board to WS2812b LEDs… what gives?!?
I also posted this in the pjrc forum, but thought you might be interested…
I tried to connect a string of 2m of WS2812b LEDS (two 1m sections of 144 leds/m) to an OctoWS2811 adaptor board via a roughly 20m length of CAT5e cable (for data - power is local to the string, ground is looped back to the Octo) and (sort of as expected) it didn’t work (sparkly / flashy Bad Things ™ happening).
But here is the really interesting part… if I put that cable between the two 1m sections of WS2812, then it worked fine. In fact, I’ve tried up to 55m (180 feet) of cable without the slightest problem (to be really silly, I even mixed the UTP and some lightweight figure 8 speaker cable!). To be clear, configuration is:
OctoWS2811 adaptor -> 1m WS2812 strip -> 55m (180 feet) CAT5e & fig 8 speaker cable -> 1m WS2812 strip
Power supplies local to the strips, common ground (0v) cable joining all entities. Voltage input measured at each strip is about 5.05v (uncalibrated cheapie DMM).
This has worked for several hours so far without the slightest glitch.
Has anyone else ever tried this?
Next thing to try - single LED adjacent to the Octo, and use that as a ‘long line’ driver. Also longer lengths of cabling, and driving multiple strips down a single UTP cable.