Minimum size of the signal wire in a string of WS2812B LEDs

Other Hardware LEDs and Multiplexing
1

I'm hand wiring WS2812B LEDs into a structure. The power for the LEDs will come from the structure, but I have to sneak the signal line in as unobtrusively as possible. I’d like to use fine magnet wire. Looking at the datasheet for the WS2812B, the input current is 1 micro amp. I assume this is the current drawn by the data line, which implies that the data wire could be very small. However the WS2812B strings coming out of china (ones with wires not strips) have three wires the same size…at least the ones I’ve ordered. I would think if china manufactures could save money by making the signal wire smaller, they would be doing it.

Specifics:
• Number of LEDs: 42, and they could be broken into multiple strands if that helps.
• Distance between LEDs: 11.5" maximum and 10.25" typical.
• Signal line voltage: 5v.

Questions:
• Has anyone tested a small wire for the signal line?
• Is my interpretation of input current from the datasheet correct?
• Is there anything beyond current that drives wire size for the WS2812B signal?
• I’ve read that the signal is amplified at each chip. If so, can I wire two WS2812B LEDs together and extrapolate the results for longer strings?

2

Not tested, but I'm sure you can use very fine wire for the data_line.

3

I would think if china manufactures could save money by making the signal wire smaller, they would be doing it.

No it is much more expensive to make if you have two sizes of wire and not one. For a start the crimped connectors would not cope with smaller wire.

Is there anything beyond current that drives wire size for the WS2812B signal?

Yes stray capacitance can degrade the signal.