OK, I have a reason for asking this question I swear, so bear with me. I've just ordered a prewired string of WS2812B pixels, but they won't be here for weeks and I'm hoping to get an answer on this before then.
Let's say I have a serial string of ten WS2812B pixels connected to my Arduino, and I use my Arduino to turn on the 8th pixel but nothing else. Then let's say I crack out my volt meter and touch the leads to the spots on that eighth pixel - VIN, VOUT, GND IN, GND OUT, DIN, DOUT. What do you expect I would see on each contact in terms of volts and amps? For this scenario you can imagine that I'm either using the 5v out on my Arduino or that I've connected the strip to a separate 5v 2a power supply if you prefer.
Second equally important question - what would I see if I touched the volt meter to the contacts on the 7th pixel, or the 6th? Or the 9th? Again, presume I only have the eighth pixel lit.
Like I said, I promise I have a valid reason for wanting to know. There are lots of guides out there about how to set up these addressable LED pixel strings, but not a lot of them get into the weeds about how they work to the point that I could find this out before I got them and tried this test myself.
You would get voltage across all of the wires at every pixel...
You have to define the total number of leds connected, which would be ten. There is still signal going to the ic, even if the LED is off.
If you decided to tell the micro controller there was only one led connected, only the closest led would be controllable, and the rest might blink randomly
Thanks QDeathStar! So, there would be nothing unique about the voltage coming in/coming out of the one lit pixel versus the unlit pixels? Each pad (including the data in/data out pad) would all be receiving the same voltage/amperage, whether the pixel was lit or unlit?
Ah, well, apparently nothing Sounds like I can't do what I was hoping I could do. I was hoping I could wire a reed switch to one of the pads on the LED pixel, so that when the LED pixel was active, the reed switch was receiving a little juice. I was hoping that the other pixels were not putting out any juice if they were not lit, at least on the data pad, so that the reed switch wired to the active pixel would be the only active reed switch.
The reason why that mattered to me would require me to explain the whole project, and I don't want to bore anyone. But, thanks again!
What you need is a strip that uses the WS2801 driver chip. That has external LEDs that you can tap the power off. Not sure if there is enough extra current capacity to power the coil of a reed switch, or how the inductive load would affect both the LED and the controller but at least you get a voltage indication of what LED colour is on.
The output at the LED pads is PWM signal. You could feed that into a buffer chip (74HC125), low pass filter the square wave to get a DC-ish level, feed that into a comparator (LM358 like on the Uno) to get a high only when the PWM level was above some limit, and use that output to drive a transistor to turn on the reed relay.
Or maybe skip the buffer, and just feed the lowpass filter output into the comparator.
Now monitoring the data lines is of no value as for one thing, the data is encoded and the "pixels" of either sort (two-wire WS2801 or 1-wire WS2812) are "smart" devices with memory, so the data is sent only when the pattern needs to be changed.
There is nothing about the supply voltage which indicates whether a particular pixel is on - nor should there be as your whole intent is to maintain the supply voltage adequate and constant. The current draw if you could intercept the connection between supply rail and pixel, would tell you that the pixel was lit - but not which colour.
The only useful way of doing this would be to use the version with separate chip and LED, and you would need extra circuitry to condition the LED voltage to control a relay.
Sounds like I can't do what I was hoping I could do. I was hoping I could wire a reed switch to one of the pads on the LED pixel, so that when the LED pixel was active, the reed switch was receiving a little juice. I was hoping that the other pixels were not putting out any juice if they were not lit, at least on the data pad, so that the reed switch wired to the active pixel would be the only active reed switch.