Pixel 101

[RoaringChestnuts]

Just have to ask... is there any thread or video for Pixel 101 ---


I would like to understand the make up of the WS2811 pixel  how the Falcon controllers control them etc....


Just dumbed down quite a bit....


Who develops this stuff and what's on the horizon for this hobby technology wise... what's beyond the WS2811, etc...

[AAH]

I don't know of a thread or video but here's a very basic runthrough.
The WS2811 pixel is a 3 channel 8 bit constant current pixel that controls each of the 3 attached leds (usually as part of rgb leds) to a current of 18.5mA per channel.
The data (like 99% of pixels) is a 5 volts signal regardless of whether the pixel is 5V, 12V or 24V
The data stream consists of a reset/latch signal and 24 bits of data that is encoded onto the clock signal. The signal for a "1" is a different width to that of a "0" and the timing between clock signals is reasonably critical.
Each pixel grabs off 24 bits and clock signals from the data stream and passes on the remaining data to the next pixel. What this means is that the reset/latch signal gets progressively wider as you get further long the pixel string.
No pixel knows what is happening to any other pixel. As far as they know they are the 1st pixel in the string as they just grab the 1st 24 bits of data that they see after the reset/latch. It is for this reason that the number of pixels per string/prop is know accurately when configuring your sequencer and pixel controller.
The dimming of the leds is controlled by the 0-255 brightness info being sent out by the pixel controller and being latched into the pixel chip. The chip then uses is own internal clock to pulse width modulate the output to the led. At 0 brightness the output is controlled at 0/255ths (times clock ratio) of a second. At 255, or full brightness, the output is on for 255/255ths of the time.
The constant current of 18.5mA is maintained by dropping voltage across the pixel chip. For a 5V pixel there might be 2V drop across the red led, another 1V across a series resistor and 2V across the pixel chip. The green and blue leds with their higher voltage drop might be 3V across the led, 1V across the series resistor and 1V across the pixel chip. When the voltage drops to 4V there is still voltage headroom to drive the red led to 18.5mA due to the lower voltage drop across the led but the green and blue will no longer have sufficient headroom to maintain the 18.5mA and as thus won't turn on as hard and be as bright. This results in the pixels appearing pink.

Hope that clears up some of the questions. If you want more info just sing out.
There's data sheets floating around for the WS2811 pixel that go into some of the technical details and clock timing etc.

[RoaringChestnuts]

I'll have to digest this but this is exactly what I needed.


The reason I was going to ask is I was curious to know if all of the data transfer could in one way or another provide diagnostics of pixel health.. meaning.. if you have pixel go bad other than visually seeing its bad that some how the data stream could report back.   I am guessing not.


I just ask because folks who have 10k 15K 20K of pixels other than something that is gravely obvious how would they know if a single pixel is bad, etc.