Flashing mirrored symmetry

Hey guys,

I’m having some flashing occurring and I’m not really sure where to start on identifying this issue. I’m powering 240 SK9822 LEDs, but only have PB driving 120 since they are mirrored and just splitting the outputs to both sets of 120 LEDs to keep frames higher. The flashing only occurs if I’m trying to drive both sets of 120 LEDs, if I disconnect one the flashing goes away. Looking for any advice on where to begin and solutions.


I’d look at power first. What do the power feed(s) look like?

There are two power zones with common ground. 2 Alitove 5v 15A 75 watt power bricks, one of the power zones back feeds the micro, each power brick has 120 LEDs on it.

I moved the power feeds one contact point forward of the first contact point on the strip to make sure I get a more reliable solder joint. When I’ve doubled up two wires on a contact point in the past it doesn’t result in as reliable of a joint a lot of times.

It could also be extra noise or signal reflection from the additional run. Try adjusting the data speed. What is the total distances to data/clk from the output, and from the ground on PB to the strips?

I have data speed at 1 MHz, dropping data speed doesn’t seem to help. What’s also strange is that if I drop brightness too low, it makes the flashing worse. This doesn’t happen if I disconnect a power zone.

From the PB, the power zone that is back feeding the micro is 1’ of cable away, while the other power zone is 4’ of cable away.

Strange! Does it happen if you have everything connected, except disconnect one or the other data+clock lines? That is, leaving all power and ground connections in place.

It sounds like something is injecting noise.

If I leave power and ground lines connected and disconnect data and clock lines to one half, there’s no flashing (on either the half that is lit that has data being sent to it or the half that has no data being sent to it and is not lit).

My experience playing with splitting a signal between a couple of 200 LED strips was that there was enough level drop or interference introduced by a Y-cable splitter to cause intermittent weird behavior. (We’re pretty much at radio frequencies – does the Y cause reflections or standing waves that occasionally add up to data errors? If it’s not plain voltage drop, would a little filtering on each leg of the Y help?)

Anyway, that project is returned to the queue for the moment, but I was thinking about trying something like this SPI Amplifier and splitter next time instead of the Y cable (unless anybody knows of something cheaper/better!)

1 Like

Seems to be something with the Y-split cable. I remembered I had a 4 output 5v signal amplifier laying around, I took apart the Y split cable and hooked it all up to the signal amplifier and that seems to have solved everything. I also ordered a pro output expander board over the weekend, and I think that should also fix this. Strange that the y-split cable introduces noise on the data channels.