Should I be able to measure the voltage on the Pixelblaze data output? I’m not sure if different patterns might change the voltage reading. In any case I thought I read somewhere that it should run at 3 volts but mine measures around 1.5 volts.
I’m having some flickering and glitchy behavior and I’m not sure if it’s a problem with the data or bad pixels. This is the 2815 Style running 12 volts. I have tried many different things to confirm I don’t have any connection issues or anything like that. The controller-to-LED length is not affecting it. I tried very short run with no change in behavior,.
I have two strings of 600 running in parallel and one string works perfectly. I’m really leaning toward bad pixels in the other string.
The voltage you read in a DC voltmeter will depend on the average number of 1s vs 0s. Anything between .1 and 4.9 is possible. An oscilloscope will show you the signal integrity - you’re looking for changes between 0 and 5V that aren’t too rounded, don’t have ripples, and don’t overshoot or undershoot 5 V.
What’s the lowest voltage you can measure along the strips themselves? 600 at 30/m in 2815 will still require power injection at several points - usually every 10m for 12V strips.
Does the glitching get better with a lower brightness slider? That’s sometimes another good indication of a power injection need. I’d the glitching point consistent across the entire length or does it move with different setups? If it affected by placing your hands on/around the signal line leading from PB to bad strip? If so, your hand’s water content is changing the capacitive coupling of the RF environment around the data line and it’s likely data signal quality related, not voltage.
A bad pixel will sometimes result in good data up to its input, then glitching suddenly after it. While a bad pixel can also screw up in other ways like making the whole strip bad under all conditions, the unchanging glitch position start is almost always a bad pixel sign.
If you have both strips running in parallel, there can be odd data reflection issues in the Y-connection of the data line. Try disconnecting the good strip and see if the bad one recovers. If so, your splitter is causing the issue and a 5V buffer (or replicating the pixels with the output expander) will probably fix your issue.
@jeff thank you for such a detailed reply. I apologize I haven’t been able to get back on this issue until this morning. I think your suggestion of data reflection issues seems to be the culprit. When I completely disconnect one side or the other, the glitching seems to stop. I’m letting it run a while to be sure but fingers crossed that may be it.
I wonder if the interference only occurs because of the close proximity of the two wires just after the split. Maybe I’ll try wrapping them in tin foil or something just to test it out… I’d love to be able to shield the wires somehow and get things working correctly for both strings to run at the same time.
Update - I think I’ve gotten rid of enough of the interference at the split to make things work temporarily. As long as I keep the brightness setting at 80% or less, all is good on both LED strips. If I go higher, I start getting some glitching on the more problematic string.
So it sounds like an output expander would clear this up?
Update number two! Seems I may have spoken too soon. Just went out to check and the problem string is glitching again. Tried bumping the brightness down to 30% and unhooking the other side but still is glitching. So I’m not sure what to try next. I don’t mind trying an output expander, if it doesn’t help it also will not hurt and I will have it for future fun.
I think an output expander is a good option to troubleshoot. If it’s a misbehaving pixel, the worst that can happen is you narrow it down conclusively.
I think with a process of elimination you can get there.
What’s your distance from controller to Y, and from Y to each first pixel?