Flickering pixels b...
 
Notifications
Clear all

Flickering pixels between segments

43 Posts
14 Users
0 Reactions
12.8 K Views
dmcole
Posts: 2196
(@dmcole)
Noble Member
Joined: 19 years ago
Dpdigital wrote:I should also note that in order to get the pixel spacing I am looking for, I have added short extension pieces of 18/3 in between each pixel, with soldered connections and covered with individual heat shrinks.

One last ground question: Have you checked the soffit outlets with an electrician's tester such as this (for illustration purposes only)? Just to make absolutely sure the ground isn't floating somewhere?

So now I'm concerned about those "short" extension pieces. You've actually never said what type of pixel you have, but with 12 volts and a Komby, we're probably talking WS_2811 ... the spec says the signal between WS_2811 nodes shouldn't be more than 10 meters (32 feet, 9.7 inches) and I doubt you've exceeded that, but just raising the issue.

The next concern would be the electrical quality of those extension pieces ... maybe the signal is somehow getting hammered going through the junction where the new power supply is getting introduced.

My last observation is that 110 pixels is really just right on the cusp of being capable of being driven by one power supply, no less three. If it had been mine to do, I might have tried it with one to start with (and if that hadn't worked, my second plan would have been to send a 16-gauge single 12+ wire down to the end).

HTH.

\dmc


Reply
Posts: 79
Topic starter
(@dpdigital)
Trusted Member
Joined: 12 years ago
dmcole wrote:Have you checked the soffit outlets with an electrician's tester....Just to make absolutely sure the ground isn't floating somewhere?

we're probably talking WS_2811 ... the spec says the signal between WS_2811 nodes shouldn't be more than 10 meters (32 feet, 9.7 inches) and I doubt you've exceeded that, but just raising the issue.

\dmc

That is a good suggestion. I will verify the line voltage grounding and report back with what I find.

Yes, we are talking WS_2811. I don't have any nodes that are that far apart. Basic node spacing is about 10 inches, but I have some wires connecting between segments that are up to about 8 feet max. Although, when I had some that were spaced at about 10-12 feet between segments, I got some erratic behavior. Probably due to the same issue we are talking about is my guess.

The reason I have 3 power supplies is more to do with voltage drop and wire distance along with power distribution for future additions, rather than number of pixels. Perhaps I am a bit overkill on my calculations and looking at it wrong. As an example, my first set of segments which runs the roofline over my garage, runs a length of 25 feet and a wiring length of about 30 feet with some connecting pieces and without the injection points. That run has 32 pixels and 3 floods. Without injection, the voltage drops to about 10v. My understanding on voltage drop for 12vdc with 18ga wire was to limit it to about 30 feet and under. Does that seem right? That kinda brings up another question about injecting power. With the power supply at the beginning of a string, how far down the line can you run injection wires from the starting point, without having voltage drop on the injection line? Still about 30 feet I would think? Total end to end distance of my roofline is about 75 feet.


Reply
Posts: 2507
(@mrgrumpy)
Famed Member
Joined: 12 years ago

My experience said I had voltage problems....my final findings were that the PIXEL wire is much smaller than my calcs, therefore becoming many resisters. 18/3 spacer wire doesn't help much when pixel wires are likely sized at #24. I needed to power inject more frequently. I finally injected at every 59 pixel (without adding ampacity - just beefy power and ground parallel cable), and at the end of my run too. Everything is now perfect. You may not need as few as 59 pixels, but that was my convenient break point. This similar problem existed on some floods too. Calcs with a good margin says they should work, but they flickered. I doubled the power injection (added ampacity here) and problems gone.

May not be your solution, but information to think about.


Reply
Posts: 79
Topic starter
(@dpdigital)
Trusted Member
Joined: 12 years ago

Ok, so after verifying the electrical grounding and integrity, everything checks out. I even tried removing and bypassing the timer switch that was running the whole circuit to see if it was causing interference. The results were the same. I added additional power injection completely to the end of the segments and still got the same results.

The wire that came on the bullet style pixels is in fact 18ga.

I'm thinking that it has to be noise or something on the data line that has increased sensitivity to multiple power supplies being used. It definitely is only affecting the last one or two pixels, and the first one or two pixels at the separation point between two power supplies.

I'm baffled!


Reply
RobG
Posts: 1255
 RobG
(@robg)
Noble Member
Joined: 14 years ago

Are they flickering when there's no signal? If yes, then it's probably power rail affecting data line (poor filtering at the output of PS.)
If flickering happens when data is present, then you can try adding series resistors on data lines to reduce reflection.


Reply
Page 2 / 9
Share: