"pebble" pixels get...
 
Notifications
Clear all

"pebble" pixels getting stuck on white

7 Posts
4 Users
0 Reactions
1,421 Views
Posts: 26
Topic starter
(@chrisgeorge0911)
Eminent Member
Joined: 12 years ago
[#57402]

I've started using 12v ws2811 pebble pixels ( https://buildalightshow.com/pixels/1325-ws2811-pebble-lights-black-wire-25mm-spacing-40-leds-per-metre-12v.html ) to replace some older element that were still using ws2811 pixel strip. 2 small candy canes (25px) are working fine.

I've rebuilt my 4 leaping arches using them, 45px each. They're driven by a sandevices e6804 (4.081), single data line to all 4 arches, and power injected for arches 3&4. I'm using FPP to drive the controllers.

However, they work fine until they go full white in my display (usually a short flash or wipe), then they freeze on that white. It's really weird, don't seem to do it on any other colour so I'm wondering if it's some pixel data issue? Pulling and reconnecting the plug resets them fine until they play white again then they freeze.

I've not had this happen before with any other bullet pixels or strips, so wondering if there is some incompatibility in my setup of the pebble pixels or something? I've set a dimming curve with max at 70% in vixen so not even driving them at max

Does anyone have any idea's what might be causing this?


6 Replies
DrNeutron
Posts: 2018
(@drneutron)
Noble Member
Joined: 14 years ago

I have the 5V Pebble pixels from RW and I can get around 150 without injection.
I would expect at worse case you would only need to inject at end of arches.
Maybe try a sequence with no injection and see how it looks.
Is it same PS for E6804 and PI? If not, are the negatives tied?


Reply
Posts: 26
Topic starter
(@chrisgeorge0911)
Eminent Member
Joined: 12 years ago

Thanks for replying.

The injection is ok. Grounds are common throughout.

Just tried with just the first 2 arches connected (ie no PI). Still got stuck in exactly the same place in a sequence.

I will try driving directly from VixenLights without FPP. Just trying to rule out bits. Update: Same behaviour


Reply
MartinMueller2003
Posts: 7614
(@martinmueller2003)
Illustrious Member
Joined: 14 years ago

That is very definitely a power problem.


Reply
Posts: 26
Topic starter
(@chrisgeorge0911)
Eminent Member
Joined: 12 years ago

So haven't totally nailed it, but updating the firmware on the 6804 which then gives the ability to increase the timing from 150ns to 300ns, and change the dimming at the controller output (to 50%) has seemed to fix it!

So could be power, could be timing... I'm tired and hungry so that investigation can wait for another day LOL.

Thanks for your help.


Reply
MartinMueller2003
Posts: 7614
(@martinmueller2003)
Illustrious Member
Joined: 14 years ago

That timing is for the interframe gap and only really affects the first pixel from the controller. The power reduction is what is making it work.


Reply
Posts: 429
(@mikekrebs)
Reputable Member
Joined: 12 years ago

The timing for the interframe gap does affect the all the pixels. But if the pixels are doing the sequence correctly until 70% white, I would say it isn't the interframe gap but something else like power.

However, any E68x that is running any newer pixels from the last 7? years should be using the V5 firmware with the adjustable interframe gap. Almost all the newer ws28xx pixels for around the last 7? years had the interframe gap changed from 50ms to 300ms. If you are running older firmware than V5 and newer pixels, you are lucky to not be having problems. Upgrade your controllers!

Also the pebble/seed pixels are relatively new in bulk (I think I first saw them in 2018) but really took off in the last year or two. There have been a few posts on ACL about problems with them. So, you should definitely ask the supplier for advice. Your power injection should be just fine.


Reply
Share: