LATEST RELEASE:  FPP 6.1 - Download from here -



Welcome, Guest.
Please login or register.
Forgot your password?

+-Site Stats

Total Members: 15474
Latest: Sobal44
New This Month: 96
New This Week: 11
New Today: 7
Total Posts: 126881
Total Topics: 15557
Most Online Today: 105
Most Online Ever: 7634
(January 21, 2020, 02:14:03 AM)
Users Online
Members: 5
Guests: 73
Total: 78

Flickering on BBB and p10 panels - resolved

Started by drlucas, May 27, 2015, 04:27:45 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.


So you have a 2nd bbb and octoscroller? 
If so then maybe  now look at the power supply s or the panels


I did get a little bit "dirty" with the power. I took the left side of the matrix and connected them to one power supply and the right side of the matrix and connected to the other power supply. 6 and 6 of the power cables that came with the p10 panels. one cable for two panels. then I took the end with the u connector on it, and put a screw in a piece of wood, then put the power output from the psu (all three of the wires that come out of the psu - maybe I should only use one of them and just) under the joined together u connectors and screwed down the wires together.


The 5V connection on the octoscroller is normally not used.  I believe it can be used to power the BBB from a regulated external 5V power supply.

Can you try running a sequence that uses just the top half of the matrix and seeing if the flicker/garbage shows up on just the top or all over?


OK - i'll do two tests tonight...i'll create a 1/2 matrix model and just test the first 40 rows of the matrix instead of all 96 and see what happens. I'll also ensure that I just have a single +5v power output going from the power supply to the group of panels. I may also just bridge the left and right sides of the board and only power from one PSU. according to the wiki, 24 panels at all white should be less than 40amp and according to the spec, I should be ok with a single power source. will report back.


did a quick test (will be late for work - doh!) but with the single power source (single wire) the problem still exists.

i'll take some better pictures and do a drawing, but with no power supplies connected, just power from from the BBB it would look like just the red is turning on the p10 panels and it is powered faintly and I can still see the strange artifacts with the circle effect. will do more testing later tonight!!


So....when I do a simple chase on my panels for channel 123 in standard mode, the first light of every panel on output 2 chases and the very last light of the last panel. it's quite messed up.

I'm going to delete the channeloutputs.json file and see what happens after a reboot.


I don't get it. But the flickering is gone. I am trying different ports on the BBB but looks like no matter where I connect the middle section it doesn't like it.

Is there a better way to debug the panel setting?

Sent from my iPhone using Tapatalk


My guess from that video sample is that you don't have the right output number configured for the middle two rows.  If you don't have the right output number configured, then all panels on that output will act like the panel they were last configured with which is normally the top left corner of the display.

I tested your sequence on my panels and did not see the flicker, so I think it was config related.  I was going to ask you to post a screenshot of the config UI because I thought you might have your width and height swapped, but I see that is correct.  what have you changed lately on the config screen?  The flicker you were seeing were actually the balls, but I think they were moving so fast that it was hard to see what was going on.  In your original video with the flicker, it looks like you had the output numbers correct because the balls were floating from top to bottom and vice versa across the 6 rows of panels.


I changed three things last night

1- moved to a single power supply
2 - reseated the BBB octo cape
3 - moved from ports 1/2/3 to 8/7/6

When I reseated the BBB, i also screwed it down on the back of the panel wood because it was dangling in thin air hanging by wires before - because of the hole in the wood in the back of the panel for access, I have limited locations to easily screw down and didn't realize that the 16pin ribbon cable was not long enough to actually get to port 1/2/3 anymore so had to move to 8. I wanted to minimize the length of ribbon cable, so I didn't even leave me 3" of slack. I'm going to likely redo my ribbon cables this weekend and clean up the mounting hardware.

For the ports on the octo, are they not top down left side is 1/2/3/4 and on the right side it is 5/6/7/8?

My assumption is that JP1=O1 , JP2=O2, etc. Is that the case?

Is this look OK from the /var/log/messages file?

May 28 23:30:50 FPP2 kernel: [   75.330999] bone-capemgr bone_capemgr.9: part_number 'CAPE-BONE-OCTO', version 'N/A'
May 28 23:30:50 FPP2 kernel: [   75.331084] bone-capemgr bone_capemgr.9: slot #7: generic override
May 28 23:30:50 FPP2 kernel: [   75.331104] bone-capemgr bone_capemgr.9: bone: Using override eeprom data at slot 7
May 28 23:30:50 FPP2 kernel: [   75.331125] bone-capemgr bone_capemgr.9: slot #7: 'Override Board Name,00A0,Override Manuf,CAPE-BONE-OCTO'
May 28 23:30:50 FPP2 kernel: [   75.331233] bone-capemgr bone_capemgr.9: slot #7: Requesting part number/version based 'CAPE-BONE-OCTO-00A0.dtbo
May 28 23:30:50 FPP2 kernel: [   75.331253] bone-capemgr bone_capemgr.9: slot #7: Requesting firmware 'CAPE-BONE-OCTO-00A0.dtbo' for board-name 'Override Board Name', version '00A0'
May 28 23:30:50 FPP2 kernel: [   75.333037] bone-capemgr bone_capemgr.9: slot #7: dtbo 'CAPE-BONE-OCTO-00A0.dtbo' loaded; converting to live tree
May 28 23:30:50 FPP2 kernel: [   75.334519] bone-capemgr bone_capemgr.9: slot #7: #3 overlays
May 28 23:30:50 FPP2 kernel: [   75.355812] omap_hwmod: pruss: failed to hardreset
May 28 23:30:50 FPP2 kernel: [   75.369602] bone-capemgr bone_capemgr.9: slot #7: Applied #3 overlays.
May 28 23:30:50 FPP2 kernel: [   75.394127] pruss_uio 4a300000.pruss: pins are not configured from the driver


Appears to be resolved as I hunted for the right port...I don't understand the output to jp mapping!!

No more flickering and the panels are all aligned!

Sent from my iPhone using Tapatalk


That's more like it!  Now you can go to work without it bothering you all day.

I do see the same "video recorder" jitter in your Vimeo shot that I saw in mine. The 8:1 timing, mixed with the 50ms framerate can make for flickers that you don't see with just the eyeballs.

Long time Falcon, FPP and xLights user


Looking good!  Glad it's working, and your port numbers match the labeling on Jon's picture and the board.



Thanks all the troubleshooting along the way to resolution. Now to start seeing what I can with with the new version of xlights and images and moving them around the matrix.

Powered by EzPortal
Powered by SMFPacks Menu Editor Mod