News:

Server migration complete, Welcome to version 2.1.1

+-+-

+-User

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

+-Site Stats

Members
Total Members: 15649
Latest: Androom
New This Month: 0
New This Week: 1
New Today: 0
Stats
Total Posts: 128374
Total Topics: 15784
Most Online Today: 37
Most Online Ever: 7634
(January 21, 2020, 02:14:03 AM)
Users Online
Members: 0
Guests: 26
Total: 26

FIXED - Panels misbehaving

Started by jnealand, December 05, 2020, 03:19:48 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

jnealand

After trying a lot of things I finally replaced the BBB and Octoo with a BBG and another Octo I had available.  Now all works great.  I'm suspected I may have just needed a reimage or evena a new uSD, but will not work on checking that about until after the seaon.  The uSD that I was using was one that has been in use for over a year and had been upgraded many times without being reimaged.

original post
-------------------------------------------------------------------------------
I posted this to FB as well and this is just a copy and paste so nothing different if you saw that post.

Cannot figure this out.  P10 4x4.  Panels are connected across, 4 panels per row.  This is a 4 year old setup.  Powered before Halloween and was seeing weird results in test mode, but each row all 4 panels were identical.  Since it seemed to be a row issue I reseated all the ribbon cables and the problem seemed to go away.  Updated FPP and ran some tests and all seemed good.  Installed it outside.  Started having issues again with all panels in a row showing identical.  It normally runs off outputs 1-4.
Swapped to outpuits 5-8
swapped to a second octoscrolla
Still no luck.
Put in all new ribbon cables and got the whole display working
BUT
the top row is very bright and the other 3 rows are very dim.
tried swapping output ports again but stays the same.
Swapped the cables for the top and bottom rows and the brightness moves to the bottom row, but I noticed at that point the at was only getting 13+/16 rows of leds on the row lit up.  Swapped back to the top and still only have 13+ rows of leds lit.  At any rate all the panels seem to be fine, but they are not getting something that would make them all be the same brightness and the row connected to output1 is not lighting all LEDs.
The BBB is powered via the barrel jack from a wall wart that came with the BBB purchase years ago.  This is year 4 for this prop and have never had issues with it since I finally got it working years ago.  I do not know what to try at this stage and am looking for ideas.  At night the effects can still be seen on the bottom 3 rows although dimmed.  Now what??  Here are a couple of photos   
https://photos.app.goo.gl/43dsRNWwKhWvtCHH7
Jim Nealand
Kennesaw, GA all Falcon controllers, all 12v Master Remote Multisync with Pi and BBB P10 and P5

AAH

That's a weird 1. A couple of tests I'd suggest.
1) swap the ribbon cable going to row 1 and 2 to see if the problem follows the ribbon or the panels. I suspect that it's going to follow the panels.
2) disconnect 3 of the 4 panels in the faulty row and try running 1 panel at a time to see if 1 panel plays up. Reconnect a 2nd, 3rd then 4th depending on results.

jnealand

The problem definitely follows the port.  And I have tried to do a single row at a time in order to see if I did have a panel problem.  Actually the problem seems to always be on port 1 or 5 now that I think about it.
Jim Nealand
Kennesaw, GA all Falcon controllers, all 12v Master Remote Multisync with Pi and BBB P10 and P5

pixelpuppy

I had a similar problem and it was the BBB.  I just moved that row to another free port. The scroller has no intelligence (most capes don't) its basically just a buffer breakout board, feeding the BBB GPIO's out to the hub75 connectors.   If you're only using 4 ports on an octoscroller you should be able to move the cable to a free port and just update the LED panel settings in FPP.
-Mark

AAH

Quote from: jnealand on December 05, 2020, 04:24:12 PMThe problem definitely follows the port.  And I have tried to do a single row at a time in order to see if I did have a panel problem.  Actually the problem seems to always be on port 1 or 5 now that I think about it.
I would try configuring your octo so that you lose outputs 1 and 5 then. You look to have enough cable so you could daisy chain out of row 2.
I had a look at the octo schematic and there is nothing unusual or common to outputs 1 and 5. All of the shared signals they have are common across all the outputs.

jnealand

I was hooked up to ports 5-8 so I moved and reconfigured row 1 from 5 to 4.  Problem still exists.  Could this be some kind of a power issue?  I did not build this setup and there are 8 panels wired together off one connection to the PS and 8 panels to a second connection.  I do have a BBG that I could free up.  Guess that will be a project for tomorrow.  I do not have spare power cables (maybe one).  Will see what tomorrow brings.
Jim Nealand
Kennesaw, GA all Falcon controllers, all 12v Master Remote Multisync with Pi and BBB P10 and P5

Stormyblade

The dim and bright rows in the panels make me think it's power related. I was having some issues with my matrix last year and it turned out that one of my ground wires for my power distro board wasn't fully seated/connected. This year I made sure I had very good connections going from my power supply to the power distro boards (I've got 3 of them mounted in the back of my matrix). I used those crimpable wire ferrule tips and made sure each connection was on the bare wire - no crimping of ferrules onto solder-coated wire. 

Just for information, I've got one 5V power supply driving all 20 P5 panels in my matrix. I mounted in a separate wall-wart style 5V plug to power my BBB and Octoscroller. Everything was rock solid for me during Halloween and it's been great thus far for the Christmas season.

pixelpuppy

#7
Quote from: jnealand on December 05, 2020, 06:40:03 PMI was hooked up to ports 5-8 so I moved and reconfigured row 1 from 5 to 4. 
Not a solution to your current problem but just for future reference, you'll get the best performance by using up the lower port numbers before using higher port numbers.  In other words, The BBB outputs scan from 1-up.  If you only use ports 1-4 then it scans from 1-4 and the next scan starts back at 1.  But if you skip ports 1-4 and use ports 5-8, the BBB scans from port 1-8 before starting the next scan at 1.

Also, if you have available ports, its better to have more rows of shorter chains than to have fewer rows of longer chains.  In other words, if you have a 4x4 matrix on an 8-port scroller, its better to chain 2 panels on each of the 8 ports instead of 4 panels each on 4 ports.  But if you do want or need longer chains because of cabling or cabinet issues, then you'll want to try and use of the lower ports first and leave the higher ports open.  (of course, skip any ports that may have gone bad)
-Mark

jnealand

The panels were on 1-4.  I moved them to 5-8 to see if iit made any difference.  But in either case it is the top row of panels that is extra bright compared to the the other 3 rows.  It has been working great on for a the past 3 seasons.  Something happened between last season and Oct 2020 when I powered it up again so that I could update FPP.  The back of the unit never came off until I started trying to troubleshoot the problems I was seeing.  Today I will retrieve my BBG that is connected to an F8-B (not being used at the moment) and see if that changes things.
Jim Nealand
Kennesaw, GA all Falcon controllers, all 12v Master Remote Multisync with Pi and BBB P10 and P5

Support FPP

+- Recent Topics

Set Time/Date by breese
February 01, 2023, 05:03:33 PM

Help How I use Orange Pi with FPP by tbone321
February 01, 2023, 02:41:58 PM

No network by Steve_63
January 31, 2023, 08:17:21 PM

F48V4-NS Pixel Controller using all 12 Differential SmartReceiver ports by BirdingPix
January 31, 2023, 02:26:30 PM

F48v4NS - Pi or No Pi? by rjhodgefamily
January 31, 2023, 09:25:20 AM

Trying to configure LOR CF50D Floods on FPP Pi Player with PiCap by JonD
January 31, 2023, 07:01:41 AM

Script/Plug-In Trigger on Data by darylc
January 30, 2023, 11:31:16 PM

Attempting to move to FPP 6.3 from 4.4 - Assigned IP is in different subnet. by Poporacer
January 30, 2023, 05:06:59 PM

Camera input for WS2811's by dkulp
January 30, 2023, 04:02:55 PM

F16v3 Won't boot or power on by rjhodgefamily
January 30, 2023, 02:22:05 PM

Powered by EzPortal
Powered by SMFPacks Menu Editor Mod