Strange 5V pixels flicker behavior

XerxesDGreat

New elf
Joined
Dec 26, 2020
Messages
11
Hey all, I'm very confused as to the behavior of the display I'm in the first stages of testing. I have a Falcon F4V3 with two differential receivers. Currently, I'm running ~300px off port 1 on the controller (run A), 540 (6 x 90 px stars) off port 5 on one differential receiver (run B), and 450 (5 x 90px stars) off port 9 on the other (run C). Run A is a house outline and I have no problems with it. Runs B and C are very similar; the first three stars are powered by one port on a power distro board, and the remainder are powered by a second port. You can see the wiring diagram for B and C in the attached image.

The behavior is that only the fourth, fifth, and sixth stars on run B flicker sometimes (looks kinda like a twinkle). However, the strange bit is that the flicker doesn't occur when running the test pattern on the differential receiver, only when doing test patterns or actual sequences from the controller. I had a similar issue with stars four and five on run C, but bridged the ground between star four and star three and that seems to have resolved it; I did the same for run B and thought it resolved things, but that was when I was testing using the test pattern from the differential receiver; the problem returned when I ran the pattern from the controller.

Some details: the two differential receivers are connected to the controller via 100ft/30m cat6 cables; there is ~10ft/3m from the receivers to the first stars, ~25ft/7.5m between the receivers and the fourth stars, and about 4ft/1.5m between the remainders of the stars. I thought it could be a power issue so I upgraded the long cable on run B to 12awg (from 14awg) and that seems to have not made any difference.

What else could cause the difference in behavior between the test pattern from the receiver and the controller? My understanding is that, since all the stars connected to one differential receiver are all on the same port they would all be affected if it's a connection issue between the controller and the receiver. What else should I test? Thanks in advance for the guidance!
 

Attachments

  • Capture.JPG
    Capture.JPG
    130.5 KB · Views: 23

OzAz

Senior elf
Global moderator
Generous elf
Joined
Dec 29, 2015
Messages
950
Location
Newcastle
Shouldn't there be a ground link between your power supply and the diff receiver, otherwise the Data line doesn't have a ground reference. Which may cause odd data issues.
 

XerxesDGreat

New elf
Joined
Dec 26, 2020
Messages
11
whoops, I realize I left the power supply unit out of the diagram; it's connected to both the power distro board and the receiver, so my assumption is that everything is ultimately running on the same ground (the power supply unit). The pixels are not directly connected to ground on the diff receiver as no power comes from those connections; only through the distribution board and power supply; could that be causing issues? And, if so, why not on the first three stars? (or is that just a "luck of the draw" type thing)
 

BAZMick

Full time elf
Generous elf
Joined
Nov 26, 2019
Messages
257
Try connecting the ground from the pixels to the diff receiver, as @OzAz said the data signal needs to reference the ground from the receiver
 

Katekate

Full time elf
Joined
Jan 1, 2020
Messages
355
Location
Portland, Vic
Shouldn't there be a ground link between your power supply and the diff receiver, otherwise the Data line doesn't have a ground reference. Which may cause odd data issues.
Differential signaling doesn't require a ground reference.
 

TerryK

Retired Elf
Joined
Feb 9, 2020
Messages
442
Location
West Central Ohio
My first thought is: if the problem is not seen with a receiver test pattern, only when from the controller then perhaps a configuration or program setting in the controller. Could it possibly be that the controller test pattern is exercising the pixels differently as well than the receiver's test pattern?

Viewing the posted graphic I noticed 2 concerns, the V- connection between star 3 and 4 which is necessary to maintain the data to V- reference between those 2 stars rather than back through the power injection cables. Also a V- connection from star 1 pixel 1 to the receiver. Again a data reference direct from the receiver rather than routed back through the power injection cables. Not to rule out voltage drop in the cables but the chance for electrical noise through those cable lengths perhaps causing problems as well.

You did not indicate the model of the power distribution unit you are using, I assume the AG and BG terminals are tied together in that unit. I also suggest checking the connections to insure they are good, that is no corrosion, tight, and such. You might enhance your drawing a bit with additional detail and re-post. Out of curiosity, if the cable length from the power distribution to the stars is about 25 feet, why is the cable from the controller to the receivers 100 feet?
 

i13

Dedicated elf
Joined
Jul 5, 2013
Messages
1,049
I'm not sure what is going on here either.

I agree with TerryK's point about connecting the negative directly from star 3 to star 4. This creates a negative loop but doing so almost never causes a problem. The trouble is that this doesn't explain why everything works in test mode.

Flicker could be caused by a dodgy Cat6 cable to the receiver board but that doesn't explain why stars 1-3 work.

Katekate is correct that a differential signal doesn't need a negative reference. This means that you don't need to connect the negative from the F4V3 to the differential receiver. You still need to connect the negative from the receiver to the first pixel but you've already done that because they're on the same power supply.
 
Top