Multiple power supplies LED strip color flickering issue

I have a weird problem I can't figure out.
I have an LED installation with a lot of strips (WS2812B). Because of the huge power consumption I installed two power supplies.
One is 5v 30a, the other 5v 40a.
Ground IS connected between both. Plus is NOT connected.
Data comes from an ESP32. I tried different codes - FastLED and WLED, both have the same Issue.

The issue:

Sorry for the bad picture quality. My phone sucks.

and idea what could cause this?
Thanks.

Hi,
Can you please post a copy of your circuit, in CAD or a picture of a hand drawn circuit in jpg, png?
Hand drawn and photographed is perfectly acceptable.
Please include ALL hardware, component names and pin labels.
Make sure you show how your supplies are connected.

Can you please post link to specs/data of the power supply.
If you are connecting both supplies in parallel, that could be your problem.
Have you split the LEDs up into to separate supply rails, with common gnd?

Can you post image(s) of your wiring of the LEDs, you should have bypassing wires to help provide current along to the end of your LED strips.

Please don't use a Fritzy image.

Thanks.. Tom... :smiley: :+1: :coffee: :australia:

Thank you for the friendly reply.

The 30a power supply is this one:

The 40a is this:

As written I'm not connecting them in parallel, ground IS connected, 5v is separated between both supplies.

I have an ESP32 and 4 clusters of LEDs. Each with its own output on the ESP.
2 Clusters are driven by the 30a power supply, the other 2 by the 40a.
ESP32 is connected to the 40a (but the problem is the same when its connected to the other one)

I have several Power injections along the length of the strip.

The only thing I can think of is an unintentional cross-connect between the supplies, which you presumably have checked for.

I checked with the multimeter and cant find any connection. Maybe I miss something. :man_shrugging:

Thats the crude shematic.
It looks a bit complicated but is actually quite basic.

ESP32 - 4 Data Outputs
Two power supplies
Some injection lines along the way

LEDs 3 in the sketch is the one doing the funky colors.
But not LEDs 4 for some reason.

I see no connection between 40A 5V - side, and gnd of Arduino, unless there should be a dot where the two black lines cross.

All black lines are connected. All grounds are one big net. :wink:
Sorry, I'm not good in schematic symbols.

By the way. I tried to connect "LEDs 3" to different outputs on the ESP32. Also changed the code to only display some of the LEDs. Still the same effect (only with less LEDs). Still on "LEDs 3". It has to be something in the wiring or power distribution. Strange thing is there is no Problem on "LEDs4" which is in the exact same power wiring.

Hi,
Thanks for the info and diagram. :+1:

Have you got the series resistor in each of the signal wires to each LED strip?

Note the suggestion;

  • Before connecting NeoPixels to any large power source (DC “wall wart” or even a large battery), add a capacitor (500–1000 µF at 6.3V or higher) across the + and – terminals as shown above. The capacitor buffers sudden changes in the current drawn by the strip.

Tom... :smiley: :+1: :coffee: :australia:

Neither resistors nor Capacitors show in your schematic, so I'd suggest adding them before proceeding.

Thanks for the tip. I added resistors and capacitors now. Still the same problem. :frowning:

I made a video to better demonstrate this strange behaviour:

  • First is only 40a power supply turned on. Everythings nice.
  • Then I turn on and off the 30a supply. Stranger things happen
  • Then I dial down the overall brightness. Stranger things become more uniform and go away.
  • When I dial brightness up again it comes back.

Clearly, when you energize the second power supply more current flows, overall. Clearly, when you then reduce intensity, less current flows. I have a strong suspicion you have a weak interconnection, or too many LEDs on one of the supplies, but we have yet to see a picture of your assembly, nor have we seen your code, either of which would have helped give a sense of the magnitude. Certainly, with 70A of 5V current available, there'd better be some beefy wiring involved.
Please provide a bit more detail.
Thanks

I solved it.

The ground connection between the two power supplies was too long.
I made a direct connection between both. And that solved the issue.

i'm still suspicious, since that reference between supply grounds should only ever see signal return currents (at most a few mA), unless you've crossed something up, but whatever, if you're happy, we move on.

How can I test for a crossed connection or a wrong wiring?
I tested everything for continuity with the multimeter. And there is current flow in all the grounds and in both the 5v circles separately.

Well, you could simply disconnect the four power supply leads, then check for continuity between them. Should be infinity between the + leads, 0 or close to it between the - leads. Then remove the wire you say connects the two, and you should have infinity between the two - leads. If not, investigate. That's what I'd do, anyway, for starters.

1 Like

Have to ask, did you find anything unexpected?

Thanks for asking. Actually not. Everything is good. Everything around 0.1 Ohm. 0.3 was the maximum I found. But that could be due to being on the other end of the line.

I will change my injection cable to something bigger. That one is the last thing I can change. Otherwise everything should be good to go.

Thanks for your help.