I’ve got a pretty standard 64x64. Standard in the way that it works without special settings with the Pimoroni Interstate 75 board.
While waiting for adapters I’ve ordered, I was now trying to wire a single board to my Pi 3B. But when I run the demo, it looks all wrong. I have triple-checked the wiring, but can’t see a mistake.
I’m using the rainbow-colored cables that come with the panels. I wonder if someone else has used the same cables and could compare the colors with mine.
Even pins
Odd pins
How’s how it looks when running Demo 0 (It’s flickering randomly all over):
Friggin hell - I cannot add more pics nor more links to offsite pics. What do I have to do to get this limitation lifted?
Here’s a link to the other pics I wanted to show: Index of /tmp/led-panels
Please show a clear photo of the panel rear side.
Please indicate the designations of all chips installed on the panel
The connector pins are standard, as I said. Otherwise, the Pimoroni module wouldn’t work, right?
There’s a 74HC245KA and several DP51260.
Does that help?
Thank you
The main driver type is DP5125. The next thing we need to learn is a multiplexer type.
Could you please to read the text on U3 chip?
Also, a picture of the whole rear view would be useful.
U3 is a 74HC04D
Fullsize pic:
The spotlight isn’t ideal for this. Gotta wait for daytime to get the labels to read better.
Sorry, please read the ID of U5…U8 chips
Got my pro camera equipment with macro lens out and finally made them all readable.
U5-U11 are RUC7258E
Thanks
Your panel has a standard type led driver and a standard type multiplexer as well. It should works with the library. You don’t need any specific settings to run it.
Perhaps the issue is in the connections. Your pictures with the wires don’t make any sense for me.
There is a connection scheme in the library Readme - use it. I connect the panels to my RPi with wires myself, without additional adapter boards - everything works fine for me.
Thanks for confirming that it’s a standard board - but I already know that, as I wrote.
There is a connection scheme in the library Readme - use it.
Well, of course I used that. That’s how I figured where which wire should go.
I was mainly asking if anyone could compare their wiring against mine, provided they’re using the same colored cable (which comes with various of these board - I have several types here, as I’ve been using them for months, just not with an RPi), as I suspect that I mixed them up even though I cannot find an error myself.
Or maybe someone recognizes the pattern for the demo 0 and can tell me what kind of wrong cabling (like mixing up certain signals) would cause this.
I hope to receive the Reichelt adapter board later today (see other post of mine) - hopefully that’ll work out-of-the-box then. That’ll still leave me with the question why my own wiring doesn’t work, though.
I don’t think that focusing on colors is the right approach. For example, I have never seen colored cables for HUB75 panels - all my cables have gray color for all wires.
The pins on the RPI have numbers, and the HUB75 connector on the panel has pin names. The table in the Readme compares one with the other. Make your own table of which pin of your RPi is connected to which pin on the panel and post it on the forum - and somebody will tell you whether you connected it correctly or not.
I’ve received the adapter boards, but I still get the same results. Which suggests the problem is not with the cabling. And since the same boards work with a Pico, it’s not them, either.
I only have a Pi 3B and a Pi 5. I booted the Pi 5 from the same SD card that I use on the Pi 3B. That booted up fine.
When I ran the demo -D 0 on the Pi 5, it showed no effects on the matrix board, but the strange thing is that I can’t then even stop the demo with ctrl-C. So that’s a no-go for testing, apparently.
What could be the problem now? Software? Or damage GPIO on the Pi?
About the software: I installed the latest bookwork with ALL programs, then turned off the audio in config.txt. That still got me the msg from the demo program that the sound driver is loaded, so I also added the blacklist as instructed. That’s all I did.
Are there known issues with using the current full bookworm system on a Pi 3, perhaps?
Or issues with the 64 vs 32 bit version of the OS?
Found this thread about testing GPIO: https://forums.raspberrypi.com/viewtopic.php?t=180505
Ran the testgpio script, which prints:
Testing...
Write 1 to gpio 17 failed.
Pull up on gpio 17 failed.
Write 1 to gpio 18 failed.
Pull up on gpio 18 failed.
Skipped non-user gpios: 0 1 28 29 30 31
Tested user gpios: 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
Failed user gpios: 17 18
But if Pin 18 (LC) isn’t working, shouldn’t I still get a fairly clear pattern if I run Demo #0? And not the total garbage that I’m showing in the pictures?
have you
- tried --slowdown-gpio set to at least 5?
- tried all multiplexers
- tried all --led-row-addr-type
Also, if you’re going to be working with these, I recommend you buy an electrodragon active-3 board (see readme) even if it’s going to take 1-2 weeks to get to you. Those boards save you a lot of trouble and have the level shifters you need
Yes, tried all those yesterday already. Will check again. Currently trying to install the Adafruit code for Pi 5 to see if that helps.
Regarding electrodragon: I am the one who wrote that I ordered the alternative one from Reichelt. I’m using that now. I wanted to get something working next week. I have made my own solution where I drive 24 panels with 4 Interstate75 that get their data over WiFi from the Pi 5, but wanted to see if I get better performance with your libs. But I’m stuck at other problems as you can see 
understood, good luck with the adafruit code, but please note that adafruit uses a different pin mapping on both pre-RPI5 and RPI5, so if you use this lib with adafruit, make sure you use the alternate mapping in the code
Thanks. Ordered a Pi 4, which should arrive Mon or Tue. Til then, I’m trying to get the Adafruit stuff installed, but am already running into problems with that, such as packages missing others, and those not installing due to the venv stuff, and I’m already close to giving up this Linux mess again.
Still, I think it’s strange that just because of Pin 18 being dead, I get nothing recognizable on the LED matrix at all, don’t you agree?
I don’t know if your pin18 is really dead nor do I know by heart what pin18 would do. Unless you are missing the clock pin which is very important, any other pin would only affect half the matrix or a single color, or some lines
Good news: I just bought a Pi 4 B locally, and now it works, with the Reichelt adapter.
So the dead pin 18 (LC) was indeed causing the trouble, it seems, or maybe even other damaged gpio pins that the pintest program didn’t find. (I’m still booting off the same card, so it’s not the software).
oh, great to hear, any idea how that pin, broke?
No idea. This was an old Pi 3 that I had laying around. Perhaps I broke it years ago. I had not done any wires gotten wrong when I did my wiring the other day, especially not on the pins 17 and 18.
