Hey, i followed the diagram from this post to set up a pixelblaze with an output expander for LED strips which are powered directly from the supply as the current will be too high for the board. It’s almost the same except I’ve added fuses to the power lines. There’s a power and a ground wire coming from the power supply to the power points on the expander (but not the channel outs! only data and ground there), and then the expander and the pixelblaze are connected via the ground, data and 5v pins. but only the output expander is powering on, not the pixelblaze. Any suggestions? Thanks
Here’s a diagram and a photo of the setup. (I realise now i could have been more efficient with my ground wiring, next time )
On your sketch, you show the data output of the expander connected to the data output (DO) of the LED strip. It should be connected to the data input (DI) of the LED strip.
In the last photo, the color coding of the wires going from the 3-way quick-terminals to the LED strip (off camera) appears to be inconsistent with the norm (black=GND, red=5V). Do you have the GND and 5V reversed?
Can you try disconnecting the LED strip and see if the situation improves?
Hi Irwin, thanks for your reply. The DO error on the sketch is not replicated in practice, i just copied it from the other thread before realising there was a separate data label for out and in.
On the colour order of the LED strip wires though - oh god, if I’ve simply plugged in my LED strip the wrong way round i’m going to be very annoyed thankyou for spotting this i will check tomorrow!
Should this be a separate issue to the pixelblaze not powering on - Is the ground / 5v connection between the Pixelblaze and the output expander meant to be unidirectional? I shouldn’t need to run a new 5v and ground wire from the supply to the pixelblaze separately to the expander, right?
If things are hooked up wrong, especially the LED strip, the results can be hard to predict. Disconnecting the LED strip completely is usually the first step to try when diagnosing problems like this. Anything that causes a voltage drop on your 5V may result in a “brownout” condition that can cause the Pixelblaze to reset or never start running.
The 5V connection is not directional, and your power wiring looks correct. It might be slightly better to eliminate the ground connections between the expander and the LED strips, to ensure that large currents don’t pass through the expander board (and possibly some strange ground loops with high LED current draw).
Based off the photo it looks right. You’d need to test what has power to see where your issue is. Do you have a multimeter? See if the PB terminals have 5v, and if the 3.3v is up on the expansion header.
You could also remove the PP and power by usb just to verify it works.
Thanks for the advice both. The pixelblaze still works fine with nothing connected and powered via USB. Adding the expander it’s also functioning. I must have spotted the error with the LED strip wiring shortly after taking the photo because it’s plugged in correctly now and the PB isn’t turning on. Same result if no LEDs connected too (with the whole board)- expander powers on, PB doesn’t.
I do have a multimeter but we’re reaching the limit of my electronics understanding here i’ve only used it for checking continuity - i’ll watch some videos! could you clarify what you mean by ‘if the 3.3v is up on the expansion header’ please?
To test for 5V, set to a voltage setting, then poke your meter probes in the top of the screw terminals (the screws are conductive). It should read about 5V (can vary a bit depending on the power supply). It should be somewhere between 4.5V and 5.5V, outside of that something is wrong with the supply or wiring. You could also do a continuity test between both terminals to make sure GND is connected to GND, and likewise for 5V, and data.
If you have a mismatch/disconnect then double check soldering on the screw terminals to the PCB, and ensure there isn’t any wire insulation going inside the terminal.
To test for 3.3V, touch the negative probe to the bottom hole on the expansion header, and the positive to the hole two up from that.
You can also do a continuity test between 5V and GND (negative) to test for short circuits. Mean well power supplies are nice and have protection circuitry that shuts off if there’s a short. I don’t think thats the problem here, since your expander is lighting up.
I spent friday evening doing a major troubleshoot and turns out i was being a bit generous when i said the output expander works fine when not connected to the rest of my circuit. I think the problem is the output expander, not the rest of my circuits unfortunately.
Powered by USB, the PB 100% works, and will control LED strips normally. With the output expander connected but no LED strip, the status and draw lights on the expander light up. As soon as I connect an LED strip, the status and draw lights turn off. Is this is expected behaviour? I can’t get the LED strip to light up at all, and I played around with the settings in browser in case, number of pixels, address of board etc, but no change.
The voltage is correct for the expander and the PB. No problems where you suggested to check continuity either.
Reluctantly I’m starting to think I’ve fried the expander somewhere with too much heat - generally I’m fairly good at soldering but I had an absolute nightmare getting some of these JST pigtails on, the solder just would not bond to some of the contacts, I had to remove and reattach several of them, couldn’t get solder to come off etc. Is the behaviour of the lights on the expander the giveaway here? Is there any way to confirm this?
This indicates something with the LEDs and/or LED wiring. Check that your positive and negative are going to the right places, and make sure your LED strips don’t have a short somewhere on their power. You can use continuity tester for that.
Usually those red, green, white wire JST connectors have 5V, data, and GND, but in really they might be wired any number of ways and don’t always have the same order in the connector. In your photo it looks reverse, but without following the wires to the strip itself I can’t be sure. Double check that your positive is actually positive all the way to the LED strip, and likewise for GND/negative.
If there aren’t any hard shorts that show up on the continuity tester, and your ± aren’t swapped, then I would measure the voltage at various points before and after connecting LEDs. See if anything is getting warm, and exactly where and when power/voltage cuts out.
Hello! Thanks so much for your patient explanations - i solved the problem in the end, I didn’t solder on the screw terminals to the expander properly
It’s amazing I’d got this far without learning how to use flux, really it’s been almost a decade since I started tinkering Anyway I’ve learnt something! And the PB is a fabulous bit of kit, I’m really excited to spend some time getting into coding patterns now the first pass of the installation is finished. Thanks for your work
Here’s a couple of pics! The design is inspired by light reflecting off the surface of water, and the piece is the focal point of a studio used for sensory rave spa events and yoga classes amongst other things
thankyou for spotting this i will check tomorrow!
i’ve only used it for checking continuity - i’ll watch some videos! could you clarify what you mean by ‘if the 3.3v is up on the expansion header’ please?
Is the behaviour of the lights on the expander the giveaway here? Is there any way to confirm this?
