Is there a way to test WS2812B strip per LED?

Hi,
I am fairly new to electronics so sorry if I make little sense. I will try to be descriptive but to the point.

Is it possible somehow to test WS2812B LEDs (60 leds per meter strip) one(or few) at a time without cutting it in pieces?

If I connect gnd,data in. 5v+ to a working strip 1st led, then it lights the number of leds I programmed.
But if I connect to a 2nd or nth led nothing lights up

Because the strip I need to debug is glued to a wooden frame and I cannot just cut 128 leds and check them all separately.

I tried searching on internet whether its possible to connect them somehow to power and lit them up, but it seems like I cannot test them without programmed controller and a data pin.
Thank you in advance for the input

Show your test code you are trying.

Use CTRL T to format your code.
Attach your sketch between code tags
[code]Paste your sketch here[/code]

It is most basic setup with Adafruit NeoPixel strandtest example, don't know if it makes any sense on copy pasting all of it here both because how long it is and how very well spread it is across the forum.

the only two things I changed in that strandtest example is data pin and pixel count.

Same code worked for me in different setups. With(out) 200-470uF 16V caps, with(out) 330ohm resistor between data pin on arduino and datain on WS8212B. With only 4 led strip and with 128 led strip.

My question here is more about how to phisically verify if every led is working without cutting the led strip.
I am suspecting that something is with data line, so how would I test/verify that.

Thank you

If you want to test all the LEDs, first turn on all Red then Green Then Blue.

Let’s say you have a 60 LED strip, number of pixels will be 60.
That’s 60 X 3 rgb= 180 separate LEDs.
Turn on 60 red then green etc.

So if I understand correctly, if something is wrong with one led somewhere in the chain, other leds that follow would still light up?
And if all of them are not lighting up, then most likely all of them are broken?
Cause I was afraid that first broken led would also break the data line and the rest of the leds would not light up not because they were broken, but because data was not getting to them

If the dataline somewhere breaks, the leds after that will not work; same for a damaged led where the data output is corrupt.

And you can not test half a strip by applying the data somewhere in the middle; you will feed the data into the output of the last led before the insertion point which is an unhealthy idea in general.

I use many of these strips.
Only had one situation where a pixel failed, all pixels after the bad one go out, you can replace a pixel if you have a good soldering work station.

However, if just the LED in the pixel fails, the remaining pixels will probably work.

Thank you guys for the input
this was the answer I was expecting based on all the searching I did, but at least now I know there are no shortcuts. Was hoping that data could go only one way and not the other.

So I started cutting leds from the first one and this first one was the faulty one.

Also when I was testing LEDs I noticed one thing.
When I was using lab power supply giving 5V output things were working smooth.
Once I placed LEDs in the table, and switched to using Mean Well APV-35-5 that is supposed to give 5V output. Interesting cause then controller+LEDs started behaving real weird. LEDs were randomly firing up and glitched in different ways.

Then I tried testing more with lab psu. Once I raised a bit over 5V it started behaving exactly the same way.

So I guess this is how this first led failed, because I had this glitching before it stopped working, but was blaming loose cable setup.

Now I just need to do some googling why is this power supply behaving like this and how can I limit/stabilize its output to 5V.

OK, now you do have the specified 1 mF electrolytic capacitor connected directly across the power connections at the start of the LED strip, don't you? And the 470 Ohm resistor in series with the data pin at the same place?

Now the Mean Well Power supplies are supposed to be good, I have no personal experance of them. However they do appear to have a part in a lot of problems we find on this forum. Given that I would steer clear of them.

Paul__B:
OK, now you do have the specified 1 mF electrolytic capacitor connected directly across the power connections at the start of the LED strip, don't you? And the 470 Ohm resistor in series with the data pin at the same place?

I do have 470uF cap between gnd and +5v lines. and a 330 Ohm resistor between led data in and arduino data pin.

Grumpy_Mike:
Now the Mean Well Power supplies are supposed to be good, I have no personal experance of them. However they do appear to have a part in a lot of problems we find on this forum. Given that I would steer clear of them.

It wasn't much of an investment, paid like 15eu for it, still I thought I could perhaps try some things before I will give up on it.

  1. currently I am feeding arduino through VIN with the same 5V that I am feeding leds. Perhaps arduino is the one being sensitive to ~5.2V and not leds.
  2. also I wonder if there is any way to drop voltage from 5.2V to 4.9

I haven't got much of success googling of what are my options, perhaps you guys could shear some light on this one?

Thanks

currently I am feeding arduino through VIN with the same 5V that I am feeding leds.

That is wrong, it will result in the Arduino chip being powered by about 4 Volts and not a stable 4V at that. If you want to do that then you should connect the Arduino through the 5V pin not the Vin pin.

Perhaps arduino is the one being sensitive to ~5.2V and not leds

If that is true then do not connect this supply to the 5V input pin on the Arduino. It looks like your supply is a killer.

also I wonder if there is any way to drop voltage from 5.2V to 4.9

Not easy but perhaps pass it through a Schottky diode first.

Grumpy_Mike:
That is wrong, it will result in the Arduino chip being powered by about 4 Volts and not a stable 4V at that. If you want to do that then you should connect the Arduino through the 5V pin not the Vin pin.
If that is true then do not connect this supply to the 5V input pin on the Arduino. It looks like your supply is a killer.
Not easy but perhaps pass it through a Schottky diode first.

oh this reply just explained me a bunch of things. Thank you!

I would think at 5.2v, both Arduino and LEDs should be fine. I run 2811's at 5.1v, and many USB chargers (which people use to power Arduino) also output 5.1 or 5.2v to compensate for cable resistance while charging. Arduino AVR processors are rated up to 5.5, and absolute max for WS2812 is given as 6~7v, so a little over 5 should be fine...

But you definitely shouldn't use the Vin pin - using that instead of 5v pin is probably the source of your problems, as your output signal might not have high enough voltage to be correctly received by the LED.... (assuming LEDs are powered from the power supply directly, not 5v pin - though that would have other problems)

so tried rewiring my esp8266 module. that now looks like this:

moving voltage from VIN to VU(the only 5V pin on a controller) was a big improvement. Controller wouldn't hang anymore. It was performing all of the strandtest patterns. However it is still glitching(once in a while displaying wrong colors/their brightness)

So next step to try is get 3.3V to a controller while feeding leds with that 5V output.

I was also wondering, in item description of Mean Well APV-35-5 it said this is a power supply for led lighting. I was wondering perhaps it is intentionally providing some sort of different voltage. I don't know, maybe some PWM controlled voltage(sorry, just shooting arrows here)

regardless of what that power supply is/does, decided to get myself an oscilloscope and a new psu since I see myself sticking with this hobby for a longer time and that will let me understand better what that thing is doing

in item description of Mean Well APV-35-5 it said this is a power supply for led lighting.

Normally simple lighting is not so critical about voltage. But you are dealing with addressable chips here which is different. I suspect the people writing the copy for the specifications have no idea about the subtleties of electronics.

Move the big capacitor to across the LED strip and add a 0.1uF ceramic capacitor across the strips power terminals as well.

I made just a quick schematic to show wiring, I have no experience in drawing/using proper symbols, so sorry if something there is misleading.

At first I ignored "for LED lighting" part cause it sounded like one of those "MS windows compatible usb cables" sales pitches. But now that I'm learning that 5V != 5V I started mentioning everything :slight_smile:

that 470uF cap I soldered physically at the begining of the led strip accross gnd and 5v.

Are 0.1uF ceramic caps for decoupling purpose and has to be next to each led?
If so, my led strip comes with surface mounted caps next to each led which I believe are there for exact same purpose.