WS2812B killed my Nano

Hello all,

This is my first post. I don't really expect a solution, but maybe I'm not the only one that has had this problem. I created a sketch heavily based on sample code to play around with my LED strip. When I was not working on it, I would just switch on my power supply and let it run for a while. Then suddenly I noticed it was frozen. Long story short, I couldn't even load a script on my Nano anymore to investigate. So I took another one (I have a few :)). It started as before, but I noticed the LEDs would not go completely. And then this one died too. I noticed that even without any LED on, it would draw .8A and when I tried to find out where all that power went, I burned my fingers on the processor chip. Well, that explains I couldn't upload a sketch.

I searched the forum I found that you should put a resistor in the data line. I did that with a new Nano and a new sketch. One that would only set each of the LEDs from 255 to 0 and the do the same for all LEDs at once. Each of the four worked, but the LEDs would not really go out. And I'm 100% sure they did before.

Suffice to say I don't dare to use this strip before I know what's wrong. Anyone has had similar problems

I've attached a photo of my setup. Two remarks on that. First I had 5V connected to the 5V-pin instead of Vin. According to the documentation, this should not be a problem. Second. I don't use a capacitor in the power line. But since I use a professional lab power supply, this shouldn't be needed.

I hope to get some replies before I return my strip to the supplier.

And then this one died too. I noticed that even without any LED on, it would draw .8A and when I tried to find out where all that power went, I burned my fingers on the processor chip.

That suggests you have wired thee power up the wrong way round or powering it with too high a voltage,

Also you should be powering the LEDs from 5V not the Vin.

You need to set your power supply to 5V and feed that into the LED strip.

I can’t see from your photo because it is low resolution but have you made the solder link next to the USB socket.

As a guess I would say your power supply is giving out too high a voltage and that is feeding back into the data pin and frying your Arduino.

But since I use a professional lab power supply, this shouldn’t be needed.

Wrong you always need a capacitor close to the strip.
Why is your photograph different from your words?

The picture is not clear where/how power is sourced. Could you draw a schematic and post, so how your circuit would be wired is shown?

Vin should only be 5V.

I agree with Grumpy_MIke and it must be something with the power. But the voltage was 100% sure under 5v (4.6v to be exact) and I do not power the LEDs from the Vin. They are powered from the crocodile clamps. The Nano is powered from the LED strip, not vice versa. And I used the 5V pin to do that first. According to the documentation, this should work, but the Vin has a power regulator behind it, so that is the better option.

I did not put any soldering fluid next to the USB port, but maybe it looks like that because it's a cheap clone :slight_smile: .

I will add a capacitor. I really thought that wouldn't be necessary when using a decent power supply.

But I learned a lot already. One of the things is to use Fritzing (strange, the resolution problem, for me the photo looks good. But, I agree, the Fritzing image looks much better).

I couldn't find a good looking power supply image yet, but power comes from the bottom "connector". The USB port is only used for programming. And yes, I will add the capacitor in my Fritzing design tomorow.

Thanks for the help so far.

They are powered from the crocodile clamps.

Bad idea. These can easily slip off and interrupt the supply while it is powered up. Not a good thing as that sort of thing can kill the strip or Arduino.
Never wire anything up while the power is applied. otherwise things break.

I'm not sure I understand. I understand that shaky power supply can cause inductive peaks, but that can always happen. I noticed it even happens in breadboard jumper cables. But, I've changed it anyway. Will have to wait with a proper test though. The LED strip seems dead :frowning: . It still gives a nice ambient light though. I try and cut it to see if maybe some part survived. But I think I'll have to wait until I get a new one. Thanks so far. At least I've learned a few things out of this.

According to the documentation, this should work, but the Vin has a power regulator behind it, so that is the better option.

The input voltage for Vin is specified as 7V - 12V; so 5V is too low and not the better option :wink:

sterretje:
The input voltage for Vin is specified as 7V - 12V; so 5V is too low and not the better option :wink:

But you should absolutely not be using "Vin" at all for any purpose. :roll_eyes:

All the logic here runs on 5 V. A properly regulated 5 V supply is what you need, connected to the LED strip and also to the "5V" pin on the Nano.