Data pin from addressable LED strip somehow frying Arduino digital output pins

Projects General Guidance
1

I made a sign that’s lit by a strip of WS2811 individually addressable LEDs which are controlled by an Arduino Uno via a digital output pin to the communication pin on the strip. The LEDs are powered by a separate 12V DC power supply which also goes out to a buck converter down to 9V to power the Arduino via the DC barrel power port.

It was working great for a couple of years then it started glitching. I tested the digital pin by writing a very simple program that simply turns the pin high then low every couple of seconds and also turns on and off the onboard LED at the same time. The pin I was using was not behaving right, going up to around 4.75V when it should be high, then going up to 5.05V when it should go low, then at the last second jumping down then back up to 4.75V all in a split second. So I tested the same for data pin 3 which was working fine. So this all implies that I somehow fried the digital pin 2.

So I just updated the code to use pin 3 instead, added a resistor going to the communication pin just in case, and made sure to insulate everything in the project box especially the 12V power bus just in case, and put it all back together and it seemed to work fine for a few hours then… The exact same glitchy behavior, it apparently still fried the digital pin 3.

So I’m pretty confused now. My only guess is that the LED strip itself has an intermittent short from the 12V to the data pin, and that 12V is killing these pins. If anyone else has any ideas on what it might be I’d be very grateful.

Thanks!

2
  • Show use good images of the wiring.
  • Show us a schematic with all components and interconnections.
  • Show us your sketch.
1 Like
3

Show us a close-up of the soldering on that part of the strip. It is a common cause to have small short between 12v & data, sometimes a bit of excess flux can cause this already. It will not only damage your ledstrip but also may damage the data output from wherever that is coming.

4

If you suspect so, put a diode in series with the data line of DPin.

1 Like
5

The voltage required for your WS2811 should be printed on the WS2811.

6

And a pulldown resistor or you won't have logic LOW

1 Like
7

Ok!

image
image180×107 1.34 KB

8

The datasheet says to add a resistor (size?) to VDD if the input voltage is below 24vdc.

9

This, from the first page, is misleading:

The chip has a built-in voltage regulator. For power
supplies below 24V, only a resistor needs to be connected
to the IC VDD pin, without an external voltage regulator.

Refer to the schematics on page 5. All power voltage options have resistances to VDD - reading between the lines, the VDD must remain around 5V, so the chip's operating current determines what resistance is appropriate for each supply voltage. I presume these resistances are present in the OP's LED string, as they appear to be necessary per IC.

All schemes shown also have a series resistance on Din, as is usual with these pixel-style LEDs.

10

And a resistor between R1 & the ledstrip. Of course the voltage drop of the diode should nopt be to great but something like a 1N4148 will work,

11

Why do you want to put that resistor in series wit the data line? Is there any good reason?

12

Refer to the datasheet for the devices.

13

The original idea is to prevent 'bounce' but a current limiting resistor will also protect the pin against an accidental short to GND. The diode only protects against an accidental short to Vcc.