First led is broken - again and again?! When I powered up my stripe

First led is broken - again and again?!

When I powered up my stripe for the first time, I damaged the first led. So, I put a 1000µF capacitor between the power supply and the first led. It worked for a while but by now, the first led is broken again. That’s very annoying since I lose more and more leds.
What are you doing to protect your leds from damage?

I don’t recall that happening to me, but then my strips are short. Do you have extra long strips, a huge amount of available current or a significant distance between the Arduino and the strip? Sure would be nice to see a ‘best practices’ bullet list for power.

@Jeremy_Hull what specification does your diode have?

it happened to me on a stripe with 240 leds and on a stripe of only 12 leds.

why does the distance between the arduino and the leds play a role in terms of damaging the first led?

You can try a 120 ohm serial resistor in the data line to limit the current in case your power line disconnect earlier as the data line.

Are you running those LEDs at 100% brightness? I haven’t have this happen to my stuff, one of the first I do in my code is set the brightness max to 16 or 32. I don’t do it for the LEDs though, I do it for my poor retinas. The damn things are bright.

If you’re connecting and disconnecting the wires while it’s on be careful to plug in ground, power, data. Disconnect in the reverse order.

It’s really bad for the leds to have data before power.

That and lack of a cap has caused issues with as few as 60 leds, in my experiences.

I use a capacitor on long runs, but its job is to act as a local reservoir for the strip. So if i run an effect that requires “twinkles” (100% white for a fraction of a second) the capacitor can supply the required current, thus preventing localised brownouts.

I have started using a 100 ohm resistor on the data line too. But that there to protect the MCU. This is since i discovered that with only the data and +ve connected, the string would light (albeit randomly), and i released that internally the silicon was sourcing its ground from the incoming negative data.

You could strap a zenner across the supply line of the strip to act as a clamp.

Assuming we’re talking about neopixel/WS2812B:

See http://forums.adafruit.com/viewtopic.php?f=47&t=51153&sid=2acf0fa29d3836898e2b85424571f89a#p258256 and http://forums.adafruit.com/viewtopic.php?f=47&t=51285#p259936

I always put the resistors as shown in the 2nd post (as close to the strip as is practical) as well as a 1000µF cap across 5V and GND.

Before I started doing this, I burned out several neopixel rings, but never since.

The key point is to prevent the data pin to the LEDs being at a higher voltage than the supply voltage of the LEDs. This is especially important if the data source (e.g. your Arduino) and the LEDs are powered from different sources.

If the data pin voltage is higher than the LED supply voltage, then the data pin will attempt to power the LED chip and pop it goes.