Problem is blowing the First pixel and subsequently losing control of the strip. I'm having

Archive FastLED Archive
21

You should look at the AdaFruit’s NeoPixel Uber Guide at https://learn.adafruit.com/adafruit-neopixel-uberguide. Everything is explained there. I lost a couple of pixels by too much voltage on the pixel, but I also lost one due to noise from a high current power supply bleeding over into the data line. As for the pixels lighting up blue, I have noticed that the PTH pixels I got from AdaFruit (SparkFun sells them now cheaper) power up blue before data is sent. Also if you must power up the LEDs while data is being sent, make ABSOLUTELY certain that ground is connected first, then +5V, then Data last. When you disconnect, follow that in reverse.
What I have found is that you can safely remove the data line while the LEDs are powered. The pixels will “freeze” in what ever state they are in. When data is connected, they resume at the point of where data is currently being sent.

Hope that helps! :slight_smile:

22

In addition to that big electrolytic or tantalum capacitor you’ll be adding for power filtering consider adding a small value ceramic capacitor too, say 0.01uF to 0.1uF. This small cap will be much more effective filtering out high frequency noise than the big cap.

23

I think I’m avoiding these issues by powering my controller from the same supply as I’m powering the LEDs, so it all turns off at once.

24

I prooved it several times @Kev_Zhu , (micro)controllers are all more or less sensitive to a higher current on an input pin than their own operating voltage. I was lucky many times too that nothing was damaged. That made me belief it is just an issue others ran into and I did not think about it anymore, which voltage I power up first. Later I was reminded by reality that it does matter.

25

Seems to be some conflicting info here… which to power up first.

Again all of mine are powered by the same battery coming from the car. I took a look at my strip and sure enough there are what look like diodes, but only at the connection points… ie every meter or or so. Weather permitting I will take a stab at it tonight.(7OCT14)

26

Where? Led power up first, controller power up second.

27

I see where I read a post wrong now… Ok so now I have a question for the group.

How do I delay start the controller if it’s power is coming from the LED strip? i.e. Everything turns on together.

28

Just to clarify the last post … the intent of my project is for the LEDs to startup with the car during a remote start.

29

I do that @Sam_Morton (the controller Arduino is powered in parallel with the strips) and haven’t had a problem.

30

@Robert_Atkins I have been this way from the start. Two blown so far.

31

There’s gotta be something else here, because I do the same thing (shared power) and haven’t had an issue

32

I really think it’s length of cable, or possibly no diode.

33

I don’t know then. My power is relatively “clean” (USB or beefy 100-200W 5V PSU), I have heard that there are all sorts of nasty spikes and things that happen when you start up a car. But if the LEDs and controller are on the far side of a DC-DC buck converter from the battery, you should be ok, depending on its spec: http://electronics.stackexchange.com/questions/19843/why-are-these-similarly-specd-dc-dc-converters-an-order-of-magnitude-different

34

This is my analysis: http://www.corbinstreehouse.com/blog/2014/03/led-cyr-wheel-ws2811-ws2812-strip-first-pixel-dyingburning-out/
(no scope testing or anything…just observation).

35

Sam, Are you using the car chassis for the power return(s) or are they wired directly back to the car battery? Have you tried reducing the brightness of the strips using the #define BRIGHTNESS to say 50, and does that make any difference?

36

@Herb_Smith I am using the car as ground to the buck converters. I added the resistor and diode tonight and I will add the caps tomorrow. So far so good with the diode in place. @Corbin_Dunn thank you for the link I read your analysis stem to stern. Btw thank you all great community here!

37

@Sam_Morton I was wondering how you are progressing towards solving the blown first pixel problem?

38

well… I tried with the caps, diode and resistor. this resulted in complete failure. I am not sure why but I suspected the resistor and removing it seemed to have worked. We have only been running a couple of days now in this config but knock wood so far so good. I have made an edict to not try and start the car while the animation is running. We are going to install relays this coming week to try and mitigate the light leakage that is occurring when the controller is off. I think after those relays are in place I will be doing a full battery of tests including car starting while running and cold start from remote with start up sequences.

39

aaaaaaand I spoke about it. It failed last night while the car was off but it was running an animation. This is super damn frustrating! I am not sure where to go from here. I guess I need to get the resistor in and make it work somehow. Maybe I read the resistance wrong…

Does anyone know why the resistor is needed? I assume LED1 is robbing the 5V from the data line for power along with the 5V and blowing it? I guess the resistor is to smooth it out?