Often happens that only the first pixel (the first 3 leds) work accordingly to what I am sending to it. The rest of the pixels appear to be full white. Sometimes if I wait long enough the strip start behaving correctly, but not always. Now, the problem is corrected if I add Tin to the connections between a pixel and the other; specially to the GND trace.
I anyway have the feeling that is not a conductivity problem… very weird…
you can see that I am working with segments, and it is always the first pixel that works correctly, and the rest are full white. When I was trying the sketch with a full 5mts strip I experienced the same.
This isn’t Star Wars where you can trust your feelings and know them to be true. You need to do the science and test your strip with a multimeter.
If there is more than an ohm or two of resistance between “ground” at one end and “ground” at the other, you have a bad connection somewhere. Same with the +5V line.
I understand… It is indeed the most reasonable thing…
I was thinking that could be something else because of its apparently random behaviour, and the fact that sometimes is fixed without me altering anything… That is why I first checked my cables, arduinos, sketches…
I measure resistance now, and normally I read 1.8 / 1.9 ohm.
Thanks JP! I think it is simp,y because I was working with only one hand. Have in mind that this is happening also with new strips, straight from the packaging.
In this video (http://youtu.be/HetsrfQmBxk) you can see a new led strip (5mts) where the first pixel works correctly and the rest are full bright. some seconds afterwards, without my intervention the white pixels start to blink until everything starts to work “normally” for just few moments. Then you can also see lost colour pixels here and there. Originally, the sketch is fading solid colours…
I dont have flux, but so far I have been normalizing the behaviour, as explained, with tin in the first join. My fear is that this led strip will be installed in an object that will be closed for ever… I am afraid this can happen after I close the object, so I might simply go over every joint… (uff)
If I were in your shoes, I would add a very small length of small, 28awg or so, stranded wire over that joint to help bridge that connection and solidify it. That is for GND, 12V and data !
Again, the f***ed up thing is that is not failing only between strip segments. It is happening also between pixels that are not segmented.
Also, it only happens after the first pixel. This doesnt mean that happens after a specific pixel, but after the pixel that happens to be the first in that particular moment!!!
well… I am not sure which one it is… sometimes “resoldering” the GND the problem is sorted out. sometimes not. I will pass over all of them… and anyway sometimes after doing this continues to fail…
Hung Vu: as I said before, the problem is not the 1st section, but any section to happens to be in first place…
Make sure GND of the strip’s power supply is connected to GND of the controller
Make sure the voltage across 5V and GND is 5 volts when measured at the far end of the strip
Make sure you’re using a controller with 5V logic level (not 3.3V) – if you’re using a 3.3V controller, you will need a level shifter to output the 5V that the LED strip is expecting.
