Hi guys, i need help yet again. I have the same “LED flicker” problem as last time but seems like for a different reason this time.
i have 180 LEDs powered by a 2 amp 5v power thingy at a brightness of 80 so i’m pretty sure they have more than enough power. The strip that is flickering is an APA104 strip. The strip that its connected to (and works just fine) is a WS2812B strip. The APA104 strip flickers compared to the WS2812B strip.I’ve had the APA104 connected to the WS2812B strip for several weeks and it worked just fine (so its not a compatibility problem). Now i cut up the APA104 and soldered it back together (as in the video). Based on my previous problem i’m guessing it could be a grounding problem or something.
Last time i had the flickering issue i was powering my controller (arduino nano clone) via a 1 amp usb battery and the strip via the same 2amp 5v power supply. Last time the problem was fixed by powering both the strips and the arduino nano by the same power source (the 5v 2 amp power supply).
Is it possible that my soldering is not good enough? perhaps the wire im using is to thin in some places?
I can provide closeup pictures if necessary. I’m just clueless as to why the flickering is happening again, and more importantly on how to fix it.
Thanks for the help in advance and sorry for the lengthy post but i wanted to provide as much detail as possible incase you guys needed it.
It looks like the APA104’s timings (looking at a datasheet), while close to the WS2812 are a bit different, in fact, the WS2812 timings that I use are close to one end of the APA104’s tolerances which, among other things, may mean that things are going to be a little bit more sensitive to line noise/timings/voltage. (I’ve seen voltage too high/voltage too low cause timing problems for the WS2812, I wouldn’t be shocked if the APA104’s had a similar problem - also, WS2812’s regenerate their data signal/timing - so the APA104 will be getting the timings that the WS2812 chips put out, which I can’t control in the library).
I just checked code in on the FastLED3.1 branch that separates out WS2811 and WS2812 timings - it looks like the APA104 is closer to the WS2811 timings - try changing your code to that, and see if it helps at all (though, I have a strong suspicion that it may not, but I don’t have any APA104’s here to test with).
Try adding a 150 to 470 ohm resistor at either end of the cable between the two strips. Could be the impedance mismatch is causing noise on the line.
@Daniel_Garcia i tried WS2811, WS2812 & WS2812B on on line #define LED_TYPE . Didn’t fix anything (as you suspected). I really doubt it’s related to code because it was working with the old code that was WS2812 (actually if i remember correctly they all worked) I’m using fastLED3 btw specifically the ColorPallete example (which ive been using for wees now). (i don’t think i updated to 3.1). You are one of the people developing the library correct? Do you need an APA104 strip? If you actually need one i’d be happy to mail you one for free. (Please don’t request one if you don’t actually need it thou. I don’t have alot of $$$ to blow on LEDs which is why im using $3 clones in the first place and i get all my parts from china) Thanks for the help. I’ll try the resister fix now.
The 3.1 branch is the only one that has separate definitions for ws2812 and ws2811, and yes, I’m one of the devs behind the library. I have some apa104s on order, not sure when they will show up though!
Do you suggest i update to 3.1? @Ben_Delarre Where would u put the resistor? the +,- or data between the two strips (probably not data i assume)?
Oh sorry. You want it on either end of the data line. Basically whenever you change the impedance of the data line you might want to try adding a resistor at either end of the added wire as this helps absorb reflections and other noise generating problems. I was getting a lot of flicker on some long data cable runs until I started doing this.
So i tried a 470 ohm and a 220 ohm (thats all i have). I put the resister on the data line between the APA104 and WS2812B. When i put the resistor there the APA104 strip would stop working completely. 
Another thing i noticed is that the first LED on the flickering APA104 strip (when it flickers) does not light up at all. The LED is not broken however as it does light up when the strip defaults to blue. The APA104 strip defaults to blue when only the + and - is connected and the data is not connected. Is it possible that a really thin wire (like hair or two thin) or a poor soldering could be the cause?
