But, having real issues driving them with Fastled.
It says they are WS2811, but all I am getting is random flickering. If I do get a constant colour, they don't update well at all - as in they just go mad.
I also noticed the first and last LED were often an incorrect colour.
At first I thought maybe a level shifting issue. Running off an ESP32, I though maybe the 3.3v data was not enough. But a level shifter actually made it worse.
This will illuminate the strip, but it then sits there and flickers horribly. The power supply is a very nice 4A 24v RS Din rail unit, so I am sure it's not the PSU.
Are you lighting up the entire 5m length? How many LEDs in total?
According to that listing, the product consumes 26W per metre, so 5m would consume a total of 130W, which at 24V would draw a current of 5.42A. Of course your sketch as it stands only lights one of the LED colours, but the PSU does not seem rated to cope with the total demand, although for that price, it should at least behave nicely on overload and just shut down or current limit. Then again, we don't know anything about how your circuit is wired or how much of the strip is being used for your test.
Obviously dodgy connectors are not going to help and I do not know enough about the problem with running FastLED on an ESP32 to comment.
I thought I had fixed it last night, and then this morning I am back with the flickering lights again.
Weirdly, if I touch the data cable (the sheath), the flickering stops. You would think that would be a ground issue, but the grounds seem fine.
Its a basic ESP32 breakout board, Use them all the time.
Tried a different one and also checked all the solder connections... all good.
Just ran it off the bench PSU and does the same flickering thing. I am stumped.
OK, so the level shifter has cleared the mad lighting effects. But, it still has issues.
The first and last LED (which in the case of this particular strip style is about 50mm long) flicker all the time.
Apparently that is a thing. Well, that kinda renders them useless for lighting effect then unless you can hide the start and end of the strip.
I added '#define FASTLED_RMT5_RECYCLE=0 ' at the very start of the code and that seems to have fixed the first led flickering, but not the last one.
These LEDs seem VERY easy to corrupt data wise.
Nice as they are, I think I will have to avoid them
IC's came this morning. That PCB of mine seems to work fine.
Still get weird issues with the first and last LEDs but from research, that seems to be a Fastled - ESP32 thing.
Only thing I can think of doing there is putting the first LED in the control enclosure where it doesn't matter, and programming the strip to show one more than the strip actually has.
Anyone know of addressable RGB or RGBW leds that can be bent/curved on the flat axis?
I theory, suppose that is almost impossible, but I need to achieve that somehow.
My only thought is to cut the strip up into individual LEDs and solder them all.... but over 5m, that is a mad amount of work.
Plus, I didn't really want the gaps between the LEDs.
