Stemma RGB "NeoPixel" breakout, when set to 0 aka off by action, instead accidentally set it to red

[tyeth]

Just came back to this rgb strip light that turns off at midnight and it's red instead of off (was white beforehand). The last value in the feed is 0, and it worked in prior runs. Not sure what's gone on and the device doesn't have serial attached. I've manually added 0 again to the feed using add data and it's turned off as it should.
I'll keep an eye out on this one as it's the light above my desk.

[tyeth]

I did not see this with just a normal tri-colour LED plugged into a second metro-s2 and StemmaRGB, possibly power source related for the 12v RGB strip

[tyeth]

I've seen some weirdness, where the chosen colour is not presented or may sometimes be presented, in two different sets of 12V rgb strips + neorgb's + power supplies + metro boards (s2 and m4-airlift). Both the power supplies were 12v (4.17A + 6A), and brightness being set at 25 instead of 255 still saw issues(maybe less).

When playing rapidly clicking the same value in the colour picker / brightness slider. Picking 0 from full white brightness gives red often, blue sometimes, usually then off after one of the strong red/blue channels failing to reset/update.

Found this which indicates there may be a marginal voltage issue for the 3v3 data signal:
https://forums.adafruit.com/viewtopic.php?p=1007627#p1007627

[tyeth]

Reconfirmed, same behaviour on all pins, 12V 6A supply, except when passed through the onboard neopixel of metro (StemmaRGB second pixel)

[tyeth]

This is a known hardware issue, where the logic signal at 3.3v is nominal, but fine if driven above 3.3v, and has been fixed in revision (B).

My first copy of rev B was batch W29440-B received 26th Feb 2025 a couple of weeks after the revision was released in the shop.

Details: https://youtu.be/VzTa1lfaCLI?si=OO9herlSkISjYwR3&t=10