๐Ÿ“ฃ๐Ÿ“ช ๐Ÿšš Adafruit will not be shipping USPS orders Monday October 14, 2019 ๐Ÿ—“. Expedited USPS orders placed after 11am ET ๐Ÿ•š Friday October 11 will go out Tuesday October 15 ๐Ÿ“ฃ๐Ÿ“ช๐Ÿšš
0

Bad neopixel ring?
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

Bad neopixel ring?

by markflip66 on Mon Jul 22, 2019 7:23 pm

I have a NeoPixel Ring - 24 x 5050 RGB LED with Integrated Drivers
PRODUCT ID: 1586
that I have worked with before many times.

I recently hooked up a new one to a adafruit gemma. I did all the safety circuitry including the resistor on the data pin and a cap across the power input. It worked for a few minutes and now it's acting like it's blown. I get all kinds of weird colors and a couple pixels dont light at all. It worked properly for a few minutes will all pixels working and displaying my code properly. Is is possible this is a bad one out of the box?

markflip66
 
Posts: 43
Joined: Mon May 18, 2015 12:42 am

Re: Bad neopixel ring?

by adafruit_support_mike on Mon Jul 22, 2019 10:06 pm

Post a photo showing your hardware and connections please. 800x600 images usually work best.

adafruit_support_mike
 
Posts: 59116
Joined: Thu Feb 11, 2010 2:51 pm

Re: Bad neopixel ring?

by markflip66 on Tue Jul 23, 2019 4:07 pm

I just turned it on this morning and the program ran as expected for a few loops then it went rouge again. I will play around with some simple code to see if I can figure this out. The funny thing is I am running this exact code on another project and it runs fine. The only difference is one is using the ring and the other is using the same number of pixels on some 30/m ribbon... I'll report back...

markflip66
 
Posts: 43
Joined: Mon May 18, 2015 12:42 am

Re: Bad neopixel ring?

by markflip66 on Wed Jul 24, 2019 3:26 am

I found the problem but I'm not sure why this is happening. The system works fine off USB power but when I run it off the converter I am using to drop my 12V supply down to 4.99V I get erratic behavior.
https://www.amazon.com/gp/product/B008B ... UTF8&psc=1
this is the link to the converter I am using.
I get rock solid 4.99V even when I change the input voltage from 12V up to 13.8V.
Any ideas?

markflip66
 
Posts: 43
Joined: Mon May 18, 2015 12:42 am

Re: Bad neopixel ring?

by adafruit_support_mike on Wed Jul 24, 2019 12:58 pm

Post a photo showing your connections when running the board from the external supply. 800x600 images usually work best.

adafruit_support_mike
 
Posts: 59116
Joined: Thu Feb 11, 2010 2:51 pm

Re: Bad neopixel ring?

by markflip66 on Wed Jul 24, 2019 1:41 pm

2019-07-24_1030.jpg
2019-07-24_1030.jpg (585.01 KiB) Viewed 330 times

here you go, the ring is on the backside of the metal plate so the +, -, and data wires go through (blue arrows). The red arrow is pointing to a 450 ohm resistor that is essentially the data wire. It's a very short distance. The wires coming in from the left are 12vdc and the cap side is set to 4.99vdc confirmed with precision multi meter.

markflip66
 
Posts: 43
Joined: Mon May 18, 2015 12:42 am

Re: Bad neopixel ring?

by adafruit_support_mike on Thu Jul 25, 2019 11:31 pm

Try taking an AC reading and a frequency reading.

Multimeters trade speed for precision and accuracy. DC voltage measurements usually take at least 200ms, which is far too slow to catch the kind of voltage spikes a switching converter can produce. An AC reading will give you a ballpark estimate of the severity of any switching residue, and a frequency reading will confirm the existence of some kind of noise.

adafruit_support_mike
 
Posts: 59116
Joined: Thu Feb 11, 2010 2:51 pm

Re: Bad neopixel ring?

by markflip66 on Fri Jul 26, 2019 9:38 pm

I turned the voltage down a bit and the problem went away. I have it set to 4.7VDC and all is good. Thanks for your help.

markflip66
 
Posts: 43
Joined: Mon May 18, 2015 12:42 am

Re: Bad neopixel ring?

by adafruit_support_mike on Sat Jul 27, 2019 12:41 am

Ah.. sounds like you hit a signal tolerance issue.

5V-CMOS devices are required to treat input signals at least 3.5V, or 70% of VCC, to be recognized as HIGH. To guarantee that, device makers set the real threshold somewhere below 3.5V. Most devices will accept 3.3V as HIGH, even though it's below the official threshold.

In this case, it sounds like the 3.3V signals didn't reach the chip's real HIGH threshold. Lowering the supply voltage lowered the threshold.. 3.3V is almost exactly 70% of 4.7V.

adafruit_support_mike
 
Posts: 59116
Joined: Thu Feb 11, 2010 2:51 pm

Please be positive and constructive with your questions and comments.