0

Troubles with possibly/probably defective NeoPixel RGBW 60 L
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

Troubles with possibly/probably defective NeoPixel RGBW 60 L

by SparklesBySaul on Mon Sep 24, 2018 3:18 am

So I'm a total N00B to Adafruit and Arduino, but I did a fairly extensive amount of troubleshooting with my parts before posting this. In a nutshell: the strip powers on and receives code, but acts a little buggy about the way it's processing said code. There's a big intro just here explaining how I arrived where I am at now. Feel free to skip to the part where it says "THE PROBLEM" to see the crux of it. Introduction:

My four supply pieces include two microcontrollers:
1 x Adafruit Trinket - Mini Microcontroller - 5V Logic[ID:1501] = $6.95
1 x Adafruit GEMMA v2 - Miniature wearable electronic platform[ID:1222] =
$9.95

And two LED strips:
1 x Adafruit NeoPixel Digital RGBW LED Strip - White PCB 60 LED/m
2m[ID:2845] = $53.90
1 x Unknown LED strip. Seems to be RGB and not RGBW. It was a gift from a friend along with a photon wifi microcontroller I've never used and since misplaced.

Aside from the very latter, the other parts were ordered in June and never used until this weekend.

So I hurdled getting Trinket & Gemma to work with my cheap Windows 10 laptop after much trial and error, frustrations, and reading. Seems to sync OK now with the latest Arduino IDE (and works alright without a USB 2.0 hub). ... Then I learned much about the frustrations of soldering. And with some luck, I managed to solder on three connections (power + 1 data pin) with the data wire magically already going the right direction as the arrows. I tinkered/had success with these two code sets the most:

1. http://aronatkins.github.io/2016/02/22/ ... _same.html
2. & the strand test: https://github.com/adafruit/Adafruit_Ne ... ndtest.ino

**THE PROBLEM**

In both instances ... the $53 Neopixel strip powers on, but instead of doing what it's supposed to do, it repeats the code per pixel for every three pixels, and then restarts. So I have a string of LEDS that are red, white and blue -- for example -- and then it'll repeat to infinity: red, white, blue, red, white, blue, red, white, blue... when the program clearly states it's the entire strip that's supposed to be either all red, all white, or all blue. (I'm specifically discussing program 1 I used here, but the strand test -- program 2 -- is showing the exact same glitch.) Almost everything else about the program is behaving normally. I can dim or brighten the LEDs or change other variables, such as the time delay.

I said almost everything because only 75 LEDs are functioning. Keep in mind: this is 2 meters a 60 LED/m strip, and I read somewhere on Adafruit.com that Gemma only supports 100 LEDs, so I cut the strip down from the 120 I was sent to 100. I inevitably cut off one more after soldering growing pains. I made sure to define the LED count in the program as 99 (and counted them by hand several times). But just for safe measure, I tried 100 and 98 as well, in case 0 counted as one of the LEDs. None of this helped. It only powers 75 LEDs and then quits programming the other 24.

The other 24 LEDs are still powered on the strip. They generally will either power off depending on how I reset the microcontroller or remain a very bright white, indicating they are not receiving programming. The whole strip I can dim to 5 or 10 for the first 75 pixels, but the 24 white ones will glow quite bright along the end of it.

I bought a Trinket and Gemma (specs above) when I originally embarked on this project - and I've since uploaded the code to both microcrontrollers. They both behave exactly the same with this LED strip, leading me to believe it is the strip that's the problem.

Meantime, the other unknown strip works flawlessly with the microcontrollers.

To note before anyone asks: no there was no capacitor etc. used with the Adafruit strip, or the other one for that matter. It's being powered directly right now from the microUSB port connected to my computer (I have a 2 AA battery pack and a 4 AA battery pack - I will attach one of them eventually when the project is finished.)

So as you can imagine - I'm frustrated. This whole project has been frustrating tbh. But I'm grateful at least the microcontrollers are working and I at least have a functioning strip for a project that's supposed to be finished in six days. I don't know if this is the right place to ask for a refund, but I'd like to either exchange or refund the Neopixel strip.

SparklesBySaul
 
Posts: 5
Joined: Mon Sep 24, 2018 2:45 am

Re: Troubles with possibly/probably defective NeoPixel RGBW

by adafruit_support_bill on Mon Sep 24, 2018 6:07 am

You need to declare the strip as RGBW - as done in the RGBWStrandTest example code:
https://github.com/adafruit/Adafruit_Ne ... ndtest.ino

adafruit_support_bill
 
Posts: 69809
Joined: Sat Feb 07, 2009 10:11 am

Re: Troubles with possibly/probably defective NeoPixel RGBW

by SparklesBySaul on Mon Sep 24, 2018 7:18 am

adafruit_support_bill wrote:You need to declare the strip as RGBW - as done in the RGBWStrandTest example code:
https://github.com/adafruit/Adafruit_Ne ... ndtest.ino


Still a no-go. I tried uploading to Gemma the full code in the link you provided, modifying the LED count and pin out.

When that failed, I tried changing the code in the the strand test in my previous posted link. The one you linked said "GRBW", so for good measure, I tried both yours and the one I used as GRBW and RGBW. Nothing worked. In the GRBW & RGBW settings - which Arduino IDE & Gemma accepts - the same 75 pixels are now stuck on at the brightest setting, in sort of an off-white, peach color, and then the remaining 23 pixels (I cut one more off, and updated the code accordingly) are still disconnected - being also stuck on but in a more white appearance.

If I change it back to "GRB" without the W, it resumes the program with the original errors I described. (Program repeating every third pixel - only functions through 75 pixels.)

SparklesBySaul
 
Posts: 5
Joined: Mon Sep 24, 2018 2:45 am

Re: Troubles with possibly/probably defective NeoPixel RGBW

by SparklesBySaul on Mon Sep 24, 2018 7:33 am

Not sure what it's worth - but both RGBW codes have a similar effect on the unknown strip. It's in the stuck on white position when uploaded. Of course, I don't know if the unknown strip is an RGBW one.

In any case - it still very much appears to me only the first 75 pixels accept programming in my Adafruit strip. (The ones on the unknown strip - even though they don't work in RGBW - are all the same color.)

SparklesBySaul
 
Posts: 5
Joined: Mon Sep 24, 2018 2:45 am

Re: Troubles with possibly/probably defective NeoPixel RGBW

by adafruit_support_bill on Mon Sep 24, 2018 7:50 am

Still a no-go.

Please be precise in your description of the problem. Changing from GRB to GRBW should have taken care of the 3-pixel repeating pattern issue.

Also please post some photos showing your soldering and connections.

adafruit_support_bill
 
Posts: 69809
Joined: Sat Feb 07, 2009 10:11 am

Re: Troubles with possibly/probably defective NeoPixel RGBW

by SparklesBySaul on Mon Sep 24, 2018 8:19 am

adafruit_support_bill wrote:
Still a no-go.

Please be precise in your description of the problem. Changing from GRB to GRBW should have taken care of the 3-pixel repeating pattern issue.

Also please post some photos showing your soldering and connections.


I appreciate your help, but "still a no-go" was obviously only one sentence (fragment) as part of a detailed description I provided you. Changing from GRB to GRBW, as I said, did NOT fix the strip / code / issue I am having. When either strand test uses GRBW or RGBW -- I repeat from the last time -- 75 pixels are stuck on, in an off-white, sort of peach color, and 23 more at the end of the strip are also stuck on, in a crisp white. If I remove the W, then the white/off-white pixel problem goes away, but the 3-pixel issue resumes.

None of this has yet addressed why both Trinket and Gemma can operate 100 pixels (and perhaps more) on an unknown strip, and not more than 75 on the $53 one I ordered from Adafruit.

There is only one soldered connection currently on the Adafruit Neopixel strip right now - the data pin. I connected the power via the other end of the strip with the wires that were already soldered on it when it shipped. For the Gemma and Trinket, I've been simply twisting the wires in their connector holes, similarly to how I'd use alligator clips - had I been wise enough to get those. I've been doing this so I can test each microcontroller on each strip. The solder you see on the Gemma is left over from my first attempt with it. Here are some photos with notes.

20180924_050033.jpg
20180924_050033.jpg (64.02 KiB) Viewed 36 times


20180924_050203.jpg
20180924_050203.jpg (92.16 KiB) Viewed 36 times


20180924_050315.jpg
20180924_050315.jpg (54.47 KiB) Viewed 36 times
Last edited by SparklesBySaul on Mon Sep 24, 2018 8:26 am, edited 2 times in total.

SparklesBySaul
 
Posts: 5
Joined: Mon Sep 24, 2018 2:45 am

Re: Troubles with possibly/probably defective NeoPixel RGBW

by SparklesBySaul on Mon Sep 24, 2018 8:21 am

This is the final photo. Your system only let me upload three.

20180924_050429.jpg
20180924_050429.jpg (60.1 KiB) Viewed 36 times

SparklesBySaul
 
Posts: 5
Joined: Mon Sep 24, 2018 2:45 am

Re: Troubles with possibly/probably defective NeoPixel RGBW

by adafruit_support_bill on Mon Sep 24, 2018 8:42 am

Changing from GRB to GRBW, as I said, did NOT fix the strip / code / issue I am having.

Since you reported multiple problems, there may be multiple steps to the solution. I believe that the change DID correct the problem with the repeating 3-pixel pattern.
Good debugging practice is the same as good scientific practice. You change only one thing at a time and observe and report the results of the change.

Since the ground wire is also the signal reference, your ground connection to the strip should be at the input end - next to the signal connection.

adafruit_support_bill
 
Posts: 69809
Joined: Sat Feb 07, 2009 10:11 am

Please be positive and constructive with your questions and comments.