max Length data cables for 144 pixel strip

EL Wire/Tape/Panels, LEDs, pixels and strips, LCDs and TFTs, etc products from Adafruit

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
yourix
 
Posts: 2
Joined: Thu Mar 27, 2014 6:37 pm

max Length data cables for 144 pixel strip

Post by yourix »

Hi

This is my third strip (the two other were burned , i removed several pixel on them but both of them finally "burn" , don't know why). I think there is much esd in our office and i suspect that.

I tried to make the data wire longer (with a 1mm electrical cable) , it never worked.
I used the condensator and the resistor as adviced , are there any reason why a long data cable could produce that ? When I use a small ground and white cable, things are ok.
Thank for any advice,

Regards

YP

User avatar
adafruit_support_mike
 
Posts: 67446
Joined: Thu Feb 11, 2010 2:51 pm

Re: max Length data cables for 144 pixel strip

Post by adafruit_support_mike »

To clarify the acronyms, ESD is 'electrostatic discharge'.. what happens when you scuff your feet on the carpet and then touch a doorknob. EFM is 'electromotive force', which refers to energy radiated by current flowing through nearby wiring.

ESD can kill a NeoPixel easily, but it wouldn't have much relation to the length of your wires.

EMF can induce current in a wire. A NeoPixel's input impedance is about 20 megohms, so even a small amount of current can generate large voltage spikes.

To demonstrate that, I put a couple of 10M resistors in a breadboard and connected my oscilloscope across the ends:
20M.png
20M.png (93.03 KiB) Viewed 243 times
The yellow trace at the bottom is a 60Hz square wave I used as a reference. The blue trace shows about 3/4v of induced voltage from the oscilloscope probe's 6" ground strap.

When I added 1' of wire between the tip of the oscilloscope probe and the resistors, this happened:
20M-and-wire.png
20M-and-wire.png (84.13 KiB) Viewed 243 times
The induced voltage jumped to about 5v.

The curve in the blue trace is typical of compact flourescent lamps (I put one near the bench as a EMF source).

The human body is a good antenna, so this is what happened when I touched the resistor lead connected to the probe:
20M-and-touch.png
20M-and-touch.png (94.66 KiB) Viewed 243 times
The voltage jumped to about 8-1/2v.


To eliminate those problems, I suggest putting a 10k resistor between the DIN and GND pads of the NeoPixel strip at the point where the wiring connects to the strip. That gives induced current in the wire a path to GND that doesn't flow through the NeoPixel.

yourix
 
Posts: 2
Joined: Thu Mar 27, 2014 6:37 pm

Re: max Length data cables for 144 pixel strip

Post by yourix »

Thank you for this information

I tried to mount the third one in a kiosk (with the resistor , the condensator) , a 5 volts 4 amp power supply and an arduino uno USB plugged on a PC .
This worked and after a fiew plug/unplugged and manipulations (open the kiosk/close it) , i got strange reactions
Sometime the strip never wake up (i have to unplug the power and plug it back), sometime the strip make random fix lights, now it is running with the animation in the arduino but the first pixel is off.
I suppose the next time i will unplug and plug it back it will probably not work the first time.

I am a bit disapointed by my neopixel strip experience, this was a POC and i am not confident in it (but I can't help thinking we do something wrong... i read everything about best practice , plugging ground before everything etc, ... ). This is the 3rd one I am ordering (about 200 $) and I think I will try to find another solution but any advice would be appreciated,

YP

User avatar
adafruit_support_mike
 
Posts: 67446
Joined: Thu Feb 11, 2010 2:51 pm

Re: max Length data cables for 144 pixel strip

Post by adafruit_support_mike »

If physical motion near the strip caused problems, there are two major options for the possible cause: physical disturbance of the wiring (which is most likely) or static discharge from external contact.

If your wiring is secure, and well removed from the areas where you get physical access to the kiosk, try connecting a 2" square of copper clad to the strip's GND rail and put that where it will be one of the first things accesible when you open things up. Make a point of touching that to discharge stray ESD any time you go in.

Locked
Please be positive and constructive with your questions and comments.

Return to “Glowy things (LCD, LED, TFT, EL) purchased at Adafruit”