Trouble with ws2811's

Archive FastLED Archive
,
1

Trouble with ws2811’s

8d31010904eec62efecd97afccfa8706.gif
8d31010904eec62efecd97afccfa8706.gif1266×712 497 KB

74631243546d9a1308ec0750b71ac941.gif
74631243546d9a1308ec0750b71ac941.gif1266×712 534 KB

4863f220406573785c700121ccc41918.gif
4863f220406573785c700121ccc41918.gif1266×712 482 KB

da5d55ea5d736147f679453044b8f31f.gif
da5d55ea5d736147f679453044b8f31f.gif712×1266 406 KB

2

I’d love some help debugging an issue.

I’m wiring up some lights in the style of Paul’s OctoWS2811 (his pins + 100 Ohm resistors near the Teensy 3). If I use his code four of my strips do the rotating rainbow properly, but four show crazy flickering.

If I use Fast SPI then the same four do the rotating rainbox, two of them flicker like mad and two don’t seem to be getting any signal at all.

I going from a 12v battery → a 3A 5V regulator on each strip. Then from the Teensy 3 to the strips using 4 strand phone cable with twin pairs ground data) to pairs of WS2811 strips.

Am I getting some sort of inducting interference from the paired cabled? Any clues?

3

That looks like what I was getting when using cat5 cable, but accidentally had two data lines sharing a twisted pair. Your 4 strand phone cable may not have the twisting to cut back on noise - maybe try flipping to cat5 (or friends, mostly you want twisted pairs). Also - on the strip end, I found that I needed to ground to both the ground that the data line was twisted with and power to get things reliable.

4

Thanks Daniel, I upgraded to Cat5e and the flickering is diminished. but it’s still there. Next I’m going to try resistors (when I switched to the Cat5 I lost my resistors).

5

Also, What kind of wiring are you running from the battery to the 5v regulators?

6

You know, I’m starting to wonder if it’s a 3.3v vs 5v signal thing?

7

Yeah. That’s what I’m thinking as well. I have some 74hct245n for pickup
tomorrow.

3.3 worked fine when the strips are very close to the teensy, but I
probably have attenuation.

8

Hrm. (With a capital H.)
I wonder… a lot of things about this. Like: if this is an issue, how many other people’s power-and-signal-related-issues need a little boost in this department. Also starting to wonder if some of the other crazy we’ve seen elsewhere is related, and which Arduino boards are “5v” vs “3.3v” and if that’s correlated. So, as I said, Hrm.
Please let us all know what you find when you 74HCT245N things…

9

This looks very similar with what I’m seeing with my POV design with the controller running at 3.3V but the strings being powered with raw power from the battery (which fluctuates between 4.2V and down to 3.0V). Despite the LPD8806 claiming they can work down to 2.7V, I get crazy flickering starting around 3.2V … consistently. So I think I’m going to have to boost everything to 5V and hope for the best. I don’t know how bad the efficiency will be but I guess I’ll just have to put up with it.

10

Problem solved. It was a 3.3 volt issue. In the end my circuit looks something like this

12v car battery -> 12v to 5v 3A voltage converter (from china) -> power and gnd for the 2811 strips
12 v car battery -> 12v to 5v 600 ma voltage regulator (pololu) -> Teensy 3 -> 74HCT245E -> data (ground from teensy to join the gnd on the 2811’s)

I’ve run the board from the teensy on a separate battery pack which also works fine since the grounds of the 12v and the teensy are connected.

I didn’t need to use any resistors.

11

Great to hear! I wonder how long before the led chip makers realize that logic in the led/hobby community is shifting to 3.3v?

12

I met someone at a party who’s exboyfriend apparently makes lights that
only need a single line in and a single line out. They control the leds by
modulating the current. Sadly I didn’t get the name of the lights…