LEDS with built in WS2811 - PICKY about power!

[Andrew]

I've got a model of 5v dc adapter that pumps out 3A, and have been buying and using them for a LONG time (3+ years).   They've been driving all manner of products for the whole time.   Funny thing is that they won't drive the ws2811.  I haven't scoped the adapter (I plan to, but things have been hectic, and I've got a workaround), but I suspect it will be either slightly under voltage, or will have messy / noisy output.   When it matters again, I'll clock it, chuck an inductor in there and try again, but for now, I've got 11 adapters that will drive 50x 60mA leds based around the ws2801 with NO problem, but with these, they don't work at all.

Fascinating!

[Patrick Cantwell]

Interesting indeed, as I've run the bare LEDs and strips off of a variety of ugly power sources with no issues at all. I even had the nerve to run the bare LEDs without decoupling caps and they were fine.. :)

[snikta...@gmail.com]

On Thursday, October 11, 2012 5:23:49 AM UTC-7, Patrick Cantwell wrote:
> Interesting indeed, as I've run the bare LEDs and strips off of a variety of ugly power sources with no issues at all. I even had the nerve to run the bare LEDs without decoupling caps and they were fine.. :)

Would the nominal 5V you would get from 4 x NiMH AAA batteries count as 5V for the purposes of running these guys? Would the nominal 6V from 4 x alkaline cells blow them up?

Also, do these controller chips drive the LEDs at a constant current or will my brightness fade with the batteries? I can't actually find a datasheet in English :-)

Cheers, Robert.

[Patrick Cantwell]

I'm reasonably certain these are constant current drivers so you shouldn't fade.

I'll try 4xAA (alkaline) cells when I get home after this weekend and see if anything blows up.

[Patrick Cantwell]

On Saturday, October 13, 2012 2:32:44 PM UTC-4, Patrick Cantwell wrote:

I'm reasonably certain these are constant current drivers so you shouldn't fade.

I'll try 4xAA (alkaline) cells when I get home after this weekend and see if anything blows up.

I got around to trying this tonight. I used 4 fresh alkaline cells and a 1M strip (60 LEDs). The batteries were reading 6.30V or so before trying anything. I haven't observed any ill effects, other than trying to go full white (0xFFFFFF) dropped the power enough that my 5V regulator for my micro didn't have enough power to provide clean 5V and my PIC hung :)

With 60 LEDs, the alkaline cells do quickly drop below 5V. It wasn't long at all before I was running the whole setup (LEDs and PICMicro) directly off the 4xAA cells. I can't easily observe any brightness change despite starting at 6.3xV and dropping to 4.80. I'd recommend 'C' or 'D' cells with any quantity of these LEDs on battery power! I'm running a continuous shift-down-the-strip type pattern and don't see any odd effects or strange things happening despite the dropping voltage. I'll leave it going with a meter attached until it finally dies, hangs, or something else goes wrong..

[Patrick Cantwell]

Also wired the whole setup to a LiPo pack I had laying around.

No discernible brightness difference. The WS2811 appears to be a constant current driver. (Which I was pretty sure of to begin with).

Setup currently running just fine down to 3.33v.

[EkriirkE]

"Bare LEDs" do not require decoupling. Those Caps are for maintaining a stable local power next to a device that fluctuates consumption.