Author Topic: Pixels take more power than I thought  (Read 4799 times)

Offline Jeffl

  • Supporting Member
  • ******
  • Join Date: Dec 2013
  • Location: Brandon, SD
  • Posts: 519
  • Kudos: 4
  • Just remember. Wireless is nice. Copper is king!
    • Christmas In Brandon
Pixels take more power than I thought
« on: March 07, 2015, 06:20:47 PM »
A 50 node string of 12v ws2811 takes 2/3 the power off (plugged into a running hub with uSC) as a 50 node string of regular LED's full on. 

And I thought I was going to get away with less power cords in the future.

Offline CaptainMurdoch

  • Administrator
  • *****
  • Join Date: Sep 2013
  • Location: Washington
  • Posts: 9,802
  • Kudos: 209
Re: Pixels take more power than I thought
« Reply #1 on: March 07, 2015, 08:00:53 PM »
Are you comparing against single colors on your ws2811 or full-on white?  Remember the ws2811 pixel is actually 3 LEDs in one piece of plastic and on some newer nodes each of those colors can pull close to 20ma for a total of almost 60ma when full-on white.  They can suck the power though.  :)  it is a good thing you are sequencing instead of leaving them on all the time.
-
Chris

Offline Jeffl

  • Supporting Member
  • ******
  • Join Date: Dec 2013
  • Location: Brandon, SD
  • Posts: 519
  • Kudos: 4
  • Just remember. Wireless is nice. Copper is king!
    • Christmas In Brandon
Re: Pixels take more power than I thought
« Reply #2 on: March 07, 2015, 08:55:12 PM »
That's the crazy part.  I was testing the pixels in an "off" state.  Just hooked to the hub.

Offline tbone321

  • Hero Member
  • *****
  • Join Date: Oct 2014
  • Location:
  • Posts: 1,583
  • Kudos: 50
Re: Pixels take more power than I thought
« Reply #3 on: March 07, 2015, 09:30:06 PM »
You may have a few "dirty" nodes in the string that are leaking excess current but lets not forget that every node in the string will be drawing power whether they are turned on or not.  Each node has its own controller, regulator, and minor support circuitry that all draws power and if you are including the uSC, that also draws power regardless of what is happening on the string connected to it. 

Offline Gary

  • Supporting Member
  • ******
  • Join Date: Jan 2015
  • Location: Chilliwack, BC Canada
  • Posts: 380
  • Kudos: 3
    • Diamond Crescent Musical Christmas Lights
Re: Pixels take more power than I thought
« Reply #4 on: March 07, 2015, 10:24:58 PM »
Hmmm... perhaps the newer versions of Falcon hardware can have a power saving mode to turn off power to ports/plugs during set daytime hours. Energy Star Certification can only be a good thing.  ;)

Offline tbone321

  • Hero Member
  • *****
  • Join Date: Oct 2014
  • Location:
  • Posts: 1,583
  • Kudos: 50
Re: Pixels take more power than I thought
« Reply #5 on: March 07, 2015, 10:58:38 PM »
I think that if one believes the static power usage is excessive, it would be a better idea to just put the hubs and strings on timed circuits and leave the primary controllers powered up all of the time.  Adding this control to the Falcon hardware just adds to the expense and complexity of the hardware and creates another possible point of failure.

Offline Gary

  • Supporting Member
  • ******
  • Join Date: Jan 2015
  • Location: Chilliwack, BC Canada
  • Posts: 380
  • Kudos: 3
    • Diamond Crescent Musical Christmas Lights
Re: Pixels take more power than I thought
« Reply #6 on: March 07, 2015, 11:38:44 PM »
Adding this control to the Falcon hardware just adds to the expense and complexity of the hardware and creates another possible point of failure.

I was thinking it would be in software/firmware.

Offline Steve Gase

  • Supporting Member
  • ******
  • Join Date: Mar 2013
  • Location: Georgetown, TX (near Austin)
  • Posts: 1,037
  • Kudos: 5
    • WinterLightShow in Georgetown, TX
Re: Pixels take more power than I thought
« Reply #7 on: March 08, 2015, 08:53:35 AM »
Adding this control to the Falcon hardware just adds to the expense and complexity of the hardware and creates another possible point of failure.

I was thinking it would be in software/firmware.
In rgb controllers the software/firmware only controls the data signal line...  power is passed directly from input to output with only fuses in between.  Consider also power injection which never goes through the controller.

a timed circuit from your power is what you need.  Remember to keep the fpp separate on a non-timed circuit.

Maybe there is an opportunity for a set of relays controlled by fpp ? ? ?
http://WinterLightShow.com  |  110K channels, 50K lights  |  Nutcracker, Falcon, DLA, HolidayCoro

Offline Gary

  • Supporting Member
  • ******
  • Join Date: Jan 2015
  • Location: Chilliwack, BC Canada
  • Posts: 380
  • Kudos: 3
    • Diamond Crescent Musical Christmas Lights
Re: Pixels take more power than I thought
« Reply #8 on: March 08, 2015, 09:39:34 AM »
In rgb controllers the software/firmware only controls the data signal line...  power is passed directly from input to output with only fuses in between.

Yeah, makes sense.

Offline Jeffl

  • Supporting Member
  • ******
  • Join Date: Dec 2013
  • Location: Brandon, SD
  • Posts: 519
  • Kudos: 4
  • Just remember. Wireless is nice. Copper is king!
    • Christmas In Brandon
Re: Pixels take more power than I thought
« Reply #9 on: March 08, 2015, 10:42:23 AM »
It actually might make sense to run your DC controllers from an AC controller that is running in a background sequence from the player.  Not sure that I will worry about it but it's something to think about.

Offline Steve Gase

  • Supporting Member
  • ******
  • Join Date: Mar 2013
  • Location: Georgetown, TX (near Austin)
  • Posts: 1,037
  • Kudos: 5
    • WinterLightShow in Georgetown, TX
Re: Pixels take more power than I thought
« Reply #10 on: March 08, 2015, 11:02:41 AM »
I don't know how well power supplies fare if you mistakenly send effects (ramps, strobing, etc.) from the A/C controller.
Also, most of my RGB gets its power from 750W computer supplies...  assuming they are somewhat loaded, that is nearly 7A -- more than a single A/C channel in nearly all controllers.

...that is why I suggested a relay.

i'm sure our board designers could think of a more elegant solution, but my jury-rigged approach built from spare parts would be:

FPP-> USB/RS485 dongle running DMX -> 3-channel dumb RGB controller -> (3) 110v relays

If I knew the GPIO pins better, I'd suggest that GPIO direct to relays would work.

Offline Bshaver

  • Developer
  • ******
  • Join Date: Aug 2014
  • Location: Denver, CO
  • Posts: 1,426
  • Kudos: 22
Re: Pixels take more power than I thought
« Reply #11 on: March 08, 2015, 11:20:46 AM »
Yes. GPIO using a SAINDEVICE 8 channel relay. Interpret the channels directly in FPP and output to the GPIO.. However; you'd need to keep the channel at 255 .. But that is easily done with a script and the /fppmm block channel mapper ! :)
Denver Colorado
45,000 blinky blinks mix of pixels and 130 AC Channels (powered by Minleon & Lynx) ALL Powered by FPP and Falcon!

Offline tbone321

  • Hero Member
  • *****
  • Join Date: Oct 2014
  • Location:
  • Posts: 1,583
  • Kudos: 50
Re: Pixels take more power than I thought
« Reply #12 on: March 08, 2015, 12:32:13 PM »
The problem with using an AC controller is that even at full on, the wave is not clean and can hammer the hell out of any inductors connected to them such as the transformers in the DC power supplies.  You can buy 120V and 240V timers that can handle well over 40A if needed and since most will only be switching on and off once per day, it makes more sense to go that route than to build a relay controller board and then have to add it to the schedule of a RPP.  All in all, 60% of the current draw of a 50 light LED string is pretty much next to nothing anyway so is cost savings really a reason to shut them down?

Offline Steve Gase

  • Supporting Member
  • ******
  • Join Date: Mar 2013
  • Location: Georgetown, TX (near Austin)
  • Posts: 1,037
  • Kudos: 5
    • WinterLightShow in Georgetown, TX
Re: Pixels take more power than I thought
« Reply #13 on: March 08, 2015, 01:56:17 PM »
The problem with using an AC controller is that even at full on, the wave is not clean and can hammer the hell out of any inductors connected to them such as the transformers in the DC power supplies.  You can buy 120V and 240V timers that can handle well over 40A if needed and since most will only be switching on and off once per day, it makes more sense to go that route than to build a relay controller board and then have to add it to the schedule of a RPP.  All in all, 60% of the current draw of a 50 light LED string is pretty much next to nothing anyway so is cost savings really a reason to shut them down?
With a large show, the cost savings might be justified.  The mentioned relay board is pretty cheap.

Scheduling would be easily done -- put an entry in the crontab of the FPP.

The other reason that makes an FPP-controlled relay most attractive (to me) is the ability to "reset" my lights remotely.  I've had situations where a pixel gets stuck, and the lights after that pixel stay on.  Usually a power cycle cleans up everything and the stuck pixel works fine... but there are other times where the stuck pixel is now a bad pixel and the lights after that pixel can't be controlled.    A power cycle will at least turn off the pixels after the stuck/bad pixel. 

With the this thread as the instigator, I think I will implement this change... :)

Offline Jeffl

  • Supporting Member
  • ******
  • Join Date: Dec 2013
  • Location: Brandon, SD
  • Posts: 519
  • Kudos: 4
  • Just remember. Wireless is nice. Copper is king!
    • Christmas In Brandon
Re: Pixels take more power than I thought
« Reply #14 on: March 08, 2015, 02:19:45 PM »
The more I hear the more I think I'll just stay with what I have always done and leave them hot.  The only difference is I'll shut them off at the end of the season in January instead of leaving them run till mid March when I usually take things down.

 

Back to top