Author Topic: Question about series wiring of pixels  (Read 708 times)

Offline jhoybs

  • Jr. Member
  • **
  • Join Date: Nov 2015
  • Location: Muskego, WI
  • Posts: 71
  • Kudos: 2
Question about series wiring of pixels
« on: December 30, 2016, 11:36:55 AM »
I'm starting to plan my conversion to pixels next year and I have a very elemental question that I can't seem to find the answer on.  I have 3 modest mega trees that I'd like to do RGB.  I'm pretty sure I can do local power injection (next to each tree), but I don't know if my data scheme is going to work.  Since my trees are modest size (11 ft), I'd hate to buy a controller for each and use all 16 outputs.

Will wiring the data lines in series work (see pic - strings folded at top)?  If it would, I'd only need to use 2 outputs of a F16v2 controller for each tree.  Because the F15v2 supports 640 pixels per output, I'd think the output would be able to handle 70 x 8 = 560 pixels?
Jim H.
Muskego, WI
Falcon F16v3, PiCap, FPPs. Lynx Expresses, xLights

Offline tbone321

  • Hero Member
  • *****
  • Join Date: Oct 2014
  • Location:
  • Posts: 1,331
  • Kudos: 35
Re: Question about series wiring of pixels
« Reply #1 on: December 30, 2016, 12:59:34 PM »
You can do that with the primary issue being the distance between the strings.  The distance between nodes for data is limited to a few feet at best and if you are connecting the strings in a serial manor for data, then the distance between the strings is also the distance between nodes from the end of one string to the beginning of the next.  Another issue with doing it that way is a kinda all your eggs in one basket situation.  If a port fails or one of the nodes fails, all of the nodes on that string from the port or the bad node also go out and when the potential is half of your tree going dark ....

Offline jhoybs

  • Jr. Member
  • **
  • Join Date: Nov 2015
  • Location: Muskego, WI
  • Posts: 71
  • Kudos: 2
Re: Question about series wiring of pixels
« Reply #2 on: December 30, 2016, 01:33:46 PM »
Thanks for the clear response, at least I'm thinking clearly.  Honestly, I never thought about a bad pixel taking out the entire channel - that probably isn't good.  The alternative is having to buy a controller for each of my 3 trees.  I have 2 additional questions:

1)  If the data line is limited to only a few feet without a pixel, would you have to place null pixels inline to more remote locations every 2 feet?

2)  My plan is to add a bunch of 2, 3 & 4 foot trees around my yard.  In this case with a bunch of remote "mini-trees", what is the best practice for creating these out of pixels in terms of wiring and the amount of controllers?

I've heard a lot of good things about the F16v2, but with my smaller trees, is there a Falcon solution for a 2-4 channel controller?

Thanks!


Offline gadgetsmith

  • Hero Member
  • *****
  • Join Date: Apr 2014
  • Location: Charlton, NY
  • Posts: 907
  • Kudos: 14
Re: Question about series wiring of pixels
« Reply #3 on: December 30, 2016, 01:41:29 PM »
Keep in mind distance between strings and/or nodes varies widely from one string to another, almost with no rhyme or reason.  Issues such as distance between nodes can easily be extended by a uamp or buffer/line driver if having issues.  I've tested up to 15ft between nodes with reliable operation, using CAT cable, which  I'd recommend for maximizing data line runs.  Pixel failure modes also vary widely, and in my experience, the most common failure is loss of a single color, although I've also experienced loss of a string beyond the bad pixel.
Personally, i think the biggest drawback is power injection. It's very doable, but more complicated (and fickle at times) than single runs from the controller.

My 2 cents.

Offline tbone321

  • Hero Member
  • *****
  • Join Date: Oct 2014
  • Location:
  • Posts: 1,331
  • Kudos: 35
Re: Question about series wiring of pixels
« Reply #4 on: December 30, 2016, 08:34:46 PM »
Thanks for the clear response, at least I'm thinking clearly.  Honestly, I never thought about a bad pixel taking out the entire channel - that probably isn't good.  The alternative is having to buy a controller for each of my 3 trees.  I have 2 additional questions:

1)  If the data line is limited to only a few feet without a pixel, would you have to place null pixels inline to more remote locations every 2 feet?


That depends on a lot of things.  The distance between a port and the first node on the F16V2 does not have that distance limit and can go much farther.  I would say that with a good cable you could easily go 50 feet or more so if you have each string in your remote locations connected to its own port, you might be able to just wire them in.  As for connecting these strings in series, you would need to test to see how far you can get.  Some can get well over 10 feet while others can't get much past 3.  You can use null nodes to extend the distance but then you need to take them into account when defining your models.  A better idea is to use a uAMP which is a falcon device that can extend the distance between it and the next node up to 100 feet and doesn't use any channel numbers.  As far as your sequence or models are concerned, they are not even there.

2)
  My plan is to add a bunch of 2, 3 & 4 foot trees around my yard.  In this case with a bunch of remote "mini-trees", what is the best practice for creating these out of pixels in terms of wiring and the amount of controllers?

I've heard a lot of good things about the F16v2, but with my smaller trees, is there a Falcon solution for a 2-4 channel controller?

Thanks!

The F16v2 is an impressive device with many capabilities but it is not exactly a low cost one.  There is another controller called the uSC which is a single port controller and is not all that much bigger than a postage stamp.  It runs on the PixelNet protocol which can be provided from either a dongle or from the F16v2.  It also uses a hub that supports up to 16 uSC's and provides both the signal and power to the uSC and the pixels.  These would work perfectly for your small trees but I don't know if they will be available in any further coops. 

Offline deplanche

  • Full Member
  • ***
  • Join Date: Aug 2015
  • Location:
  • Posts: 179
  • Kudos: 4
Re: Question about series wiring of pixels
« Reply #5 on: December 30, 2016, 08:40:32 PM »
In the F16, you can define null pixels on the string outputs, so you don't have to worry about adding those to a model when sequencing, you can take care of that on the hardware side.

There are also expansion boards for the F16, that allow you to add additional strings, which can be in more remote locations. 

If you add a sketch of what your layout would look like, you might be able to get more detailed suggestions.

Offline jhoybs

  • Jr. Member
  • **
  • Join Date: Nov 2015
  • Location: Muskego, WI
  • Posts: 71
  • Kudos: 2
Re: Question about series wiring of pixels
« Reply #6 on: February 01, 2017, 01:54:12 PM »
So now that the F16v3 is released, this is what I'd like to do:

Using a single v3:

1) Drive an 840 pixel tree with the controller mounted in the tree (ie: power and signal would feed from the center of the tree, none more than 3-4' from the start of the first string).  I'd probably use 1 or 2 outputs.

2) Drive two 180 pixel trees each around 25' from controller to start of first string.  For these trees, I'd power inject them with external supplies located in each tree.

So, from what I've read, this is doable, correct?

For reliability, what is the best way to drive the 2 remote trees from a data signal standpoint (SPT2, CAT5, null pixel, diff board - seems like overkill)?

Offline pixelpuppy

  • Hero Member
  • *****
  • Join Date: Aug 2015
  • Location: Dallas, TX
  • Posts: 506
  • Kudos: 10
Re: Question about series wiring of pixels
« Reply #7 on: February 01, 2017, 02:09:41 PM »

You could do that with a Falcon F4V2 controller ::) It has 4 pixel outputs, expandable to 12.  With this you could have 4 ports local, 4 more ports at a further location on one side of the yard and 4 more ports further away on the other side of the yard or up on the roof.
Vixen and xLights for sequencing / FPP for scheduling and playing / Falcon controllers for pixels / DIY controllers for everything else

Offline jhoybs

  • Jr. Member
  • **
  • Join Date: Nov 2015
  • Location: Muskego, WI
  • Posts: 71
  • Kudos: 2
Re: Question about series wiring of pixels
« Reply #8 on: February 01, 2017, 02:58:56 PM »
Yeah, I do know that a v3 is overkill, but I'm also planning down the road.  Next year, I was going to add 2 more 840 trees and then eventually add a bunch of 100 pixel trees.

 

Back to top