Fence Pixels

Operation Construction Issues
 

In 2013 I put up the first fence pixels. Unfortunately, it turned out I had received a bad batch. They were extremely unreliable and I had to replace several dead pixels every week. So in April 2014 I torn them all down and pitched them in the trash.

In November 2014 I put new strings into operation with the big difference being 10cm spacing where as in 2013 the spacing was 15cm increasing the total pixels to 665 instead of the previous 440.

Fence pixels 2014 - 665 individually computer controlled LEDs over 200 feet.


Fence pixels 2014 - Perimeter lighting.


Fence pixels 2013 - 440 individually computer controlled LEDs over 200 feet.

Smoothly cycling through the color spectrum. The meteor takes a fast flight around the yard.

2013 Videos:

 

The strings have been cut to lengths that fit between the fence posts and the stock connectors have been replaced with much smaller JST JWPF waterproof connectors which you can see in use here.

The 12 volt pixel power is delivered by 25 amp buck converters mounted on each corner post which are fed from a 35 volt 14 awg lead that runs along the rail next to the pixels.

Buck converters are located on the corner fence posts

The pixel controllers are model ECG-P2 made by Joshua Systems. The 5v digital signal the controller produces will not travel very far, so it is converted into a differential signal using the model ECG-PPD-D pixel extender, also by Joshua Systems.

The 35 volt/16 amp power is supplied by a model SE-600-36 made by MeanWell.

Fence control is located in the basement



12 volt / 25 amp buck converter

In April 2014 I torn down the original fence pixels. They were extremely unreliable and I had to replace several dead pixels every week. The signs were there from day one when six of the eight strings I purchased had dead pixels right out of the box.

I contacted Ray Wu in May and he was aware that his "Technicolor Pixels" had issues but he assured me the problems had been corrected and the new version would be reliable. So I took a chance and ordered replacement strings, which I tested when they arrived, and there were no dead pixels.

Now when I get a dead pixel I am replacing that segment of pixels between the posts with pixels that have been "sealed" with Loctite 406 , which is applied around the LED hat and the wire entry points.




Damage from some critter to a non-fence related coax cable

This cable is not part of the fence lighting but it does show again how critters will damage cables. This cable was lying along the foundation and was intended to be temporary but it ended up being there for almost two years. My dogs could not get to this cable so it was most likely a rabbit or mouse.

At this point I am working to get any remaining fence pixel cables off of the ground. In the process I found the data cable to the west side had a small 1/2 inch wide hole chewed in the jacket but the insulation of the conductors did not appear to be damaged so I taped it for now. This is a flood-filled direct burial cable that was lying on the ground ahead of where it is buried. Perhaps the critter decided to find something else to eat once it tasted the flood-filled gel.

Damage from my dog digging up the cables


Yellowing Plastic Domes - new on left, 5 years old on right

Over the first four years the pixels were in operation they were left on day and night. Usually this only included the pixels next to each fence post and two to four additional pixels evenly spaced out between the posts. Over time the plastic domes over these LEDs have become very yellowed, especially the pixels next to the fence posts which were always on. In June of 2018 I added a daylight timer to the software that turns the pixels off 15 minutes before sunrise and back on 15 minutes after sunset.

Exploded buck converter on 14 December 2017


Fence pixel melted down on 30 March 2015


Puppy chewing damage to connector