Pixels… Finally

The replacement pixel boards arrived Thursday. JEC even had them made with a fancy white solder mask in the hopes it would reflect better than the standard green solder mask. I rushed to assemble one Thursday night. I got it assembled but was up until 1:00AM playing with it. At first I tried the code in the DIYC file library that allows for addressing without re-flashing the PIC, but didn’t have any luck. It seemed like the firmware was working, but wouldn’t respond to any vixen commands, so my best guess is that I wasn’t programming the channel correctly. I quickly gave up and put JEC’s standard firmware on for testing purposes. It fired right up and worked as expected. The only hiccup was that I put an amber LED where a red one was supposed to go, so I spent a few minutes desoldering that and replacing it.

I got a few of JEC’s power injector boards and they made life a lot simpler for getting power to the pixel. I wired the injector to a molex connector from an old CPU fan and plugged it into an old power supply to get my 12V. Friday night I put the pixel on my multimeter to see what the LEDs I chose draw at full power. It was a little under .5A, so in theory this power supply should be able to power 50 of them, or in my case, my whole string of 30. JEC recommends anything above 20 be powered separately because the CAT5 cable can’t handle that kind of load.

I wanted to see the pixel do a rainbow of colors, and didn’t want to hand encode all the values in vixen. I’ve been working on a PHP script that will let me modify vixen sequences, so I used it to do it. I found that rather than using RGB, HSV allows you to quickly do a rainbow by holding S and V constant at 1 and varying H from 0 to 1. I also found a bit of PHP code that would let me convert HSV to RGB. Once I had the RGB values I wrote them to the vixen sequence channel data. I still need to figure out how, from an RGB value, to determine the yellow and white components of that value, so that I can introduce the yellow and white channels appropriately. I also tested out my vixen script that generates a flicker effect on given channels with the amber LEDs on the pixel. The effect was actually quite convincing and still has room for some tweaking.

Pending water tightness tests, I’m planning on using 5LB white containers that deli salads come it. I have easy access to them and they’re free so I like that aspect of them. You’ll see in the video below, that the white plastic diffuses the LEDs quite effectively and still allows enough light out to keep it bright.

Yesterday was my daughter’s birthday party, so I didn’t have time to play, but today I took some time while the other daughter was napping to shoot a quick video of the pixel in action. I was rather disappointed because there is some very noticeable flicker in the video that isn’t apparent to the naked eye. I’ll have to see if there are any other settings on the camera that will help that in the future.

Saturday I finally received the DMX2Renard converter boards from Frank. He only had 5 so sent me one he had already populated. It was kind of neat because the one he populated, he put a RJ45 plug right on the back rather than the specified jack and needing a small patch cord. I think I’m going to try that trick on the other ones I assemble. It makes for a neat little add-on module. I think that pretty much catches up on all the news. We’ve decided to go on vacation this year, so pretty much any further blinky spending will be put to a stop, but honestly I think I have enough to keep me busy and still put on a great show for 2009. I’ve still got to pay for the stuff I’ve committed to over at DLA for the wireless DMX stuff. I went back and forth whether to drop out of the coops, but in the end, to run the Pixels I’d need a DMX splitter, so I might as well keep with the wireless and then effectively have a splitter.

Leave a Comment

You must be logged in to post a comment.