BTW, there are alternate ways to mount pixels that can help to compensate for the aspect ratio. For example, instead of pointing the pixels toward the viewer and using them as point sources of light, you could turn them around and use stand-offs, and then they behave like miniature flood lights to give you more diffusion off the surface of the garage door (if it is a light color). An example is shown in the diagram on page 10 of this article:
http://downloads.eshepherdsoflight.com/Howidid-DumbRGBPixelGrid.pdf
EDIT: page 26 + 27 shows a little more about the mounting technique with stand-offs.
That mounting method is probably easier with pixel strings rather than strips, but it could also be used with LED strip. (ignore the part about the charlieplexing if you are using pixels)
don
Thanks for the input guys, guess I should explain what I am doing a bit as well. The display I am working on is for a community light festival that we have every year, the old sign as you came into the park was just a wire frame with letters on it using standard lights, was about 20' long and 4' high (The reason I am trying to come close to these dimensions) Of course after years of use it is in need of repair and after looking at cost to re-light it we looked at using pixels instead but then we took it a step further and thought what if we make it into a matrix so we can change the text, add background effect, etc.
The sign will be along the side of the road as you enter the festival area, so it will be about 100' back from the road, but we have alot of flexibility in where we actually put it to get the most effective view of it. There will be no backboard or anything, in our region we can get some pretty severe wind gusts as winter sets in so we don't want to have a huge sail out there if it had a plywood backboard or was solid in any way. What I am looking at doing is making an ABS grid 17' x 4' with vertical supports every 12", then use the silicone strip clips to attach the strings to the grid at whatever spacing we establish. I am thinking of just temporarily attaching them and testing it until we are happy with the resolution of it and adjust spacing until it looks proper. Once I have all of this we are thinking of just attaching the new grid with the matrix to the original wire frame so it shows the evolution to the more modern system and saves me from having to build a support structure as well.
The majority of things that will be displayed is just text messages, any special events, schedules, sponsor recognition and so forth, with some background effects like snowflakes, fireworks, etc. It would be a huge bonus if I could also display images, but is not a key requirement for what I am trying to accomplish.
The more input and ideas I get the better! I really want this to wow people as they enter the festival area this year.
Sounds like fun.
I would probably use the 12V strips, that way you can power it from one end only. The strips seems to have lower resistance than wire, so you maybe able to drive all 5M or 16.4 feet of LED's from the power side.
At $30 for a 5M string thats $480 dollars for the LED's. If you space the strips at 2.66 inches and use two horizontal pixels, it will give you a resolution of 75x16. This is basically two lines of text at 12 characters. Since it appears this sign is not for personal use, I would suggest going ahead and doing the 1:1 with 32 strings at 1.33 spacing. This gives you much more space for putting up messages. Basically, you get four lines by 25 characters. With xlights, you can size the characters to suit you. This will give you more pixels for your sign, and more options.
Also, if you want daylight readability, 5V is marginally brighter, but may require inserting power in the middle of the string.
Also, how far is the sign from the control computers. Just curious, most controllers do 10/100 Ethernet which should be good to 100 meters. You can use a Raspberry PI if you need a local controller.
Ron,
Thanks for the input. I have been playing a bit with x-lights trying to get resolution figured out a bit, just made 2 matrix's in the preview model, 1 using single pixel model with length 150, and the 2nd simulating using pixel length of 75 for the strip(groups of 2). There is a huge difference in resolution between the single pixel model does have much better length and resolution. Of course this is still just a model, from what I have seen so far I need to see it in real life to really know what it will look like.
As far as the pixels, I have a strip that I have been using for testing and it has no problem running the entire strip on 5V from what I see so far. I was planning on looping the strips at both ends and injecting power to each row. From the budget numbers I have for this year my limit will be 18 strips so I will have to see what I can do with that, I actually overshot my budget already, didn't account for 8100 channels, so I need 2 controllers as well as shipping on the strips adds considerably to the cost, plus I am in Canada so items from China are subject to 26% duty and tax on them as well. So $450 turns into $650 with shipping, then turns into almost $1000 by the time I get them. All the hidden costs add up quick.
The Raspberry Pi looks very interesting but still don't fully understand that one yet. I was planning to put the pixel controllers right on the display with a wireless router, then put the control PC somewhere safe, just need to fire it up and figure out the range I have first. Next year to bring it up to Raspberry Pi might be a good option, won't have time to add that to it this year with everything else on my plate for this year. Would definitely like to know more about the Raspberry Pi solution though.
