Sunday, October 12, 2008
Prototyping Begins
So this weekend was full of making some stuff to begin prototyping. To get a 12 power supply to run my LED rails I took an old computer power supply and hacked it to run without needing a feedback from a motherboard by crossing two pins on the motherboard power connector. I then put a switch in that connection so that I could turn the power on and off (without unplugging it from the wall). I then tapped into the 12 volt line going to one of the disk power lines, and made two bare wires come off of it.
I gutted the free 17 inch LCD I had got from my work, and found that it had the FFC issue; meaning that the stupid ribbons going to the LCD crystal from the logic board were these really really flimsy and flexible ribbon-like cables. Think as thin as a couple pieces of paper (arg!).
I put the backlight from the LCD on the bottom of a cardboard box.
I took my Microsoft VX-3000 web camera, took it all apart so all I had left was the tiny circuit board and the USB cable going to the computer. Then I unscrewed the lenses, broke apart the small glass IR filter on the underside of the lenses, cut a small piece of floppy disk film, put it on the camera sensor and screwed it all back together. The floppy disk film acts as a homemade, cheap IR filter for the camera (you can also use exposed negative film or a chunk of one of those memorex black cds).
I taped the web camera down onto a circular piece of cardboard, and placed it on top of the backlight that was on the bottom of my box.
I placed the LCD crystal on top of the box.
I then created one LED rail, by bending some cardboard, spacing out tiny punctured holes 2.1cm apart, and wiring two rows of 8 LEDs each.
I attached this to my 12 volt power lines, and made sure they lit up. Now you cant see IR like, but I got 850nm which is closer to red in the visible spectrum than the 900+nm IR lEDs commonly used. I chose this wavelength simply because I had read that the lower the wavelength, the brighter the illumination. And I realized also that they produce a faint red glow so I know when they are on and working without having to use my web camera to see if they were.
I placed the rail on top of the LCD panel, turned everything on and made some measurements as to the angle of my web camera lenses, so that I knew how deep I'd have to make my box at a minimum so that the camera picked up the inputs from the whole screen. Measurements and calculations for using a 17" LCD screen are seen on the following notes:
Here is what the setup looks like using the cardboard box:
(Notice the white FFC cables hanging over the box. Component boards on right side are the boards from the LCD)
(Backlight on)
(Place fingeres on and see blobs produced on the screen...looks good)
(LED rail along one side. Notice tape measure. I tested different layers and how far away from the tip of the LED that I got a recognizable blob. The best combination was anything having a piece of glass/acrylic as the most bottom layer)
So I now had numbers to work with in what dimensions I needed to make my box.
I gutted the free 17 inch LCD I had got from my work, and found that it had the FFC issue; meaning that the stupid ribbons going to the LCD crystal from the logic board were these really really flimsy and flexible ribbon-like cables. Think as thin as a couple pieces of paper (arg!).
I put the backlight from the LCD on the bottom of a cardboard box.
I took my Microsoft VX-3000 web camera, took it all apart so all I had left was the tiny circuit board and the USB cable going to the computer. Then I unscrewed the lenses, broke apart the small glass IR filter on the underside of the lenses, cut a small piece of floppy disk film, put it on the camera sensor and screwed it all back together. The floppy disk film acts as a homemade, cheap IR filter for the camera (you can also use exposed negative film or a chunk of one of those memorex black cds).
I taped the web camera down onto a circular piece of cardboard, and placed it on top of the backlight that was on the bottom of my box.
I placed the LCD crystal on top of the box.
I then created one LED rail, by bending some cardboard, spacing out tiny punctured holes 2.1cm apart, and wiring two rows of 8 LEDs each.
I attached this to my 12 volt power lines, and made sure they lit up. Now you cant see IR like, but I got 850nm which is closer to red in the visible spectrum than the 900+nm IR lEDs commonly used. I chose this wavelength simply because I had read that the lower the wavelength, the brighter the illumination. And I realized also that they produce a faint red glow so I know when they are on and working without having to use my web camera to see if they were.
I placed the rail on top of the LCD panel, turned everything on and made some measurements as to the angle of my web camera lenses, so that I knew how deep I'd have to make my box at a minimum so that the camera picked up the inputs from the whole screen. Measurements and calculations for using a 17" LCD screen are seen on the following notes:
Here is what the setup looks like using the cardboard box:
(Notice the white FFC cables hanging over the box. Component boards on right side are the boards from the LCD)
(Backlight on)
(Place fingeres on and see blobs produced on the screen...looks good)
(LED rail along one side. Notice tape measure. I tested different layers and how far away from the tip of the LED that I got a recognizable blob. The best combination was anything having a piece of glass/acrylic as the most bottom layer)
So I now had numbers to work with in what dimensions I needed to make my box.
Subscribe to:
Post Comments (Atom)
Nolan,
ReplyDeleteI've been building my own multitouch box after your pattern here (LED Light Plane). I have had some difficulty getting blobs to show up. I've tested my LED arrays and infrared camera and think they are working properly. The LCD panel works well, I just don't know what the problem could be. I'd appreciate any help you could offer. I describe the project and the problem more in depth on my blog at http://russlewis.org
Thanks,
Russell Lewis
Hi,
ReplyDeletegot a school project like this aswell and I was wondering if I could pattern my work with you. Whats your nui id so that I can see some of your post there (in case you have one)
Feel free to do whatever you like with the info you read on my blog, but if you can, reference me. My name is Nolan and my nui id is PeauProductions. Im sure you'll find plenty of posts that Ive done.
ReplyDeletehey PeauProduction u r doing good job.. i m making my own mt and following ur blog.
ReplyDeletei've to inform you that i m also translating ur blog in Hindi Language for indian reader.. i need ur permission.If u have any objection with that plz inform me on my id aatur19231209@yahoo.co.in
my blog : questsomething.blogspot.com
which software u r using for touch sensing.
m using Touchlib on my system but it is not working properly. Will u sujjest me another one. I'd appreciate any help you could offer.
http://questsomething.blogspot.com
Thanks,
Aatur Harsh
Hello reader from Singapore here! I find this really awesome as I am given a task like this for my college's final year project. This is such a good resource I can research on!
ReplyDelete