So I converted my Side DI LCD to a FTIR LCD this weekend, and am getting so so results.
Here is a video showing the view from tbeta.
So after being on this forum for months and searching through many forum posts, I still dont think I can find anyone who has created a successful LCD FTIR. Most people using an LCD give up and go with a Side DI (which is what nima I think ended up with) or Rear DI, go with a LLP, or combine these techniques to make up for each design’s possible shortcomings. Or maybe the fact is that people have created them, but havnt posted enough about them for me to remember them.
My setup works, but since there is no compliant layer, it takes a lot of force to create bright enough points to be registered as blobs. Putting a compliant layer on isnt out of the question, but for LCD setups, I dont know if anyone has created a usable compliant layer that is clear enough not to disrupt the LCD image.
My acrylic is about 6mm thick, using 48 (12 on each side) 5mm LEDs, which are pointing straight into the 90 degree sides of the acrylic. Ive seen mention of angling the LEDs or sides of the acrylic:
http://nuigroup.com/?ACT=28&fid=36&aid=48_5C8SbBuWgVIaM7gCANad http://www.outgribe.com/jason/ftir/ftir.html
But also reports that it isnt necessarily needed (any insight is welcome).
Let me know what you guys think, as I really like the control that my LCD Side DI produced, but wanted to minimize the 5mm or so height that my fingers would get picked up to a contact pickup. If I cant get this figured out, Ill probably tear some CD drives apart and put together an LLP setup....
I would turn on amplify it will amplify the blob input, and make it somewhat more sensitive. and tests that i’ve done mind you these arent done with LCD but it shouldnt matter, i’ve found out that haveing led’s on all sides and every cm makes a big difference.. the type of acrylic also make a big difference. Cast Acrylic is a lot better than extruded, but the best is Scratch resistant, and non glare acrylic, they are the best.
I didnt mention it in my first post, but I used all pre-loaded filters in tbeta to get the best blobs I could. I didnt want to get giant circles, only blobs about the size of my fingers, so setting like amplify werent necessarily turned all the way up. Im pretty sure it isnt a tbeta setting that is hindering the blobs being produced, which is why I am asking what hardware techniques (or software if there really is anything else to try) that would help an LCD FTIR setup perform better, thanks.
Can you post the raw image from the camera? It’s hard to tell if it’s software or hardware related without a picture of what the camera sees (unfiltered).
Alright, here you go:
1) Shot of tbeta (no fingers)
2) Shot of tbeta (fingers)
2) Shot from webcam software (no fingers)
3) Shot from webcam software (fingers)
(the webcam software has a purple hued image since I probably messed with the hue while in tbeta recently)
it seems as though you have a hot spot or ur loosing a lot of IR at the center of your screen that will also effect sensitivity, threshold will have to be set a lot higher just because of that giant IR spot.
Hmm..that “square” of light that you see is the LCD backlight, ill mess with the threshold some more. I thought backlights didnt produce IR, but I guess my camera is seeing it. On the lenses I have a piece of floppy disk. Just a note, when I take the acrylic off and use the setup as a Side DI, the backlight isnt a problem at all and the blobs come out amazingly.
Any comments on whether I should angle the LEDs/acrylic edges or a suitable compliant layer for an LCD setup....?
The backlight may be an issue, but part of it looks software wise. First, you don’t need highpass and smooth on. Highpass has a smoothing effect and thus turning both of them on wastes resources. Also, highpass normally isn’t needed for FTIR. I would turn that off. You can see that ur loosing a lot of the blog intensity with highpass on. As long as the backlight doesn’t change, it should be removed with the background filter. I would also turn threshold all the way down until you get noise, then turn it up just slightly above the noise level so it’s more sensitive.
Well, I turned off the highpass filter and moved the threshold down, along with moving the smooth and amplify sliders around, but couldnt produce any better results than I had with the previous settings. I noticed also that some settings made it so that the fingers pressed down directly above the backlight, would light up and produce blobs, but not produce blobs when pressed down not-over the backlight. Also, it seemed that sometimes if I pressed down right next to another finger, the first finger would “steal” the light from the other and not let the second finger light up unless I pressed really hard. Also, some blobs only were produced after a second or so of holding the finger pressed on the acrylic, as if the camera saw it, but didnt want to make it a blob yet.
Still no dragging ability and firm pressure still needed to create blobs.
Hmm...so the number of LEDS....I have 48 total, all spaced less than 3mm apart, on all four sides, that have viewing angles supposedly from 15-30 degrees
Yeah the number of leds sounds about right… but what do you mean by viewing angle.
If you are refering to the leds lens light angle then this would be the best.
Angle: Sometimes the specs will show both full and half angles. So I’m putting both, but they are equivalent.
Maximum half angle: +/- 60 degrees
Maximum full angle: (60 + 60) = 120 degrees
Anything less than +/- 48 will not take full advantage of TIR, and anything above +/- 48 degrees will escape the acrylic. In order to ensure there is coverage, going beyond +/- 48 degrees is fine, but anything about +/- 60 is really just a waste as (60 - 48 = +/- 12 degrees) will escape the acrylic.
Peau,
Just read your blog about your MT table. It is pretty much exactly what I am wanting to build. I have an old HP laptop sitting around my house that I want to salvage the LCD from. Do you think there will be any problems with that? I keep reading about the ‘FCC’ issue but don’t have a clue what it is.
As for using an LCD from a laptop, the hardest thing may be the ribbions going from the lcd to the control board, but the biggest hurdle would be actually using the LCD at all. The reason is that unlike a seperate LCD, using one that relies on the vga “control board” on the motherboad is probably impossile to use, unless you want to squeeze your whole laptop inside your box. Keep in mind, all LCDs need both a control board and a power board to drive the image. So your best bet is not to use it.
Well, my overall intention is to mount the screen in a coffee table. I think i am going to use the laptop itself as the computer, only with the monitor displaced and made MT. There is a reasonable amount of slack in the cable connecting the monitor to the screen… Now to pick out a webcam. But that’s a conversation for a different thread.
