Compliant surface

The past week or so I’ve been thinking about (and gathering the materials for) the body of the table, and a compliant surface. I’ll post more details on the table later, for now I want to talk about the compliant surface.

Tinkerman has had some very succesful results with textured silicone on drafting vellum (http://nuigroup.com/forums/viewthread/2197/). He pretty much applied textured silicone (not thinned!) to a piece of drafting vellum, let it cure, and then laminated the other side of the vellum to keep it from smudging and eroding and such, and then place the whole thing on the touch surface silicone down. Of course this technique on it’s own is useless for me, since I’m using an LCD matrix, which means that whatever compliant surface I want to use, it needs to be (near) transparent.

So I was thinking, what if I take out the drafting vellum and just apply the silicone directly to the laminating sheet, and then texture it the same way Tinkerman did (with some finely textured cloth, such as nylon). Just to clarify, this is what I’m after:

I’m not sure what the properties of textured silicon are as far as transparancy goes, but according to mr. Tinkerman himself it should be fine. I’ve been trying to find a laminating sheet that’s big enough, but so far no luck. Maybe I’ll run some tests with a smaller sheet first.

I’ll keep you updated on this and expect some more details regarding the body structure of the table soon!

First SketchUp

Unfortunately I suck at SketchUp, but today I made a little sketch during work, to try and get a clear picture of what I want my table to look like. This is what I came up with for now:

Like I said, it’s still rough and simple. The two interesting things here though are the fact that I want it to have speakers (they probably won’t be mounted like that, but oh well) for sonic feedback on touches and gestures and such. Also, I’m thinking of going with different looking materials than everyone else. So far I’ve seen a lot of futuristic, shiny, polished looking tables. I think it’s be nice if I could find some high quality, darkly coated wood where you can still see the textures.

Anyways, just a thought. I’ll try to get a bit better in SketchUp and make some more detailed graphics of what my table will look like, and what the layering on the inside will look like.

TouchLIB screenshot

So here’s the screenshot from the TouchLIB config like I promised. This shot was taken in the early evening, so there’s a moderate amount of environmental light (thus, distortion), but after calibrating and tweaking it a bit this is what I got at very light pressure.
 

 

Again, dragging is still not working well and it’s hardly possible to track the blobs when you do. So I need to either come up with a compliant surface, or I need to give the lasers that are being experimented with over at NUI a go.

The MT surface is working

Today I’ve mounted and wired my LED’s. Like I’ve said in previous posts, I’ve mounted 20 LED’s in 4 parallel series of 5, 10 LED’s on top and 10 on the bottom. At first only 1 series worked, so I’ve had to check and recheck my circuit a couple times, making it a bit time consuming. All in all it probably took me 5 to 6 hours to get it all done.

I’ve used quite a bit of duct-tape for now, I’ll probably come up with a more permanent solution later. Also, try to stay away from duct-tape because it’s very, VERY annoying to have to take it all off again to check your circuit.

Here’s some pictures:

 

This is the wire I used; standard wire really, you can get it anywhere. To save myself a lot of messy soldering, I decided to go with these connector blocks (I don’t know what they’re called in English, they’re called “kroonsteentjes” in Dutch). Just pop in the wire and screw it down; perfect!

Like I said, I used a lot of tape to make sure everything stayed where it was supposed to be. I’ll probably screw some things down to the frame later, or design the body of my table to safely house all the parts.

 

All the LED’s are now mounted; I’m ready to put in the resistors and start the wiring!

Here’s the surface with all the wiring and the resistors. It’s still pretty clean, but I don’t trust the tape to hold forever. Also, transporting it like this is tricky because it’s very fragile.

Here’s the stripped adapter wires; make sure you get your polarity right or bad things might happen (i.e. make sure that you know which one is + and which one is -). Also note that usually on LED’s, the short leg is – and the long leg is +, but on these OSRAM’s it’s the other way around.

 

Here’s the complete thing after I took it apart 3 times and tweaked things until it finally worked.

 

 

And of course, awesome blobs. I also configured TouchLIB and everything worked fine. Of course dragging is still a bit of a problem seeing as I don’t have a compliant surface yet. It needs to be a clear surface because I’ll have an LCD matrix underneath that needs to be visible, but I really don’t want to go with silicone because I just don’t like the feel of it. If anyone has any suggestions please let me know.

So, that’s about it for today. It wasn’t nearly as bad as I thought it would be, and I got everything up and running pretty quick. I’ll upload some TouchLIB screens soon, and maybe some video’s of running demos.

So again, if anyone has any suggestions for transparent compliant surfaces other than silicone, please let me know. 

The next step is to start taking apart my LCD screen (scary), find a good compliant surface to get my dragging smoother, and start designing the body! Expect some SketchUp images of the table soon.

Stay tuned for more. 

MS Surface

A big update on the MS Surface project, also featured on the NUIgroup homepage, but I thought it’d be nice to have it on here as well. This is amazing stuff:

For more video’s check out the NUIgroup.

Nice showcase

Found this on nuigroup.com.

I couldn’t find any info on what language this is written in, or any of the techniques used.. Looks like it’s a rear DI, can’t tell for sure. But still; nice!

Building the frame

Today I got started on building the frame. I cut up my aluminum profiles and drilled holes for the LED’s to sit in. I’ve got 20 LED’s total, 10 on one side and 10 on the other. Unfortunately I didn’t get around to soldering, but hopefully I’ll get a chance to do that this weekend.

Here’s some pictures:

This is the saw I used to cut the profiles at a 45 degree angle to make sure they fit together nicely.  

An hour and a paralized arm later, I was ready to space and drill the holes in the frame that the LED’s will sit in.

 

I bought these LED holders which, upon later inspection, turned out to be too big for the holes I drilled. I won’t be using them for now.

 

I mounted the LED’s without the rubber holders, and they ended up fitting really nicely. I’m considering leaving it like this, though I might end up drilling out the holes a bit bigger and putting in the holders anyway.

 

Here’s the whole thing put together; nice!

 

This weekend I’ll be sticking in the LED’s permanently and doing all the wiring. If that comes along without too many issues, I’ll have time left to do some preliminary software tests with Touchlib. 

Backtracking

Okay, so I started this blog a little late into the project. I’ve already polished my acrylic and modded my xbox 360 live vision cam to detect IR light. I’ve taken some pictures of the results and the other materials I’ve already gathered, but unfortunately I don’t have any footage of the polishing or camera modding, but the web is full with detailed tutorials and how-to’s (try www.nuigroup.com).

Here’s some pictures:

 
This is my 50cm x 35cm acrylic.

I polished the sides with fine sanding paper and a small blowtorch.

This is my powersource; a regular 12v adapter. Available at any random hardware store.

This is the aluminum profile I’ll be using. I don’t think I’ve see anyone else use this for their table, but the shape looked functional to me; the acrylic will be clamped in the “u” shaped part, while all the soldering work of the LED’s will be sticking out on the other end, supported by the horizontal part of the frame.

The frame at another angle.

 

I bought 20 OSRAM SFH485 IR LED’s, recommended all over the web as some of the most suitable LED’s for an FTIR surface. I’ll put these in 4 parallel series of 5, 2 series on top and 2 series on the bottom of the acrylic, powered by the 12V powersource with one 270 Ohm resistors at the end of each series.

 

This is my modded camera. Unfortunately I don’t have any pictures of the process of modding it, but it was pretty straightforward. The hardest part was prying it open without breaking it (note the prying marks around the edges). After that all I had to do was unscrew the insides and remove the IR filter, and replace it with 2 layers of exposed photo negative. After that I screwed the assembly back together and popped the outer shell back together. Now it doesn’t filter any IR light (a good way to test this is pointing a remote at the camera; if it washes out the image with white completely, you’ll know that your camera filter isn’t filtering IR light. If all that shows up is a small light dot, then something is wrong).
 

I’m hoping to do my soldering and construction of the frame somewhere along this week, so I can run my first software tests.

More pictures and footage will follow soon. 

First steps

I’ll be recording the process of building a LLP-FTIR hybrid LCD multitouch table on this blog, like so many others do. Currently I’m working on getting all my materials together. I’ve got my LED’s, IR cam, acrylic, power supply, resistors and aluminum frames. The plan is to use an LCD matrix from a 22″ widescreen, instead of using a projector. I chose this because even though I will have to make the screen a lot smaller than it could be with a projector, the image will be a lot clearer and it will be a lot cheaper to make. This is still a couple of steps away though, because first I have to wire my LED circuit.

Pics will follow soon.