We’ve been learning a lot from the NUI Group, so we thought we’d give back a little by sharing a new technique: getting multi-touch blobs from infrared LED ribbons. We first tried to solder our own IR LED circuits. Although we got it to work, it was challenging, time consuming, and created many potential points of failure.

Because I was doing some other research involving lighting, I was familiar with flexible LED ribbons, and had a few of these ribbons/reels on hand in the visible light spectrum. We saw that many people were testing their blobs in IR right away, but that often required a dark room, an IR camera, etc. If you’re just trying to test the quality of your surface, this complicates the situation, as poor blobs could be the result of the camera, stray IR Light, the software, etc. So what we did was wrap one of the visible white light LED ribbons around the acrylic to see the FTIR effect first-hand, with our own eyes. By testing for blobs with visible light, you isolate the effects of the surface, can quickly see the results of different techniques (silicone of various types/thickness, Sulky Solvy, etc.) Not to mention it’s pretty cool to witness FTIR with your own eyes and watch your fingertips glow

We were getting really good blobs with the white light LED ribbons, so we started to hunt around for LED ribbons in the infrared spectrum. We didn’t find a source online, but we had a contact at EnvironmentalLights.com from my prior lighting work. We spoke with him, convinced him of the value it would bring to the multi-touch field, and he actually created a new product for us: Infrared LED Ribbons. You can buy these by the foot or by the reel, and a bunch of accessories are available: wire connectors, power supplies and cables, aluminum channels, spacers, etc.

We did a quick test of the infrared LED ribbon, and it produced some really good blobs with direct touch of your fingertips. It also generated good blobs with test sheets of silicone and Sulky Solvy with a Rosco Grey screen on top. What’s great about the ribbon is that there are so many LEDs surrounding the acrylic, that it’s pretty much completely flooded with IR light, even if the sides are not super clear. And it saves so much time.

Earlier this week, we built our first multi-touch display using infrared LED ribbons, which worked quite well. We used a polished piece of acrylic from TAP Plastics, and an L-shaped strip of infrared LED ribbon with a soldered plug that attached to our power supply. This was placed inside aluminum channels (specifically sized for the LED ribbons) that were cut in the shape of a frame, with neoprene spacers to keep the LED lights the right distance away from the acrylic. We used Sulky Solvy for the compliant surface and Rosco Grey for the screen. Both of these were stretched over the acrylic to make them taught, and affixed using a perimeter of thin double-sided tape. Our IR camera was a modified Xbox Live camera, and we borrowed an Infocus DLP projector for this test.

Our next steps are to work on the housing that will hold the display, find a permanent projector, and start testing our blobs with the touchlib software. We’re using Macs, and ran into a few snags with touchlib on that platform. But we plan to make some tweaks to that (or maybe the new Mac tracker will be ready by then, fingers crossed!) Our ultimate goal is to investigate and document the human factors and usability of multi-touch interaction, which we’ll continue to share with the NUI Group community.

Thanks for posting! I (we) really appreciate it =)

Nice work so far. Many of us have searched for LED IR ribbons with no luck. It’s great to finally see something!

A couple questions, do you have any specs on the LEDs (radiant intensity, etc)? Afterall, all LEDs aren’t created equal lol. It’d be nice to see some in the lower range as 940nm can be high for some cameras.

Also, can we get more information on this “Sulky Solvy” material? What is it, what type of material is it (is it semi sticky?), and how well does it work for FTIR? This is a new compliant surface for the community and compliant surfaces are one of the things the community really needs more experiments with (especially ones that can be purchased and used as-is).

Thanks again and continue the nice work. Congrats on your first working display!

Congrats on finding a supplier willing to listen. I spent a few weeks this spring contacting everyone I found that made led ribbon trying to get anyone to take a chance & load the hopper on their machine with IR LEDS to no avail. What specific LED device are they using? That would answer a bunch questions I would have regarding the output of these ribbons...I agree with Cerupcat, availability of a longer wavelength will provide more consistent results, 940 is also above the transmissive cutoff for some brands of acrylic

I posted some information on how we chose Sulky Solvy in the “Extremely Cheap Compliant Surface Material!” thread, which is what originally turned us on to the idea a while back. We’re definitely not the first to use this material, but we did get good results with how we used it.

I’m asking our contact at EnvironmentalLights.com for the specs on the IR LED ribbons, and will post them once I have them. I’ve also asked him if he’s open to producing a different wavelength.

If you had to define specs for your ideal IR LED ribbon, what would they be?

Thanks. I must of missed reading that he used the “Sulky Solvy.” I’ll have to try it myself. I never had any positive results (like many others) with interfacing fabric, which could be the reason the Sulky Solvy got overlooked.

I’ll have to get back to you on what some ideal IR LED ribbons would be since I’m on my way out right now. It could be interesting to see a circuit built in that allows for pulsing the LEDs and possibly an input pin to sync the pulse to a camera. Not necessary, but would be a great option for some more advanced MT techniques.

Hey guys great work, got a question though what are you using to secure your IR frame at the corners the blue plastic joint?

The blue joint is a temporary solution, we’re using blue painter’s tape We’re planning to solder three of them together to form a U-shaped frame. Then we’ll probably attach a clasp to secure the last piece. This way it will be easy to open it up and slide out the surface for maintenance (or to swap it out).

Ohh cool i did something similar i welded the 2 of the corners and the last piece i used aluminum tape to secure it… i thought you had found some sort of plastic joint cap…

That projector you guys use has a nice shape, which model is it?

Edit: Found it: it’s the inFocus LP120 - can’t find it on the official websites tho.

Taha - 09 August 2008 10:41 PM

Ohh cool i did something similar i welded the 2 of the corners and the last piece i used aluminum tape to secure it… i thought you had found some sort of plastic joint cap…

Not a bad idea…

So I need a quick response on this…

We contacted Greg, who is making the IR ribbons.  He is about to submit another order for IR Led strips.  He only wants to carry one wavelength of IR, so he wants to know what the ideal wavelength is.  We have had good results with 940, but I guess some cameras have trouble with that.  What is the most versatile wavelength?

Are the cheap cams (like modified X-Box live) able to pick up 850?  What about 880?

I told him I would get back to him later today or tomorrow… so a quick reply would be appreciated.

Thanks,
Jon

A few of us talked it over and basically here’s a summary of what we agreed on:

Ideal Specs:

Wavelength: 850nm (or 880nm, but 850nm would be better) - Both of these are easily seen by cameras and visible light filters can be easily found for these wavelengths. The lower the wavelength, the higher sensitivity which will equate to the higher ability to determine the pressure.

Radiant Intensity: Minimum of 80mw. The ideal is the highest radiant intensity you can find, which can be much higher than 80mw.

Angle: Sometimes the specs will show both full and half angles. So I’m putting both, but they are equivalent.

Minimum half angle: +/- 48 degrees
Minimum full angle: (48 + 48) = 96 degrees

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.

If you have any questions, feel free to ask. We appreciate your willingness to include the community.

wow, ENVIORMENTAL LIGHTS as the best customer service. Very nice, very helpful, just messed up the credit card info but still very good. When i 1st thought about doing FTIR i fund these ribbions a while ago and asked someone but they said the 940 was to high, guess they where wrong.Good work guys, just ordered quite a bit of stuff. Ill be posting my progress soon.

When i was on the phone she says that they will be realsing some 850NM soon :]

It looks like I’ll be trying these out with my next build, I’m still working with the LLP method right now.

Purchase a load of retail ir leds $60
Spend hours drilling and soldering $$$$
Connecting up to the power supply with an excited grin $0
Look of disgust when you see the faintest Blobs imaginable #$%&@$# PRICELESS

IR Led ribbons are the work of the gods. Is there any Australian group members interested in a group buy, just went to order some but the $120 dollars shipping seems a little excessive!!!

Steve

Jon, please let us know if the suggestions for ‘Ideal LEDs’ above were made in time.

If they were and you get information on current and future LED specs (datasheets preferably), that would greatly help the community. Thanks again!