There are several core technologies available to build a multi-touch device – Frustrated Total Internal Reflection (FTIR), Diffused Illumination (DI), Laser Light Plane (LLP), and Diffused Surface Illumination (DSI) are among the most popular.
Previously I had intended to build my table based upon LLP technology as this was the most cost effective of the solutions and despite requiring a little extra TLC and setup provided very strong illumination and finger tracking. However, there are a few downsides to the technology, not the least of which is the chance of blinding yourself (no, really) with a powerful laser that doesn’t emit any visible light. This means you won’t realize you’re in trouble until it’s probably too late. The warning “Do not stare into laser with remaining eye” is ever so relevant here. If you’re careful you have nothing to fear but this is a concern. Another is that there’s no really good way to do fiducial (object) tracking with LLP. Some ideas were put forward but most are experimental and there is little software support for such a setup. Since I wanted to be able to track fiducials this was a negative, but I figured I’d get the table built first and then deal with this shortcoming later on.
Well, as my previous post pointed out, I’m starting a new build and with that new start comes a new technology choice as the basis for the build – DSI.
DSI utilizes a special acrylic with tiny micro-mirrors embedded throughout the entire surface that cause light injected into the edges of the material to be redirected out of the front and rear faces of the material. A simple writeup of the technology can be found on the NUI Group Wiki here. As you would expect a material that exhibits these kinds of properties doesn’t come cheap which is what put it out of reach on my previous build. With a larger budget this time around DSI has become a viable option. Given proper infrared illumination DSI provides solid finger tracking without the occlusion concerns of LLP and, most important of all, DSI provides a good basis for fiducial recognition and tracking.
Once I get a bit more design work out of the way I’ll be ordering the materials for this part (the actual surface of the table) – but for now I still have a PC to finish building and a projector to play with.
In the previous installment I discussed selecting a projector. This time I explain what to do when you finally have selected a projector and realize that there’s no way to get the image size you want within the confines of the enclosure you want to build. The answer is mirrors.
As I had previously stated I wanted an enclosure size of no more than 30″ high and 34″ wide and a screen size of 45″ diagonal, but the throw length of the projector I choose is 56″ to get that screen size. Here’s a step by step walkthrough of how to make 56″ of crap fit in a 30″ box.
First off how did I calculate a 56″ throw distance? I used the projection calculator found here. The results of which look like this:
It’s also a good idea to download the user manual for your projector as this will contain valuable information regarding the throw and placement guidelines. The relevant information for my projector looks like this:
The important things to note is that the suggested throw matches well with the previous calculation and that there is no “offset” in the image, the bottom of the image is projected straight out from the center of the lens and makes a nice right-triangle (the +/- (B) data is used for 16:9 mode, not 16:10 and I plan to use 16:10). Armed with this data – 56″ of throw for a 24″ high screen I can now start “folding” the projection – and I mean literally folding it. Using a method suggested frequently on the NUI Group forums I translated the projection triangle of the projector onto a sheet of paper. First I used TurboCAD to make a precise scaled triangle with one leg of 56″ and one of 24″:
I then printed this triangle out at 1/5th scale and cut it out. It’s also advisable to make a scale cutout of your projector itself and attach it to the end of your triangle but I found it much easier to play with different folds without it. Still it’s good to have.
After a while of attempting various single bounces I realized I’d need to do two bounces (much harder to get right) to get everything inside the enclosure dimensions I wanted. I spent a good bit longer playing with options here until I came up with the following:
I was able to measure (roughly) the dimensions and scale them up to get a feel for if this would fit inside my enclosure. It looked good (it is kinda hard to accurately measure something shaped like that) at first glance but I wanted to use another tool to validate. There just so happens to be a tool that can do this located here (beware this thing does some awful things to my CPU but it DOES work – eventually). After working with the tool to mimic my folding attempts I got the following results:
If you look in the lower left you’ll see the projected screen size – all I cared about here was 24″ in height, this calculator only does 4:3 aspect ratio calculations but if I hit 24″ high I’ll get a 38″ wide image. Looking at these results I can indeed use two mirrors to make a 56″ throw fit inside my 30″ high box. The projector is sized per the user manual and I tried to leave a little room to clear the cables off the back. The projecter is off to one side a bit, sticking out beyond the projected image but I plan to have a lip around the entire table so I’d just match that offset on the other side so that everything looks symmetrical – I need a place to put the computer and PSU anyway. Since I’d likely add a lip of 4-5″, I still end up with a total width of 34″ (or slightly less) which is right where I want it.
Hopefully this information helps others go from projector selection, to throw calculation, to paper triangles, to mirrors, and finally to construction. There has been a good bit of confusion on the NUI Group forums regarding this process so I wanted to detail it out a bit and show how each step relates to the next.
This is just a minor step along the way, I still need to figure out proper mirror sizes, and it’s unlikely that these calculations will all work out quite so neat and tidy when I actually drop it all into a box. This just gets me to a point where I know I should be able to make it work, and for now that’s good enough.
Next time I’m hoping to have some lasers. Cover your eyes.
Selecting a projector for a project like this isn’t as easy as it might seem. There’s a huge selection of choices available and price certainly plays a factor when trying to decide between them. One of the key requirements for a projector going into a multi-touch table is the throw length and in this case length really does matter.
While I am waiting for some other pieces to arrive I decided to try and find a suitable projector for this project. I really need to come up with a name for this, “project” isn’t very cool sounding. As previously stated one of the key factors when looking for a projector to go inside a table is the throw length. Specifically you want a projector with as short a throw length as you can get. A projector with a short throw length will create a larger image at a given distance than a projector with a longer throw length. There are “short throw” projectors available but they typically involve complex optics and have correspondingly high prices. They can also complicate your table design since the steep throw angle often (but not always) requires placing the projector in such a way that it requires significantly more space on one side of your design than the other. With all of this in mind I set out to find the best projector for my table – that I could reasonably afford.
Before you get into this too deeply there’s a few key things you need to know:
- What is your desired screen size – how big do you want your projected image to be?
- What is your desired enclosure size – how big do you want your table to be?
- What resolution and aspect ratio are you planning to run your table at – 800×600? 1024×768? 1920×1080?!
- What environmental conditions will you be using the table in – Dark room? Full sunlight? Something in between?
- Your budget – how much can you spend?
Once you have answers to those items you’re ready to go. In my case, I’d like a roughly 45″ diagonal screen size at either a 16:9 or 16:10 aspect ratio. This amounts to an image about 25″ high and 38″ wide (slightly different depending on which aspect ratio). Because of the aspect ratio I want I am looking for a projector with 1280×800 or 1280×768 resolution. I had initially hoped for a 1080P resolution but then reality hit (see the budget part). As for the enclosure, I don’t want my table to be more than 30″ high which is typical desk/table height (between 28-30″) and I don’t want it to be more than about 34″ wide – since it won’t fit through the doorways/stairways in my house if it’s wider. I imagine the table will be roughly 50″ long but that doesn’t really factor in here – the key dimensions here are the height and width. I also want the table to be usable in fully lit rooms, perhaps not full sunlight (that causes it’s own issues with IR tracking) but certainly partially sunlit rooms are on the list for me. My budget for the projector is $500 – which is , as I quickly realized, too low. More on that later.
Now that I had some parameters to work with I set about finding suitable models. A couple of great resources for this are:
Between them they cover a huge range of manufacturers and models and give pretty detailed specs on all of them.
I spent about 2 days combing through looking for projectors matching the specs I wanted. Once I had the list narrowed down a few things became apparent – my budget was woefully small, and nothing even within the realm of twice my budget was going to get me the screen size I wanted within the size contraints of my enclosure. The first of these issues was addressed by setting my budgetary expectations to a more realistic level. While there are several 800×600 and even a few 1024×768 projectors around that can be had for just over or around $500, there are not many 1280×800 projectors that can. While there are many 1280×800 projectors that list under or near $1000, none of them are true “short throw” projectors – those list for somewhere around $1500-$3500. Projector brightness or lumens is also a factor here. Some projectors output 1200-1600 lumens – this is still really bright, but maybe not bright enough for full daylight viewing or when paired with high contrast projection materials. I wanted something with 2000+ lumens. This, also, means more money.
The obvious first step after this dose of reality is to think “well, maybe I can find a used projector and save some money – I’ll hit my budgest that way”. And that seems perfectly reasonable. The problem with that idea at the level of projector I am buying is that replacement bulbs cost 1/3 – 2/3 the price of the whole projector. So if I were to find a $1000 projector for $500, but had to buy a replacement bulb (within a few weeks or even months), I might end up paying $800 in the end (bulbs typically range between $200-$300). At that point I’d be better off just springing for the new model, getting a warranty, and knowing the bulb should last a good, long time. So I upped the budget to $800-900. Really only two projectors made this cut (though there are lots of projectors in the range of $800-$1200). The first is the Dell 1609WX, the second was the Epson PowerLite Home Cinema 700 (hereafter known as “The Epson”). The list prices for these are $849 and $799 respectively. Features and specs are very similar with a few differences – here’s a comparison.
Here’s how I made my selection-
- Price – the Dell lists and pretty much sells for $849. The Epson lists for $799, but sells for as low as $749. Advantage Epson.
- Throw distance – The Epson throw ratio is at 1.5, the Dell, 1.55 (smaller is better). The Epson has a better zoom. The difference in throw distance for the screen size I want ends up being about 3″. That seems insignificant but as you’ll see later, every inch counts. Advantage Epson.
- Display Technology – The Dell is DLP, the Epson is LCD. Most would say DLP gives a better picture with greater contrast, and it’s probably true, but the rated contrast gives the edge to the Epson (but never trust those numbers folks!) so how to decide? In this case, because of the potential for strange mounting orientations LCD is a safer bet. DLP has moving parts and they are engineered to work properly witht he projector sitting on a table, or hanging from the ceiling – both cases projecting horizontally. In my case I may need to mount the projector vertically and that means possible issues with the DLP color wheel. Advantage Epson.
- Size and weight – the smaller it is, the easier to mount/position. The Dell wins here but the margin is slim, still, inches count. Advantage Dell.
- Brightness – The Dell is rated at 2500 lumens, the Epson, 2000. Since both meet my requirement of at least 2000 I needed to look a little deeper. Heat is going to be a concern in this box and honestly 2000 lumens might be overkill anyway. In most cases the Epson should run cooler since it should have a less powerful lamp = less power draw = less waste heat. Advantage (conditionally) Epson.
- Connectivity – This is pretty much a wash as all I cared about was a digital video input but since the Epson supports HDMI it does reduce the total number of cables required if I decide to use the projectors speaker (and I probably will), it’s smaller physically than a DVI connector and easier/cheaper to find right-angle connectors for. Advantage Epson.
All in all, the Epson solidly wins. So now I have a projector. There’s only one problem… To get the screen size I want requires 56″ inches of throw and I have a box no more than 30″ tall. Fun.
Tune in next time for the answer of how to fit 56″ of crap in a 30″ box.
With the decision to try and build a multi-touch table comes a lot of planning. Anyone who thinks they can just jump in feet first and be successful will probably find themselves quickly disappointed. It doesn’t mean they won’t be successful, just that the journey will probably be much harder than they anticipated.
With that in mind, I’m taking a bit of time to work out some details before making any major purchases. The first decision I needed to make is what kind of technology I’m going to use as the basis of the table. There are several ways to build a multi-touch surface, all of which can be found over at the NUI Group website. For my table I am choosing LLP, or Laser Light Plane, because of it’s ease of construction (relatively speaking) and it’s ability to provide fast and precise finger tracking.
The parts list for an LLP table is fairly short:
- Lasers – most commonly 780nm 5-25mW
- Surface material – acrylic, Lexan, glass
- Projection material – sits behind the surface material and allows an image to be displayed
- ‘Blob’ detection camera – this device detects the actual “touches” by seeing the scattered IR light from the lasers
- Projector – fairly self-explanatory
- Mirrors – used to extend the throw length of the projector, more on this later…
- Framing and other construction materials – All the junk to tie it together. I’m hoping to use T-slot aluminum extrusions for the core of this.
- A computer to tie it all together and to run the apps!
Some of these items are no-brainers – the lasers for one. The sooner I have those the sooner I can start fabricating mounts for them, figuring out how to power them, and so on. Others, such as the projector, require some careful consideration. Following entries will detail the selection process for these components. Signing off for now.