Wiimote Project

I've noticed that there is a lot of discussion on which led to buy.  It seems the issue is that all these IR Pens rely on the reflection from the display surface and so rely on the intensity and focus of the beam being very bright since most surfaces do not reflect IR very well.  I've come up with a simple and cheap solution involving a wad of aluminum foil, a clear plastic tube (like a straw) and some clear tape. Not elegant but very effective (I'll post pics soon, my phone/camera battery is low at the moment). 

Most IR LEDs are plenty bright enough for the wii mote.  The issue is that the LED is pointed in the wrong direction, i.e. away from the wiimote IR sensor .  You can check this by using IR camera monitor feature and pointing your IR pen at the wiimote.  It will see a very bright spot.  Now try again by pointing at different surfaces and see how little IR light gets reflected back.  Some surfaces okay, most not so great.  You can also check this using a digital camera.

My half penny idea is to reflect the light backwards (towards the wiimote) with the aluminum foil and not rely on the display surface to reflect the IR spot.

Cut a small section of the clear tube (about an inch should do) and wad the aluminum foil to a ball that will fit snuggly into one end of the clear tube.  Cap the IR LED with the tube and foil inside .  Tape the tube to your pen body .  Ideally, the aluminum ball should sit a bit off the led so that the light can shine on the bottom hemisphere of the aluminum wad.     The wiimote will pick up the IR light reflected off the aluminum wad like a champ.  Greatly improves the effectiveness of the wii white board setup.

An elegant solution advancing along the same vein would use a clear hard plastic tube with a shiny ball bearing instead of the aluminum wad and the plastic tube would be attached to a base switch that would activate the LED when the plastic tip/ball bearing presses a surface (this switch assembly would be similar, but smaller, than the touch lights if people are familiar with those).  It would integrate a socket for a 5mm LED as part of the switch/plastic cap assembly. 

Charlson,

I have noticed a similar phenomena myself by accident.

I have been experimenting with a home made pen using a weaker LED (only half the forward current of the Vishay LED). In a recent trial using a large screen it bummed out and lost tracking in a big way, there was a series of broken lines when I tried to draw with Smoothboard's annotate features. I turned the pen around and used it backward - result = perfect tracking.

I have used the same pen with a rear projection system, (dodgy job - grease-proof paper behind the sliding door of a lab fume hood). Again - result = perfect tracking.

We should logically be either putting the sensor on the other side of the screen (rear projection) or transmitting our signal in the right direction for the sensor - backward. My feeling is rear projection is the best way to go, as you also remove the shadowing effect of the user standing between the wiimote and IR source.

I too look forward to your pics.

Cheers
DJM

charlson
Interesting, I wonder if you could glue a small piece that just sits on the end, combine this with diffusing the whole LED with sandpaper and you might get a really good result.

benpaddlejones

I saw this post and thought I would try out your idea.  Here's the pen.



[youtube]http://www.youtube.com/watch?v=DPbej0wRKUE[/youtube]

I'm going to try the pen this week at school.  I'll let you know how it works.

So I tried the ball bearing IR pen today at school.  It works well.  The pen didn't pick up that well at times - I had a 1.2 V rechargeable in it.  When I put in a 1.5 alkaline, it seemed to work better.  With the shiny ball bearing to reflect the IR signal backwards, I wonder if it would actually be better to use a TSAL5300 or TSAL6100 - which has a smaller dispersion angle but higher intensity - what do you think?  I'll try the other diodes out.

I've included a brief clip of me writing on my board. 

[youtube]http://www.youtube.com/watch?v=loBdx6xfDjU[/youtube]

This is a great idea but it also greatly complicates the construction. Guys and gals, in the term "ir pen" the word pen is a mis-momer. We are really making ir signalling wands.

I don't think ir "pen" is a misnomer.  It all depends.  With the "tide-to-go" pen and now with this ball bearing pen, I'm really trying to get "pen" functionality.  I want to be able to do more than just click buttons and move windows around.  I would like to be able to write out an entire lesson, or a full solution to a complex math problem without having to fidget with a button, or press insanely hard into the screen.  I would like the writing to feel natural, as if using an actual pen.  I think that both models do a very good job at accomplishing this.