In December of 2010, Tron:Legacy, the long awaited sequel to TRON, a hacker favorite movie, was released. I hiked a couple miles from my in-laws house to the theater for an 11am showing. The theater only had a half dozen people in it which was perfect since I did'nt want to deal with fanboys or anyone talking. I had pumped myself up enough to just want to watch it in peace. To be honest, I was most intersted in the soundtrack by Daft Punk.
The various trailers and clips that had been released beforehand hinted at a world with a visual astehtic that closely matched how I see the world. One clip in particular really stood out for me. It was released to show the Daft Punk cameo in the 'End Of Line' club. It featured the character 'Castor', the over the top, David Bowie-esque proprietor of the club. What immediatly caught my eye was his cane. It appeared to be glowing with no decernable source. Was it picking up ambient, was it an effect? I watched the movie with anticipation to find out more than the one minute clip showed.
While the movie could have had a stronger story, the visuals and music did not dissapoint. I also finally got a good look at Castors cane. In short, I wanted one and knew I would have to build it myself.
Immediatly after the movie I began to research the cane. I guestimated from still photo's and the clips available that it was 1-1/2" diameter as this is what fit in my hand nicely in the same manner as Castor's (not having Michael Sheens hand handy for reference, I was building to my scale, not his). A few stills from different angles showed different lighting so it was not a Lightsaber sort of overlay effect and considering the other costume elements in the movie, it was surely a practical effect, meaning they were lighting it up for real on the set. At one point you can see a darker portion in the handle of the cane. This led me to think it was a single piece of bent, clear acrylic that simply had a flashlight inserted in the handle. The internal reflection would direct the light from the handle, down the length of the cane and since it was unfocused, it would 'leak'out along the way, giving it a glow through it's whole length, same principal of internal reflection as fiber optic cable
Some googling found a local acrylic supplier in Edmonton, Johnston Plastics. A few emails were exchanged and the project went on hold through the holidays.
In January I went to see the movie again, this time in IMAX 3D. Well worth it, and this time I got a really good look at the cane. I purchased an 8 foot length (only sold in those lengths) of 1-1/2" Extruded Acrylic rod, which cost about $6/ft for a total of about $50 with tax. I could have gone for a cast acrylic rod, except that cost about $25/ft with the same 8 ft minimum. Cast acrylic is much more suited to machining and likely would have been a better choice, however at that cost and the fact it was my first acrylic project, I went the cheap route.
A key piece of research came in the form of a Youtube video from Disney's D23 Armchair Archivist which was a 10 minute documentary about some of the props from the movie, including the cane. The brilliant part was that they had really good, clear closeups of the prop and it was available in 720p HD. I immediatly ripped it and grabbed some stills.
In hind sight if I had had this video earlier I would have guessed the acrylic was 1" or 1-1/4" but the 1-1/2" was still very suitable, so while not screen accurate, it would still be cool and soild as hell.
To bend the acrylic was decided to be the first step. This was the most crucial and likely to screw up. Since we had 8 ft length, if we screwed up, we could cut it off and try again.
The bend was done with a heat gun and while I only needed 7" of handle, approximatly 12" of acrylic was bent since we were not sure how it would behave and if we'd come up short.
The bend radius was determined to be about 1-1/2" which is almost exactly that of a wine bottle, so thats what we used. The rod was supported above a steel table on wood blocks, draped over a wine bottle. About a half hour with a heat gun was needed to slowly heat the rod all the way through. We tried using tin foil to make a hot box around except the foil would heat up and where it made contact, would melt into the acrylic, not a good thing.
In the end, patience and gravity was all that was needed. The 'dangling' end was held up until it was obvious that it was ready to bend, then it was slowly allowed to bend over the bottle. One problem was that we were heating it with the bottle in place, which ended up heating up and melting the acrylic to the bottle and flattening it out slightly. Also, if you try this, make sure to use the side of the bottle without the label as it has a nasty habit of catching fire and also sticking to the acrylic when you pull it up.
After the bend, we chopped off the excess 5 inches of handle and used this as a test piece. This was a great idea since we could tech techniques beforehand and not screw up our bent piece. More on that later.
Since the screen prop had a 'flat' bottom hole, I assumed they used some sort of mill or CNC machine to mill out the hole for thier light source. Not having access to a mill or CNC big enough to do the job, a drill press was the only option. A 1" forstner bit was used since it leaves a realitivly flat bottom hole, only a small depression in the middle, rather than the cone that a normal helical drill bit would leave.
Measurements against my hand and the prop photo's and video showed the light source to be about 3" deep and about 1" wide. I had a cheap dollar store aluminum LED flashlight that was nearly perfect.
Using a drill press we drilled out our test piece. Slow pressure and lots of water allowed for a very nice hole to be drilled. Too much pressure would cause the acrylic to melt rather than cut, too little would have the same effect. This is why a test piece is so very valuable, it allows you to understand what will happen since acrylic is a very wierd substance to work with.
We tried some other techniques for polishing and sanding to determine what worked best. After this we went and drilled the bent cane, which is harder than it sounds.
Getting the now bent piece square in 6 dimensions to drill a perfectly striaght hole down the center was a royal pain. Since the bend can never be perfect, it took alot of hand finessing and shimming to square up. If we had been able to drill the rod first (at 8 feet long) there was still the danger that during heating for the bend, the hole would deform and the flashlight not fit. If we had put in a place holder to maintain the holes shape, there was the possibility of melting it in place permanently.We drilled it with some difficulty. Having to stop and clean out the drilled out material meant we never got it lined up the same way twice. There is a slight detour of about 1/8" down the length.
Now that things were bent and drilled, it was nessecary to polish all the rough bits from drilling and manhandling.
Acrylic has a really neat property. When it's scratched or rough and opaque, if you run a blow torch over it, it will melt the tiny rough edges and go clear. This technique meant that if you use a very fine sandpaper and get the surface smooth to the touch, you can torch it and it will go clear almost instantly. This meant alot of hand sanding of all the machined edges. Many, many hours of sanding.
To sand the interior of the hole was tricky since it was hard to put pressure in the right place to sand. A dremel tool on an extension (like a dentist drill) with various sanding attachments allowed me to get down into the hole and sand the imperfections. Some of the difficulties in drilling meant the imperfections were deep and took alot of work to remove.
Getting the flame of the torch down to the bottom was another difficulty. Since the flame had no exit, while you were melting the bottom, you would melt everything along the way as the heat escaped. This meant you had the very real chance of overheating and melting your creation. For this I used a hot air station with the temperature tweaked to the lowest that would 'clear' the acrylic and a piece of copper pipe to direct the heat where I wanted it. The trick with this is to not touch the edges or else the hot pipe will melt the edges you are trying to repair.
As mentioned, my light source of choice was an aluminum, 6 LED flightlight from a dollar store. It was about 1" wide, 3" deep except for some grip ridges along it length. Mattf at the Edmonton Hackerspace was kind enough to run the flashlight through his lathe and mill the ridges off and giving it a diameter of 0.95" and a nice brushed finish. He also milled off the lip at the end where the LED's were to make it fit nicely at the bottom with the LED's butted right up against the bottom of the hole. The flashlight had a push button switch on the bottom and a screw cap for battery access that was left with a but on the knurling on it so as to provide some friction when it was jammed in the handle.
A thin bead of silicone caulking (clear of course) around the very back edge of the flashlight provided a tight fit that would not fall out.
After alot of work, the cane was ready. Transporting it to Defcon was a trick since it was larger than my luggage and I already had 2 carry-on bags. Fun fact: Medical aids like canes and crutches are excluded from the hand luggage requirements and no one wants to take a limping guys cane away, so I just walked it onboard in my hand. I also found out that a guy with a cool cane gets through security much faster instead of the usual treatment I get :)
The crowd at Defcon loved it. This is a crowd that appreciates blinky stuff and considering the darkness of the hallways and parties, the thing glowed like crazy and was hard to miss.
The cane fared pretty well, but there were some lessons learned.
The first was that the thing is heavy. At just over 3-1/2 feet long, it weighs about 3 pounds which is not bad, but can be tiring. Going with a 1" diameter rod instead may lighten this up and be more appropriate for someone needing more than a prop.
The second was that cheap dollar store flashlights dont deal well with high power commercial AAA batteries. One of the LED's blew right after I changed batteries on the second night. While the others stayed lit, it changed the resistance and greatly dimmed the others. This however gives me an excuse to change out the flashlight and build in some high power LED's instead.
The third is that the extruded acrylic has some intersting properties. The bend, over the several months since it was made, slowly developed some cracks on the underside. I believe that these are caused from internal stresses due to uneven cooling and the compression/tension forces generated by the bend. They actually look pretty cool and some of the smaller ones can be torched out. MAny additional ones developed from putting weight on the handle. While the cane still feels very solid (no flex at all) these are just the natural forces working themselves out. Cast acrylic may be stronger for these forces, as could more even heating/cooling such as raising the temerature slowly through the whole thing instead of the direct heat of the heat gun. Something akin to the rigs the professionals use to make displays and such.