Figured it was time I shared this here...
I've been a huge fan of Valve ever since downloading Half-Life: Uplink from a game demos website back in 1999, then running out to purchase the full game after saving up enough nickles and dimes. Since the release of Half-Life 2, I have been eying the Gravity Gun, and though I have put a lot of research into it I still feel there is too big of a gap between the player-view and world-view models to truly decipher it into a real-world prop and give it the due I think it deserves. I could always compromise on both models in effort to reach the middle ground, but with the graphics in each episode getting better as time went on, I thought it best to wait. Episode 2 had just been released with the Orange Box, so operating under the presumption that Episode 3 would be along in another year, possibly with a newly updated model for the Gravity Gun, I halted the build and put my research aside for another day. That was four years ago, and not a word on Episode 3 or Half-Life 3 since. Fortunately, the Orange Box came packaged with a new obsession: Portal.
It is hard to ignore the similarities between the Aperture Science Hand-held Portal Device (AKA Portal Gun) and the Gravity Gun; if you like one it's likely you enjoy the other. With the world-view and player-view models showing quite a bit of consistency I knew it wouldn't be too difficult finding that middle ground without huge compromises. Volpin's brilliant build came onto the scene a couple of years ago to show us it could be done and it could look fantastic, and I was getting antsy waiting for Episode/Half-Life 3, so late last summer I decided that diving into the Portal Gun seemed like the next best thing.
Diving might be too strong of a word; I certainly took my time. I spent from October till January going through screen shots, ripping models, checking measurements, researching materials, creating sketches, blueprints and planing my attack before I started building. It was importaint to me that if I took on the build I didn't just copy how others did it in the past, but came up with my own method that would hopefully stand out on its own. At Aperture we do all of our science from scratch. No hand holding!
I am pleased with the progress thus far, and hopefully we will see the end of it in the next month.
Over the months I accumulated a binder of sketches, screen shots, and measurements before combining it all into a set of blue prints I had printed out at 1:1 scale to aid with the build.
Lasers Make it Better
Since the Portal Gun is such a precision instrument, I decided to have a lot of the parts laser cut... and because laser cutting is super awesome and makes my life easier. I designed the parts in such a way that they practically snap together tongue-n-grove style. It cost a little more, but the company was kind enough to find ways to offer me discounts.
Since most of how this goes together is in my head, there was room for human error, so some parts did require a little modification to make work in the real world. The center piece shown here had to be chopped down.
Using a 4" OD tube split down the middle to close up the bottom side of the chamber:
Figuring out how to make the chamber head took a bit of thought. The multiple layers created a bit of a challenge.
It started off simple enough with my 4" and 3.25" pipe and a couple of laser cut spacer rings.
It was once I got to this part that I realized I wasn't sure where to go from here.
I thought perhaps I could achieve the desired results by layering on thin sheets of styrene with the vent patterns pre cut into them. I went as far as to have my buddy Adam rebuild it in 3D so I we could convert it to a Pepakura file to be printed, but I had doubts over how well I could clean up the inside walls once in place.
The exorcise was not for nothing, though, as the Pep file created will help me space the vents evenly and in the right angle. I had initially believed all the parallel lines were on a single point perspective and met somewhere in space, but closer inspection reveals the further to the sides they go they follow more of a downward slant.
I went back to the drawing board to figure this out. I thought that if I filled in the entire bottom area with putty that perhaps I could get my machinist friend to carve in the grove as illustrated above, then I could add the walls and fill in the gaps, but there was some debate over the structural integrity of the piece, if the putty could indeed be lathed, and if the piece as a whole could survive the process. In the end we decided not to tempt it, so I went to the wood shop with an idea.
I thought if I matched the angle of the chamber head on my bandsaw, and made some sort of circle cutting rig, then I could cut out the rings and recreate the image sketched out above.
The rig itself is just a 1/8" thick sheet of styrene with a notch for the blade and a screw for a center post. I haven't bothered to see if someone already manufactures a circle cutting rig (I am actually kind of new to power tools like this), but this got the job done well enough. The trick is to drill a hole through your piece then stick it through the center post on the rig and slowly rotate it against the blade.
Here you can see the rings cut out of styrene and wood. The center wood ring slants down to a smaller diameter than the clear pipe, so I had to cut a segment out of the pipe to make it all fit just right.
I had to make a throw away mold for the wood ring. Using ply wasn't the best idea. Through the course of the day I guess the temperature would change enough to cause the layers to expand and contract at all different rates. No matter how smooth it was on one day, it was a wrinkled cracked mess the next.
The last step to achieve my illustration was to add a slight slant towards the inside of the smallest ring, and to smooth it all out.
To make the outer layer I decided to use putty pressed into the void. Release was used as a barrier.
The Pep pattern was used to stencil in the notches on the chamber head.
The putty was divided up on the band saw and glued back into place.
And several rounds of filler primer + sanding later.
With the chamber head done I could now join it with the rest of the chamber.
The shape is pretty simple. Once the frame was together it was wrapped with a layer of plastic sign.
Once it was completely wrapped I slushed some resin around the inside for added strength and then I trimmed the edges.
Testing the rear plate to make sure it all fit together as planned.
I am using Magic Sculpt putty here. It takes longer to harden than Apoxie Sculpt, but I kind of like that since it gives me more time to work.
Testing the rear access cover. I had holes laser-cut to match a both larger and smaller drawer pullers. The smaller one is actually more accurate to the game model but may be uncomfortable to many and more difficult to find at your local hardware store.
Pepakura maker DungBeetle was gracious enough to allow me to use his pep file for this project. If I had to do it all over again, I'd probably work from scratch, but that isn't due in fault to his work but rather my inability to glue paper together correctly and have it keep an even shape.
I had to make braces in effort to even out the paper shells before filling the insides with resin.
I will un-ugly this duckling, I swear!
Braces were also built to hold the shells in the correct position relative to the rest of the device. This allowed me to fill in the areas between the shell and the device with putty to create a contour so that when casts are made the two pieces would slip right into place without having to fudge around with positioning.
Jimbo7 from gbfans.com was kind enough to machine up these parts. Big thanks to him!
At this point I had to focus on another project for a few weeks. I had planned to have the prototype completely built by the Portal 2 launch date, but unfortunately I could not make it.