I was watching the 2013 Hugo ceremony livestream when I made an offhand joke that I should have a base design in mind, just in case I was ever asked. And that would have been that, except 2 things happened:

1) Spokane won the Worldcon bid. I grew up in Spokane. I went to my first con in Spokane. I'd be a local artist! And it would be the closest Worldcon to Seattle (where I live) in over 50 years, so I was definitely going.

2) A great design idea popped into my head.

I had to make it happen. It was a moral imperative.

See, every year the base of the Hugo Award is different. The rocket is always the same, but the Hugo Committee gets to decide on their own base. For the last ~10 years this has tended to be done through an open competition instead of a direct commission. So if you want to make a base, you just have to apply, usually sometime the previous winter.

That's what I did. And on August 22, 2015, twenty seven Hugos mounted on my base were handed out. Here's how I made them.

(Note: You might want to check the tutorial on making metal gaming dice I did earlier this year, as it covers many of the same processes.)

Step 1: 3D Design

The image I had in mind was a kind of spiky, tessellated... something. Rocket blast, maybe, or the central plateau of Washington state, surrounded by mountains. I wanted to leave it ambiguous.

My first step was experimenting with various 3D design tools. I had a decent amount of CAD experience through my job, but I wanted something a bit more freeform than that. I tried Blender again, which had scared me off in the past. This time I stuck it out through the terrible learning curve and actually ended up kind of liking it. It's certainly very powerful, and having a tool that is also good for rendering scenes has been very useful when I'm applying to public art RFPs.

I included a model of the Hugo rocket in the design, so I could judge the full effect better. I later ended up getting a version of it 3D printed to go with my prototype base. I based the rocket model off the official spec sheet. The fins are a bit too thin compared to the real thing, but the overall dimensions are quite good.

Once I was happy with the design, I needed to find a way to "unfold" it into individual polygons. I had heard of the Japanese papercraft program Pepakura being used by costumers to make armor, so I tried that. It worked -- and it even let me test my design in paper first! I'm glad it did, because this let me refine the design in a very fast and cheap way. Things always look different in real life.

Step 2: Cutting and Bending Steel

Pepakura can export to DXF, so I sent that to a local waterjet provider that I work with occasionally and they cut it out of sheet steel. The edges that needed to be bent were perforated so they would bend along the correct line. I was thinking the final product would be aluminum, with TIG welded seems, so the perforations would be hidden.

The result, plainly put, was a disaster.

The perforations worked for bending, but they were super ugly and quite expensive, as having to start and stop the waterjet so much added a lot of time to the cutting. But worse, unlike when it was done in paper, things wouldn't just pop into shape. All the angles had to be bent exactly right before it would go together. And even then, the sub-assemblies had to be held in exactly the right orientation for welding. Any small errors would quickly accumulate, leaving giant gaps in the shell. Depressed, I went back to the drawing board.

The solution to the ugly perforations was suggested by my waterjet guy. Turns out waterjets can etch materials by only cutting partially through them. We ran a test of this and the results were beautiful!

Next I got a copy of the ideal base 3D printed. This was easy, since I already had the 3D model of it. Once it arrived, I was able to use it as a template for the bending of the pieces. I'd make a bend, compare it to the printed version, then tweak as needed. After a couple iterations of this they were perfect, or at least perfect enough.

Step 3: Welding Jig

That still left the problem of how to hold them for welding. After a lot of thinking, I came up with a moderately clever idea. Once again going back to the digital model, I created a shell around it the thickness of the metal I was using. Then I created a 1/4" shell around that. I cut this into 4 pieces, then added holes in it the diameter of some rare earth magnets I had on hand. These were all sent off to get 3D printed.

The result was a set of magnetic jigs which exactly fit the outside of the base. Individual subassemblies would lock into place, held there by the magnets. Since the jigs were on the outside, this left the inside completely free, making welding there was easy. This meant dropping the idea of using aluminum, but that was fine. The new etched score lines were so crisp and beautiful, I didn't mind at all.

Step 4: Base Plate

Because the bottom of the shell was open, I wanted to mount it on a plate. I decided to do this out of aluminum to get some visual contrast. Aluminum is also a material that nicely bridges both sides of Washington state, representing Kaiser (and thus hydropower) in the east and Boeing in the west.

It was easy to get them waterjet cut out of aluminum, but they needed to bolt on somehow. This meant welding nuts onto the inside of the bases, but how to align them? I ended up making this funky little X-wing shaped tool to hold them in the right place while welding them on.

The plates themselves were only lightly finished, with countersunk holes for the screws and a nice fine-grit sanding of the cut edges.

Step 5: Surface Treatment

The raw mill scale of the steel wasn't a good look, and anyway I needed to guard against rust. I ended up with a fairly simple (but very labor intensive!) process.

First the bases were sandblasted to remove all the mill scale as well as oxidation marks left from welding the underside. Note: Commercial abrasive blasting services are very cheap! Setting up your own sandblaster is very expensive and likely to be far too slow for any real volume of work. Learn from my mistake!

Next I painted a gun bluing solution onto the bases, being very careful not to get it into the seams where the pieces were bent. This took a long time to do right, but it resulted in a wonderful subtle contrast along the edges of the piece which I really loved.

After letting it sit out overnight, I would then scrub them with a wire brush, thoroughly rinse them off, and buff them with a wheel and another wire brush.

Finally I sprayed on 3 coats of lacquer, being extremely careful not to let drips form!

Step 6: Mass Production

And that was it... except once I was selected as the base designer, I had to do it again. 32 times! (Actually I got enough steel cut for 40, just to have spares. I ended up making 37 in total, not counting all the prototypes.)

This took a very long time, basically 3 months when it was all said and done. The bending and the surface treatment were the slowest and most tedious steps.

But it sure was beautiful to see them all sitting there, done!

Step 7: Final Result

It involved a lot of work and a more than a couple sleepless nights, but it all worked out in the end. I had them done over a month ahead of Worldcon, giving me just enough time for an unusual piano project before heading to Spokane.

There was a slight panic at the last minute because I sized the holes wrong on one of the nameplates before sending the file to the laser etching service, but that was easily solved with my dad's drill press. And I missed the masquerade because I was stuck in a hotel room bolting on rockets. But you know what? That was absolutely okay. This is probably the closest I will ever come to winning a Hugo myself, and I loved every minute of it.