Lamp Saga - Tumblr Posts

Chandelier Saga Part 1: Conception

A bit over two months ago, I was contacted by someone with whom I used to work, who now works with The Possible Project, a non-profit youth entrepreneurship outreach program thing. They were having a huge fundraiser gala at the beginning of October, and wanted to commission a big hanging chandelier as a focal point for the banquet. They're getting ready to build themselves a maker space, complete with a laser cutter and other tools, so they wanted the piece to show off rapid fabrication in some way, probably by having the piece made largely of laser-cut wood.

I was given pretty free range, with the understanding that I would come in every week or two and show the kids in the program what I was doing, giving them a view at how one might go about making a big project, and getting input from them about the design. The goal was to be able to have them also help with assembly, if possible.

Since I don't especially want to take up too much vertical space on people's feeds, jump the break for the early stages of my design process:

The first thing I did was to stay up for a while one night when I was already kind of punchy and just spew ideas down onto a page. Most of them were laser-cut wood ideas, though a few of them I was less sure about how to actually make them.

Having done that, I crawled the internet for precedents, and gave my first talk to the kids about what sort of things are possible with the medium. When doing something new, I always think it's a good idea to look around and see what sorts of solutions other people have come up with. It's a way of getting some inspiration going, finding ideas that you can riff on, and also get a sense for what sorts of ideas people haven't much tackled yet. (In fact, a large part of why I started working with leather was that, while working at a laser-cutting shop, I was looking around for materials that lasered well but not many people were using. And I ended up with leather.) Here are some of my favorites. Actually, one sentence is not quite enough.

And then I started sketching, both digitally and physically:

There was a major consideration that steered my thinking at this point: size. They wanted the final piece to be just about as big as could be made while still fitting inside someone's car to travel to its final destination. The thing is, the laser cutters I have access to are only so big--18 by 32 inches, to be precise. So anything large has to be constructed of smaller pieces. At the same time, I was told that the heaviest the riggers would hang something for the event was 75 lbs. That meant the ideas that involved big planes of material were right out.

So I decided to try to make it as a surface built up from small units connected together, making a skin that was light and self-supporting, without needing all the pieces to be hung off of a structural skeleton. This also meant that I shouldn't need to construct large structural ribs up out of smaller parts, which I was glad about. I was thinking about going for something built like one of the concepts I prototyped in paper, but with a more expansive and amorphous shape, and possibly with a second layer of skin inside the outer one.

But, everyone loved liked the idea with the overlapping scales and light emanating out from inside. Including me. I was a bit hesitant because I was pretty sure it would take more careful geometry shenanigans and tweaking than the other ideas, and I had only so much time to work on it. But I went for it.

What followed were quite a few rounds of strenuous modeling in Rhino broken up with cutting and assembling prototypes. But I think that's enough for this post. Stay tuned for Part 2!

Chandelier Saga Part 2: CAD

Picking up the story of making this giant lamp, we get to the hardest, most important part, but also the least photogenic. That would be, the computer modeling and prototyping.

Here's what I was going for in terms of structure:

The scales are made of 1/16" thick plywood, linked to their neighbors by bolts. The struts spanning the widths of the scales are there to pull the ends together and make sure the centers bend upwards to let enough light through.

It turned out to take something like 4 full rebuilds of the computer model before I was happy with it. The root of the problem is related to the fact that it's really hard to model how real-life materials will bend. So every few days for a couple weeks I had to re-draw all the geometry with slight differences, head in to the laser shop to cut the pieces out, and assemble them to see what it would look like.

Hit the jump for a bunch of screen shots of the different steps involved in designing the final piece.

Working in Rhino (my 3D Computer Aided Design software of choice/training/convenience), I started with an ideal originator surface, created by revolving a curve. I dropped grid points along it:

Then I drew out the base shape for the scales with points and lines. The tips of the scales are drawn out in space perpendicular to the grid point they're above. Their distance from the surface is proportional to the width of that scale:

Next, turning that into surfaces and mirroring the scales along their center-line. (Also, there were some steps involved in expanding the edges of the scales to accommodate the bolts.)

Next comes a polar array to get them all in place:

And a cut-away view to show the internal structure that would hold the light bulbs in place:

But then (after testing the floppiness of the previous prototype) I decided I should give it a bit more structure to support certain points of the surface. Figuring out the geometry for that was a pain:

With that done, time to flatten the scales:

...and lay out the sheets that I would be laser-cutting:

Then I just have to cut twelve copies of each of these sheets, plus a half dozen or so more sheets for the internal structure.

And that mostly wraps it up for the CAD part of this story. I'm intentionally glossing over a lot of the geometry headaches, because you don't want to think about it. Trust me.

Next up will be photos of construction! Though I foolishly forgot to take any photos of actually laser-cutting the scales.

I'll leave you with the front view of that prototype I showed at the beginning:

Chandelier Saga Part 3: Assembly

And here's the exciting part--putting everything together!

We have:

Pieces being cut on the laser cutter

Treating all the wood with flame-retardant

Sorting all the pieces back out (I etched a label on every piece so this would be feasible.)

Beginning to construct the skin, expanding radially from the bottom center

Assembling the parts for the support structure, along with getting the bulb sockets in place.

Attaching the substructure into the skin

Testing it out (and replacing the bulb that released the magic smoke)

Moving it upstairs and finishing construction in the living room

Just barely fitting it into the car.

And that was it! I was done! It would be hung by rigging professionals, and it would be beautiful! Photos of it installed will be coming in another post.

Chandelier Saga Part 4: The Big Day

Seeing this installed and lit up was fantastic. It was hung in a big tent for a fundraising gala banquet, and it really held its own in the big space. I don't know who it was who had the idea to scatter the smaller paper lamps around the chandelier, but it was a fantastic idea.

I went in before the event started to take photos, and they had lit it up beautifully with color-shifting spot lights. The effect was really cool. By the time the banquet started, though, they'd toned that down to let it illuminate itself. I did very much enjoy seeing the play of colored shadows across the textured surface, but seeing the light seeping from below the scales, making it glow from within as I'd hoped it would, made all that work worth it.

Everyone seemed to love it, which was awesome. The Possible Project, who had commissioned the design, made a couple of these available on their Kickstarter, and two people ordered one. So there will be a couple more of these made!

Chandelier Saga Part 5: Electric Boogaloo

And I thought I might be done. The big chandelier was finished and hung and beautiful, but I still had more work to do. The organization I designed the chandelier for was launching a kickstarter to raise money for the maker space they were going to build, and they really wanted a small version of the lamp to make available as a backer reward.

Though I was able to reuse most of the 3D modeling techniques I worked out for the big one, this one basically had to be redesigned from the ground up. I adjusted the shape to be more compact, and to look better from the side. The structure of the big one relied on the wood bending a certain amount so the scales could be bolted to each other, but that wouldn't work on a this small a piece, so instead it has a central structure with radial ribs that the scales get hung on.

This was the first iteration for the small size. It looked good, but there were two major things I was unhappy with. The big issue had to do with the attachment of the scales to the ribs. Each scale had a single slot along the center-line that fit onto a tab/hook on the ribs. This setup means that all the scales can rotate a little bit along their center-line, so they kind of rustle and jostle in a breeze. Basically, it doesn't feel super solid. I would need to adjust the attachment in some way. The simpler issue was that the scales weren't letting as much light through from below them as I would like, so I would need to angle them outwards more.