Step Three: Laminate

After the shitshow that was bending-my-first-arch I knew I needed a serious revamping.

Call the engineers.

I was, and am, very lucky to have two friends who are both talented engineers with lots of applied experience designing and building their own ambitious projects.

Briefly (or not), I’ll introduce them both with a bit of their backstories.

Willie

I’ve already made mention of Willie in a previous post, so we’ll start there. Willie and I worked together at Bike Friday, which is… I won’t even get into it. Suffice it to say, we built bikes for the same company and would, on occasion, proverbially fistbump one another’s personal projects. (His more so than mine). Willie was the engineer on Whale Wars (yes, the Whale Wars). Willie builds amazing things. For instance, this 10 ft tall rideable T-rex tricycle.

Jove

In 2012 or so I wandered into a design/manufacturing business I was curious about (I have a habit of doing that) and befriended one of their engineers (also that). I built bikes and he designed bike trailers, so we had plenty to talk about. On top of the work he did designing trailers for Burley (which coincidentally… nope, nevermind) Jove also designed and built all sorts of tools for making his own knives. He jokingly aspired to one day start his own company: Knives, by Jove. Oh, the British. Right, in case the name wasn’t a giveaway— he’s English. Though, he grew up in Ireland and when I asked if that made him Irish he emphatically declined. Anyway, he is whip-smart and quite talented. These days he’s the lead engineer and production manager for an EV company in Oregon, Arcimoto.

I had both fellas on speed dial for the duration of the project, and both visited the project on at least one occasion each. Jove had laminated his own kayak and so I knew he’d have some helpful insights:

Did you pre-soak the wood? No. Do that. How much glue did you use? Not a lot. Use a lot. Get Tite-Bond III, it’s water soluble until it sets, and then it’s waterproof, I used it on the kayak. Do I need to worry about protecting it from the rain while it sets? No, it might take longer, but honestly, it might even help to do it while it’s raining.

The rest of the process improvements came from a thorough assessment of what went wrong the first time and brainstorming ways to correct for, or mitigate, those errors. For instance, when the test arch split apart, it was (in addition to being dry, with insufficient glue) because too many of the seams were all in one place. The seams needed to be staggered (think brick wall). So I did the calculation based on the decreasing sizes (the inner pass being shorter than the outermost) and figured out where to place the joints to both maximize my useable board footage and minimize overlapping joints. Someone somewhere told me to use a scarf joint. The fuck outta here. Like I even know what that means. I’m building this shit in the front yard with borrowed tools, from a library. Butt joints. That’s what free tools get you.

I laid the boards out in the yard in descending order, labeled and positioned. The arrows were to orient which end went where, which, at first glance, might seem unecessary, but it’s actually pretty useful. Wood comes in a range of quality, so one edge might have a split or damage, or a knot or some other bullshit. Orienting all of the crappy edges to a single side is one way to minimize future work, namely sanding.

Now, get em wet.

With the wood wet, but without glue, I stacked them together in place, clamped them and lifted the whole shebang onto the truck bed. I now knew better than to attempt a “dry” bend— I needed my wood to stay straight and flat for the glue up. This maneuver was exclusively to determine if I correctly envisioned the mechanics of solo lifting a very long, heavy, floppy, and fragile item. Thankfully, I did.

Though one oversight (and this is why you do practice runs) was that I’d put all the clamps on the wrong side. Once lifting the stack onto the bed, the natural next step would be to turn it on its side and then bend it into place. Looking at the setup, you can see that if I switched all the clamps to the opposite side it would allow me to easily rotate it into bend-readiness. Duly noted.

The next improvement I made to the process was to lay out alllllll of the things in place before gluing. Once the first bead of glue is down the clock is ticking, the pressure is on and my adrenaline fight-or-flight brain goes into stress-vision. The myopia of this mode doesn’t align with locating shit that isn’t right in front of my face, and the last time (while the glue was turning bone dry, and I was panicking, and some dum dum was bothering me) nothing was where I needed it to be. It was the exact worst moment to try to hunt down all the disparate pieces needed to bend the arch. Lesson learned. As evidenced by the following:

With everything all in place, the stacking order set, the wood pre-soaked, and the solo lifting operation practiced I was finally ready for glue. Of course, first I had to undo my practice stack. For ease of transport and to minimize the precious time it would take to move it, I did my glue up right next to the truck bed. I used copious amounts of glue, which felt wasteful, but using not enough glue would result in a much larger waste of materials (the wood, primarily) and all of the time put into shaping, preparing, and schlepping the wood.

Glue on, clock ticking, but I was feeling good— I’d made enough improvements that I felt like I’d get it right this time.

Smooth sailing, so far so good, lah-di-dah, then I look back and uh-oh

The boards were separating. Fuck.

The top was the worst by far. Part of this was due to placing the seam for the outermost pass at the very peak (poor choice). Another factor, which I could and did correct for in subsequent arches, was that I’d oriented the DIY wedge clamps (the 2x4s encircling the outside of the arch) in the wrong direction. If you look at the photo below, you can see my mistake. The wedges clamp the arch in place by hammering the wider part of the wedge, and in this setup, that forced the boards apart. Simple fix for the following arches, I would just flip the wedging direction, and voila, the force of clamping would drive the boards together instead of apart.

The last factor in the separation was just a natural result of the bend. It makes sense when you think about it, each layer has to bend at a slightly different radius, travel a slightly different distance, and therefore is moving at a slightly different speed. All of those minor variations mean that there is friction between the layers as they move, not in unison, but in degrees. The water and glue could only do so much to lubricate the movement, and so the friction naturally wanted to catch and pull at the seams.

This was the hardest thing to correct for. The only solution in that moment, but also moving forward, was to go down to the very end, to the foot of the arch, stick a little block up against the offending layer and whack it repeatedly. For this arch I’d gone far enough that I wasn’t getting a whole lot of movement from my whacks. Goddamn it.

I was increasingly ratcheted into an adrenaline-addled state of panic and stress-vision. Thankfully, only one thing was going wrong, instead of everything all at once.

Not so thankfully, almost as if on cue, at the exact worst time another —another— fucking clueless little old lady wanders up and starts asking me lots of questions. Old habits die hard. I still couldn’t bring myself to be brazenly rude and just tell her to fuck off. Instead, I firmly said, this is stressful and I can’t talk to you. She paused, clearly registering that words had been uttered, but either chose to ignore me or was just a fucking moron and she kept going.

“Well, how do you…”

YOU DON’T UNDERSTAND. THIS IS STRESSFUL AND I 👏 CAN’T 👏 TALK 👏 TO 👏 YOU.

She understood that.

And then she did this weird thing, I’ll never forget, she started anxiously picking up my trash. Five or six pieces, clearly panicking. And then she just walked off with it. Presumably in search of a trashcan.

I made up my mind right then and there if this happened a third time that I would just shout NO. And keep shouting no until they fucked the fuck off. Another variation, which still makes me laugh, is that I’d shout “BOO, BOOOOOOOOOO” but very clearly in the voice of the heckler in the Princess Bride. (Stop reading this and go re-watch it. The whole thing. It’s on Disney +). For better or worse, I never got the chance.

There was only so much that my hammering could do. I’d passed the point of no return on all the joints higher up. I got some movement on the ones closer to the legs.

Despite the gaps, I’d made a huge leap in my process and results. It was with immense satisfaction that I removed the clamps and everything stayed put. No springing. No crackling. Staid. Solid. Though quite unsightly, all that glue. Blehk.

While I was wise enough to foresee the massive amounts of glue being an issue for the bed of the truck, and thereby saran wrapping the whole of it, I somehow didn’t put 2 and 2 together and realize it would also need protection on the inner and outer faces of the arch. Oops.

Thankfully the damage was in a spot that would be easy to hide.

And there it is, the first arch. First useable one, anyway.

I continued to work like a madman bending arches. On top of my full time job and a few extra curricular activities I was sleeping maybe 4 hours a night and building in every waking free moment. From the outset I knew that’s what it would take, and yet I continually underestimated just how long each step would ultimately take. That’s kind of building projects go. And it’s not unique to me, it seems a near universal failure of accuracy in projection. Maybe humans are just really bad at predicting the future. Maybe there’s an evolutionary reason for this recurring unfounded optimism. There’s got to be a name for this phenomenon, it’s so common it’s borderline pathological. Ha. Let’s call it that: borderline pathological optimism.

Case in point, bending my next arch by flashlight.

You can see that I flipped the blocking for the wedge clamps around. I also made a plastic wrapped buffer board for the outer edge and and used plastic lined scraps on the inner edge with each c-clamp. I figured that if any plastic got glued to the surface I could just sand it off. Much easier and less potential for damage than having to break off a piece of wood that accidentally got glued on.

With each successive arch I got better results. By the end, the fit up was tight, and the process well grooved.

It was so satisfying to see my collection grow.

By the time I got to the end, thankfully, I had just enough material left over to cobble together a replacement for the test arch. There was no guarantee of this from the outset, in fact, I was more than halfway convinced I’d be stuck with my shitty first attempt as the flagship arch (i.e. the only arch visible on the outside of the structure). I even came up with talking points to justify the crummy test arch— it was a testament to learning and growth, a sign of my humble acceptance of my own flaws, like the intentional imperfections of Navajo rug weavers.

I threw that shit out.

And you can see why. Having achieved something much closer to the level of excellence I’d hoped for I was happy to go with that. It was, after all, going to be THE ONLY ARCH visible to the world. Why the hell would I want the literal shittiest one to represent the project as a whole? I didn’t. And now I wouldn’t have to. Plus I could spare the bullshit talking points. Big sigh of relief.

To illustrate, here they are side-by-side:

And if it wasn’t clear enough, you can really see how stark the difference is with one atop the other.

Working 20 hours a day was starting to take a toll. My hands were going numb. And staying numb for hours at a time. I needed a break.

So I paid a visit to some of my very favorite humans, Kyle, Molly and their daughter Francis. Or Momobobotown, as I refer to them collectively.

They had just bought a ranch and (I think) this was my first chance to see it transformed into their home. After a delicious meal (they are both really good at food) I read Francis some of her favorite books and we played with her toy vegetables. Such a sweetie.

Kyle then gave me a tour of their land, 30 acres or something crazy like that. He told me all about his plans to build it out, get goats and chickens, what they’d plant and where. You know, hippie Oregon farmer stuff.

And then we went to the river and put our naked boddesses in the water. Which, on a hot summer day, felt amazing.