Will my wooden ZENSAI warp?

 Hardwoods are complicated materials. While being significantly easier to machine compared to aluminum, brass, and other common metals handled in a typical machine shop, it comes with a distinct set of challenges you must tip-toe around to get reliable parts out of them. It will expand and contract in certain directions at specific rates according to moisture in the air, which can be critical in assemblies relying on multiple different components interlocking one another.

Image from https://www.mokumoku.co.jp/

Hardwoods store/supplier not too far from Tokyo

 In the context of custom mechanical keyboard housings, there are 2 ways around this to make sure the board stays functional in various conditions.

 One method, is to kill this material property entirely. There are various ways to treat hardwoods, pre or post machining, so that they are much more dimensionally stable than their untreated counterparts. I'll avoid going in detail as this is not a field I'm familiar with, but this usually involves a chemical, mechanical, or a thermal process to stabilize the material. Although this potentially sacrifices workability and structural integrity, if done correctly, you get a neat material that not only looks pretty, but one you can do dimensional gymnastics with; allowing you to do precise interlocking joints reliably.

 The other method, is to embrace it. Forget dimensional gymnastics and tight tolerances, build a board much in the same way you build furniture; with large prescribed clearances and joints meant to function at various expansion rates. Although this approach puts a lot of weight on the design phase and somewhat complicates the final assembly process, in the end you get parts that can be milled out of less expensive and commonly available stock, doesn't depend on the integrity of a complicated process for long-term reliability, and leaves just a couple more options for the type of finishes you can apply afterwards.

some photos I snuck in at work c:

 The ZENSAI in its 12th iteration, goes by the latter. Not because treating the wood to be dimensionally stable is a worse option, but because I felt that embracing these material properties and integrating them into the design was something I could fully wrap my head around with less hassle. Many might say increasing clearance in a high-end product like this detriments its look and 'premium' feel, but I don't believe that has to be the case, considering the result.

 So, how exactly does the ZENSAI get around these challenges?

 Let’s first go through wood movement. Wood is a hygromorphic material, meaning it will expand and contract in response to changes in humidity. This movement happens at different rates in different directions, moving the least in the direction along the grain. This can also happen unevenly throughout the wood depending on which surfaces are more exposed to surrounding air, which can cause the wood to warp, and in certain overconstrained scenarios, this stress can be enough to cause the wood to crack.

 There are 4 major components in the board; top case, bottom case, external weight, and the plate assembly, including the PCB, switches, and keycaps. Between these components, there are 3 major joints; one for the 2 cases, one for the plate assembly, and one for the weight. Each of these joints have been carefully considered and adjusted to make sure they remain functional under a range of conditions, and each hardwood component has been made sure that they aren’t overly rigid on its own as a means to prevent cracking.

 The joint between the 2 cases is by far the simplest; having a generous amount of clearance on all sides, with slotted screw holes securing space in the direction across the grain. This makes sure assembly is always possible even in the worst case scenario where the 2 cases expand and contract in opposite directions.

 Same goes for the joint securing the plate assembly to the top case; leaving a generous amount of clearance on all sides. For visual consistency in the final assembly, you will want to make sure there is even spacing around the entire plate when screwing down, although this isn't critical for functionality. More clearance is left between keycaps and the blockers to make sure there won't be any interference, regardless of how you install the plate.

 The weight is where it gets a little tricky. In my previous ZENSAI 65 i5, the joint for the external weight was done similarly to everything else; with large prescribed clearance on all sides. I've received feedback on this however that it was visually off-putting, which made me reconsider how this was handled entirely. For i12, the weight has a tapered interface with the bottom case, somewhat inspired by machine tool tapers used for CNC mills with pneumatic toolchanger systems. Because the mating surfaces are at a 45 degree angle, any amount of movement in the bottom case across the grain is compensated for by how deep the weight sits in it. There is some prescribed clearance on both ends of the weight, but for the most part, the weight sits without any visible gaps.

 

 Both top and bottom cases were designed to have large cavities in the middle. From what I have observed handling raw stock, when large chunks of wood go through stress from uneven drying, often times it is too rigid to bend itself out, leading to cracks in the end grain. The goal with these cavities is to reduce overall rigidity, so some amount of stress can be compensated for by freely warping and bending instead.

 Provided all the components are properly clamped down with all screws, all mating surfaces should be able to contact correctly for a consistent typing sound. The 3 fixed bubble feet form a flat plane for the 4th adjustable foot to align to, eliminating rocking of the board, even if the bottom of the case is warped and not perfectly flat.

 One could consider these quirks as shortcomings, signs of imperfections on a less reliable material. To some level I agree. As much as I’d love to see more wooden boards on the market, there’s just so much more you can achieve with machined aluminum and brass without stressing over each and every joint. As I have spent an obsessive amount of time iterating on this board, however, it started to grow on me. I found myself enjoying having to tip-toe around these challenges, and it has become a core part of my design language over the years. As imperfect as hardwoods are, there's something beautiful about having to work with these constraints. Stressful at times, but absolutely worth it in the end.

- shoonie