October 19, 2022
For reasons far beyond my comprehension, Google Discover has been promoting my blog article about my love of building arcade sticks. It has been getting a surprising number of hits since mid-February 2022. After noticing this trend, I went over the article one more time myself.
Of course, that immediately gave me the itch of building a new stick.
Around that time, I came across the Seimitsu LSX-NOBI-01-PRO (AKA Bulett (sic) Lever). It’s a signature model designed by Seimitsu with the help of professional Tekken player, Daichi “Nobi” Nakayama. It caught my eye immediately; I really liked that it looked like a gear stick. Since I had never used Seimitsu parts before, I decided to scratch my itch by making an all-Seimitsu Tekken stick.
Since I only make arcade sticks for myself, I haven’t made a ton of them over the years; this would merely be my sixth build. But even though I am still trash at fighting games, I have become a better woodworker. So I tackled this new project with a decent amount of confidence.
Then I made some decisions that proved to be disastrous, and I completely fucked up several steps of the process. Both while building it and as soon as I finished it, I could see my mistakes, clear as day. All I could think of was that I should have been way smarter than this, and that I suck at woodworking.
During all that self-loathing, I realized that almost every one of my builds has suffered from at least one mistake. This is likely because on every new build I try to introduce at least a single novel element: different brand parts, shapes of lever, restrictor gate, button layouts, color and wood finishing techniques. I want each of my sticks to be completely unique and to play differently from one another. Also, up till now, I have always had new tools to work with on every new project. At least a part of these conditions poses new problems and forces you to adopt new solutions to old problems.
Trying new things and staying out of your comfort zone is the prime recipe for making mistakes.
The thing is, though, making mistakes is actually the prime recipe for learning. Screwing things up and trying to figure out how to fix them is a great way of honing your craft.
Mistakes are hard on your self-esteem, and the fear of making them can stop you dead on your tracks, even before you begin. How you face the eventuality of a mistake before it happens and how you handle a mistake after it happens defines the entire process. When you make things for yourself, without any strict deadlines, embracing the possibility of mistakes and throwing yourself in wild experimentation will actually make you love what you’re doing, and that love will permeate everything you make.
I need to keep reminding myself that, until it becomes an integral part of my mentality. I am not there yet. This is why I decided to write this article as an ode to mistakes. I want to go through the ones I’ve made with every build, and underline what they taught me.
This will probably be a long one; I’ve made a bunch of mistakes.
Like I mention on “Why I love building arcade sticks“, my first stick back in May 2009, was made out of an old laptop box. Of course, that cardboard build was more of a proof of concept, and the biggest hurdle then was “padhacking”. Padhacking is soldering cables on an existing gamepad PCB, in order to use it as an interface. I used an old Playstation controller PCB that I bought on ebay. Since my interface was a hacked PCB, i used barrier terminal strips to easily be able to switch cables.
Even in that first prototype, though, what became evident is the importance of the box’s three dimensions.
Its horizontal area defines how comfortable your hands will feel sitting on the upper panel. You want to give your hands enough space to rest on the stick, while minding the overall size and weight. Fitting it comfortably on your lap or finding the space on a desk is tricky if the box is huge. Height also plays a key role, especially if you prefer keeping the arcade stick on your lap; you want your elbows bent at a comfortable angle.
Later that month I made my first stick out of plywood. I used thin plywood for the top, bottom, front and back panels and thicker plywood for the side panels. The only tool I owned then was a drill, so these panels were cut to size at a home improvement store. In fact, I was lucky to find a staff member that was intrigued by my project enough to disregard the store’s safety guidelines; those panels were definitely below the minimum cut size allowed.
I used my human anatomy books from med school to “clamp” the panels and drilled the necessary holes using spade drill bits. I then assembled the box with screws and added all the electronic components… and a handle!
The plywood looked a little dull, so I thought of adding a clear finish to the box. Back then I used my own cables that were merely twisted in place on the button pins; removing all of them to paint the box and redo that process from scratch would have been quite a hassle. So, I convinced myself that I would be careful enough to apply it without getting any of it on the buttons.
But of course it did get on the buttons! And the buttons got sticky!
Back then I had no idea you could safely disassemble an arcade button. So I applied mineral spirits on the buttons with a cotton swab, trying to get the solvent between the plunger and the external housing. It took a while, but I did manage to clean them and reinstate them to their original glory. They work great to this day!
I was happy with the end result, but there were a few things that bothered me. The thin plywood made the stick a little too lightweight, while also giving it the acoustics of a cajon. And I wasn’t thrilled about the screwheads that showed on the top around the lever; I wanted a cleaner look. So I knew that if I ever made another wooden version, I would have to use thicker panels. This way I could install the lever from the inside, screwing it directly on the top panel.
What came after were a couple of gimmicky versions: the iStick and the electric junction housing stick. The first was a fun idea that came to me after taking a closer look of an ipad box. I wondered if it could fit all the different parts and cables and remain solid. The electric junction housing seemed like a quick and easy way to make what almost looked like a commercial stick. Of course, they both felt what they were: cheap. I knew that it was time to move to solid timber.
The first two sticks I made out of solid wood were almost identical; I viewed them as a two-player bundle. They both featured Sanwa levers and Sanwa buttons set on a Vewlix layout. Vewlix is a sit-down arcade machine form Taito that is the japanese standard for HD games. Its button layout is used on a number of commercial arcade sticks as well.
Still, to differentiate the two even slightly, I made one out of fir and the other out of beech.
In the meantime, I had bought a Bosch circular saw, and used a DIY guide to make cuts as straight as possible. No matter how careful I was, that setup was flimsy and my cuts weren’t always perfect. That’s when I discovered the power of wood filler, and when I started dreaming of owning a table saw.
To minimize the number of times that I would reposition the guide to make a new cut, I preplanned everything on paper. But, in doing so, I completely disregarded the wood grain direction. The top and side panels of the fir stick came out with a vertical instead of horizontal grain. This gave the top face of the stick a weird, unusual “flow”, and two lateral stripes of end grain. The beech stick had the “normal” horizontal wood grain direction on the top, but it also had end grain laterally.
Is wood grain direction really a problem?
There was no visible problem right then and there. But, years later, I decided to apply some oil on the fir stick. And those two lateral end grain stripes got really dark, sticking out like two sore thumbs. That’s when I learned about the particularities of the end grain in regards to how much more finish it absorbs.
In bigger projects, grain direction can have repercussions on stability and robustness. You don’t have to worry about that with a small box, so you could experiment and see what looks nice. In my case, it wasn’t a conscious decision. And after applying finish, I realized that it could have looked much better had I taken a different path.
And that is what I would classify as a mistake.
You can’t come back from all mistakes. Sometimes, you just have to start from the beginning, or reset to an intermediate state. In woodworking, this forces you to use new material and produces scrap, that you can hopefully use on future projects. In the case of these two sticks, I can’t remedy the grain direction snafu without completely sawing off their side panels. I can only hope to at least save the front, back and top panels.
What I could remedy was another mistake on the beech stick: I erroneously drilled a much bigger hole for the micro usb cable. I fixed it by cutting out a plug using the hole saw, and gluing it into place. I then made a new hole in the center of that cylinder and widened it with my Dremel.
I started working on a third wooden build when I discovered South Korean brand Crown/Samducksa and its beautiful bat-style levers. At around the same time I found out about the existence of silent Sanwa buttons as well. The end result was already in my head: I wanted a dark warm reddish brown, combined with white parts.
I also wanted it to have a new button layout. My research took me to the “Namco noir layout”. The noir layout is mainly used in Namco machines, and disposes the buttons in a much more ergonomic curve. But then what really caught my attention was a comment on a Reddit discussion “Namco noir vs Sega Astro City”. Astro City is a type of sit-down two player cabinet by Sega. The first comment on that discussion was “Except you’re missing the best one.” The best one was supposedly Sega Astro City player two! Apparently, the player one layout intentionally forces the player to tuck their shoulder, so the machine can accommodate two players. I took a closer look at the two layouts, printed them out and tried placing my fingers on the paper. That user was damn right!
With a clear project in mind, I got to work.
In the meantime, I had bought the entry level Bosch PTS-10 table saw. Doing exact cuts wasn’t an issue anymore! I cut all the pieces – consciously choosing grain orientation – cut the holes for all the parts and glued everything together. I had never built a box this fast.
Still, there was ample space for mistakes on the finishing part of the process. And I sure made them!
Finishing is quite a Pandora’s box, especially if you have to juggle between two different languages to gather your info. In my case, that’s Italian and English. I am still not 100% sure on how to call the product that I used on this stick. And I have the can right in front of me. It says “vernice” in Italian; Google translates that as “paint, stain, varnish” which are three completely different finishing products!
Paint covers up the wood grain completely as it forms a layer on top of the surface. Stain penetrates the wood, coloring it while keeping its grain visible, but offers little to no protection to the elements. Varnish gives a clear protective coat on top of the surface.
Since what I used gave a dark brown color, it was not varnish. Since the wood grain was clearly visible, it would suggest that it is most likely stain. Stain is supposed to offer no protection, yet the can says “offers great protection”! Apparently, there are products with intermediate characteristics as well.
So yeah. Finishes are a doozy!
So I got my can of stain/varnish/whatevs and applied it on my arcade stick with a brush. Through the videos I had watched, all brush applications had looked so smooth and perfect. Mine was gooey, uneven and it left a ton of brush marks. I hoped that the situation would improve with the next coats. While it partially did, the end result had completely covered the wood grain, giving it a dark, glossy look.
This was not what I was trying to achieve and it didn’t really look good at all. I really hated it. Moreover, I used oil before applying the stain and – once again – didn’t remove the buttons before doing so. I thought “hey, oil can’t make buttons sticky!”, but it turns out that it can dye the Sanwa silent buttons! Apparently, the plungers are made of a rubbery material; it’s what makes them silent, and it’s super absorbent. So the oil gave them a slight pinkish hue. Not horrible, not great either.
I had zero patience at that point, so I put that box to the side and made a new one. I also did a few tests with paint, and realized how I could get the result I wanted: by first applying the finish with the brush and then going over it with a roll. What the roll basically did was smooth everything out and remove all excess material.
After three coats I had the result I wanted. And I effing loved it.
I also took the opportunity to add a USB gender changer on the left side of the stick. A USB gender changer has a USB-A port on one side and a USB-B port on the other. It has a cylindrical part which passes through a drilled hole, and a face cover that secures it on the panel with screws. It is basically a way to add a proper, fixed and reliable USB-B port to the device.
While I was very happy with the result of the Korean build, I kept thinking about the previous, discarded box. I was looking at it in my garage, thinking that it was a damn shame. Its structure was legit; I had to make something out of it.
So I got my orbital sander and went to town.
There are probably different and better ways to do this, but this was my approach at that moment. After stripping it completely of all paint, new ideas came to mind. I had some plexiglass scraps lying around, and I thought that I could use a piece as the bottom panel. The entire stick could become thinner by at least 1,5 cm / 0.6 inches, almost the thickness of the current bottom panel. I used my table saw to shorten the box, and my jigsaw to cut the plexiglass to size. Then I found it’s actually easier and more accurate to do that on the table saw as well. I drilled holes on all four corners and the box was reborn.
Only problem was I had no more arcade parts.
This made me dive back in the process of looking for interesting parts to use. Now completely on the “make every stick unique” mentality, I knew I had to try out new brands. When my arcade stick monomania started, I had briefly considered buying one of the smaller Hori EX 2 fight sticks. Reading that the parts on that weren’t great drove me away from that plan. But I now knew that Hori made high grade arcade parts of its own. Furthermore, they were gorgeous, in matte black!
So I put those in the cart and thought of going big and choosing a second brand I’ve never used before. A Happ-style lever from Industrias Lorenzo and classic concave pushbuttons!
I painted the reborn box slate gray and put the Hori matte black parts in it. To this day, I think it’s one of the better looking sticks I’ve made. The plexiglass bottom made me look into cable management and I took better care of that aspect. Still, I was nowhere near the levels of some incredible cabling masters out there.
I even decided to retrofit all my sticks with the USB gender changer.
What about the IL parts then? I didn’t have a box for those. What should the chassis look like? What would the layout be?
Happ parts were and still are the classic bat lever/concave buttons combo you see on American arcade machines. They were once made for Happ Controls by a company named Industrias Lorenzo. IL now sells the same parts through their own brand, while Happ has taken a different path as Suzohapp. It seemed appropriate to look for inspiration on American fighting games. And the undisputed king of those is Mortal Kombat.
This would become an MK arcade stick.
Ever since the first installment, Mortal Kombat arcades had a very unique button layout. Contrary to its Japanese competitors, the game featured a dedicated “block” button. The MK and MKII arcades featured five buttons put in an X layout: high and low punch on the left, high and low kick on the right, and block in the middle. MK3 added a “run” button to the mix, that went lower and on the left of the earlier layout. Mortal Kombat X added two more buttons, “interact” and “switch stance”.
Ultimate Mortal Kombat 3 was the last game of the series to have an official arcade machine. The rest of the series continued on game consoles and PC. But when MK11 came out, the fine people at “I like to make stuff” built an official MK11 arcade machine for the game’s developer, NetherRealm Studios. That project was a great source of inspiration, both in terms of functionality and looks. They incorporated the two new buttons on their layout – albeit using smaller buttons for them. The top panel holding the controls was yellow and it featured the MK dragon in white.
Considering how “interact” and “switch stance” are important gameplay elements, I went on a different direction with the extra buttons. I used full size concave buttons, putting them on the sides and on the same line as the block button. I drew all the buttons closer together and rotated the entire layout a few degrees counterclockwise. This last tweak was clearly inspired by the Astro City P2 layout, in favor of a more ergonomic feel.
Even with a plexi bottom panel, the stick came out huge; the IL lever is a big boy! The concave buttons strongly reminded me of pinball machines, so I put one full-sized white button on each side panel. While these regularly work as Select and Start, they can be mapped as flipper buttons for a pinball game.
Inspired by the look on the “ILTMS” MK11 cabinet, I went for the white MK dragon on a yellow background. However, I tried to accomplish it by using wood finish alone. I applied a yellowish finish on the entire box while masking out the dragon design with a PVC stencil.
Then I spent hours playing MK9 and MKX, falling in love again with a franchise I hadn’t visited for more than twenty years.
Fast forward to the beginning of this year (2022) and the Tekken stick.
Some elements would remain the same as the Hori stick: the box dimensions, the acrylic bottom, and the button layout.
The first novel element of this stick would be its parts, all Seimitsu. I had never considered building a stick with Seimitsu parts before; I thought that they weren’t that different than Sanwa. I couldn’t have been more wrong. When the parts arrived, I realized that, contrary to the convex Sanwa buttons and the concave IL/Happ buttons, Seimitsu buttons are dead flat. And that was super intriguing.
The second novel element of the stick would be its paint scheme. I wanted to paint it red while leaving the kanji characters of Tekken’s logo – 鉄拳 (Iron Fist) – in the wood’s natural color. I tried different Asian character fonts and settled to Mochiy Pop One Regular. It had that “bold pop Japanese neon sign” feel that I really liked.
To achieve this color scheme, I would make an adhesive PVC stencil and use it to mask out the characters. Given the complexity of the characters, I worried that the stencil wouldn’t stick on a top panel featuring nine holes. I figured it would be easier to paint this while the box was glued up, but had no holes yet.
Since there were no red stains at my local home improvement store, I paid a visit to a paint store. I tried explaining what I was looking for: applying a red hue on wood while keeping the wood grain visible. The solution they offered was a mix of clear stain with just a touch of red paint.
Building the box posed no particular challenges. Once done with that, I prepared the elaborate stencils and applied them on the top panel. I then used a paint roller to apply the custom red stain as evenly as possible. Stain penetrates the wood, so in many points it went under the PVC. That completely ruined the clean stencil style I was looking for. It felt like a major not unforeseeable SNAFU. After briefly trying to remove the unwanted infiltrated blotches, I embraced the chaos. I passed diluted stain over the untouched parts of the kanji characters and sprinkled some pure stain, Jackson Pollock style. The end result looked like a bloodied MMA canvas after a busy night of fights. And it felt quite fitting; after all, it’s a fightstick we’re talking about.
I was then ready to drill the necessary holes, but it was quite difficult to do them properly. Mainly, there was no easy way to provide the correct amount of support to prevent wood blowout. I even had to do the side holes completely freehand. The result was suboptimal, to say the least; the interior looked like someone had assaulted it with a router. But that was nothing compared to the other realization that filled me with dread. The hole of the top left button was completely misaligned, in a rather obvious and ridiculously disturbing way.
I pushed forward, finished the box, cut the acrylic bottom panel, applied all the electronics and tested it. And then I put the stick aside, quite upset. I really liked how the stain looked, but the misaligned button hole was in front of me, driving me crazy. Also, the interior of the stick looked terrible with all that wood blowout. Sure, that part wasn’t in front of my eyes, but I knew it was there, and it bothered me. Compared to my previous two builds, this looked like a messy, inferior result. And it felt as if I had somehow forgotten things that I had learned along the road.
I slept on it. As the tide of disappointment receded, my mind started working out possible solutions for everything.
First, the hole. To fix it, I would have to plug it first, just like with my beech stick. I used a hole saw to cut a plug out of a fir piece of scrap I had lying around. I sanded it down to the exact diameter and glued it to the hole. After letting it dry completely, I drilled the hole again, this time in perfect alignment. The pushbutton’s outer ring completely covered the small piece of the plug still visible, and the result was seamless.
Second, in parallel, the blowout. There are more elaborate ways to do this, but I went with fast and nasty: copious amounts of wood filler. Wanting to turn this into another learning experience, I actually tried making my own wood filler. I mixed fir sawdust and glue into a putty, and applied it generously on every blowout crevice. I let it dry and then sanded down any excess. The result wasn’t great in terms of grain and color; it looked like Deadpool’s face. So I decided to hide those inconsistencies pretty much like Deadpool does: covering it in red. This was the first time I painted the interior of the box, and the result was surprisingly good.
As a final touch, I replaced the four pieces of felt at the corners with a full strip of felt. This made a nice soft frame on the stick’s base.
In the end, I believe that the Tekken stick turned out to be my most beautiful build to date.
And, as it usually happens, it has given me a number of ideas on how to improve my previous builds. A couple will receive a few mild touches, while others will likely be heavily modified.
One thing is certain: there will be fresh mistakes to fix.
