Midterm season has arrived! Along with exams in other courses, we had a project in ENGI 210: build a working mechanical model of a movement from 507 Mechanical Movements. We could start with the 2D drawing we made earlier in the semester, or choose a new movement to model. I was torn for a while, because my original slider crank mechanism was very simple and I wanted to do something related but more interesting. After perusing the site, I drew up two possibilities: 145 and a railway handcar mechanism.

I finally moved forward with the railway handcar because it had more of its motion directly translating between elements, making it more likely to function when built in plywood. This is my first time using gears, so I found Gear Generator quite helpful. I tried a 2″ and 4″ pitch diameter set, and scaled the rest of my model to match that. I already had a nice wheel 2D drawing, so I included that and laser cut an initial low fidelity prototype in cardboard. At this stage, paper straws make great easy-to-cut axles for wheels and gear shafts!

This prototype worked just well enough to show that I needed to design carefully; the mechanism has two inherent “jamming points,” when bar on the gear is parallel to the lever connection, so pushing on the lever cannot apply any torque towards rotating the gear and wheel. My best shot was to give the system enough energy that once turning, momentum would carry it past the sticking points. John and Carlos and I brainstormed with Dr.Wettergreen how I could achieve this, and I decided to make the wheel as heavy as reasonably possible and reduce friction where I could.

Enter the Quarter Inch Steel! Part of the assignment was to make one functional part out of plasma-cut metal, so I figured what better way than to make a hefty wheel for my design? I found a clean piece of scrap in the machine shop with room enough for one wheel, and ran a test cut small square in the corner.

Hmm, so despite my best estimates from the rate samples, 40 in/min was too fast to cut through. I tried 20 in/min and then it worked. I’m now ready to cut my piece, one shot… cue the random breakdown of the plasma cutter. Halfway through my cut, the machine started aiming at one spot continuously and then stopping to blow on it, not moving on. We tried turning the machine off and on, reseting, and finally found the breakaway switch (take the wand off its mount and put it back). Through many rounds of restarting, I learned the “jump to” command to start a cut from the middle. In the process, there was lots of mysterious air and bubbling, and we found that hitting the emergency stop on the machine does nothing (?!?). Apparently there’s a software bug with that? I’m just glad everything’s okay, and I could finally manage to re-jog manually (eek, estimating) and finish my cut.

I really liked working with the steel; it’s a different animal from the 1/16th inch aluminum I used on my diamond. It cam out of the plasma cutter looking like this,  a bit grungy but very satisfying to hold.

I had a blast going at it with the angle grinder – it was night so the sparks flew beautifully as I took off the slag. I was getting eaten alive by mosquitoes in the night, but it was worth it. With the steel, rather than get gashes and scratches as with aluminum, the angle grinder left a smooth, lightly textured surface. I’d thought about sandblasting the wheel, but as soon as I saw the angle grinder’s effect I decided to leave that as its surface treatment. I did the other side the same way and used the tool upright to smooth the perpendicular edge of the wheel as well.

Nick and I figured out how to operate the drill press in the machine shop and cut a hole for the half-inch dowel. I’m glad I did this rather than trying to plasma cut a perfect circle. This way I could calibrate my diameter to 15/32″in order to press fit (more on that later). Here are my three iterations of wheel. Look at the  reflections on that steel! I love how there are facets that reflect light at different angles; it means that when it turns, there’s variation instead of a bland uniformity as it goes around.

Cut to, well, more cutting. The main laser cutter, the Epilog, was tragically out of commission for a day or so there. I managed to cut one piece of baseboard running 2 speed 100 power and doing the cut twice, and it still didn’t go fully through! John learned how to use the Muse in its stead and kindly taught all the other 210 folks who were at the OEDK that evening. I cut my backboards and found them slightly distorted but workable, and by the next day the Epilog was back up and running (feels good to get back to 5 speed 100 power clean cuts!).

I had all the parts; now for the assembly. For the wheel I used a bearing with half inch interior and got a very snug press fit by subtracting 0.007″ from the diameter. I started constructing the backboard and some additional supports, paying attention to order because my backboard (2 0.2″ sheets thick) interlocked with the top sheet of the baseboard.

Before assembling the rest of the system,  I did some prep by coating my parts with a Beeswax and Orange Oil treatment. Shoutout to Liz for suggesting this- I love this stuff! It’s efficient and makes the wood glow, just wiped on and off 20 minutes later. I also put together some support structures so that the shafts wouldn’t wobble and everything would be at a proper layer height from the backboard, using wood glue to fasten these and the base board layers.

As for the wheel, a half inch dowel was too large so I sanded it down and hammered the wheel on. I used two scrap rings to measure how short to make the dowel end and have it fit the bearing width. The dowel was loose in the bearing, so I ended up veeery carefully putting epoxy on the inner surface of the bearing to help hold it.

From there, it was just connecting the pieces with, dowels cut to length and many laser-cut “donuts” of inner diameter 0.25″ and outer diameter 0.6″ covered in epoxy. It was done.

And it worked!!! I’m so happy with the movement; it’s smooth and does flow past the jamming points as it gets up to speed. I like how the mechanism makes the user pay attention to when to push on it: if you just push and pull up and down at a random speed, the wheel will alternate directions. Feel the rhythm, though, and pushing in concert with the gear’s rotation will make the flywheel spin continuously in one direction. Interacting with the lever, you can feel that the system has a satisfying weight to it. I also like that it’s representative of a real-life mechanism with a specific purpose, so it’s like manipulating a miniature piece of a handcar or pump trolley.

 

Cost Analysis: Steel plate 1/4″ would be $19.80 for a 1’x1′ square, and I used about 2 sheets’ worth of 1/4″ birch ply for this project, so $12.64. Beyond that, I just used epoxy, $4.38 for 5 packs, gorilla glue, $5.60 a bottle, and 1/4″ and 1/2″ dowels, $0.78 and $1.68 respectively. The beeswax finish would be $7.98 a bottle, slightly more expensive, but that only brings materials costs to $52.86. Time is another matter. I’m gonna estimate 6 hours of this was active machining time, $180 at $30/hour, and I worked for 25 hours in all, so $480 at $20/hour. The total would be $712.86. Yes, it does feel like 3 or 4 times as big a finished product as the other projects so far.