Formula 1 Race Track | ENGI 210: Prototyping and Fabrication

I chose mechanical movement number 55, which I used to create a racetrack moving a Formula 1 car. The user will be able to turn a wheel attached to a gear (gear B) which will cause that gear to rotate and then in turn cause the track to rotate (gear C), moving the car around the track.

Brainstorming

By brainstorming and sketching my ideas to visualize, I determined the best way to hold the mechanism would be a platform in which the gears sit. Instead of attaching them to something and preventing rotation, the gears will freely rotate, confined to the space by the fit of the internal gear in the circular cutout of the platform, with gears A and B attached to dowels that are attached to the platform. To achieve this, the platform will be two pieces of plywood thick, with one having a circular cutout. Below I also fleshed out where the car will be positioned to not interfere with the mechanism.

I clarified the dimensions of the entire piece given the dimensions of my box. The wooden platform would be 10″ x 10″, with the internal gear (gear C) having a pitch diameter of 8″ and an external diameter of 9.25″. Therefore I estimated the circular cutout to have a diameter of 9.55″ to allow for space to rotate. Gear B would have a pitch diameter of 2″ and gear A would have a pitch diameter of 4″ such that they would both mesh with gear C.

I created these gears on gear generator before modeling on OnShape. See first gear C (internal gear),

then gear B,

and finally gear A.

After talking to one of the TAs, I realized a bearing on gear B would be beneficial so that the gear spins along with the dowel as the user turns the gear. This will reduce friction and improve free rotation.

I also sketched out the concept for the steering wheel, which will be the spot at which the user interacts with the device. It will made of metal, attached to the top of the dowel on gear B, allowing the user to turn the track.

After brainstorming and formulating all of my ideas on paper, I moved to OnShape to visualize the assembly and create files that could be exported to be laser cut.

Above is the annotated three-dimensional rendering of the brainstormed sketch for the mechanism. Missing are the car and steering wheel. The car I plan to create in the woodshop from a small block of wood, and the steering wheel will be cut from metal. Below are the inspirational photos for both of those components.

Note the Ferrari is a no-brainer as the most historic and iconic F1 team. Fun fact: Ferrari only began manufacturing automobiles to sell so that they could fund their F1 team.

Prototyping

The only component of the mechanism I was asked to prototype (due to my altered timeline/project) were the gears, as there were concerns regarding the small pitch diameter of gear B.

I downloaded the files for gears A and B as .dxf’s, and then realized I would have to adjust them on Adobe Illustrator. I resized them and set all of the settings to laser cut based on the material (cardboard).

They came out looking a lot cleaner than I expected which was encouraging, as I wasn’t sure what to expect at all. I wasn’t able to cut the internal gear before class that day, but Dr. Wettergreen said this was all he wanted to see and I was good to move forward, as the gear seemed small and fragile on Onshape, but was good in reality.

High-Fidelity Execution

To begin, I imported the .dxf’s of the gears to Adobe Illustrator and added the necessary holes in the center for the dowels in gears A and B and the circle around the exterior of gear C. I cut a 1″ x 1″ test square first to ensure the settings were adequate before cutting all three gears and both layers of the platform.

I used 10 frequency, 100 power and 50 speed. However, I only run the cut three times, so when I lifted the lid I realized the laser didn’t cut the wood all the way through. I tried for a decent amount of time to rehome the machine and rerun the cuts dry, but realized that the chances of me getting the cut perfectly on top of the existing lines were slim to none. So, while frustrated at my waste of material, I recognize this as a beginner’s mistake and learning opportunity. Since then, I have run every cut until I can see the pieces drop to the bed.

I decided to recut the three gears, and placed them strategically to still use the wood I had messed up on. I spent a lot of time positioning gear C but forgot about the laser feature when making dry cuts. So, despite my attempts to jog it perfectly, it ended up cutting over a preexisting line because I was looking at the wrong point to determine where the laser would actually cut. This time though the three gears were actually cut out of the plywood as I was determined not to make the same mistake again, running through the job four times.

On the left above, you can see the line in the bottom right corner of the gear, which is from the first cut of the platform that didn’t fully slice through the wood. In the right photo above, one can see despite flaws, the gears printed out relatively well. It was at this point though I realized on OnShape, I estimated the diameters of the circles inside gears A and B, essentially estimating the diameters of the dowel for A and bearing for B. I had forgotten about this, and therefore went to check if the bearing fit within the hole on gear B, and if there were any dowels that fit within the precut hole for gear A. As expected, the fits were imperfect.

As a result, I was told to measure with a caliper the diameter of the bearing and the diameter of the dowel I chose for gear A, and then cut a bunch of test circles to see which would be the best fit. The diameter of the dowel I chose for gear A was 12.76mm, or .50236″. Considering the kerf I measured for the laser cut box project (on the same laser cutting machine) was .0068″, I added that to the diameter of the dowel (.50236″) and used .50236″ as the diameter of the smallest hole for A, and .50916″ as the largest hole for A. I used even integrals between those two values to cut 5 holes for A. I then repeated that process for the hole for the bearing of gear B. The diameter of the bearing was 21.97mm, or .865″. I added the kerf to that such that the range of diameters would be between .865″ <= x <= .8718″. Below are the calculations I did in my notes app. Note the “larger gear” is gear A, and the “smaller gear” is gear B.

After cutting out five circles for each gear, as pictured below, I stuck the dowel for A and bearing for B inside each respective hole and determined which fit was the tightest and most comfortable. I starred those diameters in my notes app, as pictured above.

I then recut gears A and B, this time with accurately dimensioned diameters of the holes in the center for the dowels and bearings. I used the same settings on the laser cutter, with 10 frequency, 50 speed, and 100 power. I ran through the job 5-6 times (I lost count!) which was successful as you can tell the gears have dropped onto the bed because they were fully cut out of the sheet.

I then did some limited sanding on the two gears to ensure all of the edges were smooth and there were no splinters sticking out.

After, I went to recut both layers of the platform and the internal gear (gear C). In my attempts to save space because of what happened the first two times, I “strategically” placed gear C inside of the square for the bottom layer of the platform. I realized after I cut it out that I needed the entire square for the platform. This cut occurred at 3:24pm and I had arrived to the OEDK around 9:00am, so really it was my fault for not taking a break.

With this cut though I successfully had gear C, and ideally the top layer of the platform as well. However, I realized after cutting this that the dimension for the circular cutout was a guess as to how much space I’d need between the gear and the platform. I had also failed to consider the kerf, and how the laser cutter itself would add some extra wiggle room between the gear and platform. I measured with a caliper the space between the diameter of the circular cutout of the platform and the external diameter of the internal gear. It came out to be 6.71mm, or .264″. I divided this value by four to find the new diameter of the circular cutout, as visually that looked to be a reasonable measurement. The gap divided by four came out to be .066″, which added to 9.25″ (external diameter of gear C) indicates the circular cutout of the platform should be 9.316″. But, to give my mind a rest considering I had begun to make careless errors, I decided to return the day after to execute this change.

It worked out well. There is a visible gap, but this will allow for free rotation while minimizing friction between the gear and platform. I did some limited sanding on all of the pieces exterior’s to get rid of small splinters and such before placing all of the pieces together. This encouraged me by demonstrating I was in fact making progress toward my final mechanism.

Beginning to consider smaller details of my project, I returned to paper to sketch and brainstorm. I measured the distance between the solid areas between the teeth of gear C to determine the dimensions for the platform for the car. I drew the mechanism at half scale to visualize, which also allowed me to determine the dimensions of the nameplate. For the platform of the car I decided on a 1.625″ x .875″ rectangle. For the nameplate, based on the dimensions of everything I drew at 1/2 scale, I decided on a 2.5″ x .833″ rectangle.

I had used wood glue the day before to glue together the two layers for the platform. To hold them together while drying, because I couldn’t find many small clamps, I used tape. But when I removed the tape after 24 hours, it disrupted the wood grain. So, I sanded it with 220 grit sandpaper, as I didn’t want to remove too much material, more smooth out what was lightly torn away from the tape.

One group had mentioned they hadn’t considered aesthetic choices before assembling their project. So, in avoiding that issue, I stained the internal gear black as it would be the race track.

When I returned for this a day later, I noticed some black was rubbing off on my hands. This was strange, as I had vigorously wiped away the stain immediately after applying. I decided to add two light clear coats to protect the stain from rubbing off, as I envisioned the track (gear C) rotating around would potentially be an issue with friction removing the stain.

Moving on, I began production of the car. While a small detail, it turned out to be a process! However, it’s perhaps the most important detail (in my eyes), so was definitely worth it. Based on the dimensions I chose for the car’s platform, I downsized that to choose dimensions for the car itself, deciding on 1.25″ x .625″ max. I went into the woodshop, chose a block of wood that was .625″ wide and 1″ thick, and made a line with pencil at the 1.25″ mark.

Using the miter saw, I pushed my wood up against the bar and brought the saw down (without turning it on) to ensure the wood was lined up correctly to the mark I made. Then, I turned it on and sliced the piece for my car. It went flying because it was so small and the saw was so powerful. Thankfully, I was wearing safety glasses and no one else was in there. But, when I retrieved it, it was a little misshapen (see below, left). I redrew another 1.25″ line and cut the piece again with the miter saw. This time, while imperfections from cutting, the pieces did not break off dramatically (see below, right). I moved forward with this block.

Next, I drew the rough shape of the car on the block with pencil as a marker for post-processing. I used the photo at the beginning that was my inspiration for this shape.

I debated how to best post-process this. There was a decent amount of wood that needed to be removed. However, given the piece flying away after being cut from the miter saw, and me not wanting to chop my fingers off, the miter saw and table saw seemed out of the equation. I also felt like hand sawing would be difficult given the size and the surface area of the clamps, and with the sanding belt machine I was worried about injuring my fingers also due to the size. I ended up deciding to use a wood file to remove material which worked well.

I clamped the piece to the side clamp at the woodworking table and used the wood file (pictured above) to file down the block. I unclamped the piece and reclamped it at different angles to remove material in the rough shape of the car. Once I’d gotten it down to the desired point (see below), I moved onto sanding.

I used 60 grit and then 150 grit sandpaper. Following the rough drawing I made as well as the inspiration photo, I sanded the piece down for 1.5 hours to look like an old F1 car. While not necessarily the most realistic, as the proportions are definitely slightly off (should be either longer or narrower — stubby now), I feel like it looks pretty decent. There is a missing chip on the front of the right side (below, left photo), but that should be covered by the wheel.

After this, I decided to make wheels. Foreseeing the necessity for wheels I kept the circles I laser cut to test tolerances for the holes in the gears. The smallest one had a diameter of .5024″. When I held it up to the car, it was far too large. Visually I determined something about half of that size would be ideal for the tires. Since I chose .507″ for the diameter of the dowel hole, I divided that by 2 (= .2535″) and laser cut four circles for wheels.

I ran the cut until I saw the pieces fall onto the bed. When I picked the sheet of wood up, three of the four circles fell underneath the bed because they’re so small. This was a bit unfortunate. I recut the four wheels. I ran four cycles the first time, so I ran three cycles this time so that they were almost all the way cut through but not enough that they’d fall underneath the bed.

Now I had five circles, which I figured would be good in case something negative were to happen to one of them, since they’re so small.

Then, I went to the wet lab to try the waterjet cutter for the first time. I had already created the Adobe Illustrator file for the steering wheel (which took longer than necessary), so I followed the instructions next to the waterjet cutter to clean up the file, resize it, automake the paths, and choose the starting point. However, because of the way I created the steering wheel in Illustrator, for some reason all of the lines weren’t connected. I kept auto-generating the paths and they were cutting out the wrong parts. I spent a while trying to figure out how to adjust lines and move the paths and such and it just was not working. So, I asked a lab tech, who helped me to add two lines to the exterior of the steering wheel and then fix the paths. I continued following the instructions to set up the machine, fill water, clamp the metal down, home the machine, zero the path, and then ran a dry cut. It seemed like it was going to work, so I ran it with the lid shut. Thankfully, it actually worked.

I followed the correct clean-up procedures on the machine (let the water drain, removed clamps, rinsed off bed and surrounding area, etc.) and rinsed off my piece. Then, since I was already in the wet lab, I decided to sandblast it then as well. It turned out pretty nice.

After finishing the metal piece, I spray painted the laser cut wheels and the handcrafted car. I thought about gluing the wheels to the car first, but then realized I didn’t want the wheels to be red — they needed to be black. So, I spray painted the bottom of the car, then spray painted one side of the wheels, then waited about two minutes, then turned the car over and spray painted the rest of it, then turned the wheels over and spray painted the other side. I left those to dry outside for 24 hours. Spray painting the wheels was kind of difficult as they’re so small that they were blowing away when I would attempt to lightly coat them with moving strokes. I had to end up positioning the nozzle vertically right above them to actually coat them.

I waited the hour indicated on the can before adding a clear coat on both the car and the wheels.

While waiting, I returned once again to the laser cutter. I had already prepared my nameplate file with the dimensions I chose. I ran a test cut to test the raster settings. The first test (square on the left) I thought was too dark. I also realized after cutting I wouldn’t want to keep running the job, as that would raster the wood again and I only wanted to vector cut again. So, I recut the test square, adjusting the raster settings (40s to 38s, kept 100p), and added three more square outlines on top of the existing square (square on right). It still felt a little dark. Also, the square didn’t fall onto the bed so I was worried about the nameplate not being fully cut out with only four cycles for the border.

I looked back at the 210 blog site for the laser cut project to see what settings other people used, but noticed they didn’t include their settings. So, I decided for the final cut to use 35s and 100p and a total of 7 border rectangles to ensure the nameplate was fully cut.

Something weird happened here with my font where it didn’t fill in. Notice the intended Adobe Illustrator file.

However, it didn’t look bad enough to the point where I felt the need to recut it. Call it a happy mistake. I also laser cut the platform for the car in this job. To match the aesthetic and the track, I decided to stain the platform black.

I left this to dry for around 8 hours and returned to clear coat it. Because the clear-coat is oil based, I only needed to wait 8 hours rather than 24. I wanted to clear-coat this to avoid the stain rubbing off, like the internal gear did on my hand. I sprayed two light coats and set it to dry for the thirty minutes. I returned an hour later and moved it to my crate.

While it was drying, I moved to the vinyl cutter. I had never used it solo before, so that day was a big learning day for me (waterjet cutter, vinyl cutter). I followed the instructions exactly, unlocking the clamps, positioning the vinyl, locking the clamps, and logging onto the software. I chose an image of the F1 logo, turned it into vectors, and sized it to be around 2.4″ long. The size I determined from my sketch at half scale such that it would fit in the upper right corner of the platform. It ended up being around 2.6″ long, but it worked nicely.

Thankfully, the cut worked the first try. But I then realized I forgot how to deal with the transfer paper. I did some quick research and cut the transfer paper to the size of the logo before using my student ID to really press the vinyl onto the paper so it would transfer without issue. I then stuck it onto the platform where I wanted it and begun to remove the transfer paper.

However, instead of the logo sticking onto the wood, the logo peeled off with the transfer paper and the remnants of the rectangular box stuck to the platform. I decided to then remove the vinyl that didn’t come off (small bits in between the F) from the transfer paper before cutting the transfer paper to match the size of the logo. I then stuck it back onto the platform in the desired spot. I used my student ID again to really press down the sticker.

Attempting to recall the lesson on the vinyl cutter from the beginning of the semester, I felt like the transfer paper was not supposed to stay on the vinyl. However, I determined instead of trying to mess with it I would ask the TAs after break.

Because I didn’t laser cut holes in the platform, I needed to drill holes so that I could attach the dowels to the platform and therefore the gears. I positioned the gears where I wanted them and drew a circle where the dowel should go. I grabbed a drill and used drill bits that matched the size of the dowels. Before using the larger drill bits though, I used a smaller bit to drill a hole in the center of each circle to ensure everything was correctly positioned. Note I secured the platform atop a piece of sacrificial wood. After drilling the small holes, I used the larger drill bits that were the size of the dowels to drill the real holes. I went through both platforms.

I then sanded the inside of the holes so that the dowels fit snugly yet in a manner that they were able to turn. Then, I laser cut another 10″ x 10″ square platform and wood glued it below the platform so that the dowels would just sit atop the platform. That way also I could wood glue one of the dowels to the bottom platform. As demonstrated below, I used clamps for 24 hours to let the wood glue dry. At the same time, I wood glued the nameplate onto the platforms.

When I laser cut the additional wood platform, I laser cut a steering wheel with a hole for the dowel and then superglued it to the metal steering wheel. I also superglued the bearing to the inside of the small gear.

I then realized the gear I thought the user would be spinning was the wrong gear. At the very beginning of the project I was told the user would spin the small gear with the bearing. However, the user should actually be spinning the larger gear in the center of the platform. When the user turns that gear, it will turn the small gear with the bearing, which then turns the internal gear and moves the car around the racetrack. To adjust for this, I laser cut another wooden steering wheel, this time with a correctly sized hole for the dowel (.507″ diameter, was figured out in test circle cuts before for gears), so that the wheel would sit atop the largest gear. I later wood glued the new steering wheel to the existing steering wheel with the metal and laser cut steering wheel with the incorrectly sized dowel hole. I used a clamp to let the wood glue dry in place. It ended up that the new laser cut steering wheel was slightly off of the original pieces, but it wasn’t so extreme that I felt the need to sand it down for hours. My friend said it looked a bit like a shadow, so we’ll stick with that.

I then set the dowels in place and put the gears on the dowels and noticed the smallest gear was too close to the internal gear, to the point it wouldn’t spin. This was frustrating, but I couldn’t just call it a day as the entire purpose of the project was to produce a functioning mechanism. So, I marked and triple checked where the gear should actually be placed, and drilled again. Note below the black circle is the mark for the correctly positioned drilled hole.

I had to redrill holes for both gears. I sanded both of their interiors a lot so that the dowels would fit snugly, yet still be able to rotate.

I then used a handsaw to cut the dowels to the correct height. The larger dowel (larger gear) needed to be long enough that the user could turn the steering wheel yet still view the project, while the smaller dowel (smaller gear with bearing) needed to be just longer than the bearing. I sanded the ends of the dowels to make them flat. I glued the larger gear to its dowel. I also glued a bit of the smaller dowel inside the hole that I drilled too close to the internal gear to cover up that mistake.

Using another bit of the smaller dowel, I wood glued it to the bottom of the car’s platform and then wood glued it to the internal gear. Note I previously wood glued the tires onto the car, and the car onto the platform.

After the large gear had dried on the dowel, I positioned it in its correct hole on the platform. I also glued the bearing on the smaller gear to the smaller dowel that had been glued onto the platform. After painting spray paint onto the ends of my internal gear to mimic the red and white curbs famously on F1 tracks, I was finished with my midterm!