Ethan's MIT Maker Portfolio

Hi! I’m glad you could stop by. I’m Ethan!

This is the beginning of the garage and, for our purposes, the start of my making journey.

Around 2015, I had graduated middle school and built an electric go kart with inspiration from my school’s electric car club. It didn’t actually move, in part because my gear ratio from motor to wheel was too high, resulting in basically no torque. I was also swimming competitively around this time which resulted in me biking in the dark after our 2hr practices were up. You know what goes well with night? LEDs, and that’s how light bike came to be.

Through these early projects, I began to fall in love with making things. It was just so fascinating! So in 2016, my family decided (via persuasion from me) to invest in a used Bridgeport milling machine and South Bend lathe (no longer with us). Little did they know what this would lead to…

To get better at using the mill and lathe, I turned to Youtube for instruction. I watched hundreds and hundreds of making videos on all sorts of projects. In many of these, it seems that the presenters were building or had built their own tools — and among these, belt grinders were quite popular. So, as my first “big” project, I decided to build one too. Drawing on inspiration from all the different videos, I came up with this design. It uses crowned pulleys in the back to keep the belt on the pulleys and a door hinge and spring assembly for fine tracking adjustment and tension. I probably went overboard with the amount of solid aluminum stock, but hey, if I have a mill, I might as well practice.

Funny story about the motor, my friend’s father owned an electronics recycling place and occasionally invited me over to scrounge for parts. Almost everything I’ve made has something from there.

It turns out, I really like making machine tools. After some browsing on Craigslist (where all the machines were coming from), I found this grinder for $50. At that price, I could probably make a profit selling the motor and scrapping the cast iron. But instead, I thought I’d improve it a bit.

The grinder did run, but the motor mount was inconvenient, and, I’ll admit, the paint was bugging me a bit.

I probably could have just slapped a new paint coat on it and been done, but that’s not my style. I took this grinder apart as far as I could go, stripped all the paint, then designed a new stand that placed the motor lower on the machine by rerouting the belt.

And this is what I ended up with! There wasn’t a straight path for a belt from the new motor location to the spindle, so I added some idler pulleys to redirect the belt. To support these, I added some braces connecting the grinder frame directly to the motor mount. Keep in mind that the vast majority of these projects were designed pencil on paper. I only really started CAD with the CNC project and more complex projects after that.

Ah, the Supermax. This is one of, if not the biggest tool repair/modification I’ve done to date. As with the grinder, this was a cheapo craigslist find. The seller said the computer controller (CNC) system didn’t work (ok, no biggie right?). When I got it home, turns out neither did the spindle or the head (nothing spun). The machine base, however, was in near perfect shape (less than 0.002” backlash), which justified fixing the rest.

First step when something doesn’t work, assess the damage and take it apart. Pretty quickly, I realized that someone must have smashed the spindle, while running, into some workpiece, frying the CNC system and breaking the spindle, all in one shot. It just so happened that someone was giving away a similar machine with a presumably functional head but shot base. The perfect match for my machine! Off to the middle of nowhere I go to remove the head in 100 degree heat, then mix and match the parts.

Well, there’s never just one thing wrong. On the first test run with the new spindle, smoke shot out the top of the motor and the machine roared like a jet engine (not good). Luckily, the smoke was just the belt slipping on some loose oil. But the roar? I had no clue.

Eventually, after a second round of disassembly, I realized that during the aforementioned crash, the motor’s keyway and bushings must have been damaged. This was causing a drive pulley to move all over the place, and, coupled with a set of bad bearings, making a ruckus.

I’m no motor repair expert, but I also wasn’t going to spend more than the machine cost getting the motor fixed. After a long session of Youtube, I worked up the courage to weld up the motor shaft, machine it down, cut a new keyway, and make a set of replacement bushings. It actually worked really well, the shaft wasn’t warped, the pulley fit, and the noise was virtually gone.

At this point, the Supermax worked perfectly as a manual mill but I never was able to fix the old computer system. Eventually, I opted to convert the machine to run on Linux CNC (a free CNC driver os) and to switch out the old motors and drivers for newer steppers. This was partially intended to be a test run for designing and setting up a CNC system (for the large CNC, next). While I started out with no clue how a CNC even works, everything turned out alright and the machine moves in accordance to the computer.

And it’s done!!! It’s been almost three years (75% of my time as decent maker) but I’ve managed to take this broken machine and turn it into a fully functional two axis CNC. On the right, you can see some projects and job shop work I’ve done with the mill.

I remember at the very beginning of this project, someone mentioned the phrase “CNC,” only to be met with my confused stare. I couldn’t have fathomed even taking the machine’s head apart, let alone converting it into a, *gasp*, CNC. But if there’s one thing in the world I like doing, it’s designing and making just about anything; and once I embark on a project, I’ll learn everything I can to make that project come true.

Next!

A long time ago, I got a CNC router for Christmas. It worked alright, one in ten times. A few years down the road (present time) I’ve gotten it to work better, but it’s never been well suited for my needs.

At first, I just wanted to modify the router: Add bigger motors, switch to lead screw instead of belt drive, that sort of stuff. But soon enough, I figured that if I’m gonna do that much work, I might as well just build a new machine altogether. Now, a CNC is a lot more complex than most projects and definitely not something that can be designed solely on paper. This prompted me to explore the free CAD software, Fusion 360. I also did a ton of research online about general and specific CNC machine tool design, informing these models. I first made a design with a moving gantry (V1) but realized that material cost would be too high. This led to V2, and V3. V3 is really where I really started thinking about manufacturability (how to join those base castings together, etc).

Originally, I was planning to make the CNC from concrete for its low cost and high dampening properties. As I did more research, I concluded that concrete exhibits low dimensional stability, and that a similar substance utilizing epoxy as the binder (epoxy granite ) is favored in the machine tool industry. I also found a variety of forum posts online about optimal compositions of epoxy granite, but no precise values. This led me to conduct my own mixture tests, using low viscosity epoxy and a mixture of sand, glass nanospheres, and gravel as my aggregates. I also tested how well steel inserts could be cast into epoxy granite blocks (pretty well actually).

I was preparing to make the epoxy granite machine when my friend’s father (where the motor came from) offered me 8000lbs of precision granite surface plates (the holy grail of precision machine tool bases). I couldn’t resist, and had them delivered to the garage. Unfortunately, this means that the machine must be designed around those new plates. Unlike the other designs, I’ve been doing the granite CAD in Solidworks. It ’s quite different from Fusion, but I really like all the different parameters, the file system, and how well it does with larger models.

Of all the projects, I think I’ve learned the most from this one. This is the project that started me with CAD, partly prompted the Supermax CNC portion, and allowed me to realize how much I enjoy the design and researching aspect of making things. My hope is to finish the granite design around winter break, and start construction in the spring.

We will now leave the world of tools behind and instead look towards electric vehicles.

One day in sophomore year, I was sitting in class and thought, “hmm, instead of making a full electric bike, could I make an attachment that converts my bike to electric when installed, and leaves it as a normal bike when not installed?” This was the spark behind the E-bike attachment project. I eventually decided to attach the system to the water bottle holder screws on my bike, add a secondary sprocket to the back well, and run a chain from that sprocket to a motor in the frame.

The design was pencil on paper, and the actual geometry was determined by trial and error (which causes the chain to rub against the frame a bit), but for my first big vehicle project, I’d say not too shabby!

As with the belt grinder, this attachment contains excessive aluminum ( See the theme here?) and can probably survive a crash at 25mph (how fast it goes) unscathed. The motor is from an electric scooter that I was given and the rest of the electronics are off ebay and amazon.

This project was my first big foray into electric vehicles, following the not so successful electric car. Some of the design choices are questionable (5 /16” plate for the body? Really?) But hey, it works. Besides, questionable choices are bound to happen on the path of learning. I think I’m at least a bit more savvy now!

You know, shopping carts are pretty slow… We can change that.

This is V1 of the shopping cart project. I think it’s the first time I welded something ( yikes!) and probably one of the most kludged together things I ’ve made. It did move, but whether that counts as function is pretty dubious. It’s like the electric car I showed at the beginning: Functional? Not really. But necessary to understand where I’m coming from.

And why is the wheel placement like that? I’ve always wanted to be able to ride the electric shopping cart as if I were just shopping (Handlebar steers, stand behind the cart). This is pretty tricky to achieve, as somehow, you have to have a pivot behind the body of the shopping cart, yet still have pressure on the drive wheels to generate drive force. I found a better solution in V2.

I basically shoved the shopping cart in a corner for two years while working on other projects. Until, one of our robotics team mentors said his work (Department of Water Management) had a bunch of broken Lime-E and Bird scooters that had been thrown in the creeks that he could give me. I couldn’t say yes any faster.

The hub motor and rear wheel of each scooter were still functional, leading to the current 5 wheel design of the cart. I’d also gotten a lot better at welding by this point, and decided to weld a whole new frame. While this is a recent project, I decided to go the old pencil and paper route and just design as I go. I also wanted to make the whole system rainproof (something the bike was not) which is the reasoning behind all the electronics boxes.

And it turned out really well! I figured out that I could steer similar to how a bike can be ridden no handed by angling a single front wheel forward. This system however was susceptible to oscillations (caster wobble) and made the cart near uncontrollable at certain speeds. To combat this, I added the bar to the front axle (shown previously) to increase the system’s moment of inertia. This worked to a degree, but still produced oscillations (now at a different speed). I then thought about how large towers combat oscillations with a liquid filled damper. The first bottle of liquid I found was a coke bottle, which, once installed, worked so well that it became the permanent damper.

I’m pretty proud of this project. I don’ t think anyone’s built a _chariot style_ e-shopping cart (although I think the first e-shopping cart is from MIT). And while I didn’t CAD this or do anything precision, it’s definitely a huge step up from my early electric vehicle projects.

Now, we arrive at the “school” projects. This was from an AP Chem extra credit assignment (build a conductivity meter). I could have done something with tin foil and a lightbulb, but I really wanted to try making glass and tungsten probes (relatively inert materials) and see if I could mimic the old style ammeters with a coil. That led to this design. I played around with fusing tungsten inside some glass tubes for the inert probes and with trying out some woodworking techniques on the frame to get that vintage vibe. I also discovered how much fun copper wire is for making little levers and stuff. The meter works well for differentiating between non-conductive, moderately conductive, and uber conductive substances but (expectedly) doesn’t tell the operator too much else. At this point in shop history, I’d gotten pretty competent at making random things; so, this was more of a “let your imagination run wild” project.

The more recent of the school projects, this was for the “build a brain” AP Psych assignment. Again, something simple like play-doh would have worked, but where’ s the fun in that? I really loved working with copper in the conductivity meter, so this time I went for a full copper frame. The sheets on top were hammered to represent each lobe of the brain, while electronics components inside represent ( in shape) each part of the brain. A switchboard at the base controls each part individually, allowing the user to associate a labeled switch with the corresponding brain part. And I couldn’t resist the idea *lightbulb* trope. So, when all the switches are turned on, the light turns on too! (via a relay). Like the conductivity meter, this was a “wild imagination” project. It was a ton of fun and something I couldn’t have dreamed of doing three years prior.

In parallel to my own projects, I was (and still am) a very active member of an FRC robotics team. We are not affiliated with a school, and when I joined four years ago, worked out of a residential one car garage limited to hand tools. Once I had set up shop, gotten and set up more equipment, build tools, and figured out how to run everything, I started inviting the team over to machine. It’s pretty nerve racking, demoing a huge machine tool in front of an audience. But watching everyone ’s smiles as they make their first cuts (as I did years prior) makes the nerves more than worthwhile. Now, I’m (along with the responsibilities of being build captain) in charge of all machining that goes on in the team, whether that be training members how to machine, or organizing 5-7 teenagers into running all sorts of machine tools simultaneously.

Given that my FRC team is also on the small side, I’ve been given the opportunity to design and machine certain assemblies from start to finish. This past year, I designed the main pivot and pivot support for our arm, using custom nylon bushings in an aluminum housing for the bearing element, and pushing for the layered polycarbonate tower as a more compliant way to support the arm (as opposed to a welded structure). I ended up machining the pivot assembly as well due to the relative complexity of the parts.

On the final days of build season, I along with another designer were dissatisfied with the current ball grabber. We bet the team that if we designed and built one that functioned better than the current system, it would be added to the robot. My contribution to that system was the frame and chain guides, using layers of polycarbonate to guide the chain smoothly around the ball, inspired by car engine chain routing. As you can see, our system made it on the robot!

Around summer 2018, I met Professor Furman of the San Jose State Spartan Superway Project in a truly serendipitous fashion. At some point, he mentioned a problem with their track, to which I sat down and offered a potential solution. That solution got me invited to conference weekly with the project and more recently, work with him and other college students over the summer. During this time, I designed and cadded a modular 12th scale track and bogie assembly, intended to be cut on the waterjet and laser cutter. The bogie is designed to be 3D printed while the protruding nubs serve as guides. This, along with the CNC, is where I really dove into solidworks CAD and where I started to dabble in using FEA for my projects. This experience is also my first time using my making knowledge and applying it to a much larger and more significant project. Here is a link to more details on the Superway and my work if you are curious: https://jacqueshariel-ango-spartansuperway.blogspot.com/

When I work, I tend to keep the garage door open . Thus, a decent portion of the neighborhood knows that a machine shop lives in my garage. One day, I got a knock on my door from a neighbor who works at the copper nanoparticle startup, Kuprion inc. Apparently, they needed something machined! Given that I’m a highschooler, I offered 1-2 day turnaround and a minimal cost. Well, it seems they liked my work and, as such, have been asking me to complete new jobs every now and then as they get their copper production setup running. So far, I ’ve made some baffles, modified their reactor lid, designed and built a reagent dispenser, and designed and built a motor adapter. Considering that four years ago, I didn’t know what a mill was, I’d say that designing and machining for money is quite an improvement! And apparently they think my work is pretty decent as well.

Well that sure was a lot of projects, right? Here ’s 16 more! Jk, I had the space so I thought I’d include an honorable mentions list. Most of these are just storage and organization for the garage. At some point, I also made copper coils for my school’ s physics department and built a venus fly trap growing apparatus for my room. I think it just goes to show how much infrastructure is needed to actually get a machine shop to function and how I’ve managed to optimize my garage for maximum space efficiency.

And this is it! The garage as it currently stands. These past four years have been a wild ride. I’ ve taken this garage from an empty room to a full fledged machine shop. I’ve made parts for a startup, worked on a college project, made custom machines, built fun vehicles, gone overboard with school projects, and hosted and mentored my FRC team. I’ve gone from not knowing how a drill press works to designing my own CNC machine tools. I’ve kindled my love of making and taken my projects to the next level. Quite frankly, I’m proud of how far I’ve come. And as the next stage of life approaches, I’m eager to make so much more!

Thank you for reading!!!