Mae Chen

B.S. in Engineering at Harvey Mudd College, expected May 2029. I build and test hardware, from satellite terminals to lunar rovers to competition robots.

About me

I’m an engineering student at Harvey Mudd College in Claremont, California. I like hardware that has to work in the real world: testing it hard, building better ways to measure it, and fixing what breaks. Since 2022 I’ve led student robotics teams, and I’ve interned at NASA Johnson Space Center and Higher Ground. On campus I lead the Vehicle Dynamics subteam of Harvey Mudd Racing and serve on the ASHMC executive board. I also founded a nonprofit that teaches robotics to kids ages 8 to 14, and I speak Mandarin Chinese at a professional working level.

Portfolio

Worm Gear Tensioning

Higher Ground · Mechanical Engineering Intern, Summer 2026

TrackPod pointing accuracy is sensitive to backlash in its elevation and skew axes. Worm gears used to be tensioned by hand (tighten a set screw until you feel resistance), which varied from operator to operator. I developed an automated, closed-loop system that reads StallGuard motor-load data to find the right tensioning endpoint. It combines two motorized gantries, an electrical interface to the TrackPod, stepper controllers, and a Raspberry Pi that guides the operator and logs the data.

Worm-gear tensioning rig
Complete setup: two stepper motors on V-slot slides adjust the elevation and skew worm gears, with a pogo-pin fixture for electrical contact.
CAD of tensioning rig
Solidworks model, including a resin-printed azimuth lock that holds the azimuth stage during tensioning.
Control enclosure
Control enclosure with the Raspberry Pi and Tic stepper controllers, which also guides the operator through the sequence.
Skew StallGuard plot
Skew-axis motor load: a learned baseline and drop threshold trigger the stop, replacing “tighten until resistance” with a repeatable measurement.

Sun Loading Experiments

Higher Ground · Mechanical Engineering Intern, Summer 2026

I ran controlled experiments to characterize how solar irradiance drives TrackPod heat gain, feeding the product’s heat dashboard. That meant building a repeatable test setup, improving temperature and irradiance measurements, and refining test articles to better represent the TrackPod. Interchangeable light-filtering covers created repeatable lower-irradiance conditions, producing data that links irradiance to internal temperature rise.

Acrylic-cover test setup
Early setup with an acrylic cover: several dome samples side by side, refined to control airflow, orientation, and measurement variation.
Final solar test setup
Final setup with ceramic tint: instrumented test articles and fans in a controlled enclosure.
Irradiance vs temperature plot
Projected air temperature and temperature rise versus average solar irradiance (linear fits, R² of 0.85 and 0.93).

Propulsion (Wheel) Motors

NASA Johnson Space Center · Engineering Research Intern, Summer 2025 · Lunar Terrain Vehicle Ground Test Unit (LTV-GTU)

Each of the rover’s four legs has independent suspension and steering, with two motors and two tires per leg. Many stators had not been clamped to the required 130 ft·lbs and slipped inside the housing, the root cause of problems that had lingered for months. With a deadline looming, my team and I rebuilt two motors in under three days. I also rebuilt a brake disk assembly after a broken caliper wire melted a brake spacer, and designed a small 3D-printed piece to hold the spring pins that keep the motor together.

Single-wheel test bed
Single-wheel test bed for calibrating and recommutating propulsion motors.
Stator windings
Leader and follower windings of a propulsion stator.
Melted brake spacer
Melted brake spacer, traced to a broken wire in the brake calipers.

Electrical and Avionics

NASA Johnson Space Center · Engineering Research Intern, Summer 2025

I redesigned the crew interface panel in water-jet-cut, brake-bent aluminum, with components angled toward the driver and passenger and mounting points matching the original isogrid panel and existing floor inserts. I also built a shielded jumper harness for the steering column, and designed 3D-printed backshells that clamp the whole wire bundle so high-gauge wires stop breaking at the contacts, in a form compact enough to fit inside the motor assembly.

Old crew panel
The old panel, with 3D-printed triangles added as an afterthought.
New panel CAD
My redesigned sheet-metal panel.
Wire harness
Jumper harness: stripped insulation, braiding, and foil shielding, then crimped each wire.

Custom Milled Tools

NASA Johnson Space Center · Engineering Research Intern, Summer 2025

Several motor casings had stuck resolver rings at the bottom of a deep bowl, removable only with a tool that mates to the ring without damaging the casing or threads. I learned to write CAM programs in Fusion and run a 3-axis Haas CNC mill, then made five tight-tolerance aluminum tools, each fitted to its component and driven by a ratchet wrench, tuning machine parameters to balance speed, tool health, and accuracy.

Toolpaths
Toolpaths for one face of a tool.
Tool mid-machining
A similar tool partway through machining.
Finished tool
The finished tool, installing and removing the threaded rings easily.

Test Driver Comfort

NASA Johnson Space Center · Engineering Research Intern, Summer 2025

Long-range test runs exposed three issues: drivers sat in the Texas sun for hours, the edgekey display was too dim to read, and there was nowhere to secure water bottles. A retractable sunshade, mounted using existing accessory-bar holes to stay minimally invasive, solved the first two, and printed bottle holders solved the third. A second version will add a deployable arm for more shade.

Driving the LTV-GTU
Driving the LTV-GTU while testing the GPS tracking system.
Rover on rockyard slope
Test drivers taking the rover up a slope in the rockyard.
Sunshade on rover
Retractable sunshade, side view.
Bottle holder
Water bottle holder for test drivers.

Goldfish: Arm and End Effector

Golden Warriors Robotics FRC #8576 · 2025 season

For 2025 I challenged my team to build something unique. My game analysis showed a mid-tier team could win by filling a niche, so I designed intake, arm, and shooter subsystems to launch 16" balls into a net 8 ft up, in an 8-week build with very limited mentorship. I served as project manager, designer, drive coach, and robot operator on a student-led team.

The arm rotates over the shooter to hand off balls, and with just a pivot and one axis of wheels (two degrees of freedom) it picks up from the ground and two elevated heights. No other team had this arm and shooter combination, so the concept was unproven until our first prototype. Computer vision lines up shots, the shooter varies speed with distance, macros sequence the subsystems, and encoders with PID control the arm and intake. The design won the Texas State Championship Creativity Award.

Goldfish robot
Goldfish, our 2025 robot. Photo: Lalitha Chakka, GWR Media.
Goldfish arm
Arm and end effector grabbing an elevated ball.
Goldfish CAD
CAD with the passive-latch winch climber.

Contact

Interested in working together or want to talk engineering? Reach out.

maechen@hmc.eduResume (PDF)LinkedIn