Design Researcher — 2024 – 2025
Magneto haptic prototyping
Haptic feedback has come to mean one thing: microcontrollers. We spent eight months asking what else it could mean, and built it out of magnets, strings, and printed channels instead.
Role
Design Researcher & Product Designer
Tools
Rhino
Fusion 360
Grasshopper
Blender
3D Printing
Mechanical Design
8 mo
analogue haptics, no motors
20+
Rapid prototype iterations
Ethos — 01
Is a computer-driven buzz the only way an object can answer you?
An eccentric-rotating-mass motor gives every event the same body: unlocking a door and losing a game feel identical. Before microcontrollers made that the default, objects told you things through their own physics — resistance, detent, release. We took that as the starting constraint. No motors, no drivers, no firmware. Whatever feedback we produced had to come from the mechanism itself.
AxLab investigates a mechanism’s possibility space before asking what it is for. The use case is an output of the research, not its brief, which keeps the mechanism strange long enough to be interesting, and makes every prototype an answer to a question rather than a demo of a feature.
We did not find the limit of analogue haptics. We just questioned what we currently know what to ask of it.
Video — 07
Motion Tracking Software
1:10
1:10
Photos
Process — 02
Bottom-up, on purpose
What I tried, in the order I tried it, and where the approach changed.
01
Start from the mechanism
No motors, no drivers, no firmware. Whatever feedback we produced had to come from the mechanism itself — magnets, strings, printed channels.
02
Sweep the variables
String length, magnet size and channel geometry were the dials. Each one changed the character of the tick, and the mechanism had to stay legible whether it was shaken, lifted, rotated or intercepted directly along a set path.
03
Track the motion, find the bottleneck
Custom motion-tracking software cut prototype cycles by 30% and exposed the real constraint. Reading relative displacement in 3D is hard, so the mechanism struggles to know what the user meant. Intent-sensing, not feedback, was the limit.
What I made — 03
Built by hand, end to end
+
CAD modeling
Grasshopper simulation
Rapid prototyping
User testing
Motion tracking software
+
3D printing
Stack
Rhino · Fusion 360 · Grasshopper · 3D Printing · Mechanical Design
Findings — 04
What the work returned
A motor is a single body of answer. Analogue haptics is a vocabulary.
8
months of analogue-only feedback
0
firmware in the feedback loop
40+
individual components over 20 iterations
Learning — 05
We didn't find a use case for magnetic haptics... yet. But...
Mechanism before use case
We let the object stay strange long enough to teach us what it is for. The brief is an output, not a starting constraint. We never found a perfect use case for this technology, but we came a step closer to understanding haptics and magnetic mechanisms.
A branch back to firmware is not a verdict
The intent-sensing problem routed one line of the research back to microcontrollers, where it became Jiawen Yao's PCB smart dumbbell, which delivers auditory and haptic feedback on a completed bicep curl. That is the right answer for a dumbbell. It says nothing about the doorframes, handles, lids and tools the analogue mechanism might still suit.
Next case study
Element Water relaunch & redesign
Harriet Fisher
Made with love.