Universal Stamp Machine

Inspiration and Concept Sketches

Prototype 1: Cam-&-Shaft

Prototype 2: Cam-&-Shaft 2.0

Driver and Metrics for Success: Prototype 1

Driver Question: What kind of cam will be able to successfully allow the follower to slide along its surface without stopping or moving out of place?

Metrics for success: 

  1. Does the follower move a lot and get pulled out of place by the rotation of the cam? 

  2. Does the follower stay parellel to the cam?

  3. Does the follower stay on top of the cam? 

  4. Does the follower catch on the cam and get stuck? 

Results:

I enjoyed this quick-build prototype, which still required careful planning and sketching. It helped me refine part dimensions, ensure smooth cam-and-follower motion, and realize the spring—not gravity—should reset the stamp

Prototype 3: Toggle Joint Linkage

Ranking/Decision Tree

Final Iteration - 3D Printing and Laser Cutting

Constraining to Same Plane

Final Steps

Results and Reflection

The stamp prints faintly, and after widening the base for my 3D parts, the placement on the card became less consistent. Still, I’m proud of the time, effort, and progress I’ve made. I learned the importance of precise math for 3D-printed dimensions to avoid reprinting or modifications, and of tackling the build step-by-step to stay on track with a complex project. Even if the machine isn’t perfect, I’m proud of what I accomplished and what I learned

Driver and Metrics for Success: Prototype 2

Driver Question: What is the ideal shape/height to width ratio of the cam so that the follower and shaft attached smoothly run against it and does not get stuck? 

Metrics for success: 

  1. Does the shaft get stuck on the cam? 

  2. Can the cam rotate in a smooth motion?

  3. Can I work the lever with one hand? 

  4. Does the follower stay in contact with the cam the entire time? Resu

Results:

Laser cutting a larger cam and adding a longer acrylic follower allowed the rotary motion to push the shaft down as intended. However, since the compression spring could only compress one inch—and my final design required two inches of travel—I realized I’d need a different mechanism. This prototype also gave me valuable insights into creating proper T-slots

Driver and Metrics for Success: Prototype 3

Driver Question: What range of distances can the joints of the bars be from one another so that the rotary motion still allows the bars to move and for the bottom joint to move down 2 inches? 

Metrics for Success:

  1. Does the bottom joint move down 2 inches? 

  2. Is the motion of the joints smooth/functional?

  3. Do the d-shafts supporting the joints stay in place?

  4. Does the spring not over-extend?

Results:

I used the drill press to make holes in the acrylic and an electric saw to quickly cut the walls and base. This prototype worked well, delivering the 2 inches of linear motion I needed. It showed me the importance of keeping joints in the same plane, minimizing horizontal movement, and confirmed that positioning the extension spring higher than the joints for reset motion was a good idea for the final build. I also realized I needed a stabilizer to guide the stamp’s downward movement.

I decided to move forward with my final prototype, switching from the cam-and-shaft model to an extension spring with a toggle-joint mechanism. I had a clearer understanding of how to design and assemble the linkages and how the motion would work, whereas with the cam-and-shaft design, I couldn’t solve how to achieve the required 2-inch stamp travel using a compression spring that could only compress 1 inch.

CAD, Exploded View of Assembly, BOM

For: Stanford University

Class: ME 102 - Foundations of Product Realization - Fall 2022

Length: 4 Weeks

Materials: 3D Printed PLA, Hardware, Acrylic, Extension Spring

Skills: CAD/Fusion 360, Laser Cutting, 3D Printing, Hardware Assembly, Iterative and Rapid Prototyping, Performance Testing

Type: Individual Project

Key Requirements:

  • One integrated device (except stamp & ink pad, which are removable)

  • Rotary input (knob/lever) → linear stamp travel ≥ 2"

  • Uses stock mechanical hardware (shafts, bushings, bearings) and one spring (compression or extension) in linear load

  • Prints consistently in same position on a 3.5" × 2" business card

  • Ink pad contacts only the stamp, stays in receptacle throughout cycle

  • Machine is freestanding, operable one-handed, serviceable, and mechanism is visible

  • Materials: acrylic, plywood, 3D prints

  • Fits within ~10" × 4" × 8"

  • $40 budget

Deliverables:

  1. Fully functioning machine meeting all specs

  2. One-sheet pictorial instructions

  3. Photo essay (20–25 slides) documenting process: sketches, prototyping, CAD, BOM, testing, reflection

  4. In-person final presentation

Context

Throughout the quarter, we learned how to model and sketch objects in CAD, create friction-fit joints in plywood, and design 3D-printed parts that achieve precise friction fits

Martian Banana Beam

Ergonomic Cheese Grater

Shroom Light