Prototyping

PollEv pinned Q&A

M E 493: Introduction to
Capstone Design
Autumn 2024
Eli Patten, PhD
Topics

> Follow-up
– S2, L5 feedback
> Prototyping
– Studio 6 plan
> Universal and Inclusive design
– Mia Hoffman
> Tech. Comms: Status Updates
– Deliverable 6
> D5, A6, A7
Visualizing the 4 Essentials of Design Thinking by Jasper Liu
Outcomes in studio 2 consistently better
than studio 0
Summary of feedback (by Claude)
top 5 Strengths:
> Engaging studio sessions and hands-on activities
– (12 comments)
> Helpful instructors and teaching assistants
– (9 comments)
> Clear explanations and lecture content
– (6 comments)
> Opportunities for teamwork and collaboration
– (5 comments)
> Practical focus on design skills and processes
– (4 comments)
Summary of feedback (by Claude)
top 5 Areas for Improvement:
> Clarity on assignment due dates and requirements
– (18 comments): see below, increasing brevity
> Streamlining Canvas organization and navigation
– (15 comments): put more RFP related links at top of modules
> Improving the Feedback Fruits system and peer review process
– (11 comments): more clear in description and adding ToDo reminder pages
> More concise and focused lecture content
– (8 comments): will continue working on it =]
> Consistency in grading and rubric clarity
– (7 comments): will continue working on it
Prototyping

> Example from Makani

> Prototype (v)
> Dimensions
> Where to start?
Makani: energy kite
(airborne wind turbine)

> Screenshots from YouTube

Feature Film
– Pulling Power from the Sky:
The Story of Makani
Initially: rapid iteration
using soft-kites + winch
Pivot: rigid wings  higher L/D  more power

9 months to build and integrate w/ motors!

One day later:
Bare minimum version
Build a prototype?
Prototype a question!

> Prototype (verb)

– To run an experiment designed
to answer a question about your
potential solution
> “Don't think of it as failure,
?
think of it as designing
experiments through which
you’re going to learn.”
– Tim Brown, CEO, IDEO
Strategic, iterative
prototyping reduces risk

> What is the next most important

thing to learn?
> What is the easiest and fastest
way to learn it?
 Consider dimensions of testing when
deciding how to prototype
Fidelity Later
prototypes

Scope

Early
prototypes
(iRobot)
Fidelity: how realistic is the test?

> Basic analytical model with many assumptions /

detailed simulation
> Cardboard / 3D printed / inj. mold.
> Realistic form and materials
> Realistic manufacturing methods
> Realistic testing environment (Agogino)
Common early questions…

> “Looks like”

> “Feels like”
> “Works like”
 Early vs. late prototyping
Early-stage prototype Late-stage prototype

Simple Complex
Inexpensive Expensive
Quick Time consuming
Surrogate materials Final materials
Lack of realism Engaging

** 3D printing can be useful, but consider the

time invested and what you get out of it **

Prototyping is highly iterative, with each successive prototype

built to answer questions that arise from the performance
successes and deficiencies of previous versions.

Fibula cutting aid in mandibular reconstruction

Scope:
What aspect of the design is being tested?

Focused Comprehensive

> Start focused

– Lower risk, less rework
– Easier debugging during integration
Scope
FAA

Boeing

Instron
How to prototype to reduce
uncertainty and risk?

> 2023 Capstone Conductive

elastomer
– Airbus – Tactile Nose Piece
> Can you use the change in
resistance or capacitance x2
across a conductive elastomer Conductive
terminal array
to measure load and angle?
Off-angle load
How to prototype to reduce
uncertainty and risk?

Conductive
elastomer

x2
Conductive
terminal array

Off-angle load
Studio 6: Prototype

> Identify focused risks

> Divide out
> Prototype/Test
What “risks” remain in your candy-bowl
preliminary design?

> 1min, by yourself

– write down all the risks you can think of that might be good
questions to ask via focused prototyping in studio 6
> 2min, share with neighbor
– How might you test this quickly?
> Report out
– Bring list with you to S6, share with team
What “risk” to prototype first?

> Core aspects of concept

> Decision that many other depend on
> Area of greatest uncertainty
> Long lead times or slow iteration
> Key, driving assumptions
3D printing – Batch runs

Batch run 1
Batch run 2
Back-up

Mon Tue Wed Thu Fri

10:00 requests requests
AM submitted submitted
Start of prints prints
studio complete complete
priority for ME 493 submitted student-led prints,
requests first come-first serve
3D printing – guides in modules

> 3 hours, 1 platter, per team per week; $4/hr on BOM

Bonus slides…
 Comprehensive Prototypes
Many comprehensive Some comprehensive
prototypes are built. prototypes build (and sold?).

High
Technical or Market Risk

One prototype may be Few or no comprehensive

used for verification. prototypes are built.
Low

Low High
Cost of Comprehensive Prototype
 Prototyping Example:
iRobot PackBot Mobile Robot
Physical vs. Analytical Prototypes
Physical Prototypes Analytical Prototypes
• Tangible approximation of • Mathematical model of the
the product. product.
• Can only exhibit behavior arising
• May exhibit unmodeled
from explicitly modeled
behavior. phenomena. (However,
• Some behavior may be an behavior is not always
artifact of the anticipated.
approximation. • Some behavior may be an
• Often best for artifact of the analytical method.
communication. • Often allow more experimental
freedom than physical models.
 Physical Prototypes

Looks-like model for customer Wheel prototype under load

communication and approval during creep testing

Sand test
Ulrich and Eppinger (4th ed., 2008) Exhibit 12-3
 Analytical Prototypes

3D CAD rendering Finite-element analysis

Dynamic simulation model

Ulrich and Eppinger (4th ed., 2008) Exhibit 12-4
Relevance to Capstone

> Testing to reduce uncertainty and risk

– Start simple, start early
> Use testing plan to develop a design schedule
– milestones
> Designing and building prototypes with clear goal/test
in mind
Testing emphasized in IDEO/HCD
Mindsets

> Creative Confidence

> Make It https://www.designkit.org/mindsets.html
> Learn from Failure
> Empathy
> Embrace Ambiguity
> Optimism
> Iterate, Iterate, Iterate
Integrating testing into schedule and
milestones

> Development is about ‘de-risking’ your design

> Don’t wait to test until the end!
– More than just “final verification and validation”
 Let’s start with some prototyping guidelines
How to construct a prototype Guidelines to maximize the effectiveness of your
concept evaluation and testing efforts:
1. Each prototype addresses a question
– Construct the simplest model possible that addresses key question
• Identify the highest risk (focus on a core function)
– E.g., How can I ensure medication will be administered only into the eye 
Prototype will focus on an enclosure that fits around the dimensions of an
average eye
2. Each prototype should produce measurable outcomes
– Ask yourself, HOW does your prototype answer your question?
– E.g., measure distance between eyelash and prototype surface using calipers,
target safe distance 1.5 cm
3. If you prototyped correctly, you should have an answer (+ / - / ∆)
– Keep an eye out for pitfalls. Here, the team found out with this prototype that
the user cannot keep their eye open!
4. Iterate based on your findings!
 Another prototype iteration
How can we deliver medication when the user has limited manual dexterity?

Alternative solution path

(abandoned after 1 week
to focus on single design)

Fork trigger prototype v1 and v3

Feedback from EIH classmates:
• Grip (7) too narrow, feels unstable
• Grip (7) too short for users with larger hand
• Front (1) too bulky

v1

Works-like prototype v2 v3
 So what do you do with a prototype?

Data collection
1. Demonstrate proof of concept (technical feasibility)

2. Quantitative measurements of design specifications (materials, loads, accuracy)

3. Observe users as they handle and use the prototype, receive feedback from
stakeholders (ergonomic consideration, size vs. feature tradeoffs)

4. Test the device in the real setting (make sure to have experimental controls)

Use the results to define more detailed design requirements (or user requirements)
and technical specifications  iterate
 Let’s start with some prototyping guidelines
How to construct a prototype Guidelines to maximize the effectiveness of your concept evaluation and
testing efforts:
1. Each prototype addresses a question
– Construct the simplest model possible that addresses key question
• Identify the highest risk (focus on a core function)
2. Each prototype should produce measurable outcomes
– Ask yourself, HOW does your prototype answer your question?
3. If you prototyped correctly, you should have an answer (+ / - / ∆)
4. Iterate based on your findings!
5. Tie it back to your need statement and core functions
 Prototyping pitfalls

1. Diving headfirst into your first good idea and not thinking through the assumptions that may later turn
out to be wrong
 Better approach: explore a range of different solutions first (can test quickly in-house)
2. Falling in love with your prototype (the “investment bias”)
 Start cheap and fast: use low-cost, readily-available materials, be prepared to break/discard/throw
away as questions are answered
3. Being discouraged by failed prototype
 Embrace the learning opportunities that lead to new insights (and eventual success)

“I have not failed. I’ve just found 10,000 ways that won’t work.” – Thomas A. Edison
Quick prototyping is about acting before you’ve got the answers,
about taking chances, stumbling a little, but then making it right.
- Tom Kelley