Open Hardware design and prototyping

Date of release: 29-08-2022
Prototyping (or experimental thinking and building of functional products) is perhaps the most relevant skill needed to design (open) hardware successfully. (Some might argue that is a fundamental skill for life

Introduction

15 years, 5127 prototypes is what it took James Dyson to create the DC01, his first upright vacuum cleaner with the patented cyclone technology. This can give you an idea of how much prototyping was needed to change vacuum cleaning technologies and set a new standard. But an open hardware project may well be a series of prototypes or even just one prototype (these are often found in hackaday.io)

What is a prototype

Prototypes are models of a future solution or product that help better define the final design specifications of the product/service, as well as communicate how the final product will be. They are used to test partially or completely ideas and designs and they have essentially an experimenting and testing purpose.

We prototype to learn fast and cheaply. To study and approach the solution iteratively, manage risks, and control failure.

Uses of prototypes

• Product/project concept validation Check that the product/project expectations and core hypothesis are objectively or empirically valid.
• Product/project concept verification Check and demonstrate the technical feasibility of the product.
• Learning:: answering questions about performance, feasibility, and desirability.
• Communication: Demonstration of product for feedback, 3D physical models of style or function.
• Integration Combination of sub-systems into system model, alpha or beta test models. These prototypes tend to be approximations that are closer to a final product.
• Milestones Sprint increment, the release of a new version of a product/project, first testable solution.

Different types of prototypes

Prototypes come in many shapes and sizes, they can be as simple as a piece of cardboard, or as advanced as a ready-to-fly satellite. A prototype is an approximation of the product. The following diagram classifies different types of prototypes using a biodigester as an example. This diagram can help you think about how can you design experiments or prototypes to approach sub-problems or aspects of your solution. A backlog of experiments or tests can be composed of focused prototypes where you develop a critical module of your solution or a functional prototype where you go after a working principle or mechanism.

Prototyping for validation vs verification

In the context of hardware design, prototypes can be useful to:

1. Validate core project /product assumptions. (In the case of open hardware these can be core design specifications that satisfy certain needs or even the purpose of the project itself)
2. Explore technical options/solutions/ideas and verify that they implement correctly the desired design specifications.

At a very high level then we can say that prototyping can be used to validate a value proposition (Is it a good value proposition?) or to verify a value proposition (Is it feasible?). In the first case, prototypes aim to test a project's theory, in the second case they are used to test the project's feasibility through experimentation and engineering.

Validation is the process of checking whether the specification captures the customer's requirements, while verification is the process of checking that the software meets specifications. Read more about it here

Overview of the testing process

The testing process involves mostly planning your experiments and prototypes based on different criteria such as resources, relevance, component dependencies, deadlines, among many more criteria.

1. Extract hypotheses from your project framing process (ideas, concepts, sketches). It might be that a hypothesis requires multiple experiments, even a cycle of prototypes that aim to test this core hypothesis.

2. Prioritize hypotheses. Which are those hypothesis that underly the core framing of your project? Which are the riskier ones, or the ones that has more weight on the overall idea? These are questions that can help you decide how to prioritize them.

3. Ideate and design experiments to help you validate these hypotheses. Before building prototypes you can create a "test card", this is way of documentation including its purpose.

4. Prioritize these experiments based on effectiveness, effort, time, completeness, and price. This can end up being the planning of a sprint or a milestone in your project.

5. Build your experiments and run your tests to answer your hypotheses.

6. Document and incorporate your learnings into your design. This may well end up being also an upgrade of your design specifications, a new version release and a changelog explaining what has changed in your latest desgin version.

Find a balance between quality, time, and budget when organizing and prioritizing your backlog of experiments using the criteria below

Image Not Showing Possible Reasons

• The image file may be corrupted
• The server hosting the image is unavailable
• The image path is incorrect
• The image format is not supported

Learn More →

image retrieved from: https://www.mistywest.com/posts/what-is-the-purpose-of-hardware-prototyping

Getting started

Our set of tools to help you get started with this process are test cards, learning cards, and a progress board.

Design and build your experiments/prototypes using a test card

At this point, you should have selected those ideas, hypotheses, and concepts that are a priority for your project currently. In this step, we will find out how testable these are. At this point the goal is to design prototypes and capture them using the test card as a documentation template..

Some examples of candidates to capture in a test card:

• An architecture or several architecture designs that you want to present to a more senior engineer to get feedback.
• A translation of a schematic into different PCB variations that fit for example a casing you have already selected from a vendor.
• A simulation or several simulations
• It can even be a set of experiments that aim to validate a core hypothesis that combines simulations, cad modeling, 3D printing, etc.

Test your prototypes and capture your learnings

At this point, you have run at least a test and you can reflect back on what have you observed, discovered, or learned through the prototype implementation and testing based on measurements. The goal here is to capture this learning and translate this into new actions (perhaps a next prototype or perhaps a document updating your new design version).

The Learning Card forces you explicitly articulate:

• Which hypotheses you choose to test
• What you observed, discovered, or learned in the field
• What did you deduct from those observations (i.e. the insights you gained from the experiment)
• How you'll act upon this learning (e.g. to improve your business model and value proposition ideas)

The progress board

This tool simply aims to structure your testing process by starting from left to right with your core ideas broken down into prioritized hypotheses (concepts, ideas, sub-problems, sub-solutions). Then we have a backlog of tests that can also be prioritized using the prototyping criteria we have shared in previous sections and ends up with learnings and knowledge that are captured in better design specifications and choices.

Exercises

Goal: The goal of these exercises is to help you set up a simple testing framework with a free, simple and available tool that is widely used by open source communities in software and hardware development.

Hopefully these exercises will help you work further in your project using an agile-like project management approach.

Exercise 1: Create a progress board

We are going to rely mostly on GitHub and markdown to do these exercises. Consider doing all the documentation using markdown in your own notes and then transfer them in github issues and the board so that you keep also those notes reusable.

Instructions:

• Create a GitHub project if you haven't for your project (remember you can use github as one of the different distribution channels for your open hardware documentation)
• Use GitHub project boards to create a progress board. Make an example that people can copy

Exercise 2: Pick up a hypothesis that is core to your project and design several experiments using the test cards

Instructions

• Document your experiments in a local hackmd so that you can reuse it later using the template we are providing below.
• Open an issue and name it based on a test design.
  Read more about how to open an issue in GitHub.
• Add the issue to your project board. Read more about how to do this step

Template

## Test name (this can be the name of the GitHub issue)
duration, deadline (these metadata can be added in GitHub issues)

Step 1: hypothesis
- We believe that .... <here goes your hypothesis>
- Hypothesis importance (score from 1 to 3)

Step 2: Test
- To verify that, we will.... <here goes your experiment design, 
you can also use sketches and images>
- Test cost (from 1 to 3)
- Data reliability (score from 1 to 3)

Step3: Metric
- And measure..... <what is your core >
- Time required 

Step 4: Criteria
We are right if...... <here goes the criteria that need to be 
met for this test to be positive>

Learning documentation

Step 5: Observation
- We observed that ..... 

Step 6: Learning and insights
- From what we learned that ....

Step 7:
- Therefore, we will ... <what you will do after capturing this insight, perhaps new prototypes or different types of tests>

Example: 3D printable hand-powered centrifuge

This exercise uses the 3D -Fuge: A New Hand Powered Centrifuge as an inspiration to demonstrate how you can design a test card and they haven't been tested, therefore the learning part is just an example for the purpose of this lesson.

References

[1] K. T. Ulrich and S. D. Eppinger, Product design and development, 3rd ed. Boston: McGraw-Hill/Irwin, 2004.

[2] “Adding issues and pull requests to a project board - GitHub Enterprise Server 3.4 Docs,” GitHub Docs. https://ghdocs-prod.azurewebsites.net/en/enterprise-server@3.4/issues/organizing-your-work-with-project-boards/tracking-work-with-project-boards/adding-issues-and-pull-requests-to-a-project-board (accessed Aug. 28, 2022).

[3] “Creating an issue,” GitHub Docs. https://ghdocs-prod.azurewebsites.net/en/issues/tracking-your-work-with-issues/creating-an-issue (accessed Aug. 28, 2022).

[4] “3D -Fuge: A New Hand Powered Centrifuge | Lab On The Cheap.” https://www.labonthecheap.com/3d-fuge-a-new-hand-powered-centrifuge/ (accessed Aug. 28, 2022).

[5] “Creating a project (classic),” GitHub Docs. https://ghdocs-prod.azurewebsites.net/en/issues/organizing-your-work-with-project-boards/managing-project-boards/creating-a-project-board (accessed Aug. 28, 2022).

[6] A. Osterwalder, Y. Pigneur, G. Bernarda, and A. Smith, Value proposition design: how to create products and services customers want. Hoboken: John Wiley & Sons, 2014.

[7] “What Is The Purpose Of Hardware Prototyping?” https://www.mistywest.com/posts/what-is-the-purpose-of-hardware-prototyping/ (accessed Aug. 26, 2022).

[8] T. Varma, Agile Product Development: How to Design Innovative Products That Create Customer Value. Berkeley, CA: Apress, 2015.

• Previous Week
• Next Week