You aren’t designing anything from scratch, and if you tried, you ought to be fired. For one, you don’t have sufficient expertise to design from scratch (neither do I). Second, your project likely doesn’t have the time and budget to design from scratch (e.g., building from scratch). But be comforted—you aren’t solving a new mechanical problem, and there are already solutions and hardware to address it. To focus on your problem, you can’t be focused on every atom of the design; you need to outsource expertise. A baker doesn’t make their own flour, a carpenter doesn’t produce their own lumber, and neither does a mechanical engineer alloy their own metal or coil their own spring—they buy it. Read this germane post from a seasoned engineer on the topic of designing a custom ball screw; it may be the most important takeaway from this post.
If you don’t know how similar problems have been solved in the past, or what exists commercially off-the-shelf (COTS) to help, or where to find either, you are going down a lonely and expensive road—frustrating managers and losing project bids in the process. Having a pulse on the existing COTS hardware and a history of how mechanical solutions have met novel problems will give you the framework for efficiently solving your problems. Senior engineers will tell you that this skill is gained via experience, but you can accelerate your “experience timeline” if you seek out experiences and reference material. A great design engineer has a lens through which they view the world, constantly evaluating and cataloguing good and bad designs, as well as hardware and vendors.
This lesson hits close to home as I intimately recall trying to redesign a drill quick-release chuck for a maintenance-of-way customer. After months of design, procurement, testing, and many thousands of dollars, I simply integrated a COTS solution that worked beautifully.
Companies don’t write press releases about wasting months manufacturing something they can readily purchase for cheap or steal the design of. A failure of this kind will be absorbed into a project overrun and attributed to “schedule delays” rather than to a gap in design vocabulary.
Let’s talk about how we can proactively build up our knowledge of great design solutions. The simplest way to build up your catalog of solutions, your mechanical vocabulary, and your list of quality vendors is to ask senior engineers directly for their wisdom. Ask your senior engineers what bookmarks they use the most, which vendors are great, what websites help them design quickly and efficiently, what books they reference, and what hard lessons they learned that are worth sharing.
I did this exercise at an internship by discovering what reference material was widely utilized at the company, procuring my own copy, reviewing it, and bookmarking any pages that my colleagues had bookmarked. It is a great gift to glimpse another engineer’s reference bookshelf. Learn how to quickly review reference material without reading it through.
It’s equally valuable to ask vendors for design input. For the benefit of gaining your business, vendors are usually happy to provide immense design expertise! If the vendor builds custom components, they’ve seen it all. If they sell mass-produced products, they know their inventory and data sheets better than you. I’ve had a vendor design a small system for me that saved me a lot of time and gained them business from my employer. A close relationship with your preferred machinist is worth at least three senior engineers in my book.
A few vendors that I couldn’t recommend more are McMaster-Carr (“… the Google of hardware, and they are every designer’s, engineer’s, and weekend tinkerer’s best friend”. – Justin Ketterer) and Misumi. You ought to stop here and read about design for McMaster-Carr if you aren’t familiar with it. Misumi will send you a paper catalog for free (also stop and request a copy of their Metric Online Product Guide). You can learn A LOT by flipping through their catalog and seeing all their components organized and detailed in context. Getting a paper McMaster catalog is easiest if you’re at a large company, or you can simply buy an older version used online—maybe your employer will let you expense it. In the meantime, you can peruse their respective websites and utilize the Random McMaster Page Chrome extension to find hardware you didn’t know existed.
“I’m convinced that 90% of any reputation I have as a rapid prototyper is due to being good at navigating the McMaster-Carr catalog. That, and developing a good working relationship with my machinist…” – Ian Rust.
On the same caliber of useful references, get your hands on Robert Parmley’s Illustrated Sourcebook of Mechanical Components! Borrow it from your library or an interlibrary loan, get work to expense it, or (last resort) view it digitally here. This book is chock-full of genius ideas, it’s easy to navigate, and it is produced for design engineers (a very rare combination).
Some miscellaneous digital references:
- Broad List
- Methods
- Automation Companies
When you find an interesting or useful website for future design ideation, make sure to bookmark it following the core principles of bookmark organization. This is well and good if you have already found something useful and have decided to catalog it. What are you to do if you have a problem you’re faced with now, but you don’t have a good starting point? In this scenario, you’re starting at a disadvantage, but there are many options.
I wish my first suggestion were more novel, but type your scenario into an LLM in plain language, and see what it suggests. I personally prefer using Perplexity AI due to their Images tab with each answer. Many mechanical engineers are visually oriented and thus will benefit from seeing answers pictorially.
I had a situation where I needed to design a custom sliding window that was mounted on a pair of vertical linear rails, was 2’ tall, and needed to open by sliding upward 1.9’. When I conceptualized the solution, I intended for the door to be balanced in all positions via a COTS gas strut, often used in newer cars’ trunk and hood lids. When I had modeled the design in CAD and had the approximate weight and travel distance, I was stunned to discover that a strut with that travel distance would be way too long for the location I had available. I was drawing a blank on how to pivot the design. Maybe a pulley and an equal mass on a string would work, like an old-school residential window balancer, but I had worries about the aesthetic impact. I typed my situation into Perplexity, and within seconds, AI had my aesthetically friendly solution: a constant-force spring. I was easily able to integrate the spring into the design, and the window actuation worked flawlessly.
LLMs are also a great option when you are visualizing a piece of hardware or mechanism, but you don’t know what it’s called. Describe it in plain language, and you’re likely to find it quickly. I was able to rediscover (I need to bookmark this) what is called a “hirth joint” in two searches on Perplexity.
Reverse image searches can help you discover information about a photograph without knowing how to search with words. For example, I can search Google with the same image of a hirth joint to get a description of the component. AI is also getting excellent at understanding and interpreting images.
If you’re taking photos or screenshots of interesting solutions or hardware, it raises the question of where to put them. For interesting screenshots, I typically put them in a running Google Slides I have called “Design Inspiration.” I have it bookmarked and easily accessible. Here is a recently added slide to this document:
If I’m researching solutions for a known problem, I will create a mood board either in Google Slides or in Excalidraw (an amazing tool).
For interesting photos, I sort them into an album on my phone titled “design inspiration.” Here are two recent additions:
Don’t forget that there are other browsers beyond your preferred browser (Firefox, Bing, Chrome, Yandex, Baidu). Consider utilizing the other major players, both domestic and international, if you need more resources. Additionally, searching in foreign languages can also help (use Google Translate as needed), as not everything is translated, and language could stand between you and finding the inspiration you need.
Expect to find solutions for your problem in industries far-off from your own. I have two laundry chutes that terminate far off from my washing machine in the basement and over a main walking space. I wanted a container to catch clothes that had a large volume but wasn’t in the way when empty. A bungee cargo net was the right solution. Had I only looked for totes or bins, I would have never ended up with a cargo net that meets all my requirements.
I also take a lot of design inspiration from farm equipment, though I have never worked in that industry. The designs are very rugged, built for cyclic loading, very user-friendly and maintenance-friendly, and space-efficient (often needing to reconfigure themselves for road transportation). All of these requirements make excellent design inspiration for my industry. Attending this year’s National Farm Machinery Expo was a very fruitful experience.
So when do you stop gathering inspiration and start designing?
- When you know you’re only procrastinating the design, it’s probably time to get moving.
- When you can explain the tradeoffs between the same few solutions you’re seeing repetitively, you’ve hit the bottom of the well.
- When you have an affordable COTS solution, it’s time to make a purchase.
- When you’ve spent “too much” of your project timeline researching (it depends on the scope of the project, but this could be from 20 to 50% in my experience).
The good news is that proactively gathering inspiration and references will decrease the time that you spend in this phase of the design, accelerating your “experience timeline.”
The goal with these references is not simply to have them but to have a broad understanding of known solutions to problems that form the backbone of your “unique” solution. If you don’t stay organized, or you don’t peruse your references often, or you don’t add to your reference library when you cross something useful, you’re to blame for creative blocks.
Homework:
- Locate a picture of a mechanism or component you’ve seen but don’t know the name of. Reverse image search the photo until you know the name and manufacturer. Bookmark the page and/or add it to a mood board or inspiration slide deck.
- Browse McMaster aimlessly for 20 minutes, navigating by category only (or the random page Chrome extension). Peruse a category you’ve never ordered from and learn what COTS products exist. Do this monthly.
- Cold-call a vendor that sells hardware relevant to your current project. Describe your project and ask for their recommendations.
- When you sit down to design a new mechanism, first write a few sentences describing the requirements and system overview. Paste the requirements into an LLM and ask it how to solve the problem with COTS components. Compare the results with your solution. Document your knowledge blind spots.
- Select an industry distinct from your own, and find a mechanism that solves a problem in your industry. Note how you would have to modify the solution to adapt it to your unique requirements.
If this topic seems overwhelming, rest assured, these skills will become habitual with practice. Focus on a few items at a time and give yourself grace as you’re learning.
In upcoming posts, we’ll discuss how to brainstorm and select a winning solution!
Sign up for email updates to be notified when other posts on ME communication are published!
