I don’t mean to brag, but I’m pretty good at screwing stuff up. Heck, even I don’t count on me to get things right on the first try. If I ever write an autobiography one title I’d consider is “Michael’s Big Book of Blunders”. And it would be a little too easy to write in that context!
But in the process of tripping over every obvious obstacle in my path I’ve learned that while failure is painful in the short term it can be highly beneficial for improvement in the long term.
There’s no shame in making mistakes because learning is an iterative process which requires the critical feedback of trial and error testing. If I’ve messed up enough times to populate an entire book then I’m probably an expert at something!
What follows is a list of engineering specific lessons I’ve learned the hard way through first hand miscalculation and errors in judgement. Hopefully you can learn something the easy way by reading about my experiences.
- Research is always step 1. Engineering is the process of making research based data driven decisions, applying creativity when necessary, and then sweating the details. Don’t skip the first part! If you think you’re the first person to face a given problem, you’re probably doing something wrong.
“For every complex problem there is an answer that is clear, simple, and wrong.” -H.L. Mencken (an American journalist)
“If I had eight hours to chop down a tree, I’d spend six hours sharpening my ax.” -Abraham Lincoln (vampire hunter among other things)
- We don’t live in a perfect world. The discrete world your 3D model lives in exists only on your computer. Once you try to bring it to life the universe suddenly conspires to introduce variations in every way imaginable.
Tolerances stack up. Failure rates of sub-assemblies multiply up in the full assembly. Statistical errors. Material & manufacturing variation. Surfaces that should be parallel aren’t. Measurement errors…even our measuring tools aren’t perfect. The more I learn about quality assurance and sources of variation, the less apt I am to make assumptions.
- Good design is the result of an iterative process. After completing the research stage you can generate creative ideas and experiment with them to see what works. Even fairly simple projects benefit from a bit of trial & error, but beware because it will likely take you longer than you expected
My latest design is NOT a toy…
…it is, however, the result of an iterative process.
- There’s always a trade-off.Gains in one area may require sacrifices in another whether its weight vs strength, quality vs cost, or speed vs accuracy. It is impossible to maximize everything! Good design is about compromise among mutually exclusive alternatives in search of a situationally optimal solution.
- Your time is often best spent doing nothing. Take breaks! Even under a tight schedule I still recommend taking some time to separate yourself from your work before you release it. Go get some coffee and let your subconscious take a crack at it for a while, then come back with a fresh perspective. Only then will you be able to see the forest for the trees.
- Do your design iterating early in the process. The effort required to fix a mistake grows exponentially with time after it leaves your desk. The decisions you make early on cast a long shadow, get your design right before anything else. If you foresee a problem, fix it sooner rather than later.
If you change it after shipping to a customer then you’ll need a taller chart.
- Know when to ‘innovate’ vs ‘integrate’. I usually try to build machines by integrating as many off the shelf parts as possible. But on the other hand, constraints can drive creative solutions, but not necessarily better ones. If the design involves Maguvyering the crap out of standardized products then a completely custom innovation might be a better alternative.
- Practice version control & data backup on everything you do. Save a copy of your work in a separate file every time you make a significant change or release something from your control. All part designs should have uniquely identifiable notes to make them traceable to their respective home assembly. Keep your database accurate and up to date!
A useful nomenclature for folder names is the date formatted numerically in year/month/day format. Today is September 20th 2015 so the folder title should be 150920. I use this for all my photos!
- Choose your data storage system carefully. When starting a company or even just a big project be careful not to let your system grow organically. You don’t want to find yourself crippled in the future because your folder system wasn’t structured for growth, large amounts of data, or use by multiple people. Refer back to the very first tip: you are not the first person to need to a data storage system!
- Sanity check your work. Periodically get a second opinion or take a step back to ask yourself if your work even makes logical sense. Use rough simple equations to make sure your more complex calculations have the same order of magnitude. Compare your results with your assumptions and check again.
- ‘Stupid problems to have’ are opportunities in disguise. The most painful problems are often the ones that could have easily been prevented. Poor recordkeeping. Excessive waste. Procedures & rules based on circumstances that no longer exist. Halting progress due to a lack of something cheap. All of these unnecessary situations are frustrating but try see them as the low hanging fruit of improving your productivity.
- Keep your bill of materials structured logically. There is no single correct way to do it, but there are many wrong If structure a BOM poorly and have to use it again for future work you’re gonna to have a bad day.
A well designed modular sand castle can have infinite tesselations.
- Aesthetics are never unimportant. This is obvious for finished goods but even if you are building something for fun or internal use its good practice to apply finishing touches to anything you put your name on. Its human nature to assume that if something looks good it also works good too. If nothing else, a fresh coat of paint can make a world of difference.
And finally, learning things the ‘hard way’ sucks, but thankfully it isn’t the only way. You don’t have to be one with scraped knees to be able to learn from the experience. Truly the most useful lesson is how to learn from other’s mistakes. Hopefully you’ve been able to do a bit of that here.
So what are some of your most memorable lessons learned the hard way?
(Final Note: I am not actually writing a book, the title image was supposed to be funny. I made the 3D book cover in about 10 minutes using http://www.recitethis.com/ & http://boxshot.com/3d-pack/ & my favorite image editor, Microsoft Paint.)