(mechanical) design errors / flaw compilation 11

[Casa250]

Hi all,
there seems to be lots of mechanical elements books on how to design systems, however, I rarely find examples of design errors and what-not-do-dos. I find that often looking at what-not-to-do is even more helpful in avoiding them.
Does anyone has good examples to show here? I am mainly looking for mechanical design flaws / errors but others are ok too.

thanks

 

[JohnRBaker]

I can recommend almost any book by Henry Petroski. He has written several books on engineering failures and how to approach designs with failure issues in mind. Now, he's primarily a structural engineer but many of his principles and ideas play across industries.

This will help you as it includes a list of his books:

https://en.wikipedia.org/wiki/Henry_Petroski

John R. Baker, P.E. (ret)
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

 

[Sparweb]

Aviation is very meticulous about learning from failures:

https://tsb.gc.ca/eng/rapports-reports/aviation/

https://www.ntsb.gov/Pages/home.aspx

A very recent example has been discussed on the forum recently: thread815-498902

https://tsb.gc.ca/eng/enquetes-investigations/aviation/2021/A21W0045/A21W0045.html

This one teaches a lot about metallurgy, quality control, and training of personnel (among other more subtle lessons).

 

[davefitz]

Prior to 1985 , most large companies had the practice of retaining key employees for 30-40 yrs so as to provide a "corporate memory" of past errors, so as to avoid repeating these errors. After the 1985 initiation of portable retirement accounts ( IRA's) and a mandate for a 5 yr minimum vesting in a pension plan, employees became disposable items and there was a loss of "corporate memory". Perhaps a good example of such errors may be the recent Boeing 737 max failure to recognize the obvious- that a single instrument's failure to accurately provide input to a critical contrtol algorithm implies a high probability of in flight disaster.

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

 

[drawoh]

Casa250,

I have a long list of things you should not do, but they apply to optics and electronics. Now that I work for an electronics manufacturer, I am developing a new list of things to not do.

If you set up a folder on your network or up on Sharepoint, can you post Lessons Learned documents? In a professional organization, this works very well. If you are working with office politicians, you could be asking for trouble. Are there some older people willing to write stuff down?

Everybody makes mistakes once. You are an idiot if you keep making them.

--
JHG

 

[JStephen]

Not exactly the field you're looking at, but still of interest- there are a number of Chemical Safety Board reports and videos relating to notable failures available online (videos on Youtube).

 

[CWB1]

I worked quite a few failure analysis contracts for a large automotive consultancy and it reinforced a long-standing opinion of mine - the most common failure by far is simply a lack of due diligence, which is why process is critical to success. Ironically, the need for me to do a thorough (not overly expensive) analysis was often questioned by folks certain their 14th guess would be correct. An Ishikawa, careful study, and a bit of testing later, the root cause was usually laziness in manufacturing or (more often) the design office.

IME the do's and don'ts of a given niche is usually a matter of efficiency and attractive design rather than simple function, and are usually fairly complex hence being rarely published.

 

[spsalso]

Similar to what CWB1 says:


Have someone else look over your work ("fresh eyes"). That was revealed to me in my first programming course in college. Besides finding your mistakes, the new person might well suggest improvements.

Don't rush it. If I put a project down, I find I'm still thinking about it, even when I'm not. Frequently during a night's sleep, or while waking up.

Just because a computer says it's OK, doesn't mean it is. Use YOUR fresh eyes to at least look over its product for things that look just plain wrong.

For my projects (usually electrical construction), I don't care where good ideas come from. Maybe from a customer. Maybe from a passer-by or manual laborer. And I always say thank you.

As much as possible, test your premises, especially if you're going where "no man has gone before". My recollection is that the Wright brothers were great experimentalists, while Langley felt he knew exactly what to build. I've wondered if anyone built a test model of that Hickory NC bridge decoration before they pushed the go button.


spsalso

 

[1503-44]

1) I find it is inappropriate to try to fix bad design and poor performance by "software updates".

2) If N things have to go wrong before failure happens, you can be sure that N-1 of those are not going to be entirely independent events.

3) Plastic is a poor substitute for metal.

4) If something causes you to hesitate, find out what and why.

5) The simplest safest way is always best.


Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[tommj]

Here are some videos that might be relevant:

https://www.youtube.com/c/BrickImmortar/videos

 

[IRstuff]

I think the OP's question is a bit of a snipe-hunt; for any given design problem, there are a few possible valid solutions and dozens of bad solutions. This is why new designs tend to be variations of a known solution, and new solutions require extreme vetting before they're implemented or fielded.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[debodine]

I am late to the party but would like to second JohnRBaker and I JRB gave a star. I own 3 of Petroski's books and would like to continue collecting them.

 

[Sparweb]

Rather than just collect Petroski's books (of which I had several) I have started giving them away to young engineers who may benefit from the tales he tells.

 

[Tomfh]

I wish we had this for structural. Not the big headline failures we all hear about, but the day to day ones that generally never get exposed, which are honestly more relevant to most practicing engineers than a space shuttle exploding or a bridge collapsing. I know in my city there are moderate structural failures on buildings happening all the time. It's generally during construction, but sometimes afterwards. It's embarassing and invariably gets sorted out with as minimal publicity as possible. I wish we could somehow compile this information as it would be so useful to see the failure statistics. The insurance companies must have some wonderful records.

 

[ProSafPlant1]

A few you might check out (can't recall which ones were better except for Petroski and Kletz books and Set Phasers on Stun):
Forensic Engineering - From Failure to Understanding
Inviting Disaster - Lessons from the Edge of Technology
Learning from catastrophes
Learning from Construction Failures
Lethal Arrogance - Human Fallibility and Dangerous Technologies
Mechanical Failure Avoidance - Strategies and Techniques
Predictable Surprise - The Disasters you Should Have Seen Coming
Set Phasers o Stun - and other true tales of design, technology and Human Error
What Went Wrong - Case Histories of Process Plant Disasters - Kletz, or
Still Going Wrong - Case Histories of Process Plant Disasters and How They Could have been avoided - Kletz
Structures - Or Why Things Don't Fall Down
To Engineer is Human - Petroski
To Forgive Design - Understanding Failure - Petroski
When Technology Fails - Significant Technological Disasters, Accidents and Failures of the Twentieth Century
Why Buildings Fall Down
Why Things Bite Back - Technology and the Revenge of Unintended Consequences
Why Things Break - Understanding the World by the Way it Comes Apart

 

[Sparweb]

Structures - Or Why Things Don't Fall Down

That one was written by JE Gordon, a UK engineer who pioneered a lot of material science, including fracture mechanics, and methods for predicting crack growth that are used today in aircraft design and certification. The life predictions establish the inspection requirements and end-of-life that we rely on now, to prevent catastrophic fracture in flight.

JE Gordon also wrote The New Science of Strong Materials (or Why You Don't Fall Through The Floor) which is still in print!

https://press.princeton.edu/books/paperback/9780691180984/the-new-science-of-strong-materials

 

[Ingenuity]

I wish we had this for structural. Not the big headline failures we all hear about, but the day to day ones that generally never get exposed, which are honestly more relevant to most practicing engineers than a space shuttle exploding or a bridge collapsing.

I recommend this book that falls into your category of day-to-day structural. I purchased this book about 25 years ago, read it cover to cover, and still go back and read chapters on occasions.

DESIGN AND CONSTRUCTION FAILURES - LESSONS FROM FORENSIC INVESTIGATIONS, KAMINETZKY, D. McGRAW-HILL, 1991 ISBN 0-07-033565-6, HARDCOVER

 

[Tomfh]

I recommend this book

Thanks. I will get a copy and read it.