The F-86 aileron bolt: if it can be done wrong
Key idea
A bolt that could be installed upside down was installed upside down, and F-86s crashed. Given enough trials, if something can go wrong, it eventually will. The lesson is in design: if we can design out the possibility of doing it wrong, we should.
Read first: this case is in What Murphy was really saying, in Factor II · Murphy’s Corollary.
Anything that can go wrong, eventually
Murphy’s Law, and the one word its popular form leaves out, are on the Factor II page: What Murphy was really saying.
Through “The bolt”: Mike’s text from the paper, p. 11.
The bolt
Paper, note 4 Chuck Yeager and Leo Janos, Yeager: An Autobiography (New York: Bantam Books, 1985), 185.
A tragic illustration of Murphy’s Law was the series of crashes in the F-86s in which a bolt on the aileron cylinder was installed upside down causing the ailerons to lock up under certain loading conditions. The incorrect bolt installations were reportedly traced to one older worker on the assembly line, who ignored the instructions regarding the orientation of the bolt because he felt he knew better how the bolt should be installed.
What it shows
- Murphy’s Law itself
- The paper uses this case for Murphy’s Law, not for one of the three factors. The bolt could be installed the wrong way, so in enough installations it eventually was. The instructions said which way it went; the design allowed the other way.
- II · Murphy’s Corollary
- The corollary is the law turned around: anything that did not go wrong could not have gone wrong. This case shows the law; the 1995 top landing and the three NASA cases show the corollary.
Questions for discussion
- Which part or step in your build, maintenance or test process can be done the wrong way and still look right?
- Where do you rely on an instruction, rather than the design, to prevent a wrong installation?
- How would you find out that one experienced person was doing a step their own way, because they felt they knew better?
- Which of your designs or procedures could be changed so that the wrong way is impossible?
- Estimate how often a step that can be done wrong is done wrong, say once in a thousand times. Using the calculator, how likely is it to happen at least once across every time that step is done in your fleet in a year?
These questions also appear, with every other case’s, in the Briefing Kit.
Cite this page
Meier, M. (2026, November 10). The F-86 aileron bolt: If it can be done wrong (Edition 1.0). meiersafe.com. https://meiersafe.com/case-studies/f-86-aileron-bolt/
Chicago and BibTeX
Meier, Mike. “The F-86 Aileron Bolt: If It Can Be Done Wrong.” Edition 1.0. meiersafe.com, November 10, 2026. https://meiersafe.com/case-studies/f-86-aileron-bolt/.
@misc{meier2026f86aileronbolt,
author = {Meier, Mike},
title = {The F-86 Aileron Bolt: If It Can Be Done Wrong},
year = {2026},
note = {Edition 1.0},
url = {https://meiersafe.com/case-studies/f-86-aileron-bolt/}
}
The F-86 aileron bolt by Mike Meier, meiersafe.com/case-studies/f-86-aileron-bolt/, CC BY-SA 4.0.