Skip to main content
Astronomy

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

15
  • 25
    $\begingroup$ @ayr because when you design something to last for 2 years, guaranteed - that means it has to be super unlikely to stop working before that. And operational time is not some fixed value, it ends when something breaks (or runs out). So you design it with that in mind - and a computer that cannot break for two years will not magically break after 2 years and a day. Same for all the other parts. So it's not as much "errors in predicting life expectancy", it's that the life expectancy the devices are designed to are a minimum. $\endgroup$
    – Syndic
    Commented Jun 19 at 9:48
  • 24
    $\begingroup$ "will not magically break after 2 years and a day" No, that's the design-goal of consumer electronics! $\endgroup$
    – TripeHound
    Commented Jun 19 at 10:10
  • 6
    $\begingroup$ @ayr It's similar to how you can often eat food that is past its expiration date. The expiration date guarantees, with some degree of certainty, that food will be edible before the expiration date. After that, it's a guessing game. The same holds true for space telescopes, but microbes cause food to "fail" in the order of days, while engineered parts in space fail in the order of years. $\endgroup$
    – AccidentalTaylorExpansion
    Commented Jun 19 at 11:46
  • 19
    $\begingroup$ @ayr: If you design something to work for 2 years with as close to 100% probability as you can reasonably get (say, 99.999%), then it will almost automatically also work for for 4 years with 99.99% likelihood, 6 years with 99.9%, 8 years with 99%, 10 years with 90%, 20 years with 80%, and so on. (Numbers are completely made up, but the principle stands.) $\endgroup$
    – Jörg W Mittag
    Commented Jun 19 at 12:34
  • 8
    $\begingroup$ @TripeHound Good jab -- with some more serious background: There is a sweet spot for consumer electronics if, say, a low single-digit percentage of devices fail while the warranty is still on. The warranty cases are paid for by the profits from lower overall production cost. Also, because most parts will be off-the-shelf, their MTBF is better known (with lower error margins) than for one-offs. For a billion dollar device, a single digit premature failure rate (if nothing goes wrong!) is unacceptable, and the error margins will be larger, so the error will be on the safe side most of the times. $\endgroup$
    – Peter - Reinstate Monica
    Commented Jun 19 at 14:45