trader_4 has brought this to us :
On Wednesday, September 14, 2016 at 6:46:41 PM UTC-4, T wrote:
On 09/12/2016 07:04 AM, Scott Lurndal wrote:
T writes:
On 09/10/2016 12:41 PM, Ed Pawlowski wrote:
tolerance for MTBF

Hi Ed,

That is not what MBTF means. If the bulb had, say 1000 hour
MBTF, it would mean that you put 1000 bulbs in a test bench
and ran them for an hour. Only one failed. MBTF does not
tell you anything about the second and so forth hours.
(I did MBTF analysis for the military.)

Yet, you can't even spell the acronym correctly.

It's not really an acronym.

And "YOU" never make a typo? Oh pardon me!

It's sure odd that you made the same "typo" four times in a row
in the same post and never once got it right. What are the odds
of that?

Use the MWBT (Mean Words Between Typos) formula [See Appendix E], just
add up all of the words in intervals between the starting typo and
between subsequent typos and divide by the number of typo events
recorded excluding the starting typo. It comes to about 14.33 WPT
(Words Per Typo) I think.

Be advised, small data sets may give suboptimum results and excluding
the 'warm-up' period and the 'past bedtime' period is recommended for
the best results.

What a buffoon. The measurement interval is exactly that, how long the
system if being measured for failures.

Not just "how long the system if[sic] being measured for failures", but
where. The idea is to exclude the outliers of 'infant failure' and
'system age wearout' by getting the data set from the steady rate of
midlife operation from the nearly flat bottom of the bathtub curve.

So we can only test light bulbs, LEDs, jet engines for a measurement
interval of one hour? Good grief

It would certainly speed up the process, and marketing would just love
the great looking numbers. No need to test them for *two* hours when
you get such good results from one hour.