[time-nuts] Reliability of atomic clocks

Florian Teply usenet at teply.info
Mon Mar 28 07:05:05 EDT 2016



Am Mon, 28 Mar 2016 03:20:14 +0000
schrieb Mark Sims <holrum at hotmail.com>:

> > Mil-spec parts would be somewhat more reliable than commercial
> > parts.
> Actually,  that is seldom true.  The main difference between mil-spec
> parts and commercial parts tends to be in the post-packaging device
> testing (e.g.. extended temperature / voltage range).  They usually
> have the same guts inside the package.    I friend of mine was a test
> engineer at a major semiconductor manufacturer and he said their
> commercial parts were actually more reliable than the mil-spec ones.
> The commercial parts were built in vastly greater quantities and
> their production flow / experience enabled them to continually fine
> tune the process and testing.  The production / test regimen of
> mil-spec parts tended to be locked-down by inflexible specs and
> procedures.  Changing anything required re-qualification of the
> entire process.
> 
In general, there are two kinds of MIL-Spec parts:
a) Genuine MIL parts that use specialized processes. And
b) Up-screened commercial parts.

The genuine MIL parts (which often nowadays don't have a commercial
counterpart as these have long been obsoleted) in terms of
reliability usually suffer from low volume. Low volume implies poor
statistical process monitoring and/or long factory storage. This can
not always be seen on the datecode as in many cases the stored entity
is the wafer, not packaged parts.

b) Up-screened commercial parts usually benefit from sufficient volume
in production to keep the process under good statistical control.

But in both cases the major effort for MIL-Spec parts is
qualification, documentation and screening. This actually makes up a
huge chunk of the parts cost in MIL-Spec parts. Due to frequent
testing, for MIL-Spec parts you'd know how they'll behave to a good
degree of precision. But this doesn't mean they would be more reliable
in the sense that they last longer. You just COULD POSSIBLY get more
detailed knowledge on when they'll likely fail under certain conditions.

COULD and POSSIBLY, because for one thing one would need to have
sufficient importance to the manufacturer to actually get the required
data, and second the derivations of lifetime often are based on
MIL-Standard assumptions, which have been established 30+ years ago and
do not necessarily cover modern technologies appropriately.

I could go into more detail, but I'm not sure this is of interest in
this context.

Best regards,
Florian


More information about the time-nuts mailing list