[time-nuts] Rubidium standard

[Steve Rooke]

2009/11/19 Mike S <mikes at flatsurface.com>:
> At 05:22 AM 11/18/2009, Steve Rooke wrote...
>>
>> The point I should have made is that most quoted MTBF figures have a
>> reasonable bearing on the lifetime of the item,
>
> But your point would then be almost perfectly incorrect. MTBFs are not meant
> to, nor do they, predict product lifetimes. They are measures/predictions of
> product reliability.

OK, I agree, the term is mean time between failures but the question
is what is the lifetime of something. This is the point we are
discussing and we really need to understand what this means. I'm not
going to quote a number of web searches, this means different things
to different people but this matter is not quite as black and while as
you think.

The lifetimes of any thing can be broadly split into two groups, there
are things that have a predetermined lifespan, like lifeforms, and
those that have no fixed lifespan. Let's look at the second broad
category, as this is what we are really talking about. So what is the
lifetime of an item in this category, is it related to MTBF or not.
Well, lets see, take some items, like consumer goods, these are
frequently thrown away when they fail as it is generally expensive to
fix them and newer models are more attractive. Now, do all consumers
wait till an item fails before they throw it away, well no, it depends
on the culture but I bet a large proportion of consumers probably do
wait till it fails. So in this case, quite a large section of what we
are talking about, the lifetime of the item is related to the time it
fails. Infant failures are obviously covered by initial warranty but
this seems to indicate that for a large sample size, the products
lifetime would be related to the MTBF. Agreed there are some
assumptions there, I expect the MTBF of a cellphone is a reasonable
amount but if you live in California and are seen with one over 6
months old you would die of embarrassment. So the lifespan of a
cellphone in California is significanly shorter than the MTBF, the
factor here is the MTBE (mean time before embarrassment).

No lets look at the corporate environment. In this environment,
failure of equipment can be very costly to a company so, if they have
any sense, they will arange to upgrade equipment well inside the time
that the item is expected to fail. Now, you may say that companies do
this well inside a shorter cycle than this as they can amortise the
cost off on tax release over a few years but I've worked in this
industry long enough to know that budgets don't work that way in the
real world and items get replaced when they fail. Again, with a large
sample size, this equates the lifetime of an item with the MTBF.

Now, there are those industries that do upgrade their things before
they fail and including those things that have actually failed and
been replaced, these things become available to people like us. Now,
for us, the MTBF is not the relevant factor here as we will repair the
thing, over and over again so it has a lifetime far in excess of it's
MTBF.

So, for the majority of cases in consumer an corporate usage, the
lifespan of any thing is related to the MTBF. For the smart set with
enough dosh, the lifespan of any thing is shorter than the MTBF, and
for the poor buggers like, at least, some of us, the lifespan is
greater then the MTBF. The point being that for the majority of cases,
the lifespan of any item is related to it's MTBF for equivalent sample
size.

Now, I know that MTBF is related to the failure rate but failures
determine the lifespan of any thing in the majority of cases.

The guys that put people into space have to really understand that
well. It's no good putting people into space for a 2 week mission in
equipment that has an expected MTBF of 1 week unless you wish the
mission to be only half completed given a sufficient sample size. It
also causes a lot of paper work and hot air to be blown which is
generally undesireable.

> "What does MTBF have to do with lifetime?  Nothing at all!" -
> http://www.ece.cmu.edu/~ganger/ece546.spring02/readings/mtbf.description
>
> "MTBF represents the statistical approximation of how long a number of units
> should operate before a failure can be expected. It is expressed in hours
> and does not represent how long the unit will last." - Learn (or review) the
> difference between MTBF and lifetime, Control Engineering, 9/24/2008;
> http://www.controleng.com/article/312365-Learn_or_review_the_difference_between_MTBF_and_lifetime.php

But all this means on whether you consider the lifespan of any item to
be greater than the MTBF, IE. you don't mind the possibility of having
to fix it a few times, or if you throw it out as soon as it fails, IE
lifespan is related to MTBF. This, of course, does not cover instances
for when there is a planned obsolescence of the item, then the
lifespan is shorter than the MTBF but the swap out cycle has probably
been planned from the MTBF in the first place.

> I don't grant Wikipedia strong authority, but it is useful, and has this to
> say: "MTBF is commonly confused with a component's useful life, even though
> the two concepts are not related in any way. For example a battery may have
> a useful life of four hours, and an MTBF of 100,000 hours. These figures
> indicate that in a population of 100,000 batteries, there will be
> approximately one battery failure every hour during a single battery's
> four-hour life span."

This is an example of an item with a specific lifespan and therefore
MTBF is not related to a single items lifespan. In this case it
determines the probability of failure of any item. Comparing the
lifespan of items having a specific lifespan with their MTBF is like
comparing apples with oranges. This is a bad example of what we were
talking about. I agree that rb lamps do have a limited life but they
can be replaced as they are just a component that fails in the actual
whole item and this is just one failure in the terms of MTBF.

> There's much more out there, if you make the effort.

And there is much more in you if you logically think about it instead
of just accepting things on the web without really thinking what we
are relating to here.

>> I felt that an example based on humans was not really applicable to
>> the real world of electronic items but that is my own opinion and I'm
>> happy if you disagree with me.
>
> MTBFs are not exclusive to electronics. Statistics, math and MTBFs are
> objective matters, so your opinion really doesn't make any difference.

Certainly statistics, math and MTBFs are objective and they have a
specific bearing on this matter. Along with this, you need to add
policy, as to how the entity using an item sees about its lifetime
related to those objectives, IE. they are related. The lifetime of
anything is a policy decision based on the items failure rate.

Steve
-- 
Steve Rooke - ZL3TUV & G8KVD
A man with one clock knows what time it is;
A man with two clocks is never quite sure.