Searching \ for '[OT]your thoughts reliability analysis' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=your+thoughts+reliability
Search entire site for: 'your thoughts reliability analysis'.

Exact match. Not showing close matches.
PICList Thread
'[OT]your thoughts reliability analysis'
2007\03\31@071836 by jtroxas

picon face
does the statement "more parts, more complex = less reliable" still holds
today electronics computers etc..... or is this just one of those myths we
sometimes hear..

I dont know if this is the exact statement nor remember what term they call
it or where it came from or who said it.. but from what I remember it sounds
something like that..

I would like to hear some of your thoughts...



2007\03\31@075415 by Jinx

face picon face

> I would like to hear some of your thoughts...

I don't think you can ignore quality of assembly and incorrect,
eg under-rated, component selection. IMO that is where a
device is likely to fail, rather than properly selected components
of (best) quality

2007\03\31@082627 by jtroxas

picon face
>>under-rated, component selection
so in a wway your saying the statement still holds true.. even though we
have better components this days... and that the more parts to use.. the
more likely youl choose an under rated component.. the less reliable...

{Original Message removed}

2007\03\31@084631 by Gerhard Fiedler

picon face
jtroxas wrote:

> does the statement "more parts, more complex = less reliable" still holds
> today electronics computers etc..... or is this just one of those myths we
> sometimes hear..
>
> I dont know if this is the exact statement nor remember what term they call
> it or where it came from or who said it.. but from what I remember it sounds
> something like that..
>
> I would like to hear some of your thoughts...

It's not as easy. A simple thought experiment: one of the common ways to
enhance reliability is to use several units, ideally all checking on each
other. This is a typical case where more parts is considered more reliable.

It's not the number of parts that much, it's how severe the damage is in
each of the failure modes and how probable these failure modes are in the
application. Adding a few parts to a circuit in strategic places can make
the likely failure modes much less severe or the severe failure modes much
less likely, and therefore increase the reliability (think PIC input
protection).

Gerhard

2007\03\31@085310 by Jinx

face picon face
> and that the more parts to use.. the more likely youl choose an
> under rated component.. the less reliable...

Well, choosing the wrong component is voluntary ;-)

I suppose the more components in a device the more chance there
is that one will fail. But it doesn't necessarily mean that one WILL
fail. There are obviously connections within a circuit, but a resistor
in one part of the circuit probably won't influence the failure rate of
a resistor way over the other side of the board. A domino effect is
always a possiblity

You might have a stereo or car that goes for years and years but
a brand new one may have troubles right from the beginning and
never be completely right

You'd have to put a figure to how long is long enough to be value
for money. If your 50 year old black and white TV broke down
today and had had very little service money spent on it, even if it
cost a small fortune to purchase, you know it was a good buy. But
it's broken down at a time when technology has moved on to bigger
and brighter things, so it may not be a great loss

2007\03\31@090056 by John Ferrell

face picon face
The time for that logic has passed.
All components cannot be treated equally and all designs/implementations
cannot be treated equally.

Most of the electronics products I have used in many years are replaced
because of better solutions rather than device failure.

I am currently replacing a Sony 13 inch TV (I once used it with a TI99
computer) because it does not tune the cable channels and it has no remote.

A few months ago I too four old computer monitors and on TV to the recycler
that were all working but obsolete.

John Ferrell    W8CCW
"Life is easier if you learn to plow
      around the stumps"
http://DixieNC.US

{Original Message removed}

2007\03\31@093011 by Rich

picon face
I think that applies more to mechanical system although it does also to
electronic systems.  However, the reliability in general tends to improve.
Nevertheless, the reliability still decreases as the number of components
increases.  The US Military has made public their studies on this issue and
so has NASA.  They are not distributed free.


{Original Message removed}

2007\03\31@095432 by jtroxas

picon face
>The US Military has made public their studies on this issue and
> so has NASA.  They are not distributed free.

yah.. I thought I remember reading something about that along time ago.. I
just cant remember where I had seen it on the net... and so I put up this
post.. I also sometimes hear some scientist on TV saying something bout it..

>>A simple thought experiment: one of the common ways to
>> enhance reliability is to use several units, ideally all checking on each
>> other. This is a typical case where more parts is considered more
>> reliable.

what about common household applience like our good old computers.. I dont
know if they have several or secondary or failsafe units that will come
alive  incase something fails...Unless were talking about something special
like multiimillion dollar aircraft.. the space shuttle or something in that
respect...

{Original Message removed}

2007\03\31@101627 by jtroxas

picon face
then maybe the more parts we used the more chances we have of using a weak
or even faulty component.. I mean manufacturers dont really test every
component resistors capasitors IC that goes out of the assembly line... I
mean they use the power of statistics basically test a few random samples
and then cross thy fingers hope that everything else it A OK.. Or do they
test all...

{Original Message removed}

2007\03\31@110658 by Tony Smith

picon face
> does the statement "more parts, more complex = less reliable"
> still holds today electronics computers etc..... or is this
> just one of those myths we sometimes hear..
>
> I dont know if this is the exact statement nor remember what
> term they call it or where it came from or who said it.. but
> from what I remember it sounds something like that..
>
> I would like to hear some of your thoughts...


Depends on what the extra parts are.

In general, the more bits you have, the more it goes wrong.  If a single
part is 90% reliable, then a system with 2 of these parts will be 90% x 90%
reliable, or 81% reliability overall.  3 parts drops to 73% and so on.

A sytem with all of it components rated at 99.999% won't have 99.999% uptime
(unless they all fail at once).  Marketing folk please take note.

On the flip side the added part could be for protection, so reliability
would increase.  Putting your camera in a waterproof case when diving would
probably increase it's reliability.  Something like a fuse doesn't really
count, not only do they occasionally fail for no good reason, the device is
still dead even if there was a cause.  That's assuming by reliable you mean
'it's working', not 'it's broken but I can fix it'.

Tony

2007\03\31@165331 by Brooke Clarke

flavicon
face
Hi jtroxas:

The concept is good, but it's not that simple.  So here are some
thoughts about reliability in general.

When designing something the "KISS" principal says Keep It Simple Stupid.

MIL-HDBK-217
http://www.weibull.com/mil_std/mil_hdbk_217f.pdf (14+ MB)
Uses a method where the stresses on each part are used to come up with a
Mean Time To Failure (MTTF) and a way to combine them into a MTTF for
the final product.  This is very important in many areas.  For example:
* A military mission that involves a jet aircraft and all the systems on
board that's going to last some number of hours.  If the MTTF of the
combined systems is shorter than the mission time then you can expect
the mission to fail.

* I knew someone who did 217 calculations as his job.  One of the things
he did was help design color TV picture tubes so that they would fail a
couple of months after the 1 year warranty ran out.

* Large volume car makers try to have their car fall apart all at once,
like at the end of the movie "Blues Brothers".  The idea is that when
the car goes to the scrap yard if it has parts that are still good then
those parts can probably be made for a lower cost by designing them a
different way.  A small cost savings times a very large number is a big
dollar savings.

Another reliability concept is that of a "single point failure".  This
is very important when designing satellite hardware.  For each and every
component you look at all the failure modes and then look at what would
happen if that mode occurred.  If a single part fails and it takes down
the whole system that's a poor design.  A common example is using a
tantalum cap from Vcc to ground as a power supply filter.  The cap fails
as a short taking out the up stream regulator and/or diodes shutting
down the equipment.  Another example is an 18 wheel truck that has a
single strap holding a bundle of pipes on the flat bed trailer.  Any
failure of the strap, either of it's end fittings or either of the hoops
it's attached to result in pipes all over the roadway.  The British
solution is known as "Belt and Suspenders".  A less descriptive term is
"an ounce of prevention is worth a pound of cure".

There's a web page that has example after example of major failures
caused by faulty "C" programs.  Things like a satellite that gets lost,
a cancer patient that gets a lethal dose of radiation, etc.  If a
computer is needed to control a life threatening procedure or something
of high value you don't want a "C" program involved and you don't want a
PC involved.  Programed Logic Controllers are good here and I expect a
PIC programmed in assembly language would come in second.

Have Fun,

Brooke Clarke

--
w/Java http://www.PRC68.com
w/o Java www.pacificsites.com/~brooke/PRC68COM.shtml
http://www.precisionclock.com

2007\03\31@171539 by Jinx

face picon face
> In general, the more bits you have, the more it goes wrong

In some cases more bits implies more functionality. Functionality
that might be excess to requirements. For example a friend's
flash nattty toaster broke down a while ago and I offered to see
if it could be easily repaired. It could, by by-passing the burnt-out
control electronics

Now it's a toaster again, but simply an element, just like the good
old days. It toasts just as well as it did but now the user has to do
their part

Functionality and the resulting complexity has two sides - it pushes
manufacturers to be creative and innovative, but it also boggles and
confuses consumers. Cellphones for example - "I just want a phone
that you can make phone calls on", or cars that you can tinker with
on a Saturday without needing a computer science diploma

So what's the answer ? Build with minimum number of components
to do the job ? Accept that complexity has a down-side ?

It's not a simple subject if you're going to figure customer's epectations
into the equation

2007\03\31@171818 by Peter Todd

picon face
On Sun, Apr 01, 2007 at 12:52:46AM +1200, Jinx wrote:
{Quote hidden}

Probably not really, don't forget the time value of money. Say the good
TV costs $5000 and lasts for 50 years. Buying a worse TV for $1000, and
having it last 10 years is actually a much better deal. As you can take
that other $4000 and do something more usefull with it. (like investing
it)

$1000 compounded at 5% over 40 years is $7030.00! More than enough to
account for the inconvienience of having to buy 4 more TV's I'm sure.

In fact doing the analysis fully:

Year 0: Starting with $5000: Spend $1000 on TV. $4000 in bank.

Year 10: We now have $6580. Spend $1000 on TV, $5580 in bank.

Year 20: $9039.60: Spend $1000 on TV, $8039.60 in bank.

Year 30: $13024.15: Spend $1000 on TV, $12024.15 in bank.

Year 40: $19479.12: Spend $1000 on TV, $18479.12 in bank.

Year 50: $29936.17 in bank....

Note that I am taking inflation into account, a %5 interest rate on your
money is artificially low, historically stocks do over %10 if you invest
them in the market as a whole. And %5 inflation is probably
unrealistically high.

I'm sure that extra $30,000 could also go a long way towards properly
recycling those other 4 TV's too, and account for your expenses in
having to buy extra 4 TV's. Not to mention, having a nice new TV every
10 years is probably much better than having a 50 year old one at the
end given we are dealing with technology.

Buy the cheapest crap you can get away with, always!

--
http://www.petertodd.ca

2007\03\31@173602 by Bob Axtell

face picon face
Peter Todd wrote:
{Quote hidden}

Remember to include taxes on the interest; that should take an edge off
of the optimism here...

--Bob

2007\03\31@173739 by Bob Axtell

face picon face
Jinx wrote:
{Quote hidden}

Or my left-handed wife, who blissfully purchased a Blackberry on Monday,
only to find after she received it
yesterday (no stores had any, so she couldn't check it out before she
bought online) that the blackberry only
moves the cursor with buttons on the RIGHT-hand side of the case. So she
simply can't work the darned
thing. And we went back to the website... It never indicates that it is
for right-handed people only....

--Bob

2007\03\31@174518 by Peter Todd

picon face
On Sat, Mar 31, 2007 at 02:34:07PM -0700, Bob Axtell wrote:
> > I'm sure that extra $30,000 could also go a long way towards properly
> > recycling those other 4 TV's too, and account for your expenses in
> > having to buy extra 4 TV's. Not to mention, having a nice new TV every
> > 10 years is probably much better than having a 50 year old one at the
> > end given we are dealing with technology.
> >
> > Buy the cheapest crap you can get away with, always!
> >
> >  
>
> Remember to include taxes on the interest; that should take an edge off
> of the optimism here...

True, but "interest" doesn't have to mean putting your money into a
taxed savings account. For instance for a busines the "interest" would
often come from *not* having to *borrow* money to finance the business,
and avoiding borrowing isn't taxed. Or you could think of it as simply
having better things to do with the money, like spending it on other
parts of the business. Similarly for your average individual excess
money will be plowed into retierment savings plans which are not taxed
until you start taking money out of them, and even then they are taxed
at the income tax bracket you have in the future.

And for that matter taxes are only charged on capital gains. So if you
have your money invested in, for instance, a mostly growth (not
dividend) oriented fund the majority of your growth won't be taxed
anyway. I used to have a fair chunk of money in such a fund and the
taxed dividends were maybe %10 of the total growth.

Anyway, point being is that interest really doesn't change things much
if you are managing your money competantly.

--
http://www.petertodd.ca

2007\03\31@174728 by Gerhard Fiedler

picon face
jtroxas wrote:

> yah.. I thought I remember reading something about that along time ago.. I
> just cant remember where I had seen it on the net... and so I put up this
> post.. I also sometimes hear some scientist on TV saying something bout it..

It depends on what you look at. If you consider components only as "black
boxes" with average failure rates (say, a resistor has a MTBF of so many
hours), then, of course, the more components, the more likely a failure.

But as I wrote before, it's not that simple. For a decent failure
probability analysis for any given device, you need to first define
"failure" for the device and your purposes. Then you need to establish the
relevance of the different failure modes of each component for that device
failure (which depends on the other components). Then get the probabilities
of the different failure modes (which also depend on the other components).
Then add that all up (considering correlations).

There's nothing that guarantees that a design with fewer components will be
more reliable. It may very well be that, for a given design, added
components increase the reliability rather than decrease it.

Look for example at PIC programmers. You want a really reliable piece? You
won't find them among the ones with the fewest components.


> what about common household applience like our good old computers.. I
> dont know if they have several or secondary or failsafe units that will
> come alive  incase something fails...Unless were talking about something
> special like multiimillion dollar aircraft.. the space shuttle or
> something in that respect...

Of course, normal household appliances usually don't have built-in
redundant devices. But think of simpler examples... A power supply with a
resistor in the input part. The resistor every now and then fails open,
with a device failure as a consequence. Somebody thought that the resistor
might be driven too close to its maximum voltage and replaces the single
resistor with three in series: no failure anymore (of this kind). More
components and fewer (not more) failures.


Then there's the question what constitutes a "component". There are ICs
with millions of transistors. There are ICs that pack together different
devices on the same silicone -- are they as many components as they contain
integrated devices? There are modules that contain several components on
thick film; are these one component (they look like one from the outside)
or several? Furthermore, you'd have to have a significant average failure
rate for each component. I don't think these numbers exist for most of the
components you'll find in a typical design, and I don't think these numbers
are very relevant; the actual failure rate depends a lot on the individual
application, so average failure rates are not really useful. I don't think
the number of components is a very relevant reliability benchmark.
Stripping out the protection diodes from a PIC's input circuit doesn't
increase reliability.

Gerhard


'[OT]your thoughts reliability analysis'
2007\04\01@051308 by Ling SM
picon face
> * I knew someone who did 217 calculations as his job.  One of the things
> he did was help design color TV picture tubes so that they would fail a
> couple of months after the 1 year warranty ran out.

Situations are even worse.

I have good reasons to suspect that my printer has an algo to
intentionally killing itself and the print head.  I shall be able to do
more test when my stock of refill ink run down.  For the time being, I
shall resist to swap "old" print head to another "old" printer.

It is quite a tough money to make in the consumer electronics areana
when compare to the fashion industry when a pair of jean easily price
higher than a MP3 player.  But...

Ling SM

2007\04\01@072024 by jtroxas

picon face
epson ink cartridge have a counter chip embedded. so you cant useother
compatible cartridge other than epson.. or refill an original cartridge when
they go out.... fortunately people found ways around this... as always..

{Original Message removed}

2007\04\01@131309 by William Chops Westfield

face picon face

> does the statement "more parts, more complex = less reliable"
> still holds today electronics computers etc.

I think that in general, that statement heralds from a time when
components had approximately equal failure rates, and it's gotten
more complicated since then, especially with parts that contain
so many individual subsystems that partial failures are possible,
and those can be a lot more complex to detect and analyze (thus
the multiple-copies with voting systems.)

Alas, with anything containing a computer, the proper maxim is
likely to be "more lines of code, more complex = less reliable."

BillW

2007\04\10@073548 by Alan B. Pearce

face picon face
>You might have a stereo or car that goes for years and years
>but a brand new one may have troubles right from the beginning
>and never be completely right

A friend of my fathers got involved with such a car. He was friendly with a
mechanic at a local garage that had sold a Japanese second hand import to a
customer. It was an exceptionally low mileage for a Japanese import, and the
reason seemed to be that it would intermittently stop dead. Dads friend got
involved, as his mechanic mate knew of his expertise in electronics, and
after several attempts at repair, they eventually nailed it to a cold solder
joint somewhere in the ignition/distributor/EMU. It would seem the original
dealer in Japan had never got to the bottom of the problem, because it was
going to cost too much to replace the necessary FRU which would have sorted
out the problem, so it got on-sold earlier than usual.

When we do failure analysis on instruments for space, one of the biggest
sources of likely failure is solder joints.

2007\04\10@074249 by Alan B. Pearce

face picon face
>Or do they test all...

Depends what specification you use. There will always be infant mortality
failures that give you the leading edge of the "bathtub curve", and for the
various MIL-SPEC devices these are filtered out at various stages, and the
higher spec you go to, the more likely they are to have been removed from
the mix by visual inspection, high temperature burn-in, or some other
criteria.

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...