On Nov 19, 7:30 pm, "Arfa Daily" wrote:
"Jeff Liebermann" wrote in message

...





"Arfa Daily" hath wroth:

Jeff and James.
Thank you gentlemen. I feel vindicated ...

Personally, I feel vindictive. Time to rant some more on the topic of
reliable products.

Once upon a time, I was doing design reviews for a manufacturer of
consumer electronics. For those not familiar with the term "design
review", it's an ordeal process, where an outside authority is paid to
shoot holes in a pre-production design, with the specific purpose of
finding screwups, catching oversights, reducing costs, and generally
making the life of the designers miserable.

State of the art in component selection and computer modeling had
progressed to the point where it is possible to predict component
failure rates given known operating conditions. The capacitor
formulas at:
http://www.illinoiscapacitor.com/techcenter/lifecalculators.asp
are what is used. Note that the lifetime for electrolytics operating
at maximum ratings is only about 1000 hrs:
http://www.illinoiscapacitor.com/products/aluminum.asp

There are similar models for semiconductors based on thermal cycles
and power components based on simple entropy. Given years of
experience with a war chest of standard components, additional data
can be added to the mix.

My part of the puzzle wasn't in the failure prediction department, but
I watched the process evolve over the years. At one point, the
manufactory could estimate the point at which the largest number of
field failures would occur and target their stocking, warranties, and
replacement products accordingly. For example, if the calculations
showed that units sold on introduction will begin to fail after 3
years, then the warranty will be less than 3 years, and a replacement
product will be available at the 3 year mark. Under ideal
circumstances, everything will fail at once. This method gets the
product closer to this ideal.

In grinding the numbers, a simple spreadsheet was used for capacitors.
I may have a copy somewhere but don't want to leak someone else's
work. If I get ambitious, I'll recreate it. Anyway, the result is an
estimate of capacitor lifetime based on known operating conditions. If
the target lifetime is 3 years (26,000 hrs), then any capacitor that
shows a calculated longer lifetime is considered over-rated and is a
potential waste of valuable pennies to the bean counters. Once
identified, the size and voltage ratings are reduced downwards to fit
the predicted lifetimes.

In the distant past, there was an advantage to using as many
capacitors of the same exact value as possible to increase quantity
discounts and reduce handling overhead. With bulk lot purchasing and
robot assembly, this is no longer the case. It's perfectly acceptable
to have a wide variety of electrolytic sizes, values, and voltages
without impacting the cost. The result are boards stuffed with 3v,
4v, 6.3v, 8v, 10v, 12v, 15v, 16v, 18v, 25v, etc capacitors with almost
identical capacitance values. Such boards are easy to spot and should
give you a clue as to the degree of penny pinching that was required
to cut costs. If you have a product on the bench to repair and a wide
assortment of capacitors are apparently dead, don't assume user abuse.
It might be that they were sized specifically to all fail
simultaneously.

Anyway, I don't have much experience fixing DVD players or with
Philips products. I do have a Philips DVDR 995 DVD recorder, which
trashes most DVD-R's it tried to burn, and now refuses to play other
home burned DVD's, although it will play most commercial DVD's.
Probably a dying laser. It also has a rather oddly designed remote
control that also eats batteries. It was given to me with the warning
"It's crap" which I have personally verified. It would be at the
local thrift shop or recyclers were it not at the very bottom of my
equipment pile and therefore difficult to extract. Caveat Emptor.

That's highly enlightening, and goes a long way to explaining the multiple
failures of caps that I often experience. It also answers for Graham, as to
why he is wrong with his contention that a cap can be run to its voltage
limit, with no ill effects. It also explains why you are absolutely spot on
about different voltage ratings of caps with the same values in the same
equipment. I have come across this many times and wondered why, when they
were using a 25v cap in one position, they then used a 16v one of the same
value elsewhere, even though the physical sizes are similar, and you would
have imagined that they could have got a better deal by buying twice as many
of a single type. Good stuff.

Arfa- Hide quoted text -

- Show quoted text -

Well, I looked at a couple of units and decided to go with the Philips
dvd player, since it had a 1-yr warranty on parts and labor and the
competition only had 90 days parts and labor. Brought it home and
hooked it up, actually followed some fairly clear instructions just to
see how good they were. Must have been written by someone with
English as a first language, because the English actually made sense,
without the usual stilted words. Turned it on and it worked just
fine.

Decided to use my older dvd at the rental and keep the new one here at
my home. BTW, the rental unit is only loaned out to ski friends who
don't trash it if they want to be allowed to rent it again. So I
don't worrry about problems with tenants like I would if it were a
regular ski rental.

Thanks for all the civil discussion this generated.

H. R.(Bob) Hofmann