FLASH, CPLD, FPGA - Future of repair?

Hi together,

looking at modern products repairing them becomes increasingly
difficult. Due to the massive use of programmable logic and
software residing in flash memory - there are lots of componentes
which can not be checked or replaced - even if they are known
to be defective...
Sometimes even the original manufacturer is not able to supply the
right chips to fix the problem. My question is how the know how
residing in such chips can be saved prior to their death?
For software there exists an archiv (bitsaver.org) to preserve
it, but how e.g. on fuse-lists of FPGAs?

Logic devices are hard to read out without the appropriate programmer.
Even if they are in a socket the protection fuse my prevent
reading... Is there any experience how long EEPROM-based FPGAs and CPLDs
really keep their programming?

The actual reason for my fear of these components is that
test equipemt, especially high quality one like tektronix or agilent,
always was an investment for many years - who is not happy about
his working Tek535 on a cold winter day? Recently there came up
relatively many of high end oscilloscopes (545xx) on eBay with
comments like "out of service", "agilent refused to repair",
"not serviceable". Since I own such a device, I suspect the
ALTERA EPM5128 or the soldered in flash memory beeing the problem.
Honestly speaking I was shocked that HP/agilent oscilloscopes are
"not servicable" only a couple of years after they have been
discontiuned :-(

Thanks god, my 54542a still works...

Erik.

 

This has been a problem for quite a while now, a number of very good Tek
scopes for example use custom IC's that run hot and are unobtainable.

 

<snippety-two>

What you describe, Erik, is unfortunately an all-too-common
problem in today's electronics world, and it is one of the major reasons
why there is such a thriving market for used (and older) equipment.

One of the biggest problems I see is that, about 20 or so years
ago, a distressing trend began where much of the electronics industry
(heck, much of the PLANET for that matter!) stopped being interested in
long-term gains and (in the case of the electronics industry) longevity
of products. They began to favor, instead, short-term profit over long-
term growth.

The result has been obvious, as you have found out with
HP/Agilent. Tektronix is no different. 10 years ago, I could buy all the
manuals (and many parts) that I needed to keep Tek's 7000 series
equipment running, and at a fairly reasonable price too.

Then, one day, availability of true "service" manuals (those
containing schematics, parts lists, and theory of operation sections)
dropped off dramatically. "Service" documentation started containing
little more than board-level troubleshooting flowcharts, use of
customized non-obtainable parts skyrocketed, and equipment was declared
(by its manufacturer) to be "obsolete" (Lord, I loathe that word...)
after as little as 2-3 years.

Tektronix is far from the only victim of the MBA-wielding profit-
hungry CEO infestation, but they are a prominent one. I will say that,
in recent times, at least one person at HP/Agilent is trying to reverse
part of the trend. They visited the mailing list for HP/Agilent users
not long ago, saying that they wanted to try and greatly expand HP's
downloadable offerings on manuals and service information for older
equipment.

How far this attempt will go is anyone's guess, but at least one
person is trying. Perhaps, in time, there will be others.

None of this addresses your problem directly, of course. All I can
say along those lines is learn to be a scrounger when it comes to used
equipment, and learn well! Such skills could really save your tail (and
a whole ton of $$ if you shop carefully).

I know. I've built my entire lab using such techniques. ;-)

Happy hunting.


--
Dr. Anton T. Squeegee, Director, Dutch Surrealist Plumbing Institute.
(Known to some as Bruce Lane, ARS KC7GR,
kyrrin (a/t) bluefeathertech[d=o=t]calm -- www.bluefeathertech.com
"If Salvador Dali had owned a computer, would it have been equipped
with surreal ports?"

 

OK, this is an interesting note - but if one does not use
the instrument regularly it should last a long time.
And knowing about this problem one could implement better
cooling by e.g. a small fan.
With EPROM and FLASH its worse: The charge on the floating gate
degrades and renders the device useless independent of using it
or not :-(

Which Tek's are likely to encounter this problem?

Thanks,

Erik.

 

Obviously - but the replacements obtained from other
equipment are old, too. I think they will last several years
but not 10 years until they "forget" the data programmed in, right?

YES - you are ABSOLUTELY RIGHT! Looking at HPs way is a interesting
example,
too.

Hey, that would be a great thing. From a developers point of view
it would be very good to know that the own work invested in the firmware
of e.g. an oscilloscope is preserved for the future. But honestly
speaking I doubt that e.g. the firmware of an oscilloscope will be
released for download??

One person my do a lot. E.g. I am fan of SGI computers and this
company had in the past some guys supporting freeware on SGI
systems - lot of this work was done in their free time. They were
really helpful regarding internals and it was a pleasure to contact
them!

equipment, and learn well! Such skills could really save your tail (and

In the past I had good contact to the electronics waste container
at Munich's university and most of my lab consists of repaired
equipment therefrom. But today I bought buy some items on eBay...

Thanks for your tips,

Erik.