QUESTION ON EMBEDDED SYSTEMS IN 199 -RESPOND

greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

NOW I have a question on embedded systems ; IF some systems recognize 99 as - end of , stop, save , do nothing , shit down etc.. isn't it POSSIBLE ? - that somewhere in the network of systems that there will NOT be a rollover to 2000 because 99 - may cause the above problems in the system that y2k fixers and programsers did not find ( or did not KNOW to check for ) and if that IS so our event horizon is in JANUARY 1999 and NOT 2000?

ex - the euro jan 1999 ; unemployment jan 4, 1999

SOMEONE , enlighten me on this - methinks TIME IS of the essence

Sean

-- sean (venturer@interlog.com), December 30, 1998

Answers

I want to know the answer, too.

-- Linda A. (adahi@muhlon.com), December 30, 1998.

I've been pondering this myself. That's why I'm going to do a shopping blitz over the next two days.

-- Anti-Chainsaw (Tree@hugger.com), December 30, 1998.

Opinion from a hobbyist-programmer-turned-commercial-software- programmer... With some experience with embedded system programming to boot...

 

We all know that the problem is the fact that a lot of controller modules, et al are using real-time clock (RTC) chips that were basic OEM models, mass-produced in ungodly numbers whose true scope we'll see soon enough.

 

RTC chips fortunately generally do not use any code to indicate end-of-program, etc. as they're designed to simply report the time when they get the right signals on the right pins. (And occasionally they can be told the time in a similar fashion.)

 

Embedded systems in misc. sensing / control / reporting / etc. applications (where the RTC chips are only used for timing down to the month or less) should generally speaking not be affected by a '99' year. It's the '00' transition that fritzes them out. Now as for any computer systems that use these RTC chips or poll them for data, that's another story entirely!

 

 

OddOne, who's debating a change of occupation since programmers might not be held in high regard soon...



-- OddOne (mocklamer@geocities.com), December 30, 1998.

I have put out a request for information on any INDUSTRIAL class PLC that shuts down at rollover. For anti-troll purposes I want the serial#s, mfr, and model# of the affected device. Then I can check with the maker and find out if there really was such a device. I don't want 'one of a kind - I made that' gadgets or VCR or microwave controllers - I just want the kind of general purpose industrial class controller that a company typically spends anywhere from $500 to $5000 per each to acquire. This has been emailed a bunch of places, posted on quite a few. Haven't gotten a single response yet that gave an actual failure. Some fairly amusing ones regarding monitoring equipment that insisted it was running in 1980 - one I can't confirm where a monitor screen could not display for some odd reason - but not a single confirmed shutdown type problem yet. If this was really a huge problem I would have expected at least a dozen replies by now, shoot fire, if it was common I would have expected a hundred or more - I was afraid my mailbox would overflow when I posted the request - now I am beginning to wonder if it is a real problem at all!

-- Paul Davis (davisp1953@yahoo.com), December 30, 1998.

The "9's = end of program" problem is a mainframe programming problem, not an embedded systems one. This is a mainframe hold-over from the days of punch cards and could vary by manufacturer. Some old mainframes used 8's and the ones I learned on used 37xxxx, the x's being the program identifier.

PLCs and other embedded systems are in their oldest forms, Z-248 microprocessor based. Most of the PLCs I worked with were 386-based devices. Most embedded, industrial and manufacturing equipment PC systems are 386 or 486 designs (My employers until June of this year are just offering Pentiums in their product lines). Like PC's, the Y2K problem will be the real time clocks' reaction to the date (keep ticking or hang-up) and the embedded PC's bios system handling of rollover.

WW

-- wildweasel (vtmldm@epix.net), December 30, 1998.



Paul, this is an honest question not intended to be confrontational: Who are you reqesting information from- the manufacturer or the end user of the plc's and what makes you think you will get an honest answer from either? Thank you in advance for your reply.

-- Bumble Bee (Bumble@icanect.net), December 30, 1998.

Moderation questions? read the FAQ