http://www.nrc.gov/NRC/reactors.html

If you go to the link above and scroll way down the page, you'll come to a section captioned, "Other Plant Information". There are a dozen links to documents in this section. Browse through the titles. See the one called, "Computer-Based Digital System Failures: December 1996 - March 1997"? Given the Year 2000 problem, sounds like it would be an informative and worthwhile read, doesn't it? At least that's what I thought.

I clicked on it. This is what I got on the page which came up:

COMPUTER-BASED DIGITAL SYSTEM FAILURES: DECEMBER 1996 - MARCH 1997 ______________________________________________________

This page has been removed until it can be re-checked for accuracy.

(Effective: May 5, 1999)

I couldn't help myself. I busted out laughing. Anybody think I should buy into the explanation for pulling that page?

After awhile, though, a slow burn of anger kicked in over all the PR, legalese, and other information manipulation that clouds the whole Y2K issue. Am I being too sensitive about this? I'm feeling like I did as a youngster when I got a "Don't worry about it. You wouldn't be able to understand," or a "You don't need to know," answer to a question.

-- Anonymous, July 22, 1999

Answers

Gosh Bonnie,

It sounds like you're really pissed. (can't believe you said that :- P) But so am I! I said last year, this is like the story of Alice In Wonderland, where we fall down that rabbit hole and spend the whole rest of our visit just trying to figure out what kind of a societal system is being run there. So strange. So many different meanings for the most basic things. Almost as if someone tells you that you must now learn to write everything without any vowels. Tough thing unless you are schooled in Hebrew first. Grrrrr.

-- Anonymous, July 22, 1999

This does not inspire confidence.

-- Anonymous, July 22, 1999

Don't worry your little heads about this:

COMPUTER-BASED DIGITAL SYSTEM FAILURES: DECEMBER 1996 - MARCH 1997 This page has been removed until it can be re-checked for accuracy. (Effective: May 5, 1999)

------------------------------

We can be sure this has NOTHING to do with the safe shut down of nuclear power plants. All of that stuff was taken care of LONG ago. They probably just had a teensy problem with their spell checker and now that everything else is remediated, they are going back and cleaning up old files. Allay... allay... allay.

-- Anonymous, July 22, 1999

Interesting, indeed.

Verified for accuracy, eh? It was accurate enough to be online for quite awhile (two years, at least).

I just so happen to have a copy of this particular document. In fact, it was included as Appendix D to the book, "Electric Utilities and Y2k". Since the book is now out of print, I suppose I could post a .pdf copy of this appendix for anyone that's interested. Here's the intro to it:

Computer-Based Digital System Failures:
December, 1996 - March, 1997

(Nuclear Regulatory Commission)

The report that follows provides a summary of NRC staff results from a 
search and review of digital system problems at nuclear power 
facilities.  It is reprinted here simply to illustrate the fact that 
there are many systems controlled by digital technology components and 
devices in power plants, both nuclear and fossil.  While this report 
covers only a four month window, a review of NRC data from 1987 to 
1997 finds one hundred eleven separate instances of significant 
digital system failures.

1.  INTRODUCTION AND BACKGROUND

In recent years, many licensees have chosen to install computer-based 
digital systems for control and protection purposes to replace 
outdated analog equipment.  As an effort to gain a better 
understanding of the overall reliability of digital systems currently 
installed in nuclear power plants, the Instrumentation and Controls 
Branch (HICB) tracks problems associated with these digital systems.  
This information can be referred to when performing reviews and 
inspections of future digital system retrofits or when considering new 
industry standards and NRC guidelines.  This report presents the 
results of HICB's search for digital system problems or failures that 
occurred during the period of December 1996 through March 1997. This 
information was obtained through interfacing with designated 
instrumentation and controls contacts in each region and reviewing 
Nuclear Plant Reliability Data System data, Licensee Event Reports, 
event notifications, morning reports, inspection reports, and Nuclear 
Utilities Software Management Group news bulletins for information 
concerning digital system problems.  This report should not be 
considered a comprehensive list of all digital system failures.

2.  APPROACH 

As outlined in "Proposed Plan to Track Digital System Failures," 
transmitted via memorandum to Jared S. Wermiel, dated December 16, 
1994, HICB has undertaken two tasks in order to identify 
computer-based digital system problems.  First, HICB maintains 
communication with designated instrumentation and controls contacts in 
each region to obtain information concerning problems associated with 
digital systems.  Second, HICB performs a quarterly review of Nuclear 
Plant Reliability Data System data, Licensee Event Reports, event 
notifications, morning reports, inspection reports, and Nuclear 
Utilities Software Management Group news bulletins for digital system 
failure information.  The problems identified during this review 
period were added to HICB's digital system failures database.  This 
database contains software, hardware, electromagnetic interference, 
and man-machine interface problems associated with digital control and 
protection systems dating back to 1993. 

3.  RESULTS

Eleven software deficiencies were identified during this review and 
are shown in Table 1.  Three of these problems were associated with 
safety-related systems.  Descriptions of the problems associated with 
the digital data processing system at St. Lucie and the digital 
adjustable speed drive (ASD) modification at Washington Nuclear 
Project 2 are provided below.  

Three digital system hardware problems were also identified during the 
review and are shown in Table 2.  These were all associated with 
safety-related systems.  The problems involving the gas turbine 
generator at Millstone and the radiation monitor remote display unit 
at Diablo Canyon are described below.  

(rest snipped)

This is an NRC study/document, not one conducted by me. Some things are just too bizarre to be made up. I'm honest to God surprised that no one picked up on this before.

-- Anonymous, July 22, 1999

Hmmmm,

Does check for accuracy, and check for politically correct, mean the same?

-- Anonymous, July 22, 1999

I suppose I could post a .pdf copy of this appendix for anyone that's interested.

I'll second that emotion. :-)

-- Anonymous, July 22, 1999

Rick, I'm smacking myself upside the head. I own a copy of your book and obviously didn't look at the appendices, or didn't remember this, it's been so long. Duh! and Double Duh!!

And here I was, doggedly searching for a back door way to access a report I had right on my bookshelf. After doing some pinpoint searches I had just come up with another access (and gave the NRC a gotcha!) when I read your post. Round and round we go..

http://www.nrc.gov/NRC/REACTOR/DIGITAL/failure9_onhold.html

The only problem is that the link at the very bottom for "Complete Table of Digital System Failures" (the study mentions it's an ongoing process) gives you the same "page has been removed" notice. But at least a little of the info is accessible.

This study was not only U.S. centric. There is a Paluel (France) plant computer problem on the chart: "Reactor's authorized operating limits were exceeded for a prolonged period of time as a result of repeated errors in calculations, control failures, and the absence of implementation of corrective actions following the discovery of the calculation errors. Source: NucNet"

-- Anonymous, July 22, 1999

I've also discovered another page with the following contents:

TABLE 1: SOFTWARE DEFICIENCIES IN DIGITAL SYSTEMS

TABLE 2: HUMAN-MACHINE INTERFACE PROBLEMS IN DIGITAL SYSTEMS

TABLE 3: ELECTROMAGNETIC INTERFERENCE PROBLEMS IN DIGITAL SYSTEMS

TABLE 4: RANDOM HARDWARE COMPONENT FAILURES IN DIGITAL SYSTEMS

This is at:

http://www.nrc.gov/NRC/REACTOR/DIGITAL/fail-dbr3_onhold.html

-- Anonymous, July 22, 1999

Bonnie, you've found all of the documents related to this issue (the text and all four tables). Ya saved me a lot of trouble - I'd suggest anyone that's interested in this material go to the links Bonnie has provided, then save the html - because my guess is that this stuff may not be available for long.

-- Anonymous, July 22, 1999

The devil is in the details, as the saying goes. Looks like a bunch of Sorcerer's Apprentice stuff to me. :-) Now, before some resident experts jump all over me for saying that, I realize we have learned and continue to learn a lot of lessons and improve the procedures in this industry. Still, this stuff is not for those who are the least bit careless or have any lapses in concentration. Gotta really stay focused in this industry, or pay the price.

-- Anonymous, July 22, 1999

Bonnie,

You are so very good at digging up the info! Have you thought of entering the legal profession? If you'd like you could kick around the idea of mounting a legal battle against this sort of deception.

Here's the newly passed "Y2K Bill"

http://universal-law.com/y2k/fed/legislation/hr775.pdf

I'm thinking it would be a good idea to begin a logbook of events, deceptions, mis-representations, names and dates. Dissappearing documents are disconcerting. Keeping copies is probably important.

Any lawyers out there who can advise on a strategy here?

The beast called electric utility industry might be in need of a tamer.

Anyone interested?

Steve

-- Anonymous, July 22, 1999

Steve,

IMHO, before any legal eagle involvement could be put together, I think we will be seeing "nationalized" control of this sector by FEMA. I know, I know, I'm just hip shooting on that statement. But, that is what I think is going to happen early next year. And you can forget *any* challenges about the way things are run if that happens.

-- Anonymous, July 22, 1999

There's some explanation and commentary of the Y2K Liability Act at The Y2K Act provides some new protections.

As I said when the bill was passed, I think this was the worst thing that could have been done right now.

-- Anonymous, July 22, 1999

Rick, I'm smacking myself upside the head.

Bonnie, you cut that out! We don't want any injury to one of the finest minds in the Y2K field.... :-)

-- Anonymous, July 22, 1999

Well, hello everyone...glad to see a lot of the forum regulars still here with the password protection (gread idea, Rick).

Okay, so I go to the links provided, and as I'm reading them, I see the problems, but it is very difficult to tell how serious the failures really are. So I went to the areas I'm most familiar with in the "Hardware component failures" section.

The first one mentions seismic issues causing relay "chatter", which could result in inadvertent trips. I have some experience in this matter, and I know that because the contacts have to close so quickly when there is a real problem, that vibration can cause the contacts to "chatter". In the protection biz, this is a "relay security" issue--a relay operating when it should not. Now, the result is that the generator trips off line unnecessarily. This is not as bad as "relay reliability," when a relay doesn't operate when it should. The software example of the annunciator panel not illuminating when it should is an example of a device lacking reliability, and this item could be serious. But possibly tripping when nothing is really wrong is not as catastrophic as one might think at first blush.

The second item mentions a remote display, so right off I think this is not a critical function.

The third and sixth items mention power supplies. Any of you who work regularly with electronic stuff know that power supply failures are a relatively common occurrence. Undoubtedly there are backup systems in place when this occurs.

I have no idea the seriousness of the fourth item, regarding steam leak detection.

The fifth item about a specific protection system is mentioned several times, so it is obviously commonly known as a problem.

Many of these problems illustrate why the power industry was so hesitant to go from electromechanical to solid state equipment in the first place--reliability and security of the systems.

Anyway, just perusing the first six items on the list, leaves me wondering what this all has to do with Y2k. Only the fourth item concerned me, simply because I don't fully understand what it means. No date related issues were raised here. It certainly does illustrate the imperfections of such equipment, but really how severe are most of these items?

What I'm driving at is this: All human built and operated systems will be imperfect, and will have failures. Is Y2k a situation that will expose enough imperfections as to cause serious problems in the power industry? I have yet to see enough problems to think that this is a reasonable possibility.

-- Anonymous, July 22, 1999

Steve - I believe the version of H.R.775 you linked was the original version. It has changed substantially since that time. The one I believe was the final compromise version (which may or may not be identical to what Clinton signed) is the following:

http://thomas.loc.gov/cgi-bin/query/D?c106:5:./temp/~c106labzt8::

-- Anonymous, July 23, 1999

We discussed the NRC links above months ago here in this forum. Why the computer based failure lists have been removed "for verification" is a mystery to me, it doesn't make much sense, this is an old report thats been on for years. The official NRC response for removing the lists: "This page has been removed until it can be re-checked for accuracy. (Effective: May 5, 1999)" Interestingly, the dated paragraph at http://www.nrc.gov/NRC/reactors.html listing "110" reactors error still remains.

I flat out do not believe the above statement, there's a lot more to this. Why the sudden interest "in accuracy" on something thats been on the site for years? In all the years I have been using the NRC site, I have never heard of such a thing.

I'm not sure what's going on here, it may not be y2k related, but something is amiss, I smell it in the air. I have a hunch, and want to check into a few things.

FYI, there is a more detailed report on Nukes at http://www.nrc.gov/NRC/Y2K/Y2KReady.html

Regards,

-- Anonymous, July 23, 1999

FactFinder,

You seem surprised by this NRC decision. Perhaps you are starting to get a wee bit skeptical about some of the things being done. In the world of the pessimists this phenomenon is known as "A Whiff Of Smoke." Watch out, it can get you running for an exit.

-- Anonymous, July 24, 1999

"These were all associated with safety-related systems."

Simply amazing. Once this document is eliminated one could make a good case that there are no problems which might inhibit the safe operation/shutdown of a nuke. Duh. One must wonder how many other documents have been redacted or eliminated that we never saw. I know, I know, stick to the facts. I would bet my life (and by the way, I am and so are you) that this document was purposely removed. Remember the little story I posted a long time ago about the dork nerd who was remediating code at a Nuke plant in Illinois? Well, no one ever, ever responded to my frequent (3 per week for a month) requests for follow up. Now, either my email was defunct or someone was told to can the story and move on. Whether or not that young man actually existed is a question we'll never answer, but the fact that his story dissapeared so suddenly is not a mystery to me.

Thanks for the info Bonnie.

-- Anonymous, July 27, 1999

Ok, Don't want to leave this just hanging. The pulling of this report from the NRC to verify accuracy is just something I have never seen them do before. My "theories":

1. Removed to avoid unecessary ammo for the anti-nuke crowd to scare the public with regarding nukes and Y2k. I did some checking, too many other digital failure reports/data left for this to be plausible.

2. Removed as an "over-reaction" due to concerns of cyberterrorism. There have been serveral articles in the technical literature regarding this issue recently, I beleive that this may be plausible. While I believe there are no valid real threats to these closed systems, facts never get in the way of the goverment ( and the publics) over-reaction to a given "threat". I offer Y2K as a prime example. Bonnie, have you seen anything on the gov't sites recently regarding cyperterrorism? Rick, have you got a copy of this full table? As best as I can remember, there was a pretty long list of computerized systems/components at various plants that had experienced problems. I rate this as a strong "maybe".

3. Pulled due to NEI influence, for same reason given i 1 above. Possible, not likely.

4. Pulled due to a utility on the pulled table claiming erroneous information. Possible - but why now, after all the years?

5. Pulled due to NRC reviewing computer failure data and being concerned about bad information getting out to the general public, ie. pulled for very reason they stated. Bwahahaha....lol

6. Pulled as a CIA experiment into the "parnoia" factor of those frequenting EUY2K and TB2000. Not likely, no testing required, is self-evident..lol

7. Pulled for reasons that FactFinder can't even guess. Most probable correct answer.

I like 2 and 7.

Any others? :)

-- Anonymous, July 28, 1999

Factfinder, there are quite a few mentions of cyberterrorism and just plain old terrorism on various government sites, including that of NERC and the NRC. (It's addressed as part of their contingency plans for the rollover and as part of plans for increased vigilance in the future. In fact the FBI and the NRC have had established communication links and protocols for this type of thing for awhile now.)

A potential reason for pulling this info (like you, I don't buy the explanation in toto)which I've been batting around, has to do with lawsuits. It's been acknowledged for almost a year now that some law partnerships have formed Y2K litigation teams. I was thinking that since establishing a prior "pattern" is often part of a winning legal strategy, that perhaps this info was pulled out of fear that it might somehow be used against the individually mentioned utilities if they did experience digital failures over the transition? Then the "check for accuracy" becomes a phrase that casts doubt upon the validity of the data and makes it less usable in a court of law?

Your ideas are just as possible, though. Somebody may have just thought to themselves, "This isn't a good thing for the public to be reading right now," and took matters into their own hands. Who knows? I don't think we'll find out the reason unless somebody out there has a crystal ball - that really works! It's a frustrating mystery, though, isn't it?

-- Anonymous, July 29, 1999