NERC Utilities Update, Revised as of November 1, 1999 is now available at the NERC website, at http://www.nerc.com/~y2k/y2k.html

Two reports are available (in pdf format):

1) Y2k Ready and Y2k Ready With Limited Exceptions List

2) Exceptions List Summary of Non-Nuclear Exceptions Reports

Of the 295 reporting organizations ) 27, or 9%, are still not ready.

For those midwesterners among us...

Minnesota Power in Duluth is on the list of plants not ready.

Although Minnesota Power is listed as "Ready with Exceptions" its website says:

"We do not expect to see interruptions to our customers' electric service due to Y2K," says Andrew Lucero, Year 2000 Project Coordinator for Minnesota Power. "Our preparations include scrutiny of sensitive systems at power plants where your electricity is generated. On June 30th, 1999 Minnesota Power filed its letter with the North American Electric Reliability Council indicating that Minnesota Power mission-critical systems used to produce and deliver electricity are Year 2000 ready." Our optimism must be tempered by the fact that we can never predict outages. Power interruptions may occur due to weather, Y2K or other problems. For that reason there is no guarantee of uninterrupted service."

http://www.mpelectric.com/y2k/index.htm

-- Anonymous, November 01, 1999

Answers

I read the NERC Utilities Update and the NEI Nuclear Power Plant readiness report (at NEI.org), and have one important question about the about the NERC study:

Out of the 3000+ electric utilities in the U.S., were only 295 included in the NERC study? If so, what is the status of the remaining 2700 utilities?

-- Anonymous, November 02, 1999

Hi F. Snyder. For clarification, none of the utilities being monitored by the NERC are "not ready". Those 27 are "ready with limited exceptions," which means that at least one mission critical device or system is not yet ready, but that because of the small number of exceptions, the utility is expected to complete remediation before the end of the year.

B. Liebenstein: Yes, www.NEI.org is a good site to take a look at, because as of yesterday, all 103 nuclear sites are reported to be Y2k ready. This is an important milestone for the nuclear industry and Y2k, because earlier projections had several plants not completing remediation until well into December, making folks nervous. Hopefully this will alleviate some anxiety.

NERC monitored the 290 or so "control areas", or just the larger utilities that would affect the grids more significantly. The smaller utilities were monitored by the American Public Power Association and the National Rural Electric Cooperative Association. Both of these organizations are reporting the same kinds of results as NERC: With more than 90% of utilities reporting, the vast majority are completed.

-- Anonymous, November 02, 1999

I live in Fargo, North Dakota. It is frequently -30 degrees below zero for a week at a time here. Sorry, but "ready with limited exceptions" is "not ready". I can understand the rational, anyway, for using this designation for fixes all ready to be implemented during a scheduled outage. (At this date I think that's weak.) However, there is no justification for the designation at this time when a company is waiting on "vendor availability". That's flat out BS. Vendor Availability isn't going to keep anyone warm, so they are "not ready". It is the height of irresponsibility to delay someone else's contingency plans by lying about one's own readiness, which, IMHO, is exactly what the power industry has done. Just because an industry organization says its ok, doesn't make it ok. How anyone in the industry can claim any sort of readiness absent integrated testing is beyond me. NERC's sampling of a small fraction of generating utilities (even if they are large ones)and ignoring the other two legs is likewise indefensible. I certainly hope there are no problems. If there are, I seriously doubt that it will cut much mustard to say a plant was "ready with exceptions" and didn't finish because it ordered systems too late for the vendor to respond.

-- Anonymous, November 03, 1999

Hello again, F. Snyder. The point of my post was not to try to convince you of the validity of the NERC information. You obviously don't believe much of what comes out of that organization.

All I'm saying is that NERC has specifically defined what it means by "ready with exceptions", but when you made reference to the NERC report, you stuck in "not ready". NERC *does* make a distinction, and since you were quoting information from that source, I think you should be consistent with the definitions they use, and then add your personal viewpoint. Perhaps I'm being a bit picky here, so I'll stop at that.

I don't want to make an attempt to change your view of the NERC information, but I am curious about your reference to "integrated testing". Would you mind briefly explaining what you mean by that? Do you mean forwarding the clocks in all devices in a given substation, at both ends of a transmission line, or between utilities, or something else?

-- Anonymous, November 04, 1999

Maybe integrated testing would avoid more problems like this one

http://dailynews.yahoo.com/headlines/local/state/louisiana/story.html? s=v/rs/19991104/la/index_1.html#4

Maybe some will explain in detail how the transmission system, tripped this nuclear power plant off-line, what sort of design flaws, glitches etc in the tranmission, nuclear generator interface could have led to this error. Could nuclear power plants once tripped off line ever be expected to get back up and running in three days. Hmmmmm ... Hmmmmm.

-- Anonymous, November 04, 1999

Just to keep you from having to look up the link here's the whole text of the article (suppose I could have posted this under more good news)

Nov 4 "Nuclear Reactor Back Online - (ST. FRANCISVILLE) -- The River Bend Nuclear Power Plant is slowly coming back online after the reactor generator suddenly shut down five days ago. Plant engineers say the generator automatically jumped off the multi-state power grid. They believe repair workers tripped it from a nearby power transmission site. While officials reported no outages this time, Entergy directors say they are concerned after a summertime record of brownouts in south Louisiana.

I guess NERC needs one more category for their nuclear plant reviews, "Ready but non-functional for undetermined reasons"

-- Anonymous, November 04, 1999

Paul, based on the description you gave, this sounds like a crew at the remote site inadvertently activated the "transfer trip" scheme. This kind of thing happens pretty frequently. A communications technician takes out the wrong comm circuit, or an electrician bumps into a relay; there are innumerable ways to cause this to happen. This kind of scheme is commonly used on high voltage lines, whereby if one end of a transmission line sees a disturbance, it automatically sends a signal to trip the other end. Disturbances on high voltage lines can cause great amounts of damage if not isolated quickly.

Here's how Y2k might cause such an event. If a protective relay at the remote (non-plant) end failed because of Y2k and tripped the local end, the transfer trip scheme, which is essentially a "slave" system, would then trip the plant end. So because of the integrated nature of that system, the problem spreads from one site to another.

These types of systems have been tested repeatedly for Y2k. Not a single protective relay in the entire world has been reported to fail in such a way as to send a trip output. But just to be sure, many companies have tested those very devices in the field.

Several utilities have Y2k tested whole substations, even whole regions, simultaneously to see if failures would occur.

-- Anonymous, November 05, 1999

Dan - I don't pretend to have any technical knowledge, and I don't know that the term "integrated testing" is the correct one. But the securities industry and, I understand, the banking industry have had some industry wide tests where they exercise their Y2k fixes together using dummy data to test their interfaces, and determine that Company B which did expansion can still exchange information with Company A that did windowing. My understanding is that the power industry has decided against specific testing of this nature because of the risk it involved. That is what I have read. On one hand, one might argue that since it is all real time, testing is taking place real time. Perhaps that is correct. One thing that seems certain to me tho, is that at the time of the April 15th "drill" there was lots of code which had not been put back into production, yet people were generally left with the misimpression that the communications drill was the same sort of industry wide Y2k testing that the securities industry was doing. I grant that anyone who knew much about Y2k knew the difference, but that is a pretty small percentage of people. I wonder now what percentage of remediated code remains to be put into production in the industry. I am not saying that more testing was necessarily practical or possible in the industry. I am saying throwing alot of new stuff on line or into production in December raises risks. Those risks are not acknowleged by NSP's advertising campaign in my community to the effect that there is no greater risk of outage on 1/1 than any other day. If there is a problem, the injury to my community will have been increased by the campaign. Its not like utilities have to prevent a "run" on power, like the banks. Its not going to hurt utilities to do the best they can, say they are, and acknowledge this is difficult stuff that carries some risk.

Snyder

-- Anonymous, November 07, 1999

Dan,

Thanks for the indepth and thoughtful response. One of the bogeymen of Y2K is the amount of information available on Y2K through the internet. It creates a real possibility as discussed before for what I would call information selection bias, especially when one doesn't have personal experience or education in a particular area. That said I still find the news sufficiently disturbing as a layperson. The Chairman of the IEEE Y2K committee in his recent letter to Ed Yourdan talks of systems of systems testing. This looks impratical to impossible to me in industries such as telecommunications, electricity and the internet. Balanced against this is an amazing reliability record for the electrical industry with heroic attentiveness to electric reliability. Finally, the extent and magnitude of this common mode failure is not clearly predicatable across industries and nations, however, there is expected to be, as many here have pointed out, significantly less exposure in the electrical industry.

-- Anonymous, November 07, 1999

Hello again, Snyder. I wanted to respond to your last post...I didn't just drop the subject.

"Integrated testing" is a generic term, and I think it does include testing from utility A to utility B. Perhaps "inter-utility Y2k tests" is more descriptive.

Nevertheless, you raise a legitimate concern. There are many Y2k issues that could affect more than one company simultaneously. Examples are OASIS systems, whereby transmission capacity is scheduled from one utility to another. Also, Independent System Operators (or ISO's) deal with multiple transactions from several utilities. Communications systems between utilites (phone, microwave, PBX, etc.), or date-sensitive monitoring devices like Remote Transmitting Units (RTU's) are also important in inter-utility operations. And PD is correct in that some kinds of grid-wide tests just aren't practical, like opening major transmission ties that could cause outages accidentally.

By the way, as of November 8, both Northern States Power and Minnesota Power are reported as "Y2k Ready" by NERC. I hope your rollover is relatively uneventful....

-- Anonymous, November 15, 1999