'notes on R.Cowles' "Will the Lights Stay On in the Year 2000?" presentation on 2/26/1999
asked in the TimeBomb 2000 (Y2000) Q&A Forum
--------------------------------------------------------------------------------
I attended Rick Cowles' presentation Friday evening, with about 60 or so other people, at a church in Arnold, MD. I suspect that there will be a good write-up in Cory Hamasaki's next D.C. Weather Report, since Cory was there and taking lots of notes.
A lot of what Rick covered is what is already on his Electric Utilities & Y2K website, euy2k.com, so I thought that I would just share what seemed to be to be New Things that I had not been aware was real common knowledge. NOTE: What follows is my impression, and I apologize for any omissions, mis-interpretations, etc., in advance. I am sure that Cory's treatment will do a better job of covering this, assuming he does a write-up.
Three Mile Island Lessons Learned
Again and again, Rick pointed out that the way that Y2K news and info is being disseminated to John Q. Public today is following the same kind of way that Three Mile Island news was handled by the industry, by the Government, and by the news media. A lot of it tends to be unreliable because it starts out as guesstimations at low worker bee levels, rises up after considerable filtering and second guessing, then branches out from there for yet more filtering and second guessing. For example, recent NERC reports tend to paint a pretty optimistic picture, yet this coming week a Senate report will state that you can expect to have power disruptions. John Q. is going to have a tough time knowing who to believe.
Rick was especially critical of the news media, hesitating at this point to even take calls from reporters anymore. He feels that they just want to get a one-liner from him so that they can represent "the other side" in an otherwise highly optimistic Y2K news story. He noted that a few days in advance of the release of the most recent NERC report, he went through the same data presented in the report, and composed a list of questions that he felt should be asked, and disseminated these questions to reporters. Needless to say, the questions never got asked. "Investigative reporting" isn't. (He did note the one and only one exception to this that has appeared in the mainstream press: the Y2K article that appeared in the January issue of Vanity Fair, which he said was well researched, and written over a couple of months.)
Y2K Plus Other Stuff is the Problem
Its not just Y2K that looms over our heads, its Y2K plus a number of other things that either will be independent events in their own rights, or have already set things up so that the reliability of being able to get clean electric power is in a fragile, weakened state.
Two events that are coming close to the same time as the righteous Year 2000 computer problem: GPS rollover in August of this year; and the "Solar Cycle 23" cosmic event that will occur early next, which is expected to play havoc with the power grid.
Events that are much more subtle, and have basically occured over a period of years within the Electric Power Industry, are the result of judicial, government, and business management decisions: De-regulation, Competition, and Re-structuring/Unbundling. The bottom line is that the industry has (like everyone else) managed to lay off all the older (higher salaried) employees that knew how to fall back to manual procedures in operating a power plant (or, for that matter, just knew how things really work). This is knowledge that cannot be replaced; going back to The Way We Were is not an option.
Embedded Controllers / Embedded Chips
The industry is heavily relying on a lot of "type testing" to determine where their embedded systems problems lie -- i.e., testing a particular component and then, on the basis of whether it is or is not Y2K compliant, conculding that all other such "identical" components are or are not. This is Wrong, Wrong, Wrong -- one has to test all of them, since "identical" components may be composed of non-identical micro-chips.
In practice, Rick has only encountered one single chip, per se, that would fail due to a Y2K problem. However, there are numerous sets of chips working together at the level of an embedded controller, that will fail if not found and fixed. Again, its a matter of "finding and fixing", and the utilities are at various stages in their progress.
Nuclear Worries
The recent event at the Penn "Peach Bottom" Unit nuclear power plant, where both secondary and primary safety monitoring systems locked up during Y2K testing (read all about it at www.euy2k.com), should serve as a big Wake Up Call. Heretofore, the Nuclear industry has assured everyone that Y2K was not an issue in terms of their safety shutdown procedures (not to be confused with day-to-day power generation, of course), yet we seemingly have a very sobering counter-example to this claim. Further, according to Rick, only a couple of days afterward, the Nuclear Industry was still going on record as claiming that Y2K was not a safety shutdown issue, as if the Peachbottom Incident had never happened! (Think: Three Mile Island Lessons Learned.)
So, Will The Lights Stay On or What?
OK, basically assuming that other things remain functional (e.g., telecommunications, which Rick confirms that the Electric Industry relies on heavily), Rick believes that there will indeed be short term power disruptions. The State of Texas, which besides having its "own" power grid, is also a model for the rest of the industry in terms of its aggressive Y2K remediation, presumably would be in the best of shape. In the case of outages anywhere, he is betting that cities will get power back sooner than rural areas. Now, understand that this might be limited power -- half of a city for 12 hours, the other half for the next 12 hours, that kind of scenario. Further, the power might be "dirty", and cause problems with devices burning out after time, etc.
Rick does not see this all starting with a bang at midnight on 1/1/2000. He believes that there is enough fault tolerance built in to the power generation systems that problems will be initially absorbed, and to an extent, perhaps even hidden from plant operator. But, as per complex systems theory, there will come a point -- by mid January, say -- when the unmistakable problems will surface. Rick expects -- again, under the premise that all else remains constant and functional, such as telecommunications -- for blackouts, "dirty" power, and the like to persist for about a year.
Thats my report. I'll post my own comments later.
-- Jack (jsprat@eld.net), March 01, 1999
...
greenspun.com |
