I read a comment on the web from a person that was supposedly knowledgeable about embedded systems in process control systems. He said that it was likely that process control systems would stop and not start again or start and not stop as they crossed the 12/31/99 to 01/01/00 boundary. If this is so, and if embedded systems are as hard to find as some comments on the web indicate, how can the power grid possibly stay up?

I believe that if the power grid stays up we will have severe dislocations, but society may survive. If the power grid goes down (and stays down, which is like if it goes down at all) then we probably loose society altogether.

Given the foregoing, I currently feel that the power grid is going down, is going to stay down, and that civilization as we know it will be gone. Does anyone agree? Disagree? Can anyone argue that the power grid won't go down? Can anyone argue that even if the power grid goes down (and by this I mean blackout of more than 85% of north America)it will come back up? (I believe that if it is down for two weeks it is down forever.)

-- George Valentine (georgevalentine@usa.net), March 22, 1998

Answers

I think we should make the distinction between "the power grid" and electrical utilities. If by "the power grid" we refer to the darn-near 100% interconnected electrical generation and distribution system in the US, yes kiss it good-bye. To me it seems the rational thing to do on 12/31/99 is for ALL utilities to disconnect themselves from each other and to depend on just the electricity they themselves produce to supply their customers. I would break this down even farther, for companies like AmerEU, they should disconnect the old Illinois Power generators/customers from the old Union Electric generators/customers. From what I've heard this would go a LONG way to preventing the cascading of failures. Also having extra workers on board that day/night would seem like a good idea. However, would the lawyers allow such a disconnect? I doubt it. Would the unions allow scab workers in the plant? HA!

Y2K is a technical problem that is VASTLY complicated by idiotic cultural artifacts. Example, if the US monetary system did NOT use fiat currency, NOR fractional reserve banking, there would ZERO discussion about financial panic, collapse, etc. Y2K is merely exposing in high relief the flaws in our society. It is simply yelling, "the emperor has no clothes" very loud and it won't/can't be shut up.

-- Ken Seger (kenseger@primary.net), March 22, 1998.

I think we must plan based on the worst case since this would give us the best chance of weathering less severe situations. Regardless of the chances we should prepare for living without electrical power for a good 6 months or more. Unfortunately this problem will occure at the onset of winter. Those who live in northern climes will be the most affected.

Here is how I look at the odds for success. If one only considers power there are three main areas these corporations must remediate perfectly: Hardware, software and host to host interconnectivity. Each of these areas must have sufficient people, money and time allocated. Because all of these must be successfully corrected it doesn't matter whether most of them are or not. That's a 512 to 1 possibility of failure.

Time to plan on no electricity from Jan 1, 2000 till who knows when.

-- Charles Polk (c.d.polk@usa.net), March 23, 1998.

Yes, there is every chance that the electrical power supply will not "collapse." (I am taking the word "collapse" to mean the loss of electricity for the majority of the population for more than 24 hous.)

Will it operate smoothly? I doubt it. Are the chances high for a nationwide blackout high if utilities do not manage to uncouple themselves from the interconnected grid that has already turned local problems into national ones? Yep. Will it happen? Probably. Will the folks at the utilities find a way to overcome the problems well enough to get power flowing again to most people in a realatively short period of time (12-48 hours)?

I believe the answer is yes.

Will civilization collapse? Only if we let it, or if we hasten the collapse along by assuming that collapse is inevitable and then reacting as if it has already occured.

-- Paul Neuhardt (neuahardt@compuserve.com), March 23, 1998.

I have been researching this question as a lay person. As a side note, it would help if the general public can just view the discussions among the power utilities. Being locked out entirely of these discussions makes research quite a lot more difficult.

My data comes from a power systems engineer who recently retired as an engineering manager working on contract from ARAMCO the Saudi oil company, for SCECO the Saudi Arabian power utility.

About twenty years ago, typical power plant excess generation capacity was 15%. Today, that figure is 8-10% if the plant is conservative, 10-12% if aggressive. "Conservative" means the plant is not expected to generate increases in revenue, whereas "aggressive" means the owners of the plant expect significant and sustained increases in revenue over time. Deregulation will bring this figure down even further.

Redundant power delivery is typically carried out by two 60% capacity lines (each line is rated for 60% of total power output of plant), or 75% if the utility has the capital resources. Identical equipment is used whenever possible to reduce maintenance costs. Most control systems associated with power delivery are located at power plants.

The critical master console at the power plant is called the Boiler Turbine Generator (BTG) control board. In nearly all cases, it offers manual overrides. Automation does not imply manual operation is impossible. Control systems typically operate very slowly because the machinery often has extremely long cycle times. For example, a turbine at a medium plant will take about 30 minutes to spin down, and several hours to spin up. Thus, total manual control is quite possible at power plants, and manual intervention by operators is even required during unusual peak load/no-load periods.

Fuel reserve capacities of power plants depend upon the nature of the fuel. The trend is towards just-in-time (JIT) delivery, because reserve stocks tie up capital. Coal transportation is difficult, so coal-fired plants are co-located whenever possible. Natural gas is transported by pipeline, so most gas-fired plants are plugged next to the pipe. Oil reserves are a minimum of several days (maybe longer in the U.S. to account for potential problems with labor strikes as another participant has posted).

With this information on hand, I need to refine it with more data. I read that an estimated 30% of the approximately 9,000 power plants in the U.S. are not prepared (or is it power _utilities_?). I need to know what the excess generation capacity of that 30% population is. I need to know what the Y2K exposure of power delivery systems are. If we can be reasonably certain that power delivery will not be irreparably damaged, and the combination of total manual control and excess generation capacity can make up at least 50% of the difference of lost power plants, that leaves us with about a 10-15% shortfall in power nationwide. Not good by any measure, but at least not approaching complete loss scale catastrophic and a viable foundation from which to rebuild the economic damage.

If things start to look really bleak next year, I suspect triage and contingency operations plans will be the watchwords of the day, and steps will be taken to ensure first the ability to distribute and then generate on a priority basis necessary infrastructure services and supplies. For example, if necessary, the U.S. Navy can commandeer supertankers---even dead-in-the-water ones due to Y2K control system problems with no manual override options (a pretty insane scenario, speaking from a systems engineering perspective)---refit the critical power and steering control systems for manual overrides, and sail them by speaker phone, sextant and compass if necessary. It won't be pretty, but it is a damn sight better than a complete cessation of civilization.

-- Anthony Yen (tyen@netcom.com), March 23, 1998.

Mr. Neuhardt, would you please tell my why you believe that the power grid can be brought up in 48 hours or less? If there is a cascade of power outages that brings down a large portion of the grid for even a little while, won't it be very time consuming to bring it back up? Look at the ice storm in Canada and at Aukland to see why I'm worried. Neither of those was coped with in two days.

-- George Valentine (georgevalentine@usa.net), March 23, 1998.

Mr Yen, I have two questions. First, do you really belive that the Navy (or anyone else) could retrofit a computer-controlled supertanker to be manually operated in any meaningful timeframe? Second, are you aware that some people on the web consider it likely that all of the nuclear power plants (about 20 - 25% of the nation's power generation) will be shut down by the NRC for non-compliance because of y2k shortly before 1/1/00, resulting in as much as a 40% shortfall in your analysis?

-- George Valentine (georgevalentine@usa.net), March 23, 1998.

Lest you think that I am looking forward to some kind of disaster, let me mention that I very much want things to continue more or less as they are. Nevertheless, I do believe that if the power grid goes down and stays down for two weeks or more, it will probably be down for a lot longer than that. I also believe that this society cannot function without electricity. And I believe that the electric utilities are not giving the "embedded chip" problem the consideration it deserves (although, like Mr. Yen, I do not have access to the closed forums.)

-- George Valentine (georgevalentine@usa.net), March 23, 1998.

Addressing Mr. Valentine's issues; thanks for the feedback. If the scenario is desperate enough, then yes, I can forsee bringing in the talent to bear on those systems and rip out their points of control enough to bring back manual operation, if it even necessary in the first place. As I mentioned, a control systems engineer who designed such a system would have had to be almost criminally negligent. It is not as if we had to create a jet fighter with fly-by-wire systems, and the cruise ships I have been on were quite capable of manual operation in the event of some accident severing links of their control systems. This jiggering will be nasty, ugly and will inevitably lead to accidents, but it is not impossible.

Nuclear power plants supply 20-25% of the U.S. _East_Coast's_ power needs. This probably makes up the majority of the 30% nationwide unpreparedness figure. Although I stress I still have not confirmed whether that is 30% of power plants or power utilities---the latter is worse than the former because utilities frequently own multiple plants. This is why confirming the ability to distribute power on such a massive scale is so critical: if we can make up the severe shortfall on the East Coast (where a significant fraction of the population is located) by delivering from other parts of the country, the impact will be kind of absorbed evenly. I believe my first-order analysis still holds, but if anyone has contradicting source data, please let me know.

-- Anthony Yen (tyen@netcom.com), March 23, 1998.

Why do I believe that power could be restored within 48 hours? Because it has happened before.

This country has experienced massive outages over large areas caused by local incidents before and recovered within hours. Less than two years ago my father in Pocatello, Idaho suffered the same outage felt by his father in Amarillo, Texas and by my company's home office in Irvine, California. The root cause of the problem: a downed transmission line in Idaho. The "cascade" effect of overloads and attempts to draw power from other areas crippled much of the western half of the country for hours. Now, if one line in Idaho can do this, how can we survive Y2K, which will no doubt affect more than one single line?

The solution remains the same. Careful, reasoned analysis quickly identified the source of the otage and the progression of failures. By bringing the downed systems back on line in an orderly, controlled manner, service to millions was restored well within 48 hours. If the reserve capacity in the generation systems is even partially as large as as has been represented here, this same type of analysis should be able to cover Y2K-related outages within that same time frame. Okay, maybe there will continue to be isolated outages and brownouts, but electricity should continue to be available to most of the population.

BTW, using this winter's outages in Canada and Maine as an example is not appropriate. Those long-term outages were due to the physical collapse of the delivery system's infrastructure. It didn't matter how much or how little electricity was generated as there were no physical lines to carry it. If we can assume that generation capacity will continue at all, and I know some would not make that assumption, then this situation is so vastly different that it does not apply.

-- Paul Neuhardt (neuahrdt@compuserve.com), March 24, 1998.

Paul, I don't think the example you sited re the down transmission line is an appropriate comparison to potentail Y2K problems. The example you gave regarding bringing the system back up was caused by a single point of failure in transmission and did not affect the generation side of the electricity equation. If Y2K is as big an issue as some believe, there will be multiple points of failure across the entire north american network.

The big question in my mind is how much generation has to go off-line before it cascades across the entire network if the different jurisdictions don't decouple from the grid pre 13/31/99?

If it does fail, I believe it will be a massive logistics nightmare to coordinate the required resources etc to bring things back up. If plants trip off line because of Y2K problems, all of that plants problems must be fixed before they can be brought back on line.

-- Patrick Galpin (pgalpin@stratum-group.com), March 24, 1998.

Yes, if multiple power generation facilities fall off-line then the problem is slightly different than a line going down in Idaho. My position is that it may not be as different as you would first think.

Assumtion: Enough power plants will be able to produce electricity to cover the loss of those that cannot. For argument's sake, say that Y2K wipes out 40% of the generation facilities of the system. If that happens all at once, 49 states and all the Canadian provences go dark. (Hmmm, wonder where all of this leaves Hawaii?). What I am saying is that the remaining 60% of the facilities have enough reserve generation capacity to keep 21st Century civilization going until the remaining problems can be licked. Maybe we don't get a long as well as we would like, but we do survive.

"But Paul, what on Earth makes you think that only 40% of the power plants will go off line? Why not 100%?" Do I have facts to support such a position? No, but so far no one has presented with me facts showing that this won't be the case. Have I seen facts showing that the power utilities are woefully unprepared for Y2K? Sure have. But none of the facts that I have seen convince me that *nobody* will be able to produce electricity. Beyond that, I will admit that only thing I have is faith that the cunning and survival instincts of the human race as a whole will carry us through the crisis.

-- Paul Neuhardt (neuhardt@compuserve.com), March 24, 1998.

Take a look at: http://www.albany.net/~dmills/fallback/fintro.htm and follow the link at the bottom of the page to "Chapter 5".

This is an article by an expert in power systems engineering who is critical of some of the comments in the Yourdon's book. Here is some of what he says:

"The point is, the calculations for regulating power generators and power systems are in-general not date sensitive. Indeed, the regulating computer programs don't even need access to the date for any purpose. Hourly, daily or seasonal variations in demand have nothing to do with it. That's terribly important."

He goes on to discuss the details of the protection and regulation functions in power generation equipment, and concludes:

"The reason I'm explaining this here is that Fallback's premise that Y2K caused problems might damage generators and other equipment, thus resulting in month or longer outages, is not a reasonable extrapolation. In my opinion, Fallback's statement, [2 day outage likely, up to one month possible], is unfounded. It shows ignorance of protection versus regulating functions."

As for the proposed "ripple effect", where failures in one part of the system propagate to others, he points out that failures occur constantly throughout the system, and it is designed to continue to function in the face of failing devices:

"I can indulge in a little hyperbole of my own. Given the many millions of components in the national power grid, and the thousands of ways that each may fail (including Y2K induced), the normal state is that the power system is challenged by failures thousands of times every day and survives most of them. Stressed by weather, especially ice storms and hurricanes, the failure rates rise millions of times higher. The ripple feedback loops are in effect continuously. I challenge anyone to come up with a substantial calculation that shows that the morning of January 1, 2000 will stress the power grid even as much as a typical ice storm."

Reading this commentary makes it clear that there is an excellent chance that the power grid will ride out the Y2K problem in good shape. And if power stays up, many of the other aspects of worst case scenarios won't be a problem, either. People can continue to go to work, gas continues to be pumped, factories continue to run.

Information from knowledgeable experts like this goes a long way to calm Y2K hysteria.

Pat

-- Pat Smith (pattsmith@hotmail.com), March 24, 1998.

Paul, as you have a Faith Position that the Human Race itself will carry us thru this crisis because of its cunning and survival instincts; do you think that it will be those all those members of the Human Race collectively living in the geographies formerly served by the now failed 40% (what is that: over 3000 plus?) of the generating stations that will be excercising their cunning and survival instincts or just as likely sub-sets of the race such as armed youth gangs for example?

If these areboth soley Faith Positions, how can either position be evaluated by anyone to help draw personal or collective survival plans? I don't know how far faith assertions get us, either individually or as communities.

-- Victor Porlier (vporlier@aol.com), March 24, 1998.

Pat, I am not sure what y2k "HYSTERIA" you are referring to. If we are both using this word the same way, I have yet to meet or read anyone in print or on this bb, who could be described as acting out of:

"a wild and uncontrollable emotion or excitement sourced in a functional disturbance of the nervous system of psychoneurotic origin".

If I have missed such articles or posts, could you point them out for me? Thanks.

-- Victor Porlier (vporlier@aol.com), March 24, 1998.

Victor, if you are willing to accept that power plants are so interconnected that a failure one place may mean failure in other, geographically distant places, why not accept that success in one place will translate to success in other places?

In other words, the same "grid" that linked power plants and distributed power before will still exit and will be able to distribute power from the 60% of plants in operation to areas normally covered by the failed 40%.

The issues really are:

1) will the "grid" still be able to distribute power generated anywhere to anywhere, and 2) will there be sufficient power to distribute.

-- Paul Neuhardt (neuhardt@compuserve.com), March 24, 1998.

Pat, I followed your link and read what the man said. I will quote part of it next:

*** To be completely candid, there is one specific kind of date sensitivity that all embedded continuous regulating systems share; electric power related or not. That is the calculation of control action as a function of elapsed time. In most cases, elapsed time is calculated from a series of pulses generated at regular intervals, like ticks of a clock. For example, a program that runs once per second may assume one second's worth of action per execution, or it might measure the time now, and subtract the time at the previous calculation. In theory, some controllers might do the subtraction with date sensitive calculations, and thus be subject to one-time massively large errors in the amount of elapsed time. Even so, most of the errors would be benign in that nothing would happen, or that independent checking algorithms or physical limits restrict the effect of the error. Nevertheless, programmers will have to dig through the elapsed time calculations to find out just how the elapsed time calculations do work.

Use your home computer thermostat as a n example. Suppose it goes crazy and orders the furnace to use zero fuel for the next minute, then returns to normal? Not much damage right? Suppose it orders minus 50 gallons to be burned or plus 50 million gallons to be burned in the next minute. The fuel pump is probably only capable of delivering 0 GPM off, or 0.2 GPM on. The number of ways the thermostat can go bad, or how erroneous the thermostat's order can be, has no bearing on the actual consequence.

Try the sanity check yourself. Think of a cruise control in your car, that adjusts the throttle every 1/20th of a second to hold constant speed. What is the worst case consequence you would expect resulting from an bogus calculation of a single 0.05 second time elapsed interval?

***

"Even so, most of the errors would be benign in that nothing would happen, or that independent checking algorithms or physical limits restrict the effect of the error" This statement, I contend, may be wrong. As I stated in my original post, someone else (and I can't remember the source or I would put a link to it) someone else said that the likely effect of y2k on embedded process control systems was to start a process and never stop it or stop a process and never start it. I don't think that these kinds of actions would be benign. Rather, I would expect them to cause the protective systems to fire and shut down the generation plant.

"Use your home computer thermostat as an example. Suppose it goes crazy and orders the furnace to use zero fuel for the next minute, then returns to normal?" It is my contention that it is more likely that the system will say turn on (or off) the heat and will not change that order unless and until there is manual intervention, that is, it will not return to normal by itself.

"Try the sanity check yourself. Think of a cruise control in your car, that adjusts the throttle every 1/20th of a second to hold constant speed." Suppose the controller had just decided to accelerate. The command would be apply more gas. There would be no countermand. Very quicly the car would attain terminal velocity, whatever that happend to be. If the dirver did not panic he could probably shut off the ignition, and as long as some other embedded system didn't override that, the vehicle and its occupants would probably survive. But that is the kind of thing that I feel can (and unless consdierable remdial work is done, will) happen.

It is my understanding that Intel 286 (and maybe 386 and 486) chips are used in embedded systems (along with many others). It is also my understanding that their real time clocks count seconds since 1/1/1980 and are formated more or less into yymmddhhmmss.fffff. This could be totaly erroneous, but if it is true, then the stop and don't start or start and don't stop scenarios seem guaranteed to happen.

If anyone can straighten me out on this, please do! I am looking for a reason to believe, but so far I have not found it.

-- George Valentine (georgevalentine@usa.net), March 24, 1998.

Pat (pattsmith@hotmail.com) talks about Dick Mills' fallback critique. That is quite old. See what Dick is saying these days.

The best reply to your comments is to let Dick Mills' subsequent words speak for itself. Note his sentences "There are many more... and The problem with..."

Add up 2 and 2 and you get a good story. Number one is this posting on the CPSR list look at the last paragraphs (entire email of Dick Mills follows below at end of letter):

>BTW. My business with Y2K work is, unfortunately, not a sign that the dam >has broken. There are a few giant companies who account for much of the >contracted services. There are many more who ought to be doing work >urgently but appear to be doing nothing. The indications are that the >triage process is taking place on that basis. Those that seek help early >get it. > >The problem with that prioritization is that it does not serve the most >critical and most needed infrastructure industries first. > > >-- >Dick Mills http://www.albany.net/~dmills

Then look at his resume at his web site and see where he is working now and who his clients are:

http://www.albany.net/~dmills/rjm.htm

EMPLOYMENT HISTORY

1997- Digital Equipment Corporation; Albany NY

Technology Consultant

I am presently working in The Americas Year 2000 Expertise Center for Digital, as a contractor from ASI. My duties include business development, tools development, and project work for Y2K projects centered around electric power.

Roleigh Martin http://ourworld.compuserve.com/homepages/roleigh_martin

-- Roleigh Martin (marti124@tc.umn.edu), March 24, 1998.

Oops, I forgot to post Dick Mills' complete CPSR posting here, since his posting went out to a popular, national listserv, it certainly was not private email -- Roleigh Martin (Dick, if you did not want this reprinted, I apologize in advance. But you did make a nation- wide posting that night.) P.S. It is really only his ending paragraph that pertains to the topic under discussion here and I quoted that before in my earlier posting.

|X-From_: listserv-reply-errors@snyside.sunnyside.com Thu Feb 26 21:42 CST 1998 |Received: from mhub2.tc.umn.edu by garnet.tc.umn.edu; Thu, 26 Feb 98 21:42:04 -0600 |Return-Path: |Received: from snyside.sunnyside.com by mhub2.tc.umn.edu; Thu, 26 Feb 98 21:42:03 -0600 |Received: (from listserv@localhost) | by snyside.sunnyside.com (8.8.8/8.8.8) id TAA16012; | Thu, 26 Feb 1998 19:40:36 -0800 (PST) |Date: Thu, 26 Feb 1998 19:40:36 -0800 (PST) |X-Authentication-Warning: snyside.sunnyside.com: listserv set sender to listserv-reply-errors using -f |Errors-To: listserv-reply-errors@snyside.sunnyside.com (Listserv processor) |Reply-To: kurlandn@att.net |Originator: cpsr-y2k@cpsr.org |Sender: listserv-reply-errors@snyside.sunnyside.com |Precedence: bulk |From: Norman Kurland |To: "Multiple recipients of list cpsr-y2k@cpsr.org" |Subject: URGENT: Webmaster Needed; Mills

|X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc. |X-Comment: CPSR Year 2000 problem discussion list |X-Info: For listserv info write to listserv@cpsr.org with message HELP |Message-Id: <1373123584835.LTK.026@cpsr.org> |X-Message-Id: <1373123584835.LTK.026@cpsr.org>

Received: 02/26 10:37 PM From: Dick Mills, dmills@albany.net

Help please. I'm nearly 3 weeks behind reading my email. It's nearly impossible to keep up. In addition, the Y2K crunch at work really hit big, so there's not good prospects of things improving within the next 22 months.

I apologize to not reading incoming email messages. I hope I didn't cause too much frustration.

I need someone else to volunteer to take over as the CPSR Y2K webmaster. I spoke to Norman by phone and he said there may be some willing volunteers. Sorry, I don't know who you are.

If you would like the job, please call me at work. (518)452-7256. I'll pass you the needed information over the phone.

BTW. My business with Y2K work is, unfortunately, not a sign that the dam has broken. There are a few giant companies who account for much of the contracted services. There are many more who ought to be doing work urgently but appear to be doing nothing. The indications are that the triage process is taking place on that basis. Those that seek help early get it.

The problem with that prioritization is that it does not serve the most critical and most needed infrastructure industries first.

-- Dick Mills http://www.albany.net/~dmills

-- Roleigh Martin (marti124@tc.umn.edu), March 25, 1998.

Pat, I would encourage you to compare Mr. Mills' resume to Rick Cowles (www.euy2k.com) and study Mr. Cowles analysis on his excellent web site. His site does an excellent job of refuting the "electricity generation and control systems don't care what date it is" argument.

-- P. Larson (ptrades@earthlink.net), March 25, 1998.

I'm the Network Operations Manager of a regional ISP in one of the three largest metropolitan areas in the U.S.

[Aside -- I was marginally disappointed to see that my quote in Time Bomb 2000 was cut between 2nd draft and printing. Ah, well. ;) ]

I don't know much about how the "power grid" or how power distribution occurs, but if it's anything like the way connectivity happens on the Internet, it's safe to assume that the lights are going out -- and staying out.

It's a very vicious circle:

Q: What keeps electrcity moving? A: Computers, either through multipurpose or embedded controllers.

Q: What powers the computers? A: Electricity.

In short, if the power goes down, all the IS/IT staff in the universe won't matter a hill of dog do-do -- they'll have nothing to do. And if they have nothing to do, nothing gets fixed. If nothing gets fixed, they's still have nothing to do.

Even assuming you could run certain key systems on backup generators long enough to make repairs, you're talking about an extended (years?) period of time, during which the people responsible for fixing this mess aren't being paid -- because their checks cut and salary records are kept by the computers which aren't working.

How long will the average person work without pay? Right up until they get hungry, that's how long.

Now the $64,000,000,000.00 question (adjusted up for inflation):

Will the power stop?

I can only speak for the area in which I live. My company is an ISP, and one of our clients is a major electric utility -- in fact, the electric utility that serves the entire nothern half of my state. In my job capacity, I got to know the head of IS/IT for the nuclear division and asked him what he thought the chances of the lights going out were.

He told me -- off the record, remember -- that their Y2K staff is small, that they don't know how many lines of code they have to fix, and that even if they found them all, it was embedded systems that were the killer. To catch them all before Year 00, they would have to check every single black box in their inventory, including those from half a century or more ago.

He absolutely guaranteed me that they wouldn't make it. He didn't know what that would mean in terms of specific power failures, but he didn't deny that there would be failures.

Basically, he said they'd find out where the problems were when they broke.

In a dark kind of way, I'm reminded of the movie Dark Star, a student film directed, produced, and acted in by a lot of people who went on to some pretty extraordinary things.

The movie is about the crew of the spaceship Dark Star, on a twenty-year mission to destroy unstable planets around the galaxy. Some fifteen years into their mission, the ship is falling to pieces (just prior to the beginning of the film, the entire ship's supply of toilet paper was lost in space), and the crew is so bored and sick of each other that they don't care.

Finally, a missile launch system malfunction light shows up on someone's console. When the Captain is informed, he shrugs and says, "Ah, don't worry about it. We'll find out what it is when it goes bad."

Unfortunately, it "goes bad" during a missile launch. While the crew is able to abort the launch, the bomb (equipped with artificial intelligence and very enthusiastic about exploding) decides that it must be God "lets there be light" inside the launch bay.

The ship is destroyed and the aforementioned Captain (an avid surfer who happened to be EVA when the ship was destroyed) locates a surfboard-shaped piece of debris and surfs on into the atmosphere, burning up on re-entry.

Why is it that I feel like the guy with the blinking red light on his console and everyone with any power to stop the coming conflagration is the captain?

John Smith

-- John Smith (pobox42@hotmail.com), March 25, 1998.

In Northern Illinois, Commonwealth Edison generates most of its electricity by nuclear power plants. Guess what the chances are that these plants will be shut down prior to Dec. 31, 1999? The third larges metro region is going to be in the dark. And these plants are old, and I suspect the computer code is pure legacy. Maybe if we are lucky, Wisconsin Power will let us plug in a very long extension cord....

-- Dennis Sherwood (emdesher@iconnect.net), March 25, 1998.

I retract my analysis, after researching further. I found some definitive information on nuclear power generation reliance. Ed Yardeni gives specific information at:

http://www.yardeni.com/y2kbook.html

citing the Nuclear Regulatory Commission (NRC) information at:

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

of which the relevant portion is reproduced below:

In 1995 U.S. net electric generation totaled approximately 2,994 thousand gigawatthours. Approximately one-fifth (22 percent) of the Nation's electricity was generated by 109 operating nuclear reactors in 32 States. U.S. electric generating capability totaled approximately 706 gigawatts. Nuclear energy accounted for approximately 14 percent of this capability. There are currently 110 commercial nuclear power reactors licensed to operate in 32 States.(1) Six States relied on nuclear power for more than 50 percent of their electricity. Thirteen additional States relied on nuclear power for 25 to 50 percent of their electricity.

I had incorrectly asserted that I had read that approximately 20-25% of the U.S. East Coast's generation capacity was nuclear. This data clearly refutes that.

If we presume at best 50% capacity of all nuclear plants, then that still leaves us with a large shortfall for excess generation capacity to make up. I'm at a loss now; I don't know how to objectively evaluate the preparedness of the nuclear plants. It is all well and good to point out that they have weaknesses, but we need statistically-valid data to make a rational analysis of the situation.

Note that for all we know, my first-order analysis as presented earlier might still be correct. We simply have no data which supports it, now that we know the percentage of nuclear-fired plants (which invalidates the relatively light importance attached to the generation capacity of nuclear plants); it is an incomplete analysis. I am now looking for data which either pushes us toward refuting or confirming the first-order analysis.

Thanks to Mr. Valentine's comments which spurred me on to go look for definitive data.

-- Anthony Yen (tyen@netcom.com), March 25, 1998.

Dennis... Don't look to plug an extension cord here in Wisconsin. Madison Gas and Electric is offering generators to local business "in case the power goes out this summer." That soon you ask??? Our nuclear power plant at Point Beach has been down for sometime now and they don't have enough power to supply us. I called MG&E today and asked if they were Y2K compliant yet and was told "no way, we have millions of lines of code to go through yet." He also told me that they were likely to miss a few embedded chips and that a home generator was a good idea.....just in case!

-- Gail (jmtomich@students.wisc.edu), March 25, 1998.

Well, this has certainly been a fascinating thread to follow during the past several days. Very civilized (unlike the flaming on many discussion groups), very thoughtful, very probing ... but also very diverse in terms of the assumptions and conclusions. I hope we can keep this going, but it seems to me that there are a couple of interesting points already.

First: if a group of Y2K-aware, computer-literate folks like us can't reach a consensus on the threat to utilities, why should we expect the general population to reach a consensus? All of this requires far too much thought and attention; it seems to me that it's much more likely that they'll ignore all of this and focus instead on the titillation-du-jour until (a) 1/1/2000 passes without any problems, or (b) the lights go out.

Second: no matter how long this discussion goes on, I don't think we'll EVER reach consensus. The situation is far too complex, and we have no historical data to fall back on -- i.e., there's never been anything like Y2K in which a pervasive, ripple-effect kind of failure might (or might not) occur. So, at the end of the day, we're all going to have to make our own personal assessment of the situation. Your assessment might not be the same as mine, and we should realize that in advance, and show proper respect for all points of view. The key thing to remember is that we're each responsible, first and foremost, for our own safety and for that of our friends and family. If we happen to convince a few other anonymous strangers on the Internet that our world-view is the right one, terrific -- but that's nowhere near as important as making whatever decisions are necessary to safeguard yourself and your family.

Paul Neuhardt, for example, is apparently going to stay in Boston, with his wife and two babies, and hope that the utility disruptions last no more than a couple of days. I know Paul, and I respect his opinions; I sincerely hope he's right, and would be thrilled if he has the chance to say "nyah, nyah, nyah" for several years after 1/1/2000. But there's no way that I can assume responsibility for the decisions that affect his family; I have enough trouble dealing with the (potential) life-and-death decisions affecting my own family. As for me: I've sold my NYC apartment, bought a house in New Mexico, and will be installing solar panels on the roof this summer. The house has six fireplaces, and I'll be stockpiling firewood during the next year. My plan is based on the assumption that we're going to have serious outages for at least a month, followed by intermittent failures, brownouts, and "dirty" power for several months after that. I may be wrong, and indeed hope that I _am_ wrong ... but the more I study the situation, the more I think I'm right.

If Y2K turns out to be a serious mess, I don't expect to profit from it; I don't even expect that my family will be completely comfortable. But I will have taken what I now consider (with full understanding that my knowledge is incomplete and imperfect) to be the prudent decisions to maintain my family's safety. They won't freeze, and they won't starve, and I hope they won't be subjected to the kind of anarchy that I think will take place in our major cities.

God bless each and every one of you -- let's keep the dialogue going, because it will provide more understanding and insight, so that we can all make whatever decisions we feel is best. I don't expect any leadership, insight, or even honesty from the utility companies or our national leaders; but I find this discussion amongst my intellectual peers to be of enormous value.

647 days to go.

Ed Yourdon

-- Ed Yourdon (yourdon@worldnet.att.net), March 26, 1998.

The issue of losing electricity is vital in my mind. I agree that on this one point hinges the amount of time it will take to recover.

What can we as laypeople do? In the matter of awareness we can ask/question/yell/prod /dig/suggest/ and just generally make a pest out of ourselves with the utilities. Let them know we are SERIOUS and the problem is SERIOUS.

Now, why do I think this will make a difference? From my personal experience. In speaking with a representative of MY local electric utility I found them just as disorganized as any other large business. One guy did this work, another gal did that stuff, and NOBODY was co-ordinating the Y2K repairs!!

To quote my source: "Nobody has realized just what all this means yet" and "I have suggested several times that we get a coordinator but it was never done".

The overall impression I got was one of desperation on some parts, and indifference on others. It is very notable that my pipline closed completely with one terse letter. It said, to paraphrase, "We are working on this and don't bother calling anymore" Asking around I heard that all the local utilities had gone quiet. Nobody was talking to anyone anymore.

I must mention one more thing. Not one person I spoke with at our utility knew what I meant by contingency planning. I found myself explaining the term on several occasions. That is NOT a good sign.

-- Art Welling (artw@lancnews.infi.net), March 26, 1998.

I'm not 100% comfortable with the answer I got from my electric utility either now that I realize that they no longer generate the power. It seems my electric company sold all it's plants when deregulation came along, and now they are simply a power distributer. They have effectively circumvented Y2K responsibility by selling the primary source of the problem.

Yeah, yeah, I'm the guy who keeps saying "Don't worry, be happy." I still don't worry about being in the dark for an extended period of time, but I do begin to question some of the economics of the situation. If there is limited power available when the lights come back on (assuming they ever go off ;-) ), who is someone like Con Ed going to ship their power to first? Their own customers or some group in New England that decided to increase their profit margins by giving up the most expensive part of the business, namely production?

This is ceratinly food for thought.

-- Paul Neuhardt (neuahrdt@compuserve.com), March 26, 1998.

Ed, I promise never to say "nyah, nyah, nyah" or even "I told you so" to you. Ever. Period.

You see, I was once thought you were an alarmist reactionary when you wrote in "Decline And Fall Of The American Programmer" that a lot of programming work was going to move off shore to India, especially to places like Bangalore. Guess who I am contracting with to do a large chunk of my new payroll system implememtation, and who my company had do a significant part of our mainframe based Y2K work? InfoSys Technologies out of Bangalore, India. You know what? These people are GOOD, not to mention cheap. I would use them again in a heartbeat.

Besides, as I make it further and further through "Time Bomb 2000" I find more and more food for thought. Yes, I'm gonna stick it out in Boston (actually about 30 miles northwest, but that's close enough) and hope for the best. But I'm probably going to have money stashed in the house, more canned goods in the pantry and more firewood piled in the side yard than normal. I might even buy that generator I've been eying since that December, 1996 snowstorm left me in the dark for three days...

-- Paul Neuhardt (neuhardt@compuserve.com), March 26, 1998.

Paul N., When you say that your "electric company" no longer generates their own power, you made me think of something. In a case like that, where does one company start and one company stop? Does your "electric company" start (ie. own the wires, relays, transformers, etc.) at the power plant gate and outward, or does it just service the billing? I believe on Cowles web site he made a point that control systems with their embedded chips are part of the problem. Wouldn't it be a kick if the power plants stay up but the distribution system goes down?

-- Ken Seger (kenseger@primary.net), March 26, 1998.

Most power companies have three division, power generation, power distribution, and customer service (for lack of a better term). The reason for this is the potential de-regulation of the electrical utilities and you being able to choose your own electric company. This means that a power company like the one mentioned above probably doesn't have a power generation division since they purchase their power. This lends itself to the power company staying up, but the generating company going down and vice versa - could get very ugly.

-- Rebecca Kutcher (kutcher@pionet.net), March 27, 1998.

As best I understand it, Massachusetts Electric ("the power company") has divested itself of all of it's generation capability. It is now solely a purchaser, distributer and reseller of electricity. In other words, the days of "retail" electricity are at hand.

I daresay this situation puts another level of complication in Y2K planning and problem resolution, if for no other reason than it introduces yet another opportunity for finger pointing when something goes wrong.

-- Paul Neuhardt (neuhardt@compuserve.com), March 27, 1998.

Paul: Happy to hear that India is fulfilling your programming needs. Not to rain on your parade, but did you catch the NY Times article earlier this month about other companies that are sending y2k remediation work to India? It seems that some of the programs are coming back with unauthorized back doors inserted for future unofficial access. It's one of those unintended consequences of y2k.

-- J.D. Clark (yankeejdc@aol.com), March 29, 1998.

Regarding John Smith's point: -Computers control power distribution,diagnose,and are used to correct,problems. -computers run on electricity -if power fails,problems cannot be diagnosed or corrected. Is it this simple? Are there manual switches that can controlled by humans? If there are;will humans know which switch to throw and when and for how long? (This talking about "humans" make me unconfortable)

-- Dr Jan S C Czarnecki (czarneck@tbaytel.net), April 10, 1998.

This is going to dredge up some old premises [promises ,?] but here's some update. Talked yesterday with the Y2K sponser at the local EMC. They are a cooperative distributing power across GA, the power is generated by their "consortium" of EMC's (the Electric Mbr Cooperatives), then distributed to the users. GA power also generates power through it own plants, then to its own distribution network, then to th eusers. Obviously, evrybody is in the grid.

Cobb EMC is close to schedule to finish testing by middle next year. Lots of programs have completed okay already, many are left. The engineer was honest, not boastful, not (to my knowledge) overly optimistic nor "politically correct" in his answers.

Good. They are aware of the problem and are working on it. Have a schedule, haven't tested everything, but know they will have problems. Power availability is the biggest problem. Next, Distibution CONTROL is the problem. (Power can get out of the generator, but each substation has to be controlled (to do it correctly and in sequence and in synchronous with other stations. Else, everything trips again, and again, ....)

They will have people at the substation to manually control (I wonder how long these can last ?) but have not exercised "blackout" conditons (need radio, communication, telephone, in-line controls an instruments, etc.) He is going to look into my suggestions of paper and desktop exercises to smoke out more problems in communication and maintenence, and will let me know results. I figure this level of preparedness means outages (of power) of probably less than 1 week until are consistently running - if the plants are operating and have coal/oil, and have trains, and ....

Nukes? I'm a nuclear engineer, and used to run the things. They will be more reliable than the rest because of their redundancies, backup controls, general robust and conservative construction, and most important, their operators know the plant systems VERY well and are VERY rehearsed in drills and emergency procedures. (If you know emergency process, then you can figure out how to make "busted" stuff work)

So (like after an earthquake) the most likely power will be from the 22% that is nuclear sourced. The rest? I'm skeptical.

-- Robert A. Cook, P.E. (cook.r@csaatl.com), September 04, 1998.

On Aug. 8, at the y2k Preparedness Expo, at Washington State's Fairgrounds, State Rep. Bill Thompson, 44th District, was asked to give a report before the 5:00 p.m. panel discussion began. He said he had been up to our State dam (Grand Coulee) a few days before and that it will NOT be Y2K compliant. He was told up there, "The power might not even make it to the grid!" Thompson also said, "This is the first global crisis to come on a precise schedule." He concluded by saying a septic tank is a plus, also have an outhouse ready with supplies of lye.

-- Holly Allen (Holly3325@juno.com), September 04, 1998.