Y2K Utility Update
------------------
Appendix A to the paper titled
"Y2K Preparation, Power, Prudence, and Paranoia"
July 20, 1999
Don Law
macadamia@bigfoot.com
Copyright 1999 by Don Law
This document may be copied and distributed freely without the need to
obtain explicit permission, as long as the document is distributed as
a whole without modification, and is used solely for non-profit purposes.
Reminder: an up-to-date copy of this appendix, the original paper,
and other utility related items can be found at this web site:
http://www.geocities.com/TheTropics/9090/y2k.html
CONTENTS
Summary
April 9 Utility test
April 30 NERC report
FPL progress
What we have learned in the past 6 months
SUMMARY
We are still six months away from 01-01-00, and there will be a lot to
learn in those final six months. However, there has been a lot learned
so far in 1999. The doomsday predictions have been shown to be in error.
The power system tests have been successful. The NERC schedules for
remedy have been kept or exceeded. There has been a sea change in the
outlook for the New Year.
The original premise of the paper still holds. There is no reason to
believe that there will be any significant power outage caused by the
Y2K bug. Please remember not to change your power consumption during
the year rollover. This will make the management of the grid easier
for the power companies.
The fact that the lights will be on after the millennium rollover does
not mean that there will be no Y2K problems. The discussion of impact on
the economy and overseas suppliers is beyond the scope of this document.
Here, we will focus only on the question of electricity.
APRIL 9 UTILITY TEST
Like most Y2K topics, the test conducted on April 9, 1999 had been a
source of much controversy. I have found a lot of misinterpretation
of what really happened, so I am including a section on the topic in
this update. In summary, this was only a test of a contingency plan,
the test was valid, the test was intended to instill public confidence,
and the test was not rigged. The test might be better called a drill.
What was really tested
The utility companies rely heavily on telecommunications to control
the power grid. A large portion of this communication equipment does
not belong to the power companies and is beyond their direct control.
The communications companies (such as phone companies) have committed
to provide uninterrupted communication into the year 2000. However,
the utility companies have adopted a "defense in depth" strategy for
dealing with Y2K. In other words, they are not putting all their eggs
in one basket. They want to have a contingency plan in place just in
case the phone company has a Y2K failure. The April 9 test was a test
of that contingency plan.
The test involved thousands of employees and a portion of the power
generating hardware in the grid. Instead of using phone lines and other
external communications systems, the power companies attempted to control
the grid using backup communications systems. These systems included
radios, satellite phone systems, and the like. There was no test of
any Y2K remedy or any advance of system dates. This was not a test of
how the power grid will react to the millennium rollover.
What was the intention of the test
NERC had two plans for the test. First, to be a valid contingency
plan test, which they rightly called a drill, and secondly, to
install much-needed public confidence. These two purposes are
described in a memo titled "Session 2-C" found at this web site:
ftp://www.nerc.com/pub.sys/all_updl/docs/y2k/planning-a-drill-4-9-99-final.pdf
The Y2k doomsday group has heavily criticized NERC for trying to instill
public confidence in the electric utility system. It is ironic that
the deep need for public confidence has largely been created by this
same group. There is nothing wrong with public confidence as long as it
is based on truth and grounded in accurate information. If the doomsday
group had not constantly blasted the NERC and the utility companies,
then there would not be such a huge need to instill public confidence.
I can hardly blame the NERC for making this one of the goals of the test.
The test was prepared for, but not rigged
One of the criticisms of the April 9 test is that the participating
companies were instructed, prior to the drill, to test the systems that
would be exercised during the drill. This was a reasonable request, since
the purpose of the test was for the process of alternative communication,
not for the individual components. I will try to illustrate this with
the following imaginary scenario:
Supposed that you are a contractor building a neighborhood of homes.
You are scheduled to have the homes completed by July 1. However, the
future owners of the homes consider telephone service to be critical.
The Phone Company is using a new switch for the neighborhood that you
are building. You decide that you want to test the switch before the
completion date of July 1, so you schedule a test on April 9. In this
test, only the phone systems are under test. To ensure that the test can
be performed, you instruct all of the subcontractors to make sure that
a working telephone is installed in every home. You ask them to make
sure that the telephone is wired properly. You don't want to have any
trouble from the home phones when you are testing the new phone switch.
You then conduct the test and the new switch works fine. However, when
you report this to the future homeowners, they accuse you of rigging the
test because you asked the subcontractors to make sure the individual
phones where in working order prior to the test. You can see that this
is an unfounded and twisted accusation.
In this analogy, the contractor is NERC, the neighborhood is the power
grid, the phone switch is the contingency plan, the phones are the
communication equipment used by the plan, the subcontractors are the
power companies, and the irate future homeowners are the critics of the
April 9 test. I hope the analogy illustrates how the accusations of a
rigged test don't make sense when you understand what was really tested.
The April 9 test was published as a success. This is a reasonable
conclusion, since the grid could be controlled without the use of normal
communications systems. However, the test was not without incident.
The test did reveal some minor problems in the contingency plan,
such as wrong phone numbers, improper training on equipment, etc.
These problems will, of course, be corrected. A follow-up test will
be performed on September 8. The important thing to understand is that
even these problems did not lead to power interruptions.
APRIL 30 NERC REPORT
The NERC continues to issue quarterly progress reports on the Y2K problem.
The April 30 report lists some key results:
1) Substantial progress has been made since the last report. Remediation
and testing progress jumped from 44% in November to 75% in March.
2) Minimal operational impact of the Y2K problem has been discovered.
With about 75% of the testing* completed in March, the finding
is that less than 3% of items tested have any difficulty at
all with Y2K. Furthermore, most of that 3% are only "nuisance"
problems, such as incorrect dates printed on reports and logs. (*
note that remediation and testing is the final of the three phases.
In March, 99% of the Inventory phase was completed, and 95%
of the Assessment phase was completed.)
3) The Y2K problem is real. Despite the good news found so far, NERC
and the power companies will not let up in the drive to complete
remediation by July 1. They are taking the problem seriously.
4) The utility companies are taking a "defense in depth" strategy.
Even though it looks like Y2K problems will not cause power
interruptions, there are still contingency plans in place so
there is no one critical link in the chain.
The essence of the April 30 report is that tests indicate that power
production will not be a problem in the year 2000. However, the utility
companies are not taking any chances and are being very thorough with
their testing and their contingency planning. For example, there will
be plenty of extra power capacity available come 2000. Quoting from
page 9 of the report: "A conservative estimate is that the generation
that will be available and Y2K ready will exceed the highest electrical
demand during the transition to the Year 2000 by at least 55%."
One of the common criticisms of NERC is that it is self-regulating.
However, most of the power companies in North America are using an
independent, external audit contractor for their Y2K programs.
90% of the utility companies are on target to be completely Y2K ready
by June 30, 1999. Of the remaining 25 companies, none of them report
exceptions beyond October 31, 1999.
The Y2K doomsayer's commonly claim that NERC will not be able to meet
its schedule. We now have over 6 months of history we can check.
I decided to compare their September 1998 schedule with what they have
actually accomplished:
The September 1998 report calls for enough remediation and testing
completed by May 31, 1999 to meet the demand expected on Jan 1, 2000.
The demand is expected to be around 40-70% of grid capacity, and 75%
had completed remediation and testing. Goal exceeded.
In September 1998, the remediation/testing was only 28% complete.
In November of 1998, it was at 44%, and in March of 1999,it is at 75%.
If you plot this out, you will see that a completion date of July is
realistic.
The September 1998 report measures the average completion date at July
1999. The April report measures the average at June 1999. The average
schedule at the power companies has been pulled in by one month.
Goal exceeded.
The September 1998 report called for Assessment 84% complete by
December 31, 1998, and 100% complete by March 31, 1999. In November
1998, assessment was 72% completed and in March, it was 99% complete.
Approximately on target.
In summary, the people that are actually fixing and testing the problem
have been able to meet and exceed their objectives so far. There is
no reason to doubt that our electrical systems will by Y2K ready by
July 1999.
FPL progress
FPL is ahead of the average power company in preparation.
The NRC audited the Turkey point nuclear power plant and five other
nuclear plants. The NRC stated these were the most in-depth assessments
they have seen. The NRC review identified no issues or weaknesses.
There are two non-nuclear generating plants here in Broward County.
The most technologically advanced, and thus the most vulnerable to the
Y2K bug, is "Unit 4" located near 441 south of I-595. In April, this
power plant was shut down and restarted with all the clocks set ahead
to December 1999. On May 1, 1999, the computers at Unit 4 rolled over
to January 1, 2000, with the plant running at full power. There was no
interruption in power generation. The press release for this event can
be found on the FPL web site, www.fpl.com.
I asked an FPL employee, Joe Nestor, for the inside story. It turns
out that the Unit 4 test results are consistent with the NERC findings.
There were a few problems, but they were only "nuisance" problems,
such as incorrect dates in logs and reports. There was no effect on
the generation of electricity.
What we have learned in the past 6 months
Proverbs 18:17 reads: "The first one to plead his cause seems right
until his neighbor comes and examines him."
We have heard from many sources predicting power problems caused by Y2K.
We have examined their premises and claims, and found them to be false
or unfounded. We have examined the claims of the people who understand
the problem and are fixing it. Their story is consistent and validated.
We've been told that "know one can know for certain what will happen."
While that may be true on a global scale, it isn't true for contained
subsystems. For example, we know exactly what will happen when the
year rolls over at Unit 4 because we have done it and observed it.
Computers are very predictable - given the same input, the will always
produce the same output.
We've been told that the rail system will fail and won't be able to
deliver coal, because the switches can no longer be manually controlled.
I personally checked this out first hand. I visited a rail switch on
the FEC rail system, and there was a giant lever for manual override.
I also visited a CSX rail switch, and it also had the manual override.
You can see the rail switches for yourself on my Y2K web site:
http://www.geocities.com/TheTropics/9090/y2k.html
We've been told that the NERC has unrealistic schedules, but when
examined, we have found that they are right on schedule, if not a
little ahead.
We've been told that the NERC April 9 test was rigged, but when the test
is understood, we can see that it is valid.
There are other non-power related quotes that are not legitimate.
One famous one is that the CEO of General Motors said that if January 1,
2000 were tomorrow that GM could not build a car for a year. The CEO
of GM never said that. This originated from a misquote of the CIO,
where he said that if nothing were done about the Y2K problem, then it
could turn into a catastrophe. The complete explanation of the misquote
is widely available in papers such as "Y2K: Who's In Charge Here?" by
Pastor John Michaels. This paper can be found on the Internet at:
http://www.calvarychapel-sv.org/y2k1.htm
I would recommend reading this 23-page sermon transcript, because it
gives an interesting picture of the Y2K doomsayers.
We've been told that Y2K problems would begin to surface on April 1
(fiscal year 2000 for New York and many companies) and on April 9 (the
99th day of the 99th year). Those dates have come and gone and no
disasters happened. We were also told that July 1, 1999 would bring a
wave of Y2K problems. It has come and gone and was nothing more
than just a speed bump.
We’ve been told that no utility company will come out and state that they
are ready for Y2K. But now FPL has come out and solidly stated that they
are ready. In the Business section of the Sun Sentinel on Friday July 2,
Sol Stamm, FPL’s Year 2000 project manager said: "When you go to turn on
your light switch at Jan. 1, 2000, the lights are going to come on."
In my original paper, I called for a test of claims against first-hand,
reliable knowledge. The claims of widespread power problems caused by
Y2K have failed the test.
In the final conclusion, it is more likely that we will have a widespread
power outage tomorrow than on January 1, 2000. While we cannot guarantee
that either of these is impossible, we should not waste excessive
time and money preparing for it. Having said that, let me add that
there is nothing wrong with being prepared for disaster in general.
I just don't recommend any special preparation for a Y2K power outage.
(
geocities.com/thetropics)