Subject: RISKS DIGEST 12.13
REPLY-TO: [email protected]

RISKS-LIST: RISKS-FORUM Digest  Monday 19 August 1991  Volume 12 : Issue 13

       FORUM ON RISKS TO THE PUBLIC IN COMPUTERS AND RELATED SYSTEMS
  ACM Committee on Computers and Public Policy, Peter G. Neumann, moderator

 Contents:
Junk Mail in Outer Space: Shuttle test mail-bombed (Peter Scott)
ATM mixup in New York (John Martin)
Computer failure helps Bakthiar murder suspect (Fernando Pereira)
Deutsche Airbus 2000 (Martyn Thomas)
Bell V22 Osprey crash (Martyn Thomas)
"Doctored" radios (PGN)
"Virus Implants in DoD Weapons" (David Risler via Jerry Leichter)
Cracker charged in Australia (Fernando Pereira)
Profitable Drug Wars -- Innocents Presumed Guilty (mauler via Charles Hoequist)
Patriot and Dhahran again (Phil R. Karn)
Re: "Traffic crystal ball" may be in your car's future (Secty Samuel Skinner,
 [editorial], via Jeff Helgesen)
Risks of Calling Reporters in Ohio: Procter & Gamble (PGN)
Risk of Power Failures in Computer Controls: 9 Mile Point (PGN)

The RISKS Forum is moderated.  Contributions should be relevant, sound, in
good taste, objective, coherent, concise, and nonrepetitious.  Diversity is
welcome.  CONTRIBUTIONS to [email protected], with relevant, substantive
"Subject:" line.  Others ignored!  REQUESTS to [email protected].  For
vol i issue j, type "FTP CRVAX.SRI.COM<CR>login anonymous<CR>AnyNonNullPW<CR>
CD RISKS:<CR>GET RISKS-i.j<CR>" (where i=1 to 12, j always TWO digits).  Vol i
summaries in j=00; "dir risks-*.*<CR>" gives directory; "bye<CR>" logs out.
The COLON in "CD RISKS:" is essential.  "CRVAX.SRI.COM" = "128.18.10.1".
<CR>=CarriageReturn; FTPs may differ; UNIX prompts for username, password.
ALL CONTRIBUTIONS CONSIDERED AS PERSONAL COMMENTS; USUAL DISCLAIMERS APPLY.
Relevant contributions may appear in the RISKS section of regular issues
of ACM SIGSOFT's SOFTWARE ENGINEERING NOTES, unless you state otherwise.

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

Date: Mon, 12 Aug 91 14:22:35 -0700
From: [email protected] (Peter Scott)
Subject: Junk Mail in Outer Space: Shuttle test mail-bombed

From _Information Week_, August 12 (who got it from _Newsday_, August 6, p.5):

SPACE HACKERS

A test of electronic-mail between earth and laptops aboard the space shuttle
Atlantis was intended to lay the groundwork for use of E-mail on space station
Freedom.  But the test is in jeopardy after 80 E-mail messages were received by
the Atlantis crew from unauthorized users.  The leak behind the E-mail address
remains a mystery.  *Junk Mail In Outer Space*, Joshua Quittner.

Peter J. Scott, Member of Technical Staff    |   [email protected]
Jet Propulsion Laboratory,  NASA/Caltech     |   SPAN:  GROUCH::PJS

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

Date: Sun, 18 Aug 91 09:58:27 EDT
From: [email protected] (John Martin)
Subject: ATM mixup in New York

The cover of the August 15th New York Daily News had a 9" x 11" photo of a man
using an ATM, and a caption to the effect of "WANTED: This man is using an ATM
card that was stolen from a rape victim 40 minutes ago."

The next day, a different man was charged with rape and robbery, and in
the August 17th Daily News, the following was printed:

"Earlier this week, police released to the Daily News and other media
outlets the photo of another man, saying that he was using a bank card stolen
from a rape victim and that they wanted to question him.

"...DeMartino said the initial picture had "a time sequence that differed
on the printout from the ATM. The bank said the error was created by the
machine downloading.

"The mixup was "a very unfortunate situation," according to Bruce Herman,
Apple Bank senior vice president and general counsel.

"There was no malfunction in the ATM system that night," Herman said. "All
relevant records and materials with respect to ATM transactions on the night
in question were made available to the police at their request for analysis
and evaluation."

Unfortunately for the man in the photo, the admission of the mistake did not
seem as well publicized as the photo.

       John Martin     [email protected]

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

Date: Sat, 17 Aug 91 11:24:20 EDT
From: [email protected] (Fernando Pereira)
Subject: Computer failure helps Bakthiar murder suspect

The AP reports from Geneva on 8/16 that one of the suspects in the murder in
France of former Iranian prime minister Shapour Bakhtiar spend Monday and
Tuesday night at a Geneva hotel, under a false Turkish identity. However, the
failure of a police computer used to check hotel registration cards delayed
until Wednesday the identification of the suspect, by which time he had already
left (This seems to imply that the false identity was known to the Swiss
police).

It is interesting how the failure of the computer system is blamed here for
something that presumably would have happened anyway if the computer system did
not exist. Or is it that the Swiss police have become so dependent on their
computer databases that they no longer use slower, more traditional sources and
methods (eg. alert hotel staff)?

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

Date: Mon, 19 Aug 91 13:07:19 BST
From: Martyn Thomas <[email protected]>
Subject: Deutsche Airbus 2000

Deutsche Aerospace has proposed a 615-passenger Airbus, according to Flight
(3-9 July). DA's executive VP for design and technology says "The tailplane
itself would be smaller, because the fly-by-wire flight control system would
allow greater inherent instability ...."

Martyn Thomas, Praxis plc, 20 Manvers Street, Bath BA1 1PX UK.
Tel:    +44-225-444700.   Email:   [email protected]

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

Date: Mon, 19 Aug 91 13:10:30 BST
From: Martyn Thomas <[email protected]>
Subject: Bell V22 Osprey crash

The Editor of Aerospace (the Asian monthly magazine) tells me that there was a
recent crash of a Boeing Bell V22 Osprey "which tipped on its side due to an
admitted 'glitch' in the lateral control system."

Does anyone have further information?

Martyn Thomas, Praxis plc, 20 Manvers Street, Bath BA1 1PX UK.
Tel:    +44-225-444700.   Email:   [email protected]

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

Date: Mon, 19 Aug 91 9:16:29 PDT
From: "Peter G. Neumann" <[email protected]>
Subject: "Doctored" radios

  Doctored radios revealed Iraqi moves during Gulf war: report

  LONDON, Aug 18 (AFP) - Radio equipment sold to Iraq before the
Gulf war was fixed so that Britain could monitor transmissions giving
the allies a crucial advantage during the conflict, the Sunday
Telegraph reported here on Sunday.
  The British manufacturers did not know that their export equipment
had been tampered with "so that the messages sent by the Iraqis could
be picked up by Britain's GCHQ intelligence nerve centre" at
Cheltenham in western England, the weekly said, quoting senior
parliamentary sources".
  "Exchanges between Iraqi commanders were picked up and then passed
on to the U.S. National Security Agency," the Telegraph quoted
sources close to the government sources as saying.
  The decision to fix the equipment had been taken well before war
broke out, but "at a time when the intentions of the Saddam regime
were of deep concern to Western strategists following the execution
of journalist Farzad Bazoft and the uproar over the supergun affair."


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

Date: Tue, 13 Aug 91 07:23:52 EDT
From: Jerry Leichter <[email protected]>
Subject: "Virus Implants in DoD Weapons"

The following message appeared recently on VIRUS-L:

       Date:    07 Aug 91 20:28:57 +0000
       From:    [email protected] (David Risler)
       Subject: Virus Implants in DoD Weapons

       From the August 1991 "Armed Forces Journal International"

       "A draft Pentagon directive that called for implanting a computer
       "virus" or software disabling mechanism in every major new US weapon
       system - one that could be remotely triggered if the weapon fell into
       enemy hands - was under consideration last December at a high DoD
       level, a knowledgeable source told AFJI recently...If that is the
       case, the device is more likely to function as a variable duration
       "enabler"...rather than a disabler that could be remotely activated to
       prevent a weapon from being used.  In all likelihood, no decision
       regarding implanting either kind of device in advanced weapons will
       come before the DARPA provides an assessment to Congress of how best
       to handle the issue.  That report is expected on Capitol Hill by
       August."

       The article goes on to say that this would be great for weapons
       exports and that EEPROMS could carry such "Trojan Horses" that could
       be activated using electrical signals.  Hmmmmmm.  Comments?

My comments:  First off, I wish people would stop applying the word "virus"
and "Trojan horse" to every new kind of software they come across.  Such
software would not spread, so it's not a virus; and there's little reason to
hide the fact that it exists (though of course the details would be secret),
so it's not a Trojan horse.  "Software disabling mechanism" is about right, of
a bit wordy.  Really, it's a lock, just like the lock on your car.  It happens
to be a "normally unlocked" lock, while most locks we deal with are "normally
locked".  The difference is understandable, given the circumstances under
which the protected devices are used.

In many ways, there is nothing new here.  All high-tech weapons already have,
in effect, a "variable duration enabler":  Their spare-parts supply.  This
isn't a particularly EFFECTIVE lock, since even in the best of circumstances
it can take quite some time for a spare-parts store to be exhausted, and
maintainers of military equipment usually prove to be very resourceful at
stretching their supply.  Besides, there's an active black market.

On a more prosaic level, it's been Soviet practice for years to build guns
with a caliber just marginally smaller than that of their expected opponents.
Soviet rifles can use NATO bullets, but NATO rifles can't use Soviet bullets -
a very effective time-independent "lock".

There have already been jokes about soldiers forgetting the password needed to
boot their tank.  I'm sure this proposal will lead to all sorts of fears about
similar problems.  However, especially if implemented with an "enabler" rather
than an external disabling signal, I see little problem from a technical point
of view - and it strikes me as a very nice safeguard to have.  Imagine if all
of Iraq's weapons had shut themselves down after 6 months.

Now, from a POLITICAL point of view, it's another question.  Would Iraq (or
any other country) be willing to purchase weapons so solidly under the control
of a potential enemy?  Certainly they'd try very hard not to.  The history of
attempts to control international weapons sales hardly leads one to be opti-
mistic that there won't be countries willing to sell unprotected weapons -
not to mention "lock removal" agents (though with computer-controlled weapons
their work can be made very, very difficult).
                                                       -- Jerry

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

Date: Wed, 14 Aug 91 09:16:38 EDT
From: [email protected] (Fernando Pereira)
Subject: Cracker charged in Australia

The AP (8/13/91) reports from Melbourne that Nahshon Even-Chaim, a 20-year old
computer science student, is being charged in Melbourne's Magistrates' Court on
charges of gaining unauthorized access to one of CSIRO's (Australia's
government research institute) computers, and 47 counts of misusing Australia's
Telecom phone system for unauthorized access to computers at various US
institutions, including universities, NASA, Lawrence Livermore Labs, and
Execucom Systems Corp. of Austin, Texas, where it is alleged he destroyed
important files, including the only inventory of the company's assets. The
prosecution says that the police recorded phone conversations in which
Even-Chaim described some of his activities. No plea has been entered yet in
the ongoing pre-trial proceedings.

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

Date:   Wed, 14 Aug 1991 17:39:00 -0400
From: Charles (C.A.) Hoequist <[email protected]>
Subject:  Profitable Drug Wars -- Innocents Presumed Guilty

The following was posted to several Usenet groups on 14 August.

Date: 12 Aug 91 13:02:36 GMT
From: [email protected]
Newsgroups: talk.bizarre,talk.politics.drugs,talk.politics.misc
Subject: "War On Drugs" Atrocities: The Forfeiture Laws

[...]
                       P R E S U M E D     G U I L T Y
                     The Law's Victims in the War on Drugs
               The Pittsburgh Press, Sunday, August 11, 1991, p.1

It's a strange twist of justice in the land of freedom. A law designed to give
cops the right to confiscate and keep the luxurious poseesions of major drug
dealers mostly ensnares the modest homes, cars and cash of ordinary,
law-abiding people. They step off a plane or answer their front door and
suddenly lose everything they've worked for. They are not arrested or tried for
any crime. But there is punishment, and it's severe.

This six-day series chronicles a frightening turn in the war on drugs.  Ten
months of research across the country reveals that seizure and forfeiture, the
legal weapons meant to eradicate the enemy, have done enormous collateral
damage to the innocent. The reporters reviewed 25,000 seizures made by the Drug
Enforcement Administration. they interviewed 1,600 prosecutors, defense
lawyers, cops, federal agents, and victims. They examined court documents from
510 cases. What they found defines a new standard of justice in America: You
are presumed guilty.

  [The articles included some real horror tales.  Part One is in
  RISKS-12.13LAW in the RISKS archive directory on CRVAX.SRI.COM.  PGN]

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

Date: Thu, 15 Aug 91 02:42:59 EDT
From: [email protected] (Phil R. Karn)
Subject: Patriot and Dhahran again                 [and Karn is up late again?]

Army Records Say Computer Shutdown Might Have Averted Scud Disaster
By ROBERT BURNS, Associated Press Writer

 [A few excerpts by PGN from a lengthy AP item presumably from 15 Aug 91]

  Army investigators concluded that the exact reason for Patriot's failure to
shoot at the Scud will never been known for sure. But they said the most likely
explanation was a previously unknown glitch in Patriot computer software.  Army
technicians had determined as much as two weeks prior to the attack that the
Patriot computer was vulnerable to losing track of incoming Scuds when the
computer was kept running for long periods, according to internal Army reports
released in response to a Freedom of Information Act request by The Associated
Press.  Tragically, no alert bulletins were sent to Patriot operators in the
field because the technicians viewed this as a minor problem of less importance
than other Patriot improvements they were working on. The technicians did not
think Patriot computers would be kept running for more than several hours at a
time.  [Earlier reports in RISKS noted that the spec called for only 14 hours
of operation, not 100, and that the clock was drifting...]
  ``Had rebooting, or shutting off the system, occurred, it would have
decreased the chance that the inexact (computer) calculation would have
occurred,'' said June 14 memo signed by Lt. Gen. Ellis D. Parker, director of
the Army Staff.
  A previously classified internal Army memo dated Feb. 20 described a series
of software improvements to the Patriot computer, including a change that was
designed to avert the tracking problem under circumstances of long continuous
operation.  But in follow-up memos intended to be seen by users of the Patriot,
no mention was made of the tracking problem or of a need for periodic computer
shutdowns. The technical specialists simply thought they had found a way of
improving further on the accuracy of Patriot, not correcting a potentially
fatal error.  ``No significance was given this change because no prior related
tracking problems had been seen,'' said an Army Patriot program office report.
``Therefore, no alarm or urgent notification was transmitted to the field.''

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

Date: Wed, 14 Aug 91 13:44:45 -0500
From: Jeff Helgesen <[email protected]>
Subject: Re: "Traffic crystal ball" may be in your car's future (RISKS-11.53)

The following letter from Samuel Skinner, US Sec. of Trans., appeared in
Tuesday's (8-13) Chicago Tribune. It's in response to a negative editorial
regarding ADVANCE, a system for traffic control being worked on in Chicago. I
could not find the original editorial.

 -\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\-

                New Traffic Technology Worth The Risks

WASHINGTON---"The smart car will never succeed," Eric Zorn proclaimed recently
in this "Hometowns" column, as he and many others in the Chicago area began
speculating on the effectiveness and long-term benefits of a new technology
that's come to Chicago.  I beg to differ with Mr. Zorn---as was evident several
weeks ago when I helped to announce the new ADVANCE project, an intelligent
vehicle highway system [See RISKS 11.53, 24 April 1991: "`Traffic crystal
ball'" may be in your car's future" -JH].

Intelligent vehicle highway systems, or IVHS, consist of many advanced
technologies which, in combination, will help us to ease congestion and improve
highway safety, mobility, and driver convenience.  Unfortunately, Mr. Zorn
cheated himself and his readers by simply concentrating on the "smart car"
portion of IVHS.

Just as important is the "smart highway" part of the equation.  This concept
uses advanced technologies to better manage traffic throughout an area,
providing a safer and smoother trip for all drivers. Actually, IVHS
technologies already are in use. Features such as cruise control and anti-lock
brakes provide "smart car" capabilities to today's drivers. In the
not-too-distant future, crash avoidance devices will detect the presence of an
obsticle or other vehicle and alert the driver to a possible collision.

The "smart highway" is also a reality in many areas. For example, computers
already are used to automatically change traffic signal timing and control the
flow of traffic onto freeways to help reduce congestion. The ADVANCE IVHS
project goes a step further by providing individual drivers with route guidance
information to make their trips safer and faster.

If the U.S. is to stay competetive into the 21st Century, we must invest in
innovative ways to move goods and people more safely and efficiently. Being on
the cutting edge of technology means taking risks, and frankly, some IVHS
technologies may not work. However, there are strong indication that many will.
Even today, one Japanese firm claims that they are selling 2,000 navigation
devices per month in Japan. If projects like ADVANCE are a success, consumers
will be buying these products from American companies and not their European
and Japanese competitors.

The long-term benefits from ADVANCE and other IVHS projects will not be known
for many years. It is also impossible to predict consumer acceptance of
advanced technologies in automobiles or elsewhere. One can only wonder what the
early reaction was to those car radios the Mr. Zorn urges us to use. In 1926,
Lee de Forest, the man who invented the cathode ray tube said, "While
theoretically TV may be feasible, commercially and financially, I consider it
an impossibility..."

Only time will tell the full measure of success for ADVANCE and other IVHS
technologies. But if we never start, we will never know.

Meanwhile, we at the Department of Transportation will continue working on
innovative solutions to assure America's transportation future. For as the
Tribune's own editorial put it so well, "...if we have learned anything this
century, it is that the future is limitless and its way paved with new
notions." Clearly, the Congress shares this view, as both the Senate and House
versions of the Surface Transportation Bill now being crafted include
substantial increases in spending for research, development and deployment of
IVHS technologies.
               Samuel K. Skinner U.S., Secretary of Transportation

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

Date: 16 Aug 91 09:30:23 U
From: "Peter G. Neumann" <[email protected]>
Subject: Risks of Calling Reporters in Ohio

SEARCH FOR NEWS LEAKS SPURS OHIO PHONE SWEEP
By RANDALL ROTHENBERG, c.1991 New York Times News Service

  Law-enforcement officials in Ohio have searched the records of every
telephone user in southwestern Ohio to determine who, if anyone, called a Wall
Street Journal reporter to provide information that Procter & Gamble said was
confidential and protected by state law.  The investigation goes far beyond
examining the telephone records of current and former employees of the giant
consumer products company, an inquiry the Hamilton County prosecutor's office
confirmed on Monday. The Journal reported the scope of the investigation
Thursday.
  The prosecutor, Arthur Ney Jr., acting on a complaint by Procter & Gamble,
ordered Cincinnati Bell to turn over all the telephone numbers from which
people called the home or office of the reporter, Alecia Swasy, from March 1 to
June 15.
  The situation began sometime before June 17 when Procter & Gamble, which
makes Tide detergent, Crest toothpaste and other familiar supermarket products,
asked the Cincinnati police to determine whether current or former employees
were leaking confidential corporate information to The Wall Street Journal.
  On Monday the newspaper reported that the company had been bothered by two
news articles published on June 10 and June 11 written by Ms. Swasy, a reporter
based in Pittsburgh who covers Procter & Gamble. The articles cited
unidentified sources saying that a senior executive was under pressure to
resign from the company, and that it might sell some unprofitable divisions.
But a spokeswoman for Procter and Gamble, Sydney McHugh, said Thursday that the
company ``had been observing a disturbing pattern of leaks'' since the
beginning of the year. She refused to elaborate, but said the decision to
pursue legal action was reviewed at several levels in the company and was made
by Jim Jessee, a corporate security officer.
  Two Ohio statutes protect the unauthorized disclosure of trade secrets. One
makes it a felony to transmit formulas, customer lists or other tangible pieces
of information that would be valuable to a company and its competitors. But
another, broader law makes it a misdemeanor to disclose ``any confidential
matter or information'' without the company's consent.
  The Cincinnati police approached the Hamilton County prosecutor's office,
which sought and received from a grand jury a subpoena for telephone records.
  A copy of the subpoena, dated June 17, was given to The New York Times by
someone involved in the case who insisted on anonymity.  The subpoena ordered
Cincinnati Bell to ``identify all (513) area code numbers that have dialed''
Ms. Swasy's home or office telephones in Pittsburgh during an eight-week period
that started on March 1.
 Cincinnati Bell serves 655,297 telephone numbers in the 513 area code, in an
area covering 1,156 square miles, said Cyndy Cantoni, a spokeswoman for the
company. In the company's entire jurisdiction, which also covers parts of
Kentucky and Pennsylvania, about 13 million toll calls are placed in an average
month, she said.
  Ms. Cantoni said she could not comment on what Cincinnati Bell turned over
to the authorities, but said the company routinely complied with subpoenas.
Under normal procedure, the company's computers would have automatically
searched its customer list and printed out only the originating numbers, and
not the names or addresses, of calls to Ms. Swasy's numbers, Ms. Cantoni said.
  The Wall Street Journal, which is published by Dow Jones & Co., reported on
Monday that neither Ms. Swasy nor executives at the Journal were informed of
the subpoena by the authorities.
  Neither Terry Gaines, a first assistant prosecutor, nor Ed Ammann, a police
department colonel involved with the investigation, returned repeated calls to
their offices.
  Alan F. Westin of Columbia University, an authority on technology and
privacy issues, said the legality of the Ohio authorities' search for the
Procter & Gamble whistleblower may depend on how the investigation was pursued.
If Procter & Gamble turned over the names and phone numbers of present and
former employees to the police and the police matched that list against the
numbers they were given by the telephone company, the rights of other,
uninvolved parties may not have been violated, Westin said. But if the police
learned the names of people unaffiliated with Procter & Gamble who called the
Journal's reporter, he said, or if they turned over a list of numbers to
Procter & Gamble for research, some Ohio residents' Fourth Amendment
protections may have been sullied.  ``When technology allows you to run
millions of calls involving 650,000 telephone subscribers through a computer in
order to identify who called a person, potentially to find out whether a crime
was commited, you raise the question of whether technological capacity has gone
over the line in terms of what is a reasonable search and seizure,'' Westin
said.

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

Date: 15 Aug 91 09:19:19 U
From: "Peter G. Neumann" <[email protected]>
Subject: Risk of Power Failures in Computer Controls

POWER SURGE CAUSES FAILURE OF SYSTEMS IN NEW YORK NUCLEAR PLANT
By KEITH SCHNEIDER, c.1991 N.Y. Times News Service

  WASHINGTON A power surge at dawn Tuesday knocked out instruments that
operators used to control the reactor at a nuclear power plant in upstate New
York and caused the failure of a succession of systems that monitored the
plant's operations.
  Workers at the Nine Mile Point Nuclear Station, on Lake Ontario about 6
miles from Oswego, were never in danger from a release of radiation, said
Niagara Mohawk Power Corp., the plant's operator and co-owner.  But the
problems at the Unit 2 reactor, the newest of the plant's two reactors, caused
Niagara Mohawk to shut down the plant and declare the second-highest level of
alert possible under federal rules.  And the Nuclear Regulatory Commission said
the plant could not reopen until an investigation into the events, which began
Tuesday, was completed.
  Niagara Mohawk lifted the alert at 7:45 p.m. Tuesday. It is only the third
time that such an alert, known as a site area emergency, has occurred at an
American nuclear power plant, the NTC said.
  Parts of the monitoring systems were unaffected by the loss of power,
enabling the operators to oversee the safe shutdown of the reactor.
  According to the NRC, the operating record of Nine Mile Point's two nuclear
reactors since the late 1980's has ranked among the worst of the 111 licensed
nuclear reactors in the United States.  For three years until its status was
changed in June, Nine Mile Point was on the agency's list of problem plants.
  The emergency was declared after one of three transformers at Unit 2 failed
at 6 a.m. The failure caused a powerful surge of electricity to rush back into
the plant, tripping the circuit breakers in the main turbine and five of the
plant's internal power systems.  The turbine shutdown caused the nuclear
reactor to automatically begin to shut itself down, plant engineers said.
Manual shutdown procedures also were started, they said.
  Four of the internal power systems that failed provided electricity to
critical gauges, safety monitors, the plant's main computer, and monitoring
equipment in the main control room.
  Some of the most important gauges operators use to control the reactor were
knocked out, including the one showing the position of control rods in the
reactor and another that measured the power of the reaction.
  Another system of emergency indicators that failed were annunciators, a
series of playing-card-sized windows at the top of the control panel that flash
and sound an alarm when equipment or processes are functioning improperly.
  Their function is similar to that of red warning lights on an automobile's
dashboard, serving as a first line of warning that can be verified by a gauge.
  The failure of many primary gauges, the main computer and annunciators meant
that if the reactor were an automobile, operators would have been driving with
a sheet across the windshield.
  Niagara Mohawk and the NRC said they considered the incident to be serious
because the power systems had been designed so they would not fail. Each had
backup batteries.
  In the event of a main electrical failure, circuits were supposed to
automatically shift the systems to battery power. The plant's engineers
determined Tuesday that the power surge destabilized the circuits that needed
to be stable for 4 milliseconds to work properly, said Gary Grant, a senior
reactor operator.
  ``Nobody anticipated this transformer failure and all this happening at the
same time,'' said Grant.
  Nine Mile Point is one of 37 nuclear plants in the county manufactured by
General Electric Co.
  Lynn Wallis, a spokesman for GE in San Jose, Calif, said Tuesday:
  ``The NRC has evaluated our design. They are licensed and they are safe.
That's all I can provide. You ought to talk to the utility and the NRC.''
  The Nine Mile Point Nuclear Station generates 1,705 megawatts of electricity
for upstate New York residents from two boiling-water nuclear reactors.
  Unit 1, a 615-megawatt reactor, began operating in 1969 and was not affected
by the incident Tuesday.
  Unit 2, a 1,080-megawatt reactor that began operating in 1988, is owned by
Niagara Mohawk and four other utilities, including Long Island Lighting, New
York State Electric and Gas, Rochester Gas and Electric and Central Hudson Gas
and Electric.
  The NRC describes a site area emergency, one of four categories of alert, as
one in which there are ``actual or likely major failures of plant functions
needed for protection of the public.''
  Only twice previously have site area emergencies been declared by utilities.
There has never been a general emergency, described by the government as an
actual or imminent degradation of the nuclear reactor core, though if the
system had been in place in 1979, the Three Mile Island accident would have
qualified.
  Last year, Plant Vogtle, a nuclear generating station owned and operated by
Georgia Power, 26 miles southeast of Augusta, declared a site area emergency
after the plant's main power supply failed and backup diesel generators were
turned on, the NRC said.
  In 1982, a steam generating tube ruptured at the Ginna nuclear plant,
operated by Rochester Gas and Electric 20, miles northeast of Rochester, and a
similar emergency was declared because of the threat of a worse accident caused
by the loss of coolant for the reactor core, said the NRC.
  The NRC said Tuesday that emergency incidents at nuclear
reactors in the United States were declining, indicating an
improvement in management and operations since 1979.
  Last year, the number of unusual events, the lowest level of alert, declined
to 151 from a peak of 312 in 1985. In 1990, the number of alerts, the second
lowest emergency event category, was 10, about the same as it had been for a
decade.
  Niagara Mohawk said it took just 22 minutes for the plant to restore power
to the control room monitors.

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

End of RISKS-FORUM Digest 12.13
************************