ML20064L117

Requests Specific Listed Info Re 820125 Steam Generator Tube Rupture

ML20064L117
Person / Time
Site:	Ginna, 05000000
Issue date:	02/05/1982
From:	Udall M HOUSE OF REP., INTERIOR & INSULAR AFFAIRS
To:	Palladino N NRC COMMISSION (OCM)
Shared Package
ML20064J857	List: ML20064L117
References
FOIA-82-309 NUDOCS 8202260148
Download: ML20064L117 (2)

Text

) '+ 'h cTv.s cvv.rs ccmc.n cs:s

(

c..as co-rv.,

..m.,....,.

u,.,..,..

o.

su ~a v u mu

..................s

- _.... mur.

"'.'71" '.S.' '; '.". :. ". ^..

'~~,'.~~';.'21' M...<,...u,.

COMMITTEE ON INTERIOR AND INSULAR AFTA1RS.

,s.oc.u,. ro.~,cio.

• -o n.

m e r. s i.... e-...-..

....e..-.,.......-.

o.....cv..n. ".

U.*i. HOUS C OF' REPiiELE N T ATIVCS

".E " C T'.f

.u.

,,,c,-

wn.

.. e...C~'.~". ;...

1~.*J *. ".C ~.": 7.'*6.,.

= H '"c 7 c". o c-m'5

.. r.. um,,s m

......m...

..............n,......

....,,.e..n,.

a.m.,, _ sm

,,..u. ~.....

n.

..... n.

e.

.~._.m e.... u... x...

-.,...... ~..

...s....

.e.C5

40...

4 L.., ar. we, pg.s.vy. rY, D.. &.

..m.......

,....x....u.

, u. o... n..

.u.u,.

m._....~

. "."m m u,..,.,O,""..c.:.

February 5, 1982

. ~.........

. r.e D.. 44 v.8,

..m.-m..

The Honorable Nunzio Palladino Chairman United States Nuclear Regulatory Commission Washington, D.C.

20555

Dear Mr. Chairman:

As a follow-on to the February 4 briefing, I would appreciate the Commission 's. answers 'to the following questions in addition to the information.recuested by

~

Mr. Markey.

1.

What is the primary significance of the Ginna incident?

2.

What was the leak rate through the break as a fun'etion of time?

3.

What triggered the steam generator tube rupture?

4.

Had there been indications of leaking steam generator tubes prior to the rupture on January 25?

5.

What was the cause of the PORV's apparent failure to clo jse? Does the apparent failure of the PORV to close cause d5'ub t ab^ou t the~ adequacy of the industry's program to test uch valves?

T What woul'd the course of the incident have been had the Q

PORV block valve failed to close partially or fully

\\.,

following failure of the PORV to close fully?

ej 7.

Did the procedure for responding to a steam generator J

tube rupture contain instructions for actions to be taken in response to development of a steam bubble in the reactor L9 pressure vessel?

8.

.Was there a need during the incident to take actions not S

specified in the plant's written operating and emergency p

procedures?

Were the emergency procedures in pla.ce at Ginna consistent with Westinghouse guidelines as dise,ussed in the January 28 memorandum from Mr. Speis to Dr. Mattson?

9.

Had a water level measuring device been available, would it have assisted the operators 'in determining 'the size of the steam b~ubble in the pressure vessel and otherwise in

/g.

1. ', v '#

, Chairman Palladino February 2 1982 i

10 What consideration has been given the potential for radioactivity escapi'ng PWRs via a path including bre,aks in steam generator tubes and a stuck open safety valve.

_;r 11.

Is it generally agreed that if a leak' had developed in

~

both steam generators, the operators would have been able to institute the " feed and bleed" process described in Mr. Speis ' January 28 memoran,dum. _

12.

how many steam genera' tor tube ruptures per year of the

~

Gi'nna magnitude or greater do you expect?

13.

What is the likelihood of-several steam tube ruptures occurring at one time?

What is the maximum number of simultaneous or near simultaneous steam generator tube ruptures that are considered design basis accidents following which the a can be brought to a safe shutdown condition by following plant operating and emergency procedures?

14.

Did WASH 1400 or more recent. risk assessments determine the probability of occurrence of events in which one or more steam generator tube failures are followed by various, combinations of PORV, block valve, and safety valve failures?

15.

How long did it take to reach cold shutdown?

Is this a period longer than desircable?

What was the reason for.the period'being longer than normal?

What kinds of malfunctions during the extended cooldown period might have led to a significant release of rddioactivity 'to the environment?

16.

Did any part of.the reactor pressure dessel cool at a rate in excess of that stipulated in the plant technical specifications?

17.

Was there a capability at Ginna to remotely vent the-reactor oressure vessel high points?

Does the Commission beli' eve that conditions might develop in PNRs calling for the use of. remotely cont. rolled valves for the purose of venting steam?

18.

At any point during the Ginna event, did the steam generator containing the-ruptured tube control the primary system pressure?

Are operators at Ginna and other PWRs

' trained with respect to actions to be taken when a steam generator. controls primary system pressure?

Sincerely, g

1

^

M x S ~.

Chairman

~

e

.... ~.......

p gana

! [7.,,, m*'o,,

UNITED STATES j

g NUCLEAR REGULATORY COMMISSION q.'-'

- C WASHINGTON. D. C. 20555 c

}

%, D Lj MAR 191982 1

J

?

MEMORANDUM FOR: Thomas A. Ippolito, Chief Operating Reactors Assessment Branch i

Division of Licensing

-i

?j FROM:

Keith R. Wichman, Section Leader i

Engineering Section 1

Operating Reactors Assessment Branch 3

Division of Licensing i'

SUBJECT:

MEETING WITH WESTINGHOUSE ON STEAM GENERATORS 1

Attached is a summary of the subject meeting that was held on March 2, 1982 in Bethesda. Westinghouse presented their views with respect to steam generator tube degradation and steam generator tube. rupture i

accident management. A list of attendees is shown in Enclosure 1 to the summary.

,?)

p/

',1

.s

. /

s.

i.

>l U C '% A p.ak u laiv' fw Keith R. Wichman, Section Leader i

Engineering Section j

Operating Reactors Assessment Branch i

Division of Licensing 1

l j

I0 p'

s/

l

}'f h

((

w --... -

..a...__

\\%

3

\\

\\>

MAR ]p me, MEETING

SUMMARY

DISTRIBUTION NRC/PDR G. Lear Local PDR l

N. Hazelton TIC /NSIC/ TERA V'. Benaroya H. Denton Z. Rosztoczy

/

E. Case W. Haass

~

D. Eisenhut D. Muller R. Purple R. Ballard' B.J. Youngblood W. Regan A. Schwancer R. Mattson F. Miraglia P. Check

+

J. Miller

0. Parr-G. Lainas F. Rosa w R. Vollmer W. Butler J.P. Knight W. Krager R. Bosnak R. Houston 1

R. Schauer W. Gammill R.E. Jackson L. Rubenstein OIE(3)

T. Speis ACRS (16)

W. Johnston R. Tedesco S. Hanauer N. Hughes T. Murley V. Wilson F. Schroeder D. Skovholt~

i M. Ernst NRC

Participants:

- K. Kniel G. Knighton s

G. Lainas A. Thadani j

S. Hanauer D. Tondi j

L. Shao J. Kramer C. McCracken D. Vasss110 S. Reynolds P. Collins W. Johnston D. Ziemann-T. Speis F. Congel 1

K. Wichman J. Stolz D. Eisenhut M. Srinivasan R. Mattson W. Minners M. Williams C. Berlinger H. Conrad E. Adensam

{

J. Laaksonen J. Mazetis i

f S. Newberry Westinghouse Electric Corp.:

W. Koo 1

E. Murphy D. Rawlins P. Matthews P. Rahe, Jr.

S. Pawlicki J. Esposito W. Hazelton

0. doodruff C. Cheng D.. !!ali nows ki s

a 4

-.__:.t__._

1

,m 4

I

SUMMARY

OF MEETING WITH WESTINGHOUSE REGARDING STEAM GENERATORS HELD ON MARCH 2,1982 A meeting was held with Westinghouse representatives on March 2,1982, in Bethesda, MD.

The purpose of this meeting was to have Westinghouse present their views with respect to steam generator tube degradation and steam generator tube rupture (SGTR) accident management. A list of attendees is shown in Enclosure 1.

Westinghouse presented the steam generator configurations of various s

models with feed ring design and pre-heater design and identified the major differences among the models, especially with respect to the feedwater flow path.

It was indicated that out of the five SGTR events 1

experienced by Westinghouse steam generators in the past seven years, 1

only two SGTR events are corrosion related. Two SGTR events are considered preventable because one event was caused by the presence of a foreign i

object and one was due to excess tube ovality in a tube fabricated by a foreign supplier. The latest SGTR event is still under study. Westing-house also indicated that with 708,000 steam generator tubes in service, j

approximately 18,000 tubes (2.6%) were plugged and 45% of those plugged tubes were in four plants.

q Steam generator tube degradation was classified into two groups (large leak and minor leak) based on the potential magnitude of primary coolant leakage. The types of tube degradation with the potential of causing a large leak are those at the U-bend apex area, those resulting from the presence of foreign objects, and IGA / SCC above the tubesheet. Other types of tube degradation resulting from denting; IGA attack in the tube sheet crevices and in the rolled tube; pitting; thinning; wear at anti-vibration bars (AVB) and the preheater baffle areas will in general, result in small leaks. Westinghouse considers the types of tube degradation 1

with the potential of causing large leakage as mentioned above are control-j l abl e. Leakage from the U-bend apex can be prevented by plugging since 3

experience has shown that this type of degradation usually occurs only 3

in the tubes of first or second rows. Degradation due to the presence of foreign objects can be prevented by stringent administrative control of tools and materials used during secondary side maintenance and visual

,j examination aided by advanced optical equipment for areas not directly I

accessable.

f In addition, Westinghouse recommended the following methods to control crevice and sludge pile corrosion, (1) reduce the operating t'emperature for plants when such corrosion is evident, (2) sleeving, (3) reduce containment inventory by sludge lancing, isothermal soaks or depressured flushing.

e

. In addition, Westinghouse recommended the following methods to control crevice and sludge pile corrosion, (1) reduce the operating temperature for plants when such corrosion is evident, (2) sleeving, (3) reduce -

containment inventory by sludge lancing, isothermal soaks or depressured

]

flushing.

A brief status summary, based on inspections of 51 operating Westinghouse plants, was presented on the tube denting problem.

It was indicated that 25 plants showed various degrees of denting activity; the rest did not show any sign of denting. Among the 25 plants, 12 plants are considered active, the other 13 plants are stabilized. Of the 12 active plants, i

Westinghouse rated denting in three plants as extensive and the other nine plants, minor.

Westinghouse recommended frequent monitoring of hydrogen content in the l

primary water as a means to estimate the extent of ongoing tube denting j

activities. This is based on the theory that the corrosion process 3

associated with denting generates substantial amount of hydrogen in j

forming magnetite.

The guidelines for SGTR emergency response (EGR), which are sponsored by I

the Westinghouse Owners Group were presented. The basis for SGTR EGR's j

are operator intensive and include operating experience obtained from P

SGTR events. The status and issues in SGTR ERG review pertaining to 3

pre-TMI guidelines (pre 3/28/79) and post-TMI guidelines were outlined.

i Items, issues, and guidelines, developed or to be developed, in Phase I ERG (11/81) and Phase II ERG (6/82) wert: discussed. Westinghouse also identified the areas to be emphasized in post Ginna Review of ERG which i

are, (1) a continuing effort to develop optimum ERG's and (2) review of ERG training methods.

]

Westinghouse concluded that the steam generator tube degradation problem is under control as demonstrated by the decrease in the number of tubes required to be plugged in recent years. Westinghouse also emphasized j.

that out of the 18,000 plugged tubes which is 2.6% of the total tubes in j

service, 45% are in four plants.

Since the information presented by Westinghouse was proprietary, Westing-house agreed to document the information pursuant to 10 CFR 2.790.

tc t

t

]i Keith R. Wichman Operating Reactors Assessment Branch j

Division of Licensing i

1

Enclosure:

j 1.

Attendance List

)

]

1 i

L

. :a.....::.~u.r=............

.a..-... a

. :. ;.. -}

t,..

• ATTENDANCE LIST

(

MEETING WITH WESTINGHOUSE MARCH 2, 1982

.y i

NRC Participants 1

G. Lainas, DL S. Hanauer, DST i

L. Shao, RES C. McCracken, CMEB S. Reynolds, Region I r

W. Johnston, DE

]

T. Speis, DSI K. Wichman, DL T

D. Eisenhut, DL R. Mattson, DSI j

M. Williams, DL H. Conrad, CMEB 3

J. Laaksonen, DSI j

J. Mazetis, DSI 4;

S. Newberry, DST

'l W. Koo, DL l

E. Murphy, DE P. Matthews, DE S. Pawlicki, DE W. Hazelton, DE j

C. Cheng, DE

'i

.j Westinghouse Participants J

2 D. Rawlins i

P. Rahe, Jr.

.]

J. Esposito

'j

0. Woodruff J

D. Malinowski ACRS Participant 1

E. Igne

!)

i ii*-

/V

. g, a

UNITED STATES

~

(".

I' t -

~

.i-S NUCLEAR REGULATORY COMMISSION j

WASHINGTON, Di* -0555

%"...../

OFFtOE OF THE March 22, 1982 CHAIRMAN The Honorable Morris K. Udall, Chairman

?

Comittee on Interior and Insular Affairs United States House of Representatives E~

Washington, DC 20515

'.c.

.~

Dear Mr. Chairman:

s i

~

This is in response 'to the questions posed'in your February 5,1982 letter

+

relative to the recent event at the R. E. Ginna Nuclear Power Plant.

Our responses to your questions are enclosed.

ds a consequence of this event, I, too, have questions on the incident and

~

-1 its generic implications and have, on January 29, 1982, requested the

.g -

staff to establish a Task Force to review and evaluate the Ginna incident.

_ 3.-

An interim report from_ that effort is expected to be~ completed this month and may provide detailed answers to some of your questions. The remainder of your questions are addressed in the enclosure.

Sincerely, s

Nunzio Pa adino Chairman

Enclosure:

Responses to Questions

~

cc:

Rep. Manuel Lujan 4

e kl9,0(L.

.,, yl i er

'd V hl"/

l

/2.

3,

~

• QUESTION 1.

What is the primary significance of the Ginna incid'ent? -

' '~

y O

y ANSWER

~

The primary significance of this event is that it apparently occurred without advance warning and challenged the ability of the plant and operators to respond in a safe manner.

It also points out the inadequacies in the steam generator inspections; i.e., the licensees do not inspect the secondary sides of steam generators, with the exceptions of.a few plants that have suffered extensive tube denting and support plate cracking.

The safety objective in such an event is

~to prevent' fuel damage and~ to allow only minimal releases of radio-active materials to-the environment.

The tube failure, whether it be i_.

the result of chemical or metallurgical reasons, or some type of

~

mechanical unloading mechanism, has not yet been detennined.

The

~

~ failure-m6de and the plant and. operator responses will be addressed in the NRC Task Force 45-day report.

q t

e a

e 4

.en

&9 e

i O

e6 9

l l

e S

6

,a

.s

(*

y

s. -

What was the leak rate through.:he break as a function of time?

00ESTION 2.

ANSWER The attached graph (Figure 1) is our preliminary estimate of the leak rate as,* a function of time, calculated from infonnation provided by the licensee.

o o

=#

e

=o M

as 4

l- ;; -

3 m

e D.

g F

e S

e i

t l

j

..,x,. ;..,

FidVRE 1.-

G.ticialip 3

2 Pre li m ino.ry E stimmte o f i e

• k h.t e n. Ti rn e ' -

eaa-

't.

R 600 E

a.

4 f

nn

.v..

4J d..

aC 2co i<'

2&

g e

i.o g

i I

t i

I I

I*

-2o0 :

Io il i t, I"

ct TIME

'(

hou.rs, EST 1/z rho

s QUESTION 3.

What triggered the steam tube rupture?

4 JANSWER The licensee is continuing his inspection of the steam generator to determine the cause of the failure.

This' inspection will include removing sections of several tubes, including the ruptured tube, for laboratory examinition.

e

• e

1. 's 9

?**

e g

g D

.3 m-'e-9 s I

w

~

e.

m7.

~~

I e

h 3

p g 9

w. [

ep eg

'~

s e

9 e

9 O

m o

e G

e E

O 9

e O

~

l

OUESTION 4 Had there been. indications of leaking steam generator tubes prior to the rupture on January 25?

ANSWER A table of the history of steam generator tube inspection and plugging through May,1981, which includes leakage experience, is attached as Table 1.

Although preliminary information from the licensee stated thht the failed tube was not leaking immediately prior to the tube rupture, whether there was any indication of such leakage will be addressed in our 45-day report.

5 e

= --

e.

9

~...

5~

s m

g e

C e

s.

1 M

e*

J

=

(

7.

~

0 T ABL. 1. 9. '.. D<..

t a.:

o.

c

/..

STEAM GENERATOR TU.DE IllSPEC. TION AND PLUGGING llISTORY.

.,s.'

No. Defects

• No. Tubes Primarftg l

No. Tubes Secondary Total,

Type of Requiring Plugged / Sleeved / Pulled Inspected Leakane, gpm Defects Degrad.

Repair A

"D Date A

D A

D

A/B..llot _

A/D Cold -

.1/--/--

--/--/--

1 0

0 0/--/---

0/--/--

1**'-

~~-

~

l

~~--

~

13/--/ 2 0/--/--

0 0

.n Factory'.

19 0-14/72 1050/----

.----/----,

2 0

2/- /--

0/--/--

19

.O a.

)3/74

'3259/1090

. 516/ 51 G... ----

2. t0 a

13/75 2174/1931 442/.442' O.0050.A S/G

,46 11 b/a 46 11 46/--/ 2 11/.--/ '-

11/74 1707/ 672 430/

39 -

2/--/--

0 2-0/--/--

39 2'

" 39 /--/--

2/--/- <

31 /7 6

---

/

53

---

/----

0.091 D S/G 0

2 a..

.5/--/--

'39 2

a 4

.100/1025 --/

75 0.'099 D S/G

.0 15 b

0 15-0/--/--

3192/3247 3192/3247

--~~

13 1

'a 13 1

13/--/--

1/--/--

]2/76 b

5

,-- /-- /- -

5/--/-,

04/76 D'4/77 '

2003/1525 260/ 260-5 0

--/--/--

0/--/--

---

/.300

---

/----

0.012 D'S/G L-8 b/a 14/--/ l; 1

15

'b 1

15 -

1/--/--

- 6/--/--

07/7'7-

@l/78

---

/ 350

---

/----

0.060 8 S/G 6

--/--/-.

@4/70 2049/1714 325/ 375 6

b/a/c 13

--/--/--

63/2-/--

02/79 2049/1714 325/ 375 13 c/a 1

31-d'/ c '

1 13 1/--/--

20/--/ 3 12/79

---

/1200

-.---/----

0,007 D S/G 2

--/-- / -

. 0/ r/--

04/00-

'3139/3102 325/ 375 3'

6-

--/--/ -

1;M/ 3_

c.

11/00 3130/3151 325/ 375' 15 c/a T27 T

122/--7'T,106/21/7 05/01 3138/3141 325/ 400 T2f T2T or.

or or or 3.7%

3%

3.7%

3.5%

or or 3.7%.3.9%

l i

.?

Type of Degrad.

a

~Hastage.

ili'~ Cracking '.

lc - ID' Cracking

,d - IGA

, e - Pitting.

f - Fatigue Cracki.ng (D&W),,

"5

,g

. Erosion / Corrosion (B&W)

• =

r o

' QUESTION 5.

Whdt was the cause of the PORV's apparent failure to close?

Does the apparent failure of the PORY to close cause doubt

-about the adequacy of the industry's program to test such valves?

ANSWER Ginna's power operated relief valve (PORV) uses pressurized " control" air to remotely operate the valve.

Control air is routed through solenoid pilot valves which in turn pressurize one side of a flexible diaphram

~

in the PORY's valve operator and simultaneously vent the other. The differential pressure across the diaphram causes it to flex and in turn moves the valve's stem and disc.

If shut, the valve is held ' shut by an internal spring and air pressure.

If open, the valve is held open by air

. pressure alone.

Based on information provided at an NRC meeting with the licensee on February 10, 1982, the failure of the PORV to close resulted from a failure of a solenoid valve in the control system of the PORV. The

~: '

"a -

failure was related to a modification of the solenoid valve that was trade 1.

specifically fo'r the Ginna PORV control system.qThe function of the failed solenoid valve is to open and relieve air pressure, thus permitting J=

the PORV-to close when signaled to do so.

At Ginna, the failed solenoid valve is physically located within the pressurizer enclosure some distance from the PORV and is not considered to be a part of the PORV itself.

y The PWR utilities, in response to one of th'e Comission-approved NRC Action u

Plan Items (i.e., Item II.D.1, HUREG-0660), funded the Electric Power a'

Research Institute'.{EPRI) to conduct qualification testing of full-size prototypical PORVs and safety valves.

The testing.of th~e PORV's in the EPRI program was completed ~as of the end of A.ugust 1981. The PORVs were tested for open and closure capability <for a variety of fluid conditions, proposed by the utilities and EPRI as generically representative of the types of fluids PORVs could be exposed to under transient or accident

.. conditions.

The NRC staff reviewed and comented on the EPRI program as it was being formulated.

During this review, and during the development of the Action Plan requirement, the problems of including PORV control systems in 'the program were specifically discussed.

In addition.to the enormous complexity involved in includi.ng as many contiol systems in the test program as there are ;PWR plants, it was also recognized that the PORV. control system elements are not directly furnished by the valve manufacture'r with the valve. ' For these reasons the PORV control system was not included in the generic PORV However, the lessons learned from the malfunction of the air--

test program.

operated control system for the PORY will be factored into a current

/

evaluation which is assessing the need for improving air systems serving components and systems important to safety.

In addition, the potential for accidents. or transients being made more severe as a result of control system' failures or malfunctions is being addressed in Unresolved Safety Issue A-47,

" Safety Implications of Control Systems."

G G

2

>P OUESTION 6_.

What would the course of the inciden1 have been had the PORY

~

block valve failed to close par-ially or fully following failure

~

of the PORV to close fully?

ANSWER Had the block valve failed to close after the PORV stuck open, the additional-coolant loss from the primary system would have caused the primary system pres-sure to continue to decrease below approximately 900 psi.

As the pressure in the reactor system decreased, the combined leak flow (through the valve and the rupture) would decrease and safety injection flow would increase jntil the flows

/ indicate were approximately equal.

Analyses by Westinghouse in WCAP-9600

=

The

'that the reator system pressure would stabilize at approximately 700 psi.

pressure would then remain relatively constant until the operator took action If the block valve to depressurize the plant with the intact steam generator.

i.

were only partially closed, the combined leak flow and safety injection flow.

I,'

C~2~ would equalize at' a pressure between 700 psi and the 1300 psi which was reached at Ginna a.fter the block valve was fully closed.

Additional leakage out of PJ-the reactor system thro' ugh the br6 ken tube in theli.solate.d'steain' generator v6uld not' occur'for the case 'of the block Valye stuck fully open'sincef the'prip.ary '

~

~

system pressure would be less than the affected' steam generator' pres'sure, if the block valve were only partially closed, th.e' reactor' system'might be pressuriz'ed so that the leakage would be less than that which occurred vitth the~ block yalye ' 7'-

G'.

fully closed.

~

The effect on core coolant inventory of a combined PORV leak and steam generator

~

~

tube leak would be similar to' a postulated break in the' reactor coolant hot leg.

with an equivalent break size of about 2 -1/2 square inches ~.

The consequences of this event on core coging would be bounded by the. spectrum of small break; analy performed for Ginna.

These analyses demonstrate that the' core is adequately protected by the emergency core cooling system in the event of a small break

~

LOCA.

The staff preliminarily concludes,' based on the discussions above, that the effect of the block valve failing to close or leaking during the event;at Ginna would have been a decrease in coolant loss through the steam. generator' tube

~

and an increase in coolant loss through the PORY.

Since coolant loss through the PORV is confined within the. containment building and coolant loss through the broken tube may be released through the secondary system safety valves, Vff-site doses, would probably have been lessened had the block tvalve stuck open at Ginna.

Small break LOCA analyses for Ginna indicate that the core would be adequately cooled had the block valve failed to close.

However, the dual break situation would have been more complex for the operators to diagnos'e and would have introduced the added difficulty of more water and radioactivity

• being released inside containment.

1/ Report on Small Break Accidents for Westinghouse NSS System, WCAP-9600, Westinghouse Electric Corporation, Jun.e 1979.'

t I Letter from 1.eBoeuf, Lamb, Leiby & MacRae, Attorneys for Rochester Gas' and Electric Co,rporation,' to L. Muntzing, U. S. AEC, transmitting small break LOCA analyses for Ginna, September 6,1974

\\

6 s.

~

00ESTION 7.

Did the procedure for responding to a steam generator tube rupture contain instructions for actions to be taken in response to development of a steaa bubble in the reactor pressure vessel?

ANSWER Based on preliminary information froc the licensee, we understand that the Westinghouse Guidelines were used (Revision 1, April 1980) for developing 3

plant-specific procedures and did not contain specific instruction for responding to a steam bubble in the reactor pressure vessel head area; therefore, they were

~not included in the Ginna procedures. However, based on special training and

,5 their knowledge of the TMI event, the operators were able to recognize the existence of the steam bubble through observation of the rapid increase in pressurizer level and reactor. vessel head temperatures in conjunction with reactor coolan't system pressure which indicated saturated steam conditions existed'-in the head area.

Furthermore, readi.ngs from the core exit and yesse.1 upper head thermocouples in conjunction wi.th the prim,ary system pressure ^

confirmed that the steam bubble was confined to the head apen.

~

A full review of the Ginna procedures is being conducted., a,nd the'results wi.ll be included in the 45-day report.

~

9 4

9 D

S e

4 m

e 6

et eS e

46 e

M s

. l

+

e v

-n,.

n y

~

~

Was there a need during the in'ident to take actions not specificd QUESTION 8 c

in the plant's written operatir.; and emergency procedures? Were the emergency procedures in place at Ginna consistent with Westinghouse guidelines as discussed in the January 28, memorandum f rom Mr. Speis to Dr. Mattson?

]

ANSWER Plant operator response to the event, including the use of procedures, is being f':

reyiewed and the results will be included in the 45-day report. The emergency procedures in place at Ginna were based on the Westinghouse Guidelines Revision I dated April 1980.

"The discussion of the event in the subject January 28, 1982 Speis memorandum I

concerned proposed Westinghouse guidelines dated September 1981 which are currently

_(

under review by the NRC staff.. Further discussion is provided in encl.osure 8-1 (SECY 82-58) dated February 10,- 1982).

1 sue.

e

',~.

e e

e k

m E

e s

O 9

O

s t

QUESTION 9.

Had a water level measuring device been available, would it

~

have assisted the operators in determining the size of the steam bubble in the pressure vessel and otherwise in bringing the plant to a stable condition?

ANSWER

~

There are several types of water level indication systems being considered by industry and the NRC staff with respect to assisting the operator in making determinations of inadequate core cooling.

Some of these systems include level indication in the reactor vessel head region.

Had such a measuring

~.

Idevice been installed, it likely would have been an aid to the operator.

The operators, however, did use the available instrumentation (pressurizer level,

-p-reactor coolant system pressure, and vessel upper head thermocouples) in making

{-

determinations 'of the existence of the steam bubble in the reactor vessel head.

M Furthermore, the core exit thermocouple' readings in conjunction with the reactor ' coolant pressure confirmed that the steam bubble was confined to the a

reactor vessel head area and took actions accordingly.

W

-4

.[?

?

e 9

g a

6 e

k as 6e 9

I ca mu=

em>~

4 S

i l

5. '

00ESTION 10.

What consideration has bee'n given the potential for radi~oactivity escaping PWRs via a path including breaks in steam generator tubes and a stuck open safety valve?

ANSWER

.i Steam generator tube rupture accidents are one of the class of design basis ac-cidents considered ~by appli~ cants and staff in each review of PWR license applica-tions. The staff's Standard Review Plan, NUREG-0800, describes the criteria and

. procedures used at Section 15.6.3, " Radiological Consequences of Steam Generator Tube Failure (PWR)".

.2.

The analysis focuses on the potential release of radioactive noble gases and radiciodine both pre-existing in the reactor primary and secondary coolant,. and T;-

generated concurrently with the accident. The former case uses the maximum activity levels permitted by.the plant's proposed Technical Specifications.

The latter case

~

postulates activity ' released from the fuel as a result of the accident, including

_~

the potential for fuel failures.

The steam generator tube failure is assumed to be a double ended rutture of a single tube for purposes of calculating the rate of transfer of primary coolant to the secondary side of the affected steam generator.

Flashing of the primary v

coolant is assumed to occur in this process with subsequent atomization and trans-fer of activity to the_ steam phase.

Radioactivity entering the steam generator from the primary system is assumed to leave the steam generator, become airborne immediately, and be transported directly to the atmosphere via leakage paths not mechanistically specified.

Such leakage could be through a stuck open~ safety valve, an open atmospheric dump valve, or through the condenser vent system.

- For FSAR safety analyses, such releases are assumed to occur during the first 30 minutes of the event, after which credit for operator action is allowed to terminate releases.

Exclusion area boundary and low popirlation zone boundary doses are calculated and cor. pared with the thyroid and whole body dose guideline values cited.in 10 CFR. Part 100.

Conservative values of site specific atmospheric dispersion character-istics are used in these calculations.

The system response to the event postulated ~in this question is not covered by the Standard Review Plan.

However, it is being considered in the preparation of new emergency procedure guidelines per TMI Action Plan item I.C l.

ea e

4 e

a e

S e

e

s

>0 QUESTION 11.

Is-it generally agreed that if a leak had developed in both steam generators, the operatcrs would have been able to institute the

" feed and bleed" process described in Mr. Speis' January 28 memo-randum?

ANSWER' Had a leak developed in the second ("A") steam generator at Ginna, the need to institute the '" feed and bleed" process to assure continued core cooling would have depended upon the leak size and total leak rate of primary coolant out of f

the primary system.

It is uncertain whether the operators would have been able to institute " feed and bleed" for reasons described below.

The primary concern associated with two leaking generators is that in order to I.-si -

use the steam generators to cool down the primary system to the residual heat removal (RHR) system entry level, the primary system pressure would have to remain slightly higher than the pressure in both_. faulted generator secondaries during cooldown.

This would result in continued leakage of primary coolant to the secondary system.

Primary coolant would have to be replaced by the high pressure injection (HPI) system which pumps water from the refueling water storage tank (RWST) into the primary system. Thus, there is an arount of leakage that eventually affects the ability to cool, the plant to RHR entry conditions

~

prior to depleting the RWST. This behavior is different than other small loss-of-coolant accide'nts in the prirary system.

In those accidents leaking water will accumulate in the cor.tainment sumps.

Once the RWST level drops to a preset value, the pump suction is switched from the RWST to the sump and sump water is recirculated through the core.

Decay heat is ultimately removed by the containment heat removal system.

For larger tube leaks in both steam generators, which might deplete the RWST in-ventory prior to RHR entry conditions being reached, the operators would be expected-to open all PORVs, essentually the same effect as causing a small break LOCA inside containment. This would rapidly depressurize the primary system (as well as remove decay heat) to below the faulted steam generator secondary pressures.

In parallel with this action the operators would isolate both steam generators. Prirary coolant makeup.would be acconplished with the HPI pumps.

At G'inna, a two-loop 1300 MWth' Plant, there are two PORVs manufactured by Copes-

~

Vulcan with a relief capacity of 179,000 lb/hr. steam.

Although neither the staff nor the licensee has perforred any detailed calculations, scoping estimates indi-cate that the Ginna plant can remove decay heat by the " feed and bleed" process.

It should be pointed out that in order to establish " feed and bleed," the operator m'ust first establish PORY operability.

In the case of Ginna, this involves reestablishing the air supply to the PORY which was initially isolated on low pressure safety injection actuation.

4 e

e

~

ANSWER 11 (CONTINUED)

At Ginna, there are procedures in place which instruct the operator dn how :

to reset the safety injection signal in order to enable reestablishing the air supply necessary for PORY operability.

This procedure was, in fact, I

u_ sed in reestablishing instrument air which allowed the initial operation of fthe PORV at Gin'na during the tube rupture event.

I 'dditionally,'there is a backup nitrogen -system which is nanually controlled A

from the control room which can be used to actuate the PORVs in the absence

~

of normal instrument air.

b '1 '

It is noted that failures in both steam generators are not required in the, design basis for PWRs.

Furthermore, existing emergency procedures, such as those at Ginna at the time of the tube rupture accident, do not provide the

~

operators with explicit guidance on how to cooldown the plant with ruptures

-in cultiple steam generators.

However, as a result of the TMI accident, the

~-

staff's TMI Action Plan item I.C.1 requires the industry to upgrade emergency operating guidelines and procedures to cover nultiple failure events.

One of the specific events cited in NUREG-0737 is tube failures in multiple steam

r _.

generators.

Significant resources to the upgrading of guidelines and proce-dures have been allocated by both the industry and the staff.

We anticipate r._

approving the new emergency procedure guidelines by the end of FY 82.

If this goal is met, upgraded procedures should be implemented at all operating plants by FY 83.

g D

,4 9

e

s s.

i OUESTION 12.

How many steam generator tube ruptures per year of the Ginna magnitude or greater do you expect?

ANSWER There have been four steam generator tube failures of this type (greater than 50 gpm) at pressurized water reactors in the U. S. to date.

The facility, date of the event and estimated leakage rate is as follows:

Pl ant, Date Gallons / Minute ((Maximum)

Point Beach Unit 1 02/26/75 125 Surry Unit 2 07/15/76 80 o

Prairie Island Unit 1 10/02/79

~

390

~

Ginna 01/28/82 700 The above data indicates that for all 48 PWRs licensed to operate in the U. S.

(as of February 1), about one tube f ailure.has been occurring every two years since 1975. The leakage rate from the Ginn~a failure is approximately the maximum possible for a single tube failure; therefore, leakage much in excess of this amount is not expected.

The technical resolution of Unresolved. Safety Issue A-3,4,5, " Steam Generator Tube Failure," is in its final stages of development and includes consideration of recommendations for improvements in inservice inspection, steam generatos secondary water chemistry monitoring and turbine condenser inspection.

These improvements when corpleted should lessen the overall problem of tube corrosion. However, these changes, when implemented, are not expected to elimi.nate totally the possibility of future tube failures.

e d

4 w

Y e

s e

l

r 0'UESTION 13.

What is the likelihood of 'several steam tube ruptures occurring at one time? What is the maximum number of simultaneous or near simultaneous steam generator tube ruptures that are considered design basis accidents following which the reactor can b.e brought to a safe shutdown condition by following plant operating and emergency procedures?

-y.

i: ~

ANSWER

~

Il _;

. Experience to date indicates that multiple tube failures is a low probability event.

( _.

As was discussed in our response to question 10, the steam generator tube rupture that is postulated to establish the design basis for the plant is the

,.~.

equivalent of a double-ended rupture of a single tube.

For des.ign base purposes, this is considered to encompass a spectrum of smaller leaks in 'either single or multiple tubes.

__~ g. _

It is our belief that plants can most likely accommodate a larger number of

~ tube failures, 1/ without exceeding the capacity of the.ECC systems and without g'

. leading to core damage.

Consequential radiological releases would also be 4:e~-

- calculated to increase.

However, the radiological source would still be.

due to the induced primary coolant activity and not from fission products

]'

released due to gross fuel failures resulting from the event.

In addition to the tube rupture used for establishing th' plant design basis, e

emergency operator guidelines and procedures presently being, upgraded as a

.. result of the TMI-2 accident will address nethods for canaging ruptures in

.J.

multiple tubes and multiple generators.

(See response to Question 11 last paragraph).

II We interpret the second part of the question to mean tube. ruptures alone, not to be concurrent with or as a consequence of design base accidents (either primary system loss of coolaht accident or main steam line break).

The tolerable number of tube ruptures concurrent with or as a consequence z.

of design Basis sccidents is rather small, dependent on the plant thermal hydraulic design and the design basis accident in ques' tion.

However, we expect the tolerable number of tube ruptures would most probably be'much larger for more likely accidents or transients.

A

QUESTION 14. Did WASH-1400 or mare recent risk assessments determine the

' ~ ~

probability of occurrence of events in which one or more steam generator tube failure (s) are followed by various combinations of PORV, block valve and safety valve failures?

ANSWER i

Steam generator tube rupture alone has been considered in PRA's as one of several types of small-break accidents to which pressurized water reactors may e

be subject.

Multiple tube ruptures and ruptures in more than one steam generator have not been considered in PRA's nor have combinations of other

~

. conponent failures such as those identified in the question been considered.

~

. We are now taking a more careful look at these scenarios.

k wel w m

e a

f Y

e 4

S e

.4 iY 9

m e

C

=

.as 8

,e g

g

~.

z 00ESTION 15. How long did it take to reach cold shutdown?

Is this a period longer than desirable? What was the reasons for the period being longer than now.al? What kinds of malfunctions during the extended cooldown period might have led to a' signi-

~

ficant release 'of radioactivity to the environment?

ANSUER The time The plant was in cold shutdown the day following the vent (6:53 p.m.).

- from reactor trip to cold shutdown was 33 hours3.819444e-4 days <br />0.00917 hours <br />5.456349e-5 weeks <br />1.25565e-5 months <br /> 25 minutes.

In

.The period from reactor trip to cold shutdown was no longer than desirable.

f act, there was no urgent need to reach cold shutdown conditions, especially after the steam generator tube leak had been terminated (equalizing primary' pressure with This the f aulted steam generator) and the plant was in a stable shutdown c'ondition.

stable safe shutdown was reached about two and half hours after the reactor trip.

~

In general, it is expected that cooldown with.a ruptured tube in one steani

[

This slower generator would be significantly slower than a normal cooldow the pressure in the ruptured steam generator to minimize or terminate reactor Since the direct release of steam from coolant leak flow through the rupture.

the ruptured steam generator is to be minimized (the steam would contain radio-active products from the primary system), depressurizing the faulted steamIn GiWid, j

gerierator must~ be'liy other less direct means.

the ruptured tube was drained to'the reactor coolant system through the r'uptured

~'

Additional cooling and depressurization was provided by cold auxiliary tube.

The length of time for the feedwater which replaced part of the drained water.

cooldown was primarily governed by the management's desire to go slowly and The time to reach cold shutdown was consistent with the plant!s cautiously.

. condition and, therefore, no longer than desirable.

If there had been no steam release from the ruptured steam generator in the early.

stage of the event, it is reasonable to expect the coold6wn period would have.

For a large initial; steam space in the ruptured steam generator,

.been longer.

a limiting factor for steam generator draining is need. to keep the stea generator tubes covered.

rapid condensation would occur resulting in a rapid d leakage back through the ruptured tube..

During most of the extended cooldown period at Ginna, the ruptured steam ge was isolated and its pressure was significantly lower than the safety valve set Th'e All other steam valves from the steam generator were secured.

reactor coolant' system was controlled similar to a nonnal cooldown, except for pressure.

measures (increased letdown, boration) to accommodate the' leak flow to the pr n

system coming from the secondary side.

As indicated in the response to-Question 10, potential releases of radioa to the environs during the short term or-lung tenn most directly relate to. -Such le additional malfunctions in the faulted steam generator.

through a stuck open safety or'. relief valve flow path or th releases are assumed during the first 30 minutes of the event, after which c vent system.

for operator correction is allowed.

s O

s

) e

~

00EST10W 16.

Did any part of the reactor pressure vessel cool at a rate in excess of that stipulated in the plant technical specifications?

ANSWER Analysis of infonnation to detennine specific cooldown rates is being conducted and will provided in the 45-day interim report.

i..

c

[

2:X-l: ^

K-

[ 'a 4

7 y

e e

0 O

% e S

9 b

i.

d 5

e.

e e

e O

g O

qq O

e O

4

L*.

Was there a capability at Ginna to remotely vent the reactor..

D -s the Commission believe that QUESTION 17_.

pressure vessel high peints? conditions might. develop in PWRs

~

controlled valves for the purpose of venting steam?

ANSWER -

The physical capability existed at Ginna but was not declared operational The reactor vessel head vent system, since. staff review is not complete.

including associated hirdware and control system, has been installed at Ginna.

Before the staff authorizes use of the installed vents, we will review not only the design but also the associated procedures which are to spec We expect to complete procedural reviews

_ 'and when not to vent.

our review of transients and accidents.

in FY-1982, and finish designs for all PWRs in FY-1983.

Engineering reactors), high point vents are required

~

This requirement was added for the purpose of providing a vent path for c

non-condensible gases that could accumulate in t.he primar head.

might accumulate in the vessel upper head after saturation conditions ar core cooling conditions.

in parts of the vessel, it is not expected they would be used for this purpose, S a

nor is it recommended that they be used to vent steam.

of Westinghouse and Combustion Engineering reactors does no y,

.~~

to continued core cooling.

lets, it would most likely condense as it came into contact with subcooled wate If, for any reason, the water exiting the' core was saturated, the steam would enter the hot. leg pipes and travel to the steam generators,,

exiting the core.

it,would be condensed.

For events such as the one at Ginna, the rethod preferred for removing s accumulates in the upper head of the vessel is to restart a reactor coolant pump.

The pump will force subcooled water into the upper head region and cond The operators at Ginna demonstrated the capability to do this i

steam bubble.

following the formation of a steam bubble in the upper head.

In PWRs with once-through steam generators (OTSGs) (i.e., B&W reactors), a bubble in the upper head of the vessel has the poten Pursuant to item II.B.1 of the THI Action Plan these plants will eventually have high point vents installed on the top of the hot leg inverted

~

condensing.

In addition, some utilities with B&W reactors will install vents on U-bends.

the top of the vessel head.

Analyses by B&W have indicated that int. 'rruption of natural c e

i the interruption of natural circulation will ultimately produce thermal-hydrau temporary phenomenon.

The staff is conditions in the primary system which restore natural cir

~

the relevant thermal-hydraulic phenomena.

D 6

6 6

O.*

AN5!!ER 17 (CONTINUED)

B&W has recently recomended use of the hot leg high point vents to vent steam.

which may accumulate during the recovery phase of a small break loss-of-coolant accident (SBLOCA).

During the accident phase of a SBLOCA, B&W has recomended the " bumping" of the reactor coolant pumps to sweep any steam trapped in the hot leg high points into the steam generator.

The use of the high point vents to vent steam in B&W reactors, as well as the acceptability of the B&W calculational models to properly predict the thermal-hydraulic behavior of the primary sytem under two-phase conditions, is under At this point in the review, it is our preliminary con-active staff review.

clusion that the use of the vents in B&W reactors to remove steam which accumu-la.tes at primary system high points may be the preferred method of steam removal if a reactor coolant pump cannot be restarted and run continuously.

m t:i 31

'q~

tr_.

• L o

A.

~

%9"5-h-

P b

w

~

b O

D D C es ee y

w

• e e

~

~

g- * -

QUESTION 18 At any point during the Ginna event, did the steam generator containing the ruptured tube c;ntrol the primary system pressure?

Are operators at.Ginna and otner PWRs trained with respect to actions to be t'aken when a steam generator controls primary system pressure?

ANSWER In order to prevent further contamination and to aid in cooling down the faulted steam generat'or, a feed and bleed operation was used.

This operation consisted of.providing feedwater to the faulted steam generator in order to

~ ~

maintain level within a desired band; a steam bubble in the steam generator was maintained during this period.

As a result of primary system pressure control by the operators through the use of the normal charging / letdown -

systems and by controlling the cooldown rate through the "A" steam generator,

~

~

4 primary system pressure was decreased in a controlled manner.

Pressurizer

- level was maintained and primary system pressure was controlled by the pres-.

However, during this period the plant was controlled in: such a surizer.

manner as to result in an inflow of water from the "B" steam generator to the

~

~

primary system through the ruptured tube.

This area, specifically during the early part of the transient, is being reviewed iurther and the results will be included in the 45-day report.

f

=, ~

The operators at PWRs are trained to maintain control over both primary and,-

secondary system pressure following a steam generator tube rupture. The goal is to minimize flow between the two systems by maintaining the two systems

~

within 50 psi of one another.

1

~

t 9

9 9

e o

g a

e 4

• m i

• e p

m 9

s 8

4 m

E ENCLOSUF.E 1 TO PAGE 8

~

_g

,~

(

)

~

February 10, 1982

\\

/

SECY-82-58 POLICY ISSUE

\\

~

(Information) c The Commissioners For:-

~

William J. Dircks, Executive Director for Operations From:

}

Subject:

MEDIA ATTENTION T A PRELIMINARY EVALUATION OF OPERATOR ACTIONS DURING THE GINNA EVENT To inform the Commission of.'the pr'eliminary evaluation.

Puroose:

On January 28, 1982, three days after the st'eam generator -

tube rupture event at the Ginna reactor, the Reactor Systems

-[

. Discussion:

Branch in the Office of Nuclear Reactor Regulation completed The ' purpose of the a preliminary evaluation of the event. evaluation was

' response at Ginna to the recently proposed Westinghouse emergency procedure guidelines for steam generator tube A copy of~ the resulting staff memorandum, rupture events.

Some of the results of the preliminary is. attached.

evaluation 4were briefly noted by NRR (Roger Mattson) at 28, 1982. A copy of the Comission briefing on January the menorandum' was also provided to Region I' staff at that

~

briefing.-

C A story on.th'is memorandum a'ppeared in'the February 8,1982 New York Times (also attached). There are two erroneous First, it fails impressions left by the Times article.

to note that in the memorandum the operator actions were '

~

. compared to new, not existing, emergency procedure guide-The new guidelines lines for Westinghouse reactors.

re currently under review' by the Reactor Systems Branch.

They were not being used by the Ginna' ope a

1

[

approved and implemented at operating plants.

Second, the Times _ article makes the memorandum appe'ar t

~

. conflict with other.statenents by NRC that the.Ginna operators

('

~

(

f.

~

~

1 Conta ct:

R. Mattson,.HRR 7 g'

[!

49-27373

%~','

L G

h~

_2 m

1

+

v p.';

(

The Commissioners '

~

~

performed well.

The memorandum'siys "it is premature t,o judge whether (a) the operator actions were correct, incorrect' or could stand improvement, and (b) whether dv.

the (new), emergency guidelines are correct or not."

The memorandum was speaking for the Reactor Safety staff in the Division of Systems Integration of NRR,wMeh, at that time, had not received a copy of the. actual'Ginna

-7 procedures or a written chronology of the eve.nts of (35 January 25.

Obviously it was premature,then for that staff to make a judgment on the correctness of operator '

'i-s A

actions..They offered none.

Neither did they contradict

~'

~

'[-

the Regional Administrator'.s statements tha't were based i

i on first-hand access to the necessary infomation. By failing to make this distinction, the Times article implies a division of. opinion between the Reactor Safety i.

staff of the bisision of Systems Integration and Regional Administrator when, in fact, none exists.

• - ~ -

~~

A-l *-

\\

William J. Dircks q-O-

Executive Director for Operaticas x

g. ?:.,

Attachments:

,(

1.

1/28/82 Staff Memo 2.

2/8/82 Times Article e

k O

s 8

.~,

b I

5 g

q 1

e

(

9 e

g

)

s. -,

a

.--. ~.,,'...*n.y.

(2

• 4 k.*

UNITED STATES h-kE'*7 ',

?!*

],{.f,. (' $_

NUCLE AR REGUL ATORY COMMISSION

p,

.e, mmat o u. o. c. 2em o

5..; % :

.,3

%.% ',f-JAN 2B n -

MEi10RANDUM FOR:

Roger Mattson, Director, Division of Systems Integration FROM:

Themis P. Speis, Assistant Director for Reactor Safety, DSI PRELIMINARY EVALUATION OF OPERATOR ACTIONS FOR GINNA SG

SUBJECT:

TUBE RUPTURE EVENT Based on the chronological listing of the Ginna events you provided us on 1/26/82,

" ~~. J which we understand was provided by R. Starostecki of Region I, I have asked Jukka 4. ~. Laak'sonen and Brian Sheron to compare the operator actions a e _i the Westinghouse Emergency Operator Guidelines for Steam Generator Tube Rupture

f. _.

Ev ents. This co'mparison is provided in Enclosure 1.

• These guidelines are called L T.;

EDI-0 and.E0I-3. We used thn 12 test version presently under review by the staff as part of TMI Action Plan Itan I.C.i.

However, the technical guidance is generally D

the same as the earlier versiens the staff reviewed and approved for the pilot

..s...

monitoiing program for NTOLS. However, we do' not know'if the eneroency procedures I'D ff in place at Ginna at the time of the accident were derived from or were consistent

~

with these cuidelines.

'Our preliminary conclusion is that the operator act'ed properly in u. sing the PORY

~

SI termina-to depressurize the RCS to the pressure of the faulted steam gensrator.

tion was also accomplished consistent with the guidelines, although it may have

'been' delayed, longer than necessary and resulted in a brief discharge of'the "B"

~

(faulted) generator to the atmosphere. The fact that the PORY stuck open complicated the scenario by producing a rapid depressurization which led to flashing of the upper head fluid. We also note that the SI pump was i estarted at 11:15.a.m., which re suited in a second lifting of the "B" generator safety valves. The reason for this

. ' ~

action is unclear.

e..

One observation we drew from this action is that. operators appear to be very hesi-tant to terminate HPI when they are allowed to, or even are supposed to. We point -

this out since, for the pressurized thermal shock. issue, the industry has tried to. -

convince us that operators wouTd always terminate HPI before the primary system was, unacceptably repressurized.

Another observation is that the operators tr'ipped the.,RCPs according to present instructiops, and restarted the A Loop RCP when allowed.

A discussion oi) the RCP

t

.~

. trip ' criteria is provided in Enclosure 2.

'A number of other preliminary observat_ ions were made of the Ginna event Which, I believe,' warrant further investigation..These are:

i r

.~

.CCNTACT:- J. Laaksonen, x29400 t

~

'B. Sheron, x27626

~

e p

8 (D

};_,-I..

' 2-M 2 B 19E

Rooer l'.attson t

. Stratification of the faulted steam oenerator_ - The faulted steam generator 1.

There war some-was being cooled ano depressurized by tne primary system.

evidence, based on thermal-hydraulic conditions of the system, that i;ignifi-cant stratification of th,e secondary coolant in the faulted generator ' occurred.

This resulted in the water in close proximity to the tubes cooling down, but

7. -

leaving a layer of hot water "in,sulating" the steam in the steam dome fro,m the Thus, depressurization of the faulted generator seemed to proceed cold water.

slower than expected.

It is not yet clear what safety significance.is as-9 sociated with this phenomenon.

- x_

si

.s

. 2.- Additional coolant system failures

~

m. ; _. __.

5

,. a)

Leak in "A" Loop SG If a leak also developed in the "A" loop steam generator, then primary P "'

coolant would continue to leak to the secondary, unless both SGs were Decay heat removal would thin 'need to be accomplished tiy P

~

isolated.

" feed and bleed". (HPI-for coolant'additiont PORY for coolant discharge).

(Westinghouse, in their latest guidelines, recommends cooldown using the Ei steam generator with the lowest level and probably the smallest leakage.)

• ,1-Y.

b) Stuck-open secondary side safety / relief valve

^-

. A stuck-open secondary side safety / relief valv.e in the faulted generator would produce a direct path for primary coolant to enter the atmosphere.

~

Moreover, the present emergency procedurss probably do not address this scenario.. The primary coolant would have to be made up with HPI water.

If the leak was not stopped, or additional cooling water supplies wera not made available, then eventually all of the liPI water from the refuel-ing water storage tank would be exhausted and a net lo'ss of primary co

~

Without corrective action, core uncovery would eventually,

f-would occur.

occur.

7-3.

Loss of steam-jet air' ejectors 4

The loss of the steam jet air ejectors due to low "A"-loop SG pressure.(< 150 psi) produced 'a loss'-of-condenser vacuum and required decay heat re I

Reasons why the A loof pressure was dropped so steaming to the atmosphere.

low, the reasons why the air ejectors were lost, and the significance of this in the course of the accident will have to be addresse3.

I.believe.it is pre.ature to judge whether (a) the operator actions were correct, incorrect *, or could stand improvenent,.and (b) whether the-energency guidelin This is :because -they are designed to cover a multitude o,f are correct or not

. scenarios in. which the Ginna accident was just one.

O

+

g e-4 g*

m.

~

e r

a

~.. ;;

t-

• Roger Mattson gg g g 1*-[.

e RSB has been addressing a similar scenario in' response to an AEOD concern.

RS will continue to investigate the above areas of concern, as well as any others brought to.our attention, and recomend action as appropriate.

p Z

MLv,

\\

~

Themis P. Speis,' Assistant Director "for Reactor Safety,

Division of Systems Integration

Enclosures:

C -. ~ As stated

• e -

s N

1 T-cc: H. Denton E. Case '- ~

D. Eisenhut L:,.

' S. Hanauer-R. Vollmer H. Thompson

~~

C. Michelson.

~

G. Lainas G. Holahan r

T. Ippolito ACRS(1)

L. Rubenstein W. Houston e

g 9

e e

e e

ed e

O s

e en e

e 0

S g

e g

o

.e M

D e

e o

e g

e e

e n

S

• e B

9 e

e e

e

t

!.. c:Ptnanne:. nu)NoAlt-Y OFinus Pcavent,v,v.t.cn picsu 0>.y 17.> e po,a.n.g py w w. ' p. a s a e i x, n i F A L' u i p, %. A t s j n t.

l.

I I

,1

. - cha!,116.t 'RAcA):f oFr 30dA/. oc ct., eses Pohv J

" 5'P ern vcPssmte s.duc -Lo ST. Cotu

\\

5J,slent.reptuuaiw 2co Psand opa,ur.arrs.

\\

on n

\\

,Pn,n%n-:'IsIesn anu.1Ltiea t.bn1

\\.

scm 9)

I

\\.

l

/

s se i

p s

~c:

N c.

4

\\.

v-N

. g%

.... - - - -s t

%u Os

~7?NE->

\\.,a.d,a\\ suhss S a.stlSe%vsar

-- - - QN'*'y 7s bh 5

l a su,se

,..q>ected) (:oc wi.ie. bin.ed on

..;..t p u a ta u -Ve ( % u\\... 'n e b 4 Nca> ch,,.5}drb ih. cu.-

3 5i;u;,,..,..jer;, (,s,,e.ww.

. t.,,-

f.-

.s,.. a.,..v.

,.. v..,..

q

. REMARKS.,'Iliese; remarks ara b enericN Emergency Operator Guidelines for Steam TIMl!

EVENT /0PERATOR ACTION

. Tube Rupture Eyents presently under st.aff review.

d

~

~ ~ ~ ' ~ '

i 9.25 5N.bIeruptureinSGB,'.findicated by

~

' ~ '

charging pump high speed, SG high level, steam /feedwater mismatch, air ejector radiation

- rip on. low' pressure, SI signal on Reactor Tripping 'on low pressure and fast depress

..a u.

m. o....

gw pressure, containment isolation, result-ization. indicate a severe rupture; Westinghouse r

.ing.in, loss'of instrument air and normal for a double-ended one tube rupture do not show charging flow and RCP seal flow as large depressurization.

9:29' Pressurizer level off-scalo low, RC pressure RCP, trip cciteria are met, the operator should tr~

decreased to 1200 psi the pumps manually 9:33 Operator trips RCP's Correct action, delayed 4 minutes The event has been diagnosed and the faulted SG h, 9:40 Operator closes main steam isolation valve' been identified from increasing level and pressur

~

at the faulted SG B and, the first corrective action is as told in the guidelines.

It is unclear if the auxiliary Flf to faulted SG is stopped and if all isolation valves

. Jw onam lines are closed.

9:46 RC pressure 1200 pst RC average temperature

<ea'ctor coolant temperature indicates that fast cooling in; compliance with the guidelines is in 475*F progress.

b53

.Codd SG A at 540 osi, level.76%, steam The status of SG A and the loop' A is as told in

~

dumped to the condenser, natdral circulation) guidelines. The initial cooldown er the RCS is al:

in. loop A, faulted SG B at 826 psi, level completed.

At this stage the gperator should st<

depressurizing the RCS to th6 SG D pressure to s:

89%

Because the RCih are f f the the tube leakage.

However, PORV is not ava$lable

.should be used.

because of isolated l'nstrument air system.

Faul SG pressure is surprisingly low and indicates lav proper isolation.

At this stage; the operator is able to depressur Operator resets SI, pumps still on, instrument the RCS, but he does not do that. Thus, the lea

'9:57 cir re-established continues and SI is needed to keep the RCS invez em m c L*

jnu.

tvtnl/UI'LRAIOR ACTION Charging pumps restof ttd; RCS at 15J0 ps,ig',e.. ; REMARKS"'

'.M'

' '

• Restart of charging pump's bercre stopping.i.1 ii -

i

,M 10:0T pressurizer level lh%,

leak is'.'not in compliance.wlth the guidelines or i

SG B level 100% narrow range, 400 inche,k icak increases becsuse the charging pumps incrca

, ;.:,

• RCS pressure; 'SG'B.. filling up is a conseq\\iedce

. ide range w

i

'. failure tb depressurize.}he RCS..

10:07 Operator opens the PORY manually The action is taken somewhat too late 1.0:08 Operator opens the PORY manually for a second If the PORY is opened on time there is no rea c.

time, PORY sticks open multiple PORV operation.

PORV block valve shut,RCS pressure 800 psi, Correct action to shut block valve.

Drop in Rr 10:09

ressurizer level offscale high.

pressure to 800 psi flashes hot liquid to steam RPV upper head, pushing water into pressurizer 10:10 SI pumps increase the pressure to 1300 psi After having r!.osed PORV.the operator is instru.c see if the system repressurizes.

If. pressure o creases 200 psi, then the 51 can be turned of f.

I to stuck open PORY, the ability of the operator follow this guidanch is questionable.

=_.

10:10

'T incore = 458*F Adequate subcnoling Auuusperic relief valve'Un au B isolated manually Pr:ecaution against opening 10:30 TPV head = 525'F (p, sat =850 [isi) ildt clear if upper head bubble condensed t t t Temperature supports existence of bubble 10:40 -

Saf.ety valve of the S(i B blows. operator Safety valve blowing is a direct result of secu secures the SI to reset the SI the SI about 30 minutes too late 10:42.

RCS at 800 psi There is obviously almostno steam bubble in the ssurizer and the pressure decreases drastically a some. water is dra~ined from the system.

Pressure stabilizes to a value where the water in the ur head starts to flash.

It is unclear how primary is lost because the faulted SG D 'should' be at ab safety valve set pressure (above 1000 psi).

s

~

n....

10:50 RC~ pump' seal return relief valve lift and dis-Reason for relief valve opening unclear, posslid

. related to the containment isolation charge to the PRT s'

o

.... ~,

..e.,...-...... c. c _..

RENARkS N' fIME ltVENT/0PERATOR ACT10t(.

l'off nJ.

~

'10:57 PRT ruptura disc blows, total loss PRT cverpressure results from combined.

~

of coolant

• 8000 gal seal., return relief valve operation 4

i il:15 SI punip~ started,SG B safety valve Reason for.SI start unknown, S1 re-Initial. inn lifts at 1035 psi (set pressure c'riteria are not met (loss of sub -cool ini, 1005 psi) -

pressurizer level <20%); wheri did the safety.

i

' ~

res6t?

11:15 T cold on operable loop was higher than core Reason unknown exit TC's 11':22 RPV head at 525*F, 97'F subcooling, pressure 1035 psi

_e.

11:29

.e

'RCP in loop A restarted, core exit and The guidelines tell to start at 1 cast one nr.P upper head thermocouples equalize a.t af ter the initial plant stabilization and SI 450'F termination, restart may be delayed because n problems with seal in.jection 12:00-Pressurizer level 80%

A 12:05 Honnal letdown established Nomal letdown /cliarging is told to be establi as the next step after SI termination 12:30 Slow cooldown started, RCS'at 923 psi 6:40 Level' in SG B re-established, SG cooled SG B depressurization is obviously no problerr by adding cold auxiliary.feedwater because there is not much steam lef t 7:05 RilR initiated, RCS at. 200 psi, 330'F 4

O o

ENC!.05 Uki'-2.

.c-7, i

. There are a number of questions that can bh raised bf f.helump tr p" issue.

- -r

~

. Th' RCP trip criterion for W plants is that the operators sho'uld trip ali e

RCPs if SI has actuated and the primary system pressure has dropped below a specified value.

This value (or rather the method by which' this value is

~

' obtained) was worked.out and. mutually agreed upon between W and the staff

f.

during the B&O Task Force after the accident at-TMI.

Essentially, the value is the set pressure of the. secondary safety valve pl6s~' adders'to account

~

-2 h c._h_.'for pressure drops back to the pressure gauge in the control room.

Typical '

M w-

s.._ c--

[h.[ values are expected to be between 1350 and 1450 psi.. A. more detailed dis-cussion of this pressure setpoint derivation is provided in Section 7.2.2-of

,

e r

,.;~.,.

NUREG-0623 (Attachment 1).

2 L:' ' It is noted that the charging flow at Ginna is isolated on an ECCS signal.

{ince -

V', - ;

seal injection flow to the RCPs is from the charging flow, it too is lost

' _U during an ECCS signal. -Thus the pumps would be required to be tripped

.following the ECCS signal, regardles's of the pump trip criteria..

For any steam generator tube iupture which depressurizes the primary system

~

to below the RCP trip p,ressure, there would be a need for the, operator to use the PORV to aid depressurization, since the primary system pressure will

?

For stabilize slighty above the faulted. steam generator secondary pressure.

{

a, smaller leak, the RCPS would not be tripped 'iixnedi.ately. and, the sprays w remain available to aid in the depressuri.ation.

For..CE and B&[ plants, the

[

1600-1750 psk)'. Thus RCP trip criterionis on low pressure SI actuation (around Both CE and B&W were

[~

earli'er RCP 1 rip would be expecte'd fortboth these' plants.

asked infomally.t,o adapt the y low pressure criterion,'but.bpth declined. -

~

Section 6.0 of NUREG-0523 reco:nnended that the-industry develop RCP trip criter

.=

which dinimized RCP -trip for non-LOCA transients (see Attachment 2-) and also reco:aranded t' hat procedures and training be initiated for handling non-LOCA events

...m which produce ECCS actuation and pump trip;' including instructions for:

2.;.

. a) tripping RCPs; monitoring and ini'tiating natural circulation; b) a pressure control without pressurizer spray; c) d) HPI termination; e) RCP restart criteria.

Atta'chment 3 provides section 7.2.5 of NUREG-0523.

The reasons for requiring RCP trip are as follows:

e continuing 'RCP operation For cer.tain small breaks in the p.rimary systent, 3

e

y. -, c f

will " pump" water out of the break, and produce a. greater coolant p[{ g IC -

inventory loss than if the pumps were tripped.

For agsmall break in the primary system; including steam gene

'T o

[- ' -

rupture, the coolant will.become two-pha'se and could evolve to a.

Wa are no,t: aware of 'any RCP's that ~have been significant void' fraction.

designed to. operate for extended periods of tirre in a highly void Continued operation in a highly voided system could result in exc

~

~

vibration and possible seal failure, or worse.

During the initial phase of many transients and accidents, the sy

, e' may resemble those of a small break,or a steam generator tube Eatiy RCP. trip with restart instruction is considered the most pru course of action.

k and

~

models which predict system performance' with a small brea u Analys.is h5 system void the RCP's operational still have large uncertainties when t Thus we do n.ot have a high confidence that pump fa fractions are high.

=

during high void conditions does not le'ad to unacceptable co

• e 4

e

3 3:,i ' '/.T Short-h., Eecuirements ine f ollo. :..; cescr10e the short-term receirements for pump trip for each of the reactor vendors.

- 7.2.1 Control F.com Ooerators IE Bulletins79-05C and 79-06C (item 1.3) require that two licensed operators be in the control room at all times (three for a dual control room) and that one of the two operators be designated to trip the reactor

.~~

coolant pumps should the f acility undergo a transient which r'esults in a The

--- safety injection actuation signal due to low primary system pressure.

-E 5

' designated operator'may perform any normal or routine control room duties at all other times.

The licensee should confirm that an operator is

-j

.),

designated to perform this' action on each shift.

6 d.7.2.2 Westinohouse-Desioned Plants I

y

... I'.

For the s...rt-term, the staff has adopted the following position for manual pump trip requirements on Westinghouse-designed plants.

y gj.y

~u -e 5taff Position on Pumo Trio for Westinohouse Plants

• " ' ~ ~

-N D We require that the reactor coolant pumps be tripped at a system pressure

~ ~

4'-

determined in the following manner:.

,. Q l (1)

Secondary System Pressure - Based on the number and size 'of the c

secondary system safety valves, the secondary pressure will be fj established by determining the pressure setpoint for that v,alve in -

^

';P Ai

• I which the calculated steam relief is less than 60 percent of the valve's relief rating.

If the calculated relief is grea'ter than 60 percent of the rated capacity, then the,next highest pressure 2

setpoint should. be used.

g Primary to Secondary Pressure Difference - To account fo/ the pressure

'd{

(2) gradient needed f or ' heat removal, pressure drop between the steam

~

generator and safety valves, uncertainty of the safety valve setpoint, pressure dro;i'from steam generator to measurement location, etc., the

-j primary pressure for RCP trip should be the se.c'ondary pressure as i

established by (1), above, plus 100 psi if the calculated adjustments i

~

are 100 ;isi: or less.

If the adjustment are determined'to be greater

.,l than 100 psi, the larger value should be used.

Instrument inaccuracies appropriate to that time in the loss-of-coolant'

~

(3) accident should be added to the primary pressure established in (2),

The resulting pressure is the indicated pressure at which the

'. 3 above.

j operator should trip the RCPs.

Combustion Enoineerino-Desianed Plants 7.2.3 Combustion Engineering has recommended that r'eactor coolaht pu=p trip'be manually initiated by the operator on receipt of. reactor trip ~and safety injection actuation signals.

Combustion Engineering is also evaluating the capability of their' pla'nts. to accomodate a pump trip on reactor trip and a lower system pressure by a method similar to that established for Westinghouse as specified in Section 7.2.2, above.,,-

The staff will accept the pump trip baseif on reactor trip and SI actuation fcr the short-term, since SI actuation pressure is. approximately 1550 to e

t

.i

V.

~.

g

[ 4. '

LCCA c2n result in a greater ens inver. tory loss, fro: the syru-than if tne p:::ps were trippec.

' = -

(2)

The ability to correctly represent the, thermal-hydraulic behavior in key components within the primary system during a small break LOCA with the reactor coolant pumps running is questionable.

Moraover, it is unclear at this time which models clearly, result in conservative, bounding calculations:

This is substantiated by the variety of different models used to represent the various primary system components in vendor analyses and the differences in the limiting small break

• c.

analyses.

It is our conclusion that this uncertainty in thermal-hydraulic modeling presently precludes the use of these models for

_f quantitative determination of small' break system behavior.with' the 2-

~. coolant pumps running.

In particular, we cannot accept their use to

'?.

substantiate allowable modes of pump operation during small braak Ch LOCAs. -

g ;-2 (3)

It.is our coh'clusion that for the pumps running c-ase, insufficient M.

integral. system experimental data presently exists to substantiate

=-

Ec the quantiative results of the analysis codes.

Moreover, we do not

~

believe any proposed testing can be performed on a schedule compatibi f

with tha't necessary for short-term resolution, which includes the 7

addition of equipment necessary to assure automatic tripping of the coolant pumps for small break LOCAs.

$9"-

(4)

From items '(2) a'nd (3), above, we find th'at tripping all of the reactor coolant pumps during small break LOCAs is required at this time, and that this pump' trip should be automatically initiated from equipment.that is safety grade.to the extent'possible.

, y

~

(5)

The impact of an early pump trip on non-LOCA transients is not predicted to lead to unacceptable consequences.

However, tripping the reactor coolant pumps for no'n-LOCA transients can aggravate the consequences of these transients and extend the time r'equired to bring the plant

~

into controlled shutdown condition.

For BAW plants, tripping of the reactor coolant puinps curing severe overcooling events increases the potential for interruption of natural circulation due to steam forma-tion in the coolant loops.

Therefore, we conclude that the criteria and requirements for rea'ctor coolant pump tTip to be esGb'lished"fRin'~ item (4) aojve slio u1 3 ~

~

~

rainimiTe', to i.fie extent practicab_le, the,pf{bab'il ty of,i~n,itiatj'g a_

reactor coolant pump trip for non-LOCA transients.

(6)

The staff recognizes the potential desirability of running the reactor 4;

coolant. pumps to provide forced circulation during small break LOCAs and we encourage the continued exploration by the industry of.means by which this could be accomplished?

For example, an increase in HPI

~ capacity or two pump operation as proposed by Combustion Engineering' j

s are a step in this direction.

1 (7)

We will, require verification of sma.11 break models with the pumps -

running against appropriate integral s'ysteins experimental. tests.. In particular, we will require.that the PVR vendors 2nd.fue1 supplier's j

O

A

. li N P!ic fcr E.Y '.'*.T a i cOr.* 2 red' 10.9 !.*

• b*.i t.n pre s s uf es of abDut

%,. **, MOD to i'.d.

. i r 1: - t E s -i np.:'.it ciar. s.

1. is excec.ed that the pressure '

t

.used f cr ;;; ;

ri; ty '.' r-i g.:.;c will' f all r pr:Usately in the rancs of

._the saf ety injection acteat. ion tressure 107 bcth CE and B&W plants.

f.2.4 Embcock and Wilcox-E asi:nad' Plar.ts EaDcock & Wilcox is also recom.anding that. for the short term, pu p trip be manually initiated on automatic actuation en low pressure of the safety injection system.

]n addition,' Babcock & Wilc x ano their plant owners '

are examir.ing the possibility of a short.erm manual trip requirement based on subcooling rather than automatic S1 actuation on low pressure only. 'The staff agrees in principle with this a.proach, but final approval

'must wait until the details of such a method have been formally submitted

- and evaluated.

The staff finds the present short-term recuirement for manual trip 'on

~

5 automatic SI actuation on low pressure acceptable.

ELW SI actuat. ion setpoints are between 1500 psig and 1650 psig and are considered consistent i

g... ;

with the setpoints at which the pumps would be tripped for both Vestinghouse g,.2 and Combustion Engineer.ing.pl.a,n..ts.

7 F.2.5 Trainino Cuidelines and Emeroency Procedures

~

IE Bulletins79-05C and 7S-06C (items 3 and 4) requested the' Westinghouse,.

~

, Combustion Engineering, and Babcock & Wilcox plant licensees to:

p (1) Develop new guidelines for LOCA and non-LOCA events based on LOCA.-

5.

analyses and RCP trip requirements, and

2. '

l (2)

Revise emergency procedures and train all license.d operators and

~

senior reactor operators bas,ed on these new guidelines. '

s

. In general, the licensees have identified guidelines, procedures, and training for loss of, coolant events in their responses to these items.

This effort on LOCA events was already in progress at the time the bulletin was issued.

~ Secause of i.he potential for initiating ECCS by other depressurization events such as overcooling because of a malfunction in the secondary system, the operator would have to trip the reactor coolant pumps before

[-

As a result, he could make a determination abo ~ut what event is occurring.

i f._

we require that the licensee have procedures and operator training to handle non-LOCA. events which may also have ECCS actuation and reactor j

i coolant pump trip.

The procedures for these non-LOCA events should include inst _ ructions. on.,>'

tripping the reactor coolant pumps, monitoring and initiating natural f.

}

. circulation, pressure control without the pressurizer spray, HPI termina,-

tion criteria, and reactor coolant pump restart crit.eria.

The licensees should confirm that these procedures for non-LOCA events are in place and

's f

the operators have been trained in their implementation.

~

e.

a.

e j

o

,a 6

j THE NEW YORK Tl.'.15S. MONDAY. FEBRUARY E.19C

~

. By MATTHy.W L WA! D ' '.

-set =stohave made thdrjob t=: rec ffi.

a.

a t

sr.x. % m %, p f"f'ased inAIE1NGTON, Feb.7 - RadoadMty wa*

lr.a's~tIrEstea:5 gtderaten. Ai Gb=a Yer eEa=ple* st.es the leak began

. n--.ar a_l.o $e a*.:mphert d=ing the recen-

"a.nd at 48 cSet p! cts of its type artrmd aM the res=:r sh= dew =, a.=$c sp.

.e.. at the Reben F Gi:na plan' the cr=try, radmacdve water is c:::u. te= sute.ad:aDy bests cJ.'.g valves because cperaten vt e "too' late,",in turning j latad around the bot ura.=fu= c=e, and asdshut*.ingdownpa sof theplantc:aw.

• EU emerii;escy F::ps.a:=rd:g to a prelimi.I then pc=;m3 through thmsamds of car. sidered ac:~"n1 in an.e=crgency.'

, nary evaluation bya staff me=ber cf the row t:bes, wtere 21 g'ves c!! Its heat. 0=ect these parts provided

e. 4

. dear Regulatery Cc--Mx,. *./ A.N. u

• r 0= side the tubes, cleas water is bcDg airto operata a valve that ca:be used to A setood re! ease oc=:rnd 35 r=! utes later '

bio staa2, which is med to n= a tur. nduce the pnssure of the rad:meuve' water, s useful step ti there is aleak..

Watse cperaten restaned the pumps for hine.

• • t=known" rtas=s.the study fe:.=d.

I na,. rad: active water, at te= pes-

~ Sh:tDownP:!rof Pu=ps w W degnes Mehest, is Also as a safety step, the cperators ei Chrcoolegy c! Act! dent n a pa r of pumps Bat, a==ng

'I _ I. '. 'hhe ev$=ati5 said it was5prt=ature" tt Pm lana that f

Fe if judge whetheractio=s bycfe!aten wert "cor.

%e t:bes b=st, aEcwing'the pressur.

• b i.nct, in=rrtet or could sted != prove =ent."

bd radioactive water to sqdrt into the udthatbyb E*3 ""..:

r i.However,its'cuther and other c%iefals of the clean steam,,at 700 to 100 Esll=s per T ~, c6-A!::i-->*4 d Ltday'that the opera.

=5:te. Re]etsed I..a its press =1:ed De opemen nnasted me airUneto the valve that relieves m a on the envtzt==mt, the water t=ned to steam ton had huded tht a=idest we!!. A spokes;

• r:an for the 7tochesti;r Gas and EleMc Com.

n the steam gueraterThe press =, rahcunwater, but dd nr. cpu &.

g

.es

• puy Judy Housto'.iald that the t5.11ty had gam too ata,==-ts the eve===.ew a.

,is talen n

y

s

• ray =.=i =a *=a= c=uie==

tab==t b= Den,..

sr u

Certain safety st

, it is clear b ret. thg com=c tary an the cLJogy.

con =nctmit.

-# ^ ~ ~7 l. De evaJustics also pdnts to proble=s in tht-respect, aggravate [the problem. Au:o.The vaave was then closed, b= the pns.

2

=aticany, the reactor's c=ntrol syste.n s=e was stilltoohigh.,,,.,

,,I safety,5yste=s cf the plant, a %.*esthghouse de,

One r=L-me later, the c sign in en===n use, which made the acddent se= sed the 1:ss of water in the radoac. c;cned the valve a seccind ti=perattes c, and it y hard er.to.c=tro!. Qse tsajoi plant'gc=pongr. :

pve Icop and t=ned on e=ergency stuck

.The analysis rt= arts that!!

-, used to rtgulate the pressure in tiie reactor was

• b ps to add water. In ::ai;y cases,this disabled by a ce=bbati:n of safety aetions, ac.1 a -*-=*y step to ins =t that the the' had been opened ce time, -

' csmiing to a cht:= legy of the acddest that it -

cere'r-='-= c= vend wi$ water.If the thee would have been'no ressan to open

$2 of thievaluati:n,.E'*:..KQ'P.

'l c=re is t=:cvered, as happssed at Three itasecond ti=e....r.; i,.,WS.p s.

• .] The evaJustice, whIrXis stamped *l draft,"j Rile kled, the resulthz heat will dam.

The halyds'saysithat th'e d. *in of.

. -M*.". i{ $ si. e evets in'the G1=na a:ddest raises the -

ate it. 4.'F* i;is dh onli=igtr.$. ~'

l was wrinen 1;y The=is Pi Speis, assistamt d' a

'?

• Pu pcssfb!11ty.cf other co b'.ati=s of fad.

re:tcr for ::a=:r sdety cf theT>ivisic s of Sys. !

Jtemi Integration.:The dvision is studying the :

$tsu te F5e Island case,the ac. =esthatcouldbemore 5,-rious. ' /

t was cc=pomded by the opera.. For example,'the chronology raises.

problem ci autc=ati: reactor systems that sn. I cide. dedsles to t=n o!! the emergency

.the question of why passee dr: ppd so.'

.teract m t= expected ways,ed cause problems ten

)

, in emergendests --,yr..r.st:c.frr.p g rt w r.

pumps. At G!==a, the pu=p"s were left,. sharply;,whm, the, safety veye en the a

1

The acddet en Jen. 25 at the plast l16 = Des

astoeger.Tur

5gthemou mayhave ster = geseptor aDowed the fint >.

i =::thest of Rechester, risulted in the t=c=n.

' 5eendelayedIc gerthesnemnyand lease to the: atmosphere, i= ply 5g that,' '

. tz Ded rtJease cf a s=all:==.=t of rsdoactivi.

resdted in a brief <"M = ge" to the at.' the valvt may have stayed opes s11gh0y 3

H

! ty, alth60gh such nJeases have bees avoided

= sphere, the asalpis said.The valve. 1=ger than it.was supposed.to. A=ceg

' during simild acddents al ether picti.rThe that aDewed the radioactive steam to es. the paints that the a alysis says " war.

?d.

f a:iount.o! radiatico released was act danger.

cape cpened the !1rst ti=e as a "d!nct rut further Isvestigatian". art what

cus, ac.mrding to'planf.and ctr-mRsiod effi.

result" of shuttkg 'down1he,pu=ps would happen if such a valve stuck in 4 ~... w,

"aber.:t :C m1=utes too iste,".acardi:g the opo pedti=n and produ::ed "a drect

, eggis, ; >y,q'- e s.

PaB IW.Fy "y coola:t to e=ter the at.

==:phere.,'According to the ama}p!s,

~

i AemMing to k'*.SM!s's mimo',"the'G!r i tothecaj sis. *:.

4 :.

I y

A seccod release came whes the

' a t:! dent rdses'the sp-cter of f ar mere serious PjF18y' gyed a

{ypn:::

b-ps to be

{

' aciidents,sc=e c! which are noi E.t.iF,2tMy ~

~

g p

curnnt emergency procedures. For exar ple, ac im art "us.h" is such. a situation, t=less 'the lesk 4

I rerssrks~aec:=panying the chrono!cgy of Rceald C. Hay =es, a ec-+sles c,.,

codd be stepped er a new supply of g l'

events i= ply that a cru:Ja! safety vaJve :may

• ** Y s:e e m ce day cf

. I*I Y=*t. sai 's.at even though re-water could be !==d, the emergeey tmeny have nuck cm r.nd if it had stayed' the a pu=ps would' c==tinue to add. water.

cpe a-.'"g to the esalysis, !t !s n: dcar 3.a--d g the p*gs r-s'dtd b a nJease em e w pg gg e how damage to the rescier s c:re would hav of radaoactJvity, it was a."censerva-bu-st tube er othedeak, tened hto beg yreve=ted.

tive step, b use the cpe: at.'.n s...-. d hd te dm',

In addices, the hes,tancy of the cperamrs to

=rn c5 certain saf ery pu=ps is this case rals es,

thought it would redu:= the pc=!b!Itty cf t=t!! ce emergency system had no i

coreda= age.

=en water left, a=d "a net h=s cf p:b the F::ib0ity that b t=:re cases, whert such -

..The a=dysis said that "cperr.mn ap-.

=zry c:clant would occur. Without cort a shut.c!!.vodd be racre !=ponant, the pu=ps parto be ver/ hes! tant" to turn c5 ce rective actics,. con ta=:Gery wuld 2:dd beleft en end press =e wodd rise so high p:=;c "when Ibey are 1Scwed :, er evetually cc=:r.".

The result of "cc:t t::>cuvery"!s cert 1

cat the reatter vessel !!self, which holds the eyes s:p;osed 10.".....

fuel.might be cracked ' ';.

In additics to these ac:.!c=s by the da:. age, the =:c stricus ki-d cf sed.

l

Mr. Speis's memorardu= steps shcri c!

cperaers, the dedsn cf the r:ac=r dest. ;

~,

- ' ~

i hferr"g to the actiers cithicperzten as er.

re s. Ecwever.his chr=olcry of the accident n::es dedsi:r.s r:ade by cperatr.:s th.at re::dted in 2e leak ru.ing ! ester, ed b a saferyvalve "bloe.:g" twi:t ed aDori g radicactive

. te2= to stat hto the at=esphere.

I

[

~2Rar::id_er,itr a.-_,at9:25 A.M.in cne c! Gin.

..