ML20235C154

From kanterella
Jump to navigation Jump to search
Second Supplemental Rept to ACRS on Brunswick Steam Electric Plant,Units 1 & 2
ML20235C154
Person / Time
Site: Brunswick, 05000000
Issue date: 10/08/1969
From:
US ATOMIC ENERGY COMMISSION (AEC)
To:
Shared Package
ML20235B311 List: ... further results
References
FOIA-87-111 NUDOCS 8709240369
Download: ML20235C154 (9)
v • d • e


Text

.____-_ _

l 1

-1 6

_n -n s n i nor n ii u j U :fi e ItL U00 UT.I  ;

D6cket Nos. 50-324 October [,1969 and 50-325 1

]

1 Second Supplemental Report to ACR$

BRUNSWICK STEAM ELECTRIC FIANT UNITS 1 & 2 U.S. Atomic Energy Commission Division of Reactor Licenstag p g llN O l' C% V "Lt4IJM= hd* E" ,

ofTICE > .

SURNAME >

DATE> .. . .

v.L aavsameset painvies omcs a toea e-se Foran AEC-)IS (Rev. 9-53) AECM 0240 8709240369 870921 I PDR FOIA '

PDR MENZ87-111 --f

hh M

i substantial' leakage in the core standby cooling system (CgCg) suction lines 1 of the Brunswick plant following the design basis loss-of-coolant seeident  ;

could lead to the loss of torus water required to provide effective core '

cooling. In our original report to the ACRg, our positium was that the design should be modified by providing isolation valves both inside and out- l side the containment or by other means such as enclosing the section piping i out to and including the first isolation valve. The ACRg ta its letter of May'15, 1969, stated that a short run of pipe, estremely conservative design, remote operability of the first isolation valve, inservice surveillance and leak detection to be a suitable exception to the general rule. We have con- J cluded that the applicant's design approach, as described in Amendment No. 8, 1 does not meet the conditions noted in the ACRg letter. We, therefore, conclude that a suitable design change such as a snard-pipe or jacket sho61d be installed on the core standby cooling system section lines of the Brunswick plant from the torus liner out to and including the first isolation j valve. i I

I J

t

> EMMMYLlSitx OFFICE > . -. . . . . . . . .

SURMA 00E>

... .m .

DAfr > .. ...

posus ABC.He (Rev.9 53) ABCM 0340 e.a. eovanneast essaries emcs essa e-see. set

[

s

~ . . . - . . . -

33.; .;_ -. . , _ _

-8FFlMMISMNtr i

1.0 INTRODUCTICII AND SIDMART This Second Supplemental Report to the ACRg discusses the background and bases for our conclusion a design modification such as a guard-pipe or jacket should be installed on the core standby cooling system (CSCs) suction lines for the Brunswick plant from the torus liner out to and including the first isolation valve. This report supplements our previous reportf of September 3,1969 on this subject. This requirement would be consistent with our general approach that, insofar as practicable, multiple protection should be provided for both active and passive failures in systems or com-ponents that must function to protect the health and safety of the public.  ;

Cross leakage in the C8CS suction line and/or in its isolation valve following i

a loss-of-coolant accident could potentially lead to a loss of recirculation l d

flow and result in offsite doses that exceed 10 CFR:.100 limits. Properly l l

installed guard-pipes or jackets around the CSCS suction lims from the torus I l

liner out to and including the first isolation valve would provide protection against flailure of the suction lines themselves.

In its May 15, 1969 letter on the Brunswick Plant, the ACRS stated:

" Engineered safety systems that are required to recirculate water after a loss-of-coolant accident should be designed so that a gross system leak will not result in critical loss of recircular*on I or in loss of isolation capability. The Committee believes that exception to this general rule may be made in respect to a very short run of pipe from the torus to the first valve, if extremely con-servative design of the pipe (and its connection to the torus) is used and suitably remote operability of the valve is provided.

The design of these systems also should provide adequate leak detection and surveillance capability."

omcE > . . . . . . . . . . .

f . _ ,

q ...p p.g .

7 --

$URNAME > -

V Mft . 6 9. .

t nTI yq b .

j 1

DAtt > . . . .. . . . . . . . . . . . . .. I Forma AEC-5 A e (Rev 9-$3) ABCM 0240 m a somessat remnme ernes nas e-o au j

=__~x_=__=--- - - -- - - - - - - -.

--.A

I 1

-0RB~AL !SLONE I 1

1 2

We agree with the ACRg that exceptions may be made in the design of the ,

C8CS suction lines provided adequate compensating design features are provided. Se ACRs stated in its letter what design features would, in its view, make the applicant's design concept acceptable. %e applicant's design approach does not provide these features. As a result, we conclude that the use of guard-pipes or other appropriate measures should be required to serve as a second barrier against failure of these lines.

2.0 BASES FOR FROTECTICII AGAINST FASSIVE COHp0 EFT FAILURES In our view, protection against single failures in either active or passive components in the emergency core cooling systems (ECCS) should be required for the long term recirculation cooling phase. h e ECCS must be designed to perform its function throughout the post-seeident {LOCA) recovery period, which may be af long as 6 as 12 months. Significant uncer- -;

tainties exist in the knowledge of potential accident and post-accident conditions and capabilities of the ECCS. %e ECCS should, therefore, be designed to accommodate without serious consequemees a single failure any-where in the system.

Se consequences of e gross failure in any ame of the ECC8 subsystems could be serious if the failure were not promptly isolated. %ese consequences  ;

include possible loss of torus water for those pomys in the remaining ECCS subeystems that ary othervi6 afunctional, and possible offsite doses that i l

exceed 10CFR100kimits. %ese two topics are discussed further in the

,gubsequent parageAphs.

}fM lQ If sunmir

. L m .s u m{_ !V a_ . ..je. a...g~

DAtt > - . . . . . . .

V .- l j

r- Asc-m ta .e.333 4 cu em . u.........-...

- . ~ . . . ..

4

w... - , i n i' ( i .LV -  !

Unisuink vek t0LI I

3 la the Brunswick design, as depicted in the attached sketches, the ICCS pumps are located in. five separate compartments, each with its own susp.

3 The two Israest compartments, each with a ecstained volume of about 81,000 ft ,

house the two RER pumps. The sump pump has a 50 spa capacity, but its drive l motor and actuating circuitry is fully dependent on the availability of 'offsite 4 power. A major leak in one of the tsCS section lines (or valves) that could j mot be isolated, would cause torus water' to flow int.o the compartment of the affected subsystem. Since the sump pump capacity is about 50 spa, with ex-ceseive leskage the affeeted eespertmest eould be filled and would paseibly overflow via the compartment stairwell, depending on the torus internal pressure )

The initial (or pre-LOCA) water inventory of the torus is 87,600 f t3.

A torus internal pressure greater than 26 pais could cause the compartment to overflow via the stairwell. Continued leakage could cause the overflow water to spill into the second RER compartment via its stairvell. If this were to occur, it could lead to flooding of the redundant RER sahepstenst

)

In addition to the direct effect of flooding, the HPSE requirements of the RER pumps and the core spray pumps which are shown to be 32 and 33 feet,-

respectively, in Table VI-2-4 and VI-2-5 of the PSAR might not be maintained in the event of excessive leakage. The nominal static head from the torus water level is shown to be about 18 feet in Figure V-2-7. To maintain adequate HPSE to avoid cav&tation inPEER and core spray pumps, the torus water tegerature 1

l nrrt"61 LI IIM A0 V etiniinntuv6ur.i-ppgfy q omcr > . . . .

SURNAK > . -

DATE > . . . . . . . . . . . . . . . . . . . . ...

Foems ABC-He (7.s .9-53) ABCM 0240 e.s os.sm.esat menne emcs i tese e-as. eiv ,

. . , ~ ..

.{

, . . ~ , , , . . -,

1 I.W..EIPill llE ()gn.

1 J >

4 aust be less than about 185'F, or the primary containment must be pressurized at about 7 psi above the saturation pressure of the torus water (the precise value would depend on, the actual terms water level). Since torus. water at a temperature less than 185'F eennot be assured throughout a post-seeident ,

i recovery period,' effective ECCS operation depende en maintaining a positive I

pressureoftheprimarycontainmentunf1pooltemperatureisreduced.

We have also considered potential offsite doses assuming gross leakage in one' 6f the CSCS suction lines, following a LOCA. A number of variables, such as the time af ter the 14CA for. the leakage to occur, the fraction' of torus  !

~

water 1eaking into the affected C8CS pump coupertment, the torus internal 4 lu.hc9<. _'

pressure at the time of the lokage, and the decontamination factor associated l 8

with the performance of standby gas treatment system and plant stack, sffect significantly the potential offeite doses. However, since leakage from the CSCS lines in effect would be a breach of containment, the resulting offsite i doses. could be well above the 10 CFR 100 guideline doses.

3.0 IVALUAfl g__o_t g BRE L33mtgMicLJtE81 5  !

The ACRS letter dated May 15, 1969, on the grenswick plant cited a number of conditions which, if satisfied, would obviate the need for protection against a leak in the auction Tide of the engineered safety systems. These conditions included: 1) a very short run of pipe from the torus to the first .

l valveg 2) extremely conservative design of the pipe (and its connection to the torus); 3) suitably remote operability of the valves and 4) adequate leak-detection and surveillance capability.

{

==> DEri 9 I .l,M,E,, nw u ggg_

vi s ,' , , . ,

J

-r , . . . . . . .

l

. -c-m . .. 4

  1. wsasvina. w w a ervi g,1i,g f L 11L - r')

5 In the applicant's proposed design, the pipe length from terus liner to the first valve is 18 feet, of which 12 feet is through a guide pipe in the concrete wall of the torus and, therefore, inaccessible for inservice inspection. As presently designed, the guide pipe does not prevent or contain leakage from the CSCS piping.

The applicant also proposes an upgrading of the existing USAs 531.1.0 Code of Power Piping and Valves (1967) including additional analysis, and quality control of materials and fabrication in addit &on to that required by the code. We agree that. the proposed additional requirements represent a conservative design. l The first valve on the line between the terms liner and the CSCS pump has been made remotely operable by the ese of a remotely controlled motor driven valve.. We find this meets the requirements as stated in the ACRS letter.

The proposed leak detection system includes the use of levnt switches and sump pumps. New leaks will be detected when the sump pump is observed to operate more frequently and/or for longer duration than usual. The sump pumps and level switches rely on availability of offsite electrical power.

To determine the leakage source, the CSCS isolation valves will be alternately  ;

41osed. We find the proposed leak detection system inadequate in terms of ability to detect small leaks and in terms of reliability because the system will be unable to identify the location of leaks that occur in the 18-feet run of pf pe upstr9am of the CSCS suction line isolation valve and the system would be ineffective in the absence of offsite power.

OFFICE > - - - - -

g p-p... p 7.g ,

,w,,

us a iv n , ,p m ,,

M> - - - - - . . . . . . . ._ II( _ _

, acn.u... ocumo . , . . . . . , _ , , , _ , , . . , ,

l4

- a,cc,h

, , , ,i,ia . ._M I N..Y.

6 We are not able to condlude that the conditions specified in the ACRS .

letter have been satisfied, because we are not sure that 18 feet should be considered a very short run of pipe, because of the difficulties in inservice inspection, and because of the limitations in the capability for leak detectirra.'

Therefore, we conclude that additional protection should be provided to protect against the consequences of leakage in the CSC8 suction lisas, i -- 4.0 GUARD-FIFE DESIGN _ APPROACB One design approach to permit isolation of pipe b-eaks in lines which penetrate the contalement is to use two valves in series; one inside and one _

outside of containment. In some instances, this design approach is, not consideri practical. The approach chosen by most FWR applicants is to place' a pressure-tight jacket around the ECCS suction lines from the centstanent suay out to and including the first isolation valve. Representative plants using this godsd-pipe approach include Pacific Gas and Electric's Diablo Canyon plant, Sacramento manicipal Utilities District's Rancho seco plant, and the Duke Power ,

Company's ocones plant.

An appropriate guard-pipe design e'en allow access for inservice inspection ,

equal to that which would otherwise be available. Moreover, leak detection can be made anch more effective, with greater sensitivity and reliability, than that of the proposed Branswick design, which uses sump pumps and level switches.

Some of the disadvantages of the guard-pipe approach include possible degradation of the line and valve because of additional welding requirements' and possible differential expansion between the line and the guard-pipe.

Al Igor nu u

- D.E. "lm'P

. nt uat UII .I i / K, . -

omer >

I . ..... . . . --

suaune > ... .. .... .. , , , , , . ,

oAre > . . . . . . . . . . . . . . . . .. . . . . . . , . . . , . . _ ,

Fone ABC.H8 (Re.9 53) ABCM 0240 gs ...,, ,,,mvine errics i sees e see.en g p/ue.p r*.pge-.- -r*4 34.w-a- -e.-- a => 8-w =ne"'-r * -- ----l----------I

, I I

l DCC o r tin u l 71: Pu utn A'. L iD . L ILI -

3 .7 Proper design attention is required to minimize this kind of degradation and the design of the line and valve should have sufficient margina to

\ accommodate the residual amount of degradation.

We conclude that the additional protection that would be obtained by use of the guard-pipe, as a recond barrier against possible loss of effective emergency core cooling water and as a second barrier to the release of fission product activity directly from the primary system to the reactor building, outweigh significantly the disadvantages associated with l

potential system degradation. l l

)

l

~"

Fr

. l'

,u A' I0.f1M V

.n . s v . vu o i Il#4 L

_QBL/Ef#,9 , DRL. ,, DRL o m er > ~'

DFktg :emm FSc er EXR5 Eels sURNAW > ,,,K((9[ {,

10/g/69

,, ,,,,,,,,,,10/[/69 10/8/69 narr ,

r-- ac-n. ta.. . u> ucu ou.

E ' '" ~lli

_ 7 7_7 - -

,

Retrieved from "https://kanterella.com/w/index.php?title=ML20235C154&oldid=3226780"