ML19308B715

Responds to NRC Requesting Responses to ACRS Questions on OL Application for Pebble Springs Facility

ML19308B715
Person / Time
Site:	05000514, 05000515, Crane
Issue date:	11/30/1977
From:	Williams J PORTLAND GENERAL ELECTRIC CO.
To:	Varga S Office of Nuclear Reactor Regulation
References
TASK-TF, TASK-TMR NUDOCS 8001160783
Download: ML19308B715 (62)

Text

,.

e l'ontT sx:: G:N :::.si. Es. :r ritic C o31 :*.sxy c.: -

PCs TLANo. Qh tson G7205

.cets.. a:. vm:.

I i

9 Peame S9n.4km i-li Q4CL.E vr0 i OOT-

^

c._-w.

.. e m m - n 1-N s-ove _mber 30, 197.7.<1 y@

EPshble' Sprin^gkNuclear Plant E Docke ts "50-5'14 50-5

/9 &

s Director of Nuclear Reactor 'tegulation

[

c.

ATTN:

Mr. S. A. Varga, Chief

't

-9

~

Light Water Reactors Branch 4 h

1 Division of Froject Management U. S. Nuclear Regulatory Cocnission p

Washington, D. C.

20555 3s

Dear Mr. Varga:

N p ~m

v.,

.e. -e w,wy_.,_ ___.. _,.

-., y m~ t f, Asireques ted, injyourile t te r - dated Novemb,er, 21',:1977i we are enclosing our responses to 26 questions 'riised'by"the ACRS on the Pebble Springs application.

Sincerely, s

Joseph L. Williams Executive Vice President Enclosure (43 copies) c:

R. Muller (ACRS) 5/ enclosures M. S. Plesset (ACRS) 1/ enclosure S. H. Push (ACRS) 1/ enclosure Dr. Fred D. Miller, Director 1/ enclosure Oregon Department of Energy m 174M 8001160 783

_.. w-.

u E

O PORTLAND GENERAL ELECTRIC COMPANY __

PE3BLE SPRINGS NUCLEAR PLANT Resoonses to ACRS Questions QUESTION 1 Provide the interpretation used in design, of GDC 19 and Reg. Guide 1.75 (IEIf. 334).

The less conservative interpretation of GDC 19 does not; allow common damage in control room.

shutdown capability RG 1.75 permits convergence of total plant down to spacing measured in inches (with some form of panel or plate type of barrier) to a few feet of open space.

More conservative interpretation of CDC 19 would require (as IAEA does) that safe shutdown can be accomplished if the the control room (and presu= ably any other given safety " space") is subject to common damage within that space.

Use of the less conservative interpretation of these criteria results as a "sof t" design with extremely heavy requirements on "adninistrative control". If the design is " soft" describe the correspondingly "hard" administrative controls.

Response to Cuestion 1 The7 ebble Springs design conforms to General Design Criterion 19 P

of Appendix A to 10 CFR Part 50 in the following manner:

The control room consists of two distinct areas - the i

control operating area and the supporting cabinet and I

The central operating area provides the main f

rack area.

operator plant interface and includes the main control

' benchboards and panels.

f t

e e

Q 1

The " convergence" of safety trains permitted by Regulatory Guide 1.75 is made use of in the main oper-ating control boards to facilitate proper operator j

surveillance and control of tha plant. All manual operations are initiated at the main control benchboards.

All automatic initiatione and logic operation are initiated in physically and electrically independent cabinets in the supporting area of the control room.

The minimum sep~aration between redundant safety-related k

cabinets is 4 ft.

In accordance with General Design Criterion 19, an auxiliary shutdown panel is provided outside the control room to enable prompt hot shutdown of the reactor.

Controls are also available at the various necessary switchgear to enable subsequent cold shutdown through the use of suitable procedures.

The response to Question 13 reviews the potential consequences of a postulated fire in the control room.

~

In the main control board area where convergence to Regulatory Guide 1.75 separation exists, common fire danage to redundant safety-related manual push-button modules would not degrade the capability to safely shut the plant down from the auxiliary shutdown panel.

In the supporting cabinet and rack area behind the main control boards, cazmon damage to any two adjacent rows of safety-related cabinets would not degrade safe plant shutdown capability from the auxiliary shutdown panel.

y Such widespread damage in this area is considered unrealistic since all cabinets are provided with internal fire detectors; the control room is continually

=anned and a fire would be expected to be suppressed in its incipient stages.

i I-k.

t o

In summary, the Pebble Springs design represents a conservative interpretation of General Design Criterion 19.

Even though fire protection design features and proca-

.dures would greatly limit the potential for common 1

damage in the control room, a very vide range of hypothe-tical common damage even'es may be accommodated without degradation of safe plant shutdown functions.

OUESTION_2 Clarify the rationale used for location of straight sections of main steam and feedvater lines in respect to potential damage to safety equipment. Is it assumed that such pipe sections are infallible?

Resoonse to Question 2 Straight sections or main steam and feedvater lines are not

. considered to le infallible. The criteria used to determine the locatons for pipe breaks is in accordance with the guidelines of Branch Technical Position ME3 3-1, in which the pipe break loca-tions will be at high-stress areas. The high-stress areas are usually at the pipe fittings. An analysis for potential damage to safety-related equipment due to pipe whip, jet impingement, flooding, compartment pressurization and environmental conditions is performed for each break location.

?-

The main steam and feedvater lines in the Containment penetration are's are straight sections and the piping stresses in these lines are kept below the "no break" limits in Branch Technical Position

• MEB 3-1.

However, a nonsechanistic pipe break is assumed in these sections. The compartments around these lines are designed for overpressurization, flooding and environmental effects due to the noemechanistic break and safety-related equipment within these compart=ents are designed for the 32vironmental effects.

-3 m --,,__

F Q

~0UESTION 3 Does the design accommodate potential for inadvertent flooding

~

from vessel and piping failures within " safety" structures or in such areas where safe-shutdown equipment is located?

Resoonse to Quescion 3 The Pebble Springs design accounts for inadvertent flooding from vessel and piping failures within safety structures or in such areas where safe shutdown equipment is located. The design bases used are as follows:

The type of failure, location, and orientation of complete pipe ruptures or through-wall leakage c' racks are determined in accordance with Branch Technical Po9ition MEB 3-1.

The flooding resulting from a single passive failure in one safety-related train of equipment shall not affect the other train of safety-related equipment. Also, the flooding resulting from the failure of non-Seismic Category I equipment shall not affect safety-related equipment. Flood protection shall be provided by either isolating the piping and vessels, by adequate physical separation from safety systems and components or by enclosing the piping and vessels with structures suitably designed to protect adjoining safety systems or structures. In certain cases credf.t will be taken for manual action in mitigat-ing the effects of f1'oeding, where several hours are available before unacceptable damage could occur, and where the required operator actions may be accomplished in a straightforward manner.

4

F o

Walls, vicerti ht barriers, panels, penetration sesis, k

and other compartment closures designed to protect safety-related equipment from damage due to flooding shall be designed to with' stand the effects of the SSE, including seismically induced wave action inside the affected compartment.

OUESTION 4

~

x What is stress-level 'and maximum local deformation in steam-generator tubes and tube sheet as result of Post-LOCA flooding of tube-side of superheat section of steam generators? Would some tube failures at this point in time seriously affect core cooling?

Resoonse to Question 4 To investigate thermal concerns in the steam generator, B&W has postulated a hot leg loss-of-coolant accident (LOCA) under condi-tions that would result in worst-case tube stresses. Refill of the steam generators was calculated to begin at about 2 minutes after event initiation. The steam generator refill rate was maximized by assuming that all emergency core cooling system (ECCS) injection operated and, at the time the generators began to fill, no credit was taken for boiloff through the break.

Credit was also not taken for heatup of the fluid as it passed through the piping and 40*7 water was assumed to quench the tubes.

The'results showed that even during this unrealistic hypothetical transient the tubes retained their integrity..

The maximum calcu' laced tube strain for this transient was 0.3" which is approximately equal to the serain at the 2% offset yield point for Inconel. With a pressure differential across the tubes of about.800 psi ac the time of the largest temperature difference the burst strength margin is still greater than 3.

=

-S-

~

lh&6 leu '2lN K b f M /X f

A.prelininary assassnent hos Indicated that th2 double-ended

?

O

~

[-

rupturo cf up to,3 tubes during o LOCA 9::uld nat sariously impair the capability to reflood and cool the core in accordance with

~

Mk1 the ~ conservative requirements of Appendix K to 10 CFR Part 50.

W*

Ah.,

_ QUESTION 5,

j Wat is the maximum secondary system pressure developed after turbine trip with first subsequent random failure being loss of main feedvater flow control leading to flooding of super-heat section of steam generators.

Assume turbine trip without bypass (loss of condenser vacuum).

Response to Question 5 The maximum secondary side pressure developed, assuming turbine l

trip without bypass and a subsequent loss of main feedvater flov control, is equal to the setpoint of the main steam safety valves.

There are tvo banks of safety valves. The "high" bank se tpoint is

! p' about 1315 psia which include: 3% accumulation.

'Ihe maximum f

allowable steam generator pressure is 1375 psia.

f QUESTION 6-

~~

i Does applicant know thatitime-depende[t levels ~will occur in; t.

[hressurlhek steam generator and reactor vessel after a rela-

..... ~_...

tively(smal1~ primary coolant break' which caus'es coolant to approach or eved partly uncover fuel pins?.Wat does operator)

..~..

r.

[do' f in respect to lIni:erpreting level in pressurizerf}

During primary system refill from high pressure injection pumps 1

.. - - ~..

]

there is {somel period when neither condensation nor naturalj

.,. ~.,.

y

.j j

!. convection;is:present; to effect heat transport to secondary i

v

.- s..

l side. l Row is transition to_ natural' _ convection,v..ithout Assistance!

p.

c, s.

s

}

ifrom primarv coolant pumos obtained.'

j

{

6 g>.e.

,a

[,,, 1 e-*=

= = * * * -

^

w R

d'

l y.

REspins2 to Quastica 6 w

g '. ~ ~

nere are two overriding concerns with any LOCA:

.. ~

T (1) Initial temoval of ; fuel-stored heatQ c

s b.E',:

-r y.

a c,c

/f4 (2) Continuous removal of core fission product decay heat.

a.;..

.e For small breaks, fuel-stored heat is removed during the first few seconds of blowdown.

We B&W ECCS system, using internal vent J

valves, precludes the interruption of decay heat removh1 for all accidents within* the frangel.of re5atively. small breaks (break size j

!<0.01, f tf); Break location, ECCS injection, coolant phase separation, Reactor Coolant System (RCS) mixture icvels and steam generator conden-sation have been considered in arriving at this conclusion.

As we understand the question, the concern is related to possible linterruption.of steam condensation within~ a steam generator due.

m

to ' re fillingl of" the ~ primary sys tem.-

In general, such a'. situation can dI occur only at extended times during the final recovery stage of a LOCA when steam condensation is no longer required. However, even if this situation occurred earlier in time, the performance of the vent valves vould be to equalize water levels between the hot,and cold regions of the primary sys tem, thereby assuring continuous fluid coverage of the core with no adverse consequences.

This is substantiated by a more detailed examination of the fluid conditions during a relatively small LOCA.

Such an accident can be viewed as a very si;ov transient during which, at any particular time, the system is, not ceaningfully different from steady-state conditions. 'Ibe RC'S can then be properly described as a sealed mano-For the B&W system, because of"the vent valves, this manometer

~

meter.

is double looped as illustrated in Figure 6-1 with important volumes identified by letters.

I b,

.___...,_.___q n

1

C *.

Macy experiments hcv2 be2n rua which shevi thht cs long cs a finid (quality less than, say 70%) covers the core, no adverse ' core.

~

(tempeEatura excursion can' occur at decay l heat power levels.

nus, 4,y M.,,..~e the design problem associated with small LOCAr is to achieve

,Mfy' steady mass and energy "aalances which assure ~ that the core remains

-,G covered. nis means that mass injection equal to mass loss, and energy removal equal to decay heat is achieved.

For a spectrum of break sizes appropriate for relatively small LOCAs, conservative I

analysis assures that no uncovering of the core occurs prior to y

achieving excess mass injection. nus, any concerns with very

$f small-break LOCAs deal with the energy balance once excess injection has been achieved.

/

lFor certain small break's, the. steam generator would act as an

energy re:noving devi,:e.

Energy removal occurs through a three-step sequence:

initially, a solid flow-forced convection process vould control heat removal, later a two phase natural circulaton procesa involving both convection and condensation heat transfer would control, and finally a pure condensation mode vould result.

In this latter mode, fluid has fallen to approximately*1evel B on Figure 6-1.

As steam is produced in the core through boiling, it travels through D F, and C and is condensed in the lover regions of H.

Concerns over the impact of-noncondensible gases.have been examined for this phase and the following points apply:

(1) Insufficient noncondensibles are available in the initial RCS fluid to block the flow of steam at G (this is a 3-f t diame ter pipe).

(2) Heat transfer coefficients with,noncondensibles l

present are sufficiently large to condense steam in the lover regions of H.

Even ff the heat transfer vere comentarily inadequate, this vould merely cause a pressure increase and resultant temperature increase until the temperature difference compensated for the lower heat trans fer coefficient.

~

Ie

-g.

l u.-

w*

.e.

-

• w ww -

e-

- * - * * * - - " < - - - * * - ~ ~ = * * * * *

^^* ' -

"" '"*~"""* '

0 9

j j

la

's j

i (3) The open manometer paths D, T, C, H, and 3 assure

_c a+~-

that hydrostatic balances exist between regions H l

~

bh and A, and between regions K and A.

If these r~+

"$? '*'

balances do not exist, fluid movement will occur to

^;i% ~

41?

produce them.

,a mz, After excess mass injection is achieved, the RCS starts to refill.

l During refill, a ri' sing vater level in region H may eliminate condensing heat transfer. Note that"a rise of level in H also

~

means a rising level in K and A. ' Thus, no immediate core concern exists. Steam pochets vill be formed at J and C.

If the level continues to rise, 'a two phase mixture vill be forced into D and F.

This vill occur through the necessity of maintaining a hydro-static balance with H.

However, if condensation ceases, the energy balance is no longer maintained. As energy is not being adequately removed from the system, the system must repressurize.

Two mechanisms are now possible:

(1) The break flow increases until it removes enough energy, or the break allows removal of enough mass to reestablish condensation, or (2) R$ pressurization continue's; until energy removal is brought about through the pressurizer relief valve path E.

Most likely, mechanism (1) vill repeat for several cycles prior to 4

mechanism (2) occu'rring. In any case,luncovering of. the: core can-4 (not take place [ Ap' ain, if 'the core fluid. level is lowered, then the fluid level in H must be lov and conde,nsation is a credible phenomen,en.

~

The flow pattern in D, the horizontal section of the' hot leg, -Es of interest during repressurization. This is illustra'ted in Figure 6-2

[

along with the pressures within the system. The following hierarchy

~

i l

of pressures exists:

Po<P3<P1<P2 h

_9

sainly because of static balances and the two phase nature of the fluid in the core and hot legs.

QUESTION 7 What is the particular design of the start-up piping and pumping system for Pebble Springs? Does it involve operating with a liquid-solid second&ry system? Has the Staf f performed a safety analysis of this system?

Resoonse to Ouestion 7 The current Pebble Springs design does not involve operating with liquid-solid secondary system during heatup. A vacuum hestup scheme recently selected as an NSS option is now used for startup.

The system is illustrated in the attached sketch (Figure 7-1).

During heacup and operation of the integral economizer once through steam generator (IEOTSC), two possible overstress conditions require particular attention. One is excessive compressive loading of the steam generator tubes caused by too large a temperature difference between the steam generator tubes and the steam generator shell.

The second is possibly overstressing the shell-to-lover-tube-sheet veld by adding feedwater that is too cold. In order to avoid these conditions, two functions must be available. These functions are (1) the ability to draw a vacuum in the secondary side of the steam generator to permit boiling at low temperatures and consequent shell condensation, and (2) the ability to heat and maintain the temperature of the feedwater before it enters the steem generator so that the maximum allowable temperature difference is not exceeded.

Drawing a vacuum,and the subsequent lov temperature boiling will allow heatup of the plant without exceeding the maximum allovable tube-to-

hell differential temperature. Maintenance of the required te=perature in the main feed line is accomplished by a circulation loop for each y

steam generator. This loop will draw fluid from the four steam generator.

r e

drains (at the outlet primary temperature) and return it to the feedvater line. The loop's function is to circulate flow during heacup and maintain the temperature of the feedvater line when feed-Without this circu-water, is not being added to the steam generator.

lation line the cold fluid in the length of pipe between the main feedvater isolation valves and the steam generator would permit addition of cold feedvater to the generator.

The vacuum heatup procedure is described below:

The steam generators are filled to a presclected level at cold conditions. Prior to ecmcencing Reactor Coolant System heatup, a vacuum is drawn in the condenser by operating the vachum pump. In the steam generators a vacuum of at least 25 inches Hg is established by opening the main steam isolation valves and the condenser dump valves. This permits boiling in the steam generator at 134*F, thus allowing heat transfer to the steam generator shell as the steam condenses which keeps the shell temperature close to the tube temperature (which is at Tavg on the primary side).

Af ter the starting of the reactor coolant pumps the circulation of primary fluid will cause heat transfer to the secondary fluid in the steam generator. Boiling and subsequent heating of the steam generator will start as soon as the temperature reaches 134*F.

The heating of the steam generator shell down to the steam outlet vill follow due to condensation of steam. The pressure vill stay about 5 inches Hg absolute until the shell temperature reaches 134*F.

During this process, steam, 3

along with noncondensible gases, will be drawn out of the steam generator. As the shell and the piping temperatures increase, noncondensibles would have been drawn out al=ost completely. At this stage, the condenser i

du=p valves vill be closed and the secondary pressure vill start increasing gradually to keep in step with the l t

I 1

r l

saturation temperature. With the secondary system isolated from the condenser, it.will heat up along with the primary system with a few degrees temperature lag. The condensa-tion built up in the annulus, the steam lines, and the steam line drains will be removed by opening drains to the condenser.

During heatup, the steam bled from the steam generator will result in a level decrease in the steam generator. Feedvater from the condensate system vill be added to maintain steam generator level.

When the primary temperature reaches about 250*F, the feed-water has to be heated before addition to the steam generator.

Recirculation is started that draws hot water from the four steam generator drains and returns it to the, feed-water line near the feedvater isolation valve to maintain the temperature of tne feedvater line between the steam generator and feedvacer isolation valve. A second recirculation loop (clean up loop) is established using a condensate pump, startup feed pump and auxiliary steam to maintain the temperature and quality of feedvater within specified ILaits for feeding the steam generator.

Heatup of the primary and secondary systems will continue until the primary system reaches 550*F.

The recirculation loop with other inter-connections can be used for wet lay-up during shutdown and also for blevdown during heatup and ou power operation.

The scheme outlined above permits heating of the secondary system within accepta'ble temperature limits (ie, without exceeding allowable stress ILaits). The pressure on the secondary side is always the saturation pressure corresponding to the temperature.

The system is not cempletely liquid filled and thus is not subjected to overpressure during heatup.

OUESTION 8 Can the plant obtain access to the low pressure RER system from the high pressure condition using oniv safety grade equipment?

Resoonse to Question S

~

B&W does not design the NSS such that the plant can obtain access to the low pressure DHRS from the high pressure condition using only safety grade equipment.

The capability of achieving and maintaining a safe shutdown condition per General Design Criterion 19 of 10 CyR 50, Appendix A, is incor-porated into the B&W NSS design. B&W defines the safe shutdown condition as removing decay heat with the steam generators while keeping the reactor 1% Akfs suberitical and a reactor coolant tempera-ture near the saturation temperature corresponding to the pressure setpoint of the main steam code safety valves. The B&W safe shutdown condition (or rather, safe hot shutdown condition) is considered a safe condition because (1) the reactor is kept suberitical, and (2) core cooling is maintained. In addition, the B&W safe shutdown condition is a stable, desirable condition for the following reasons:

(1) All condition II, III, and IV transients (except LOCA) terninate in the safe hot shutdown condition.

(2) No condition II, III, or IV transients require forced cooldown of the RCS from hot shutdown to a cold shutdown condition.

1 (3) It requires the least controls, equipment and operator actions.

(4) It is a very stable condition in which many RCS parzneters are inherently controlled.

(5) No significant changes are required to RCS conditions such as a

boron concentration, pressure, temperature and water inventory..

9 8

'yer :ner= ore, B&W recommends that plant cooldown shculd be conducted us:an; fz=iliar and routine procedures and equipment, which provides -

ope stional flexibility and reduces the possibility of operator er _ _. Therefore, following an event that takes the plant into a saga hot shutdown condition, it shou'id remain in this condition unt-1. it can' be cooled down, if desired, using procedures and equip-me:: as close to routine as possible. Since the NSS is designed to mal =tain a safe hot shutdown condition, there is no need to accelerate a p'_ ant cooldown. All systems and equipment required to achieve and sai=tain a safe hot shutdown condition are safety grade and can perform their functions with loss of offsite power.

Beczuse B&W designs for a safe hot shutdown rather than a safe cold shu: sown, all the equipment and controls required for going to a safe

~

cold shutdown are not safety grade, and the equipment and controls i

tha: are safety grade as such because of some other safety function.

Tab

• e 8-1 identifies the functions that are essential for cold shut-dow= and which of these can be accomplished using safety grade equipment and controls. The essential, functions to be performed are discussed below; the listing corresponds to Table 8-1.

i System Function f

i RC 1.

Neither method is safety grade; however, two diverse methods are available: Should power be lost, the spray valves can be manually operated and the electric-operated relief can be remote-operated by battery.

RC 2.

Natural circulation is always available.

7 RC 3.

Both methods are safety grade.

SP 1.

One method is safety grade (the Seismic Category I source of feedwater is from t

the backup service water spray pond).

k l l

.w

- e 9

Svsces Function MU&P 1.

One method is safety grade.

)

DER 1.

Although this function can be performed using only safety grade equipment, it is desirable to use some nonsafety grade equipment and controls, such as the DER

, cooler bypass valve and the venturi bypass valves, y

DER 2.

Neither method is totally safety grade since manual valves need to be operated. These include the spray valves and the auxiliary spray line reactor building isolation valve. The equipment and controls that are safety grade are such because of some other safety functions. Consequently, they may not be as effective as desired for cold shutdown.

CA&BR 1.

One method is safety grade.

CC'J 1.

Component cooling water to the DER pumps and coolers is safe ty grade.

w e

i l

i l

OUESTION 9 Defend the rationale of having only two " active" service systems The which perform continuing or long-term safety functions.

first " accident" is the fa*. lure of one train thus destroying

" normal" redundancy. Dependence on a single system in terms of consequence of f,ailure of that remaining system is ei,sential to understanding intrinsic risks of such designs.

Describe each such system and consequence of total f ailure of services provided by that system as a function of time. Only

" active" f ailures beyond first failure need be considered.

Possible examples of such systems are:

Battery (DC power system) (consider parasitic loads) 1.

On-site AC power system - assuming prio,r loss of of f-site,

2.

AC system 3.

Service water system 4.

Ccaponent cooling system 5.

Environmental control (HVAC) systems

" Redundancy" may be expressed in terms of time to restore service by any means whatever before undue damage ensues.

1 Resconse to Question 9 The design of providing two " active" service systems that perform continuing or long-term safety functions is in compliance with the General Design criteria in Appendix A to 10 CFR Part 50.

The design philosophy recognizes the fact that equipment is subject to failure and, as a minimum, the plant =ust be able to survive an incident (or single failure) without endangering the health and safety of the public.

i \\

~

i

)

The definition of single failure used for Pebble Springs is as 1

follows:

A single failure means an occurrence that results in the loss of capability of a component to perform its intended safety functions when called upon. Multiple i

failures resulting from a single occurrence are con-sidered to be a single failure. Fluid and electrical systems are considered to be designed against an assumed single failure if neither (a) a single failure of any active component (assuming passive components function properly), nor (b) a single failure of a passive component (assuming active components function properly), results in a loss of the safety function to the nuclear stema electric generating unit.

These criteria are applied to the service systems of safety systems, with the exception that the single failure is limited to an active failure in the short term following an initiating event.

During the long term, a single failure can be either an active or passive failure.

a.

In many Pebble Springs systems, the degree of redundancy provided for active components extends beyond that strictly required to meet the single failure criterion. Examples of these design features are:

(1) A third, spare component cooling water pump.

9(2) Four auxiliary feedwater pumps (two electric, two diesel driven).

(3) A third, spare makeup (HPI) pump.

3 e

e (4) A backup vital battery charger in addition to the two normal vital battery chargers for each load group.

(5) Four independest Engineered Safety Features (EST) batteries are provided to supply the four vital 125-V DC and 120-V instrument AC buses.

In tt.e event of a loss of offsite power, the standby diesel generators would provide power to these vital buses via the battery chargers for each channel. Only if a standby diesel ' fails, would the capacity of the related ESF batteries be required. There are no parasitic loads, as only Class IE loads are powered from these batteries.

In addition, Pebble Springs is being designed to be capable of maintaining the reactor in a safe shutdown condition, with the core covered, for up to 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> following a total loss of off-site and on-site power (ie, failure of both standby diesel generators).

The postulation of this situation is worse than the assumption of an " accident" being the failure of one train, and an " active" failure occurring within the other train since a double failure of the standby diesel generators would have to occur.

OUESTION 10 What are off-site dose levels resulting from Steam-Cenerator c'ube7 failure, associated with loss of off-site AC power due to upset from turbine generator trip? What is probability of such a grid failure following turbine trip?

?

l !

l

Resoonse to Ouestion 10 PGE simulations of grid stability deal with the modeling of physical phenomena in the time domain via algebraic and differen-tial equations. Since PGE does not use reliability models or probabilistic methods to assess the possibility of loss of off-site AC power at a nuclear generating plant, an exac.t probability of grid failure due to turbine trip cannot be quoted. However,

' x ve can give some qualitative measures of the probability of this occurrence for the Pebble Springs plant.

The offsite transmission systems are designed to remain stable for certain credible disturbances. Included among the disturbances considered in transient stability studies for nuclear plants

,are:

(1) The response of the interconnected system when the plant under stody is tripped off-line.

(2) Loss of the largest nearby generator.

as (3) Loss from a fault of the most critical transmission line.

These investigations result in systems designed to remain stable for these disturbances. Studies on Pebbie Spring conducted in 1974 investigated, mmong other things, these three classes of disturbances.

For the network studied, the system was stable and the integrity of offtite AC power was maintained. No remedial action was necessary to achieve this performance. Specifically, when the Pebble Springs

.i unit was tripped off-line, the voltage in the. vicinity of the off-1 l

site power source substation dropped less than 1.0%.

Additionally, no large power swings were observed. This is indicative of the strength of the nearby network and the ability of the Pacific North-vest hydro spinning reserve to quickly accon=odate any generation /

~ load i= balance. It should be pointed out, though, that the network

)

l.

-.g----

w,-.,

w

.__w-.

,,,=-w-n,

-9

-y-

=,-.,, = -,

I as planned today, in the vicinity of the Pebble Springs plant, differs from that system investigated in 1974. However, the difference. is that additional lines have been planned.

In 1974, three 500-kV lines out of Slatt Substation (which Pebble Springs feeds into) were specified. Today, four 500-kV lines are planned. There are additional 500-kV lines specified in the nearby network over and above what was foreseen in 1974.

All this would be U2* the direction of increasing the stability mar 2 n of the grid.

i An article entitled " Loss of Electric Power Coincident with LOCA" in the January / February 1977 issue of Nuclear Safety cites Federal Power Co= mission (FPC) transient stability information for power plants east of the Rockies. They give a probability for loss of offsite power caused by a LOCA induced generator trip of 10-3 Additionally, they point out that this number would be lower for a plant that had a high transient stability limit due to high installed transmission capacity and extensive grid inter-connections. This vould be the case in the Pacific Northwest.

~

The strong 500-kV interconnected network and the large hydro generation base, which has the ability to absorb reasonable load and generation imbalances, increases the stability limit.

Therefore, given the following:

.(1) That transient stability studies performed for Pebble Springs in 1974 showed that the integrity of the offsite AC power source was maintained when the Pebble Springs unit quas tripped; (2) That the 500-kV network planned today

r the vicinity of Pebble Springs is stronger than that t)anned in 1974; (3) That the generation / transmission system in the Northwest is " tighter", ie, grid interconnections are more extensive, y

than the systems east of the Rockies for which the FPC cites a statistic; __

o (4) And that the 500-kV and 230-kV transmission Itnes provide two separate preferred sources to each unit, we can conclude that the probability of loss of offsite AC power (a grid failure) due to a turbine trip at Pebble Springs would be well below 10~3 Offsite thyroid doses from a steam generator tube rupture accident with coincident loss of of fsite power are estimated to be 40.3 rem the exclusion area boundary and 10.4 rem at the outer boundary at of the low population zone. These doses are well within the guide-lines oI 10 CFR Part 100.

Assumptions used in this accident analysis were as follows:

Time Sequence:

O min Tube rupture occurs Low pressure trij occurs 6 min Loss of offsite power occurs 6 min Operator initiates RCS cooldown at 15 min Isolation of steam generators at 22 min ne Leakage and venting parameters:

Reactor coolant leakage to affected steam generator prior to isolation 79,440 lb Steam vented to atmosphere from affected steam generator 284,830 lb Steam vented to atmosphere from unaffected steam generator

~154,020 lb Steam generator partition factor 0.1 Main condenser partition factor 0.1 i

i L

Source term parameters:

Reactor coolant dose equivalent I-131 concentration 60uci/g Secondary coolant dose equivalent I-131 concentration 3.4 x 10-39Ci/g Preaccident, affected steam generator Preaccident, unaffected steam generator 7.9 x 10-4uci/g 60 Spiking factor included in source terms Radiation dose parameters:

Fifty percentile I/Q value 3.'63 x 10-' sec/m3 Exclusion area boundary 9.42 x 10-5,,efm3 Low population zone m3 see 3.47 x 10 -4

/

Breathing rate Dose conversion factor for I-131 1.48 x 106 rem /Ci The source tern parameters are based on anticipated Technical Specification requirements and are probably conservative by a factor of four based on observed iodine spikes at Westinghouse and B&W reactors.

Following a steam generator tube rupture, an estimated 6 percent of reactor coolant flashes to steam after leaking into the secondary

.sys$ en.

Hence, a steam generator partition factor of 0.1 (10 percent carryover of radioiodine) is considered to be a conservative value. l l

I

OUISTION 11 Are any special precautions taken for storage and handling of hydrazine?

Resoonse to Question 11 Special precautions are taken for personnel protection when handling hydrazine. The hydrazine purchased for the plant will be a 35 by veight aqueous solution, which can cause irritation to the eyes and skin. Those handling it vill be required to wear goggles, rubber gloves and pr,tective clothing.

One is the Hydrazine is stored in two places within the plant.

300-gallon hydrazine storage tank, which is in the Auxiliary Building and contains a 35 aqueous solution for the Containment Spray System.

It has a nitrogen blanket and is not vented to the room, so no vapors in the room are expected except during filling and draining operations which will occur only on a very infrequent basis. The normal ventilation in the area provides five air a.

changes per hour. The 35% aqueous solution has no flash or fire point and does not represent a fire hazard. The other storage area is in the Turbine Building where the feedvater chemical injectica hydrazine storage tank is located. This tank contains about 1000 gallons of diluted hydrazine solution, with a 1% to SI hydrazine content. The tank is vented to the roem, but the ventilation system, which provides two air changes an hour, will prevent migration of any hydrazine vapors.

7 OUISTION 12 What is status of investigation of merits of a primary vessel coolant level indication systep for use in post LOCA cooling for small breaks?

b -

Response to Question 12 B&W is no longer considering the use of primary vessel coolant level indication systems. Present analyses show that adequate system protection is provided by existing equipment and sensor design. For the specific case of small breaks in the primary system, please note the response to Question 6.

OUESTION 13 The fire protection system may be characterized as a "hard" or "sof t" system in respect to independence or dependence on fire detection and extinguishing systems.

In a local sense, in what particular locations is this plant dependent on administrative protection and early detecting-extinguishing techniques to protect vital shutdown system from fire damage? Is complete burnout assumed for local plant space or area such as one spreading room?

Resoonse to Ouestion 13 Generally, Pebble Springs design features include separation of redundant safety-related components by three-hour fire barriers, in addition to appropriate provisions for fire detection and suppression. In certain cases, it is not feasible to separcte redundant components by fire barriers. These areas have been analyzed to verify that suitable detection and suppression systems

' art provided and other design provisions incorporated to assure safe plant shutdown. These areas are limited to *he following described below:

Control Room Separation we in excess of Regulatory Guide 1.75 guidelines is provided throughout the control room, D

s except in the main control boards where the convergence of redundant cabling occurs down to the separation limits of Regulatory Guide 1.75.

It should be noted that.

the plant is designed to be capable of safe shutdown out-side the control room per General Design Criterion 19 as described in the response to Question 1.

A fire in one of the main control boards, if unsuppressed, could affect push-button modules for redundant components.

Such a fire, however, would not affect the automatic operability of those systems required to initiate rapidly following a postulated loss of offsite power (ie, auxiliary feedvater and related systems). Further-more, this postulated unsuppressed fire vould not affect 4

the capability to operate all safety-related loads frem,

the remote shutdown panel or locally, as appropriate.

Therefore, while rapid fire suppression would be expected for a control board fire, complete board burnout would not impair safe plant shutdown from outside the control room.

A fire in the area behind the main control boards could potentially involve an Engineered Safety Feature Actua-tion System (ESFAS), Reactor Protection System (RPS), or Solid State Interposing Logic System (SSILS) cabinet, all of which are safety-related. Cabinets of different safety trains are separated by at least 4 f t (horizontally) in this area, and SSILS cabinets containing logic cards for redundant safe shutdown systems are separated by at least 8 ft (horizontally). A fire in this area could conceivably involve SSILS or ESFAS cabinets. Such a fire, however, would not affect the ability to manually actuate the equipment at the auxiliary shutdown panel or i

locally, as appropriate, but could conceivably af fect the aute=atic start logic for the affected equipment. In the l

unlikely event of a loss of offsite power, concurrent with... -,

s-

.i

V F

a control room fire involving a SSILS or ESFAS cabinet, the affected auxiliary feedvater train may not be immediately available to restore feedvater flow. Physical separation, however, is provided between redundant SSILS and ESFAS cabinets, as noted above, and automatic fire detectors are provided inside the cabinets. Thus, at least one auxiliary feedvater train and related services will always be available.

Since the control room area is continually manned, a postulated fire should be suppressed in its incipient stages, providing assurance that the fire would not affect redundant cabinets and therefore would not impact on safe plant shutdown.

Containment Inside the Containment, separation of redundant ecmponents by 3-br fire barriers is not possible. Separation of redun-dant components is maximized using the following general concepts:

(1) Widely separated train A and 3 cable penetration

areas, (2) Widely separated train A and 3 piping penetration

areas, i (3) Providing cable separation greater than Regulatory Guide 1.75 requirements where practicable.

0 i

The limiting fire in the Containment is a reactor coolant pump (RCP) lube oil fire. These pumps are provided with automatic suppression systems and detectors. A postulated fire would be expected to be suppressed in its incipient However, a more conservative analysis has been made stages.

assuming a much larger potential for fire damage than that based on suppression system actuation. This analysis veri-shutdown fies that an RCP fire would not af fect safe plant even using very conservative damage assumptions. Highlights of this analysis are:

(1) The safety-related level, pressure and flow transmitters associated with the RCS, and pres-surizer and steam generators are all located outside die missile barrier which bounds the steam generator cavities. A postulated pump fire will not influence these transmitters.

(2) The failure cf the non-Class II power cabling to the RCP will not affect safe plant shutdown.

(3) Only one safety-related reactor power range fluz channel passes through any single steam generator cavity. Loss of this channel will not degrade the Reactor Protection System (RPS).

(4) Cabling associated with RCP temperature sensors enters the steam generator cavities. However, even I

if a pump fire is assumed to damage a significant l'

number of these temperature signals, there is sufficient redundancy and backup' protection avail-able to ensure that no degradation of the RPS trip function will occur.

L l.

b -.

e (5) The pressuri:er heater banks enter the pressurizer two locations, separated circumferential1y by at Considering the RCP automatic suppres-120 degrees.

sion system and other preventative measures noted, a pump fire would not be expected to affect both heater bank locations. One bank in each heater penetration area vill also be routed in conduit.

(6) The letdown coolers and associated isolation valves are located in the north steam generator cavity.

The letdown flowpath is normally open, and any indirect effects of an RCP fire causing an interruption in power to the isolation valves would not affect the flowpath.

(7) In gene *ral, only cable trays supplying power to components inside the specific cavity are routed in that area (along the missile barrier inner vall). The only excep-tions to this rule are train A cable trays, associated with the Containment air coolers, which travel through both steam generator cavities. This routing is selected to air maximi =e the separation from train B Containment cooler trays, which are located entirely outside the missile shield. Distance and s'hielding provided by the steam generators and the pressurizer would generally limit the impact of an RCP fire on any area cable trays.

However, even if degradation of loads in cable trays in the area of an RCP is assumed, no degradation of safe shutdown capabilities would occur.

(8) The pressurizer electromatic relief valve and associated motor-operated isolation valve are located in the south cavity. In the unlikely event of inoperability of these valves due to a f' ire, the pressurizer safety valves, which require no motive power, would provide RCS over-b pressure protection.

. l

-..,,,,,y

o (9) The pressurizar spray valve and associated motor-operated isolation valve are located in the south cavity. Inoperability of the spray function vould not affect a safe plant shutdown.

(10) The north steam generator cavity contains both decay heat removal (DER) letdown lines from the RCS (with four motor-operated isolation valves). A single RCP fire would not be expected to influence both letdown lines. However, even if failure of all four isolation valves is postulated, the plant may be maintained in a safe hot shutdown condition, indefinitely.

Subsequent to fire suppression, a DER flow path could be established by manually operating the DHR isolation valves.

Auxiliary Shutdevn Panel The auxiliary shutdown panel is only utilized to shut down the plant if an event, such as a control room fire, forces an evacuation of the control room. The auxiliary shutdown panel contains redundant instrumentation and controls for safe shutdown equipment and contains the transfer switches used to shift control from the control room to the shutdown panel. A postulated fire in the auxiliary shutdown panel could damage the transfer switches such that equipment control from the control room could be degraded. The separation internal to the shutdown panel and rapid fire extinguishment vould be expected to limit the effects of a postulated fire to transfer switches of only one safety train. To provide assurance that a postulated shutdown panel fire could i

not degrade control room equipment operability to unacceptable levels, the design will include an inter-Posing-3-hr rated fire barrier tatveen safety train A b

and B transfer switches and other ccaponents whose failure could degrade control roem control.

Cable Soreadine Rooms The Pebble Springs design employs two separate cable spreading rooms, each separated from oth'er plant areas by three-hour fire barriers and provided with automatic water spray systems and detectors. This design is not dependent upon administrative protection or early detecting-exti,nguishing techniques to assure safe plant shutd own.

Considering the flame retardant characteristics of cable and the speed with which a fire can be expected to propagate, horizontally, a postulated fire would not be expected to involve an entire cable spreading room. How-ever, even as a bounding analysis assumption, the complete lors of cabling in one cable spreading room would not affect safe plant shutdown.

OUESTION 14 As a general principle why is the design heavily dependent on the component cooling system for safe shutdown rather than using

~

the presumably more reliable service water system? Both concepts are used in the industry.

Resnonse to Ouestion 14 A closed loop Component Cooling Water System (CCWS) has been selected to supply cooling water to components required for safe i

^

shutdown for the following reasons:

(1) The closed loop CCWS provides an intermediate 7

barrier between the Nuclear Steam System (NSS) components handling potentially radioactive l

fluids and the Ultimate Heat Sink (URS) open to the environment. This closed loop design reduces the possibility of radioactive leakage to the environ-

=ent.

The Service Water System (S*45) is open to environment through the non-Seismic Category I reser-b voir or the Seismic Category I spray pond and, hence, I

is not as suitable.

i

a.

l 1

(2) Demineralized water with corrosion inhibitors is used in the CCWS to minimize corrosion and fouling of water passages in components of the NSS and other systems required for safe shutdown. This reduces the maintenance requirements for the components and increases their availability.

Service water is taken from the Columbia River (which provides makeup to the reservoir), spray Tunes, or reservoir and is not as high purity as CCW.

1 Ihose portions of the CCWS required to perform a safety function are designed to Group B or C quality standards and Seismic Cate-gory I requirements. A single failure in the CCWS will not affect the system safety function. A third, 100% capacity, spare CCW pump is also incorporated in design to provide a degree of system reli-ability beyond that required to meet the single failure criterion.

QUESTION 15 As an example of equipment separation which may be overlooked,

\\

describe the separation of the compressors for safety grade

=

air cooling systems.

1 Response to Question 15 The compressed air system serves no safety function. The safety grade air-cooling systems do not utilize pneumatic controls. In applications where air operators are used on safety grade valves.

the failed position of the valves has been selected to obviate

~

compressed air in an accident scenario. On 'these bases, the air compressors do not need to meet the separation criteria applied to safety-related equipment.

OUESTION 16 3

Describe the inlet-air protection system for the main control room. -

n--

-.. s.

j

Mat ' dose level would be imposed on operators af ter a LOCA with

" realistic" releases (not TID) to containment but with a single failure being that of electrical blowout of an intermediate size penetration (say 10" dia.)?

Resoonse to Question 16 The inlet, air openings for the main control room are located in the north and south walls of Control Building structure as shown on Figure 16-1.

The separation shown prevents both air inlets from being affected by a common missile or fire. Each of the two air inlet openings are missile protected.

Radiation doses to control room operators have been evaluated for a design basis LOCA assuming a 10-in. diameter open penetration in Containment for the duration of the accident. It is assumed that 100 percent of the inventories of noble gas and iodine isotopes in the fuel assembly gaps are released into the Containment at the time of the accident, with 50 percent immediate place out of the iodine isotopes on internal Containment surfaces. The Control Room Standby Cooling and Filtering System is actuated, and the normal control room ventilation systems are deenergized and isolated by an ESFAS signal prior to release of activity from the Containment.

Resultant doses to control room operators during the 30-day period following this postulated accident are 23.2 rem thyroid and 17.5 rem

. ggpma whole body. The limitations of General Design Criterion 19 of Appendix A to 10 CFR Part 50 do not apply since a Containment failure of this nature is not included in the plant design basis.

None.theless, doses of this magnitude would not i= pair the ability of the control room operators to maintain the plant in a safe shutdown condition.

Calculational codels used for this analysi; are described,in Pebble y

Springs PSAR Sections 15.13 and 15.36.

Assumptions that were utiliaed include the following: -

PSAR Table 15.3-6 CAP activity source terms Iodine percentages Elemental 91:

4:

Organic Particulate 5:

Contaimment penetration leak rate 1700 / day Containment spray removal rates 42.00 hr-1 Elemental iodine 0

hr-1 Organic iodine Particulate iodine 0.45 hr-1 Control room standby ventilation syscem data Outside air intake flow (filtered) 1000 cfm Recirculation flow (filtered) 2500 cfm Unfiltered inleakage flow 10 cfm 3

Control room volume 63,600 ft Iodine filtration efficiency

~

99:

4 Control room I/O values 0-2 hr 3.6x10-3 sec/m3

~ 3.6x10-3 sec/m3 2-8 hr 2.6x10-3 sec/m3 8-24 hr 1.1x10-3 sec/m3 24-96 hr 2.4x10-4 sec/m3 96-720 hr 7

Control room occupancy factors 0-2 hr 1.0 2-8 hr 1.0 8-24 hr 1.0 24-96 hr 0.6 96-720 hr 0.4 5

Respirator protection factor (self-contained breathing apparati kept in control roem) 50.

CUESTION 17 Describe electrical protection for power-carrying penetrations subject to in-containment faulting during LOCA.

Include penetra-

. tion for main coolant pumps. Describe protection in context of both overcurrent trip and ground fault (arcing) protection to prevent electrical-burnout and thus loss of mechanical integrity of the p.enetration. Include penetrations handling non-safety grade power circuits.

Response to Question 17 (1) 12.47-kV & 4.16-kV Systens The only circuits into the Containment at these voltage levels the four (4) retctor coolant pumps on the 12.47-kV system.

are There is a 15.0-kV feeder breaker for each motor with two (2) motors on each of two (2) buses. The 12.47-kV bus supply braakers act as backup to the feeder breakers.

Each feeder is protected by (a) an instantaneo,us current relay on each phase for short circuit protection, (b) a time delay current relay on each phase for avsrload protection, (c) an instantaneous or short time delay ground fault detector (the ground fault is limited to a maximum of 1600 amperes on the 12.47-kV system), and (d) instantaneous differential relays around the motor vinding. All of the above relays will trip the feeder breaker involved.

Each 12.47-kV bus supply breaker is protected by (a) a time delay overcurrent relay on each phase and (b) a time delay neutral overcurrent relay. All of the above relays will trip the supply breaker involved.

k. l

• 't.*

-ml

The supply breaker relays will be coordinated with the feeder breaker relays to ensure the proper sequence of operation and all relays will be set to operate before the penetration is damaged. In the event that the supply breaker relays cannot be coordinated with the penetration damage curve a second breaker with relays will be provided between the feeder breaker and the penetration. The relays vill be set to ensure that either breaker will protect the penetration. The need for the additional breaker can only be determined after the damage curve for the penetration is available.

The feeder and supply breakers are provided in the nor=al course of the auxiliary system design, are non-Class IE, and are located in the Turbine Building, a non-Seismic Category I structure.

Separate non-Class IE battery sources are provided for the feeder breakers and sunoly breakers or the feeder breakers and the additional breakers if they are required.

(2) 480-Y Load Center Systems The circuits into the Containment from the 480-V load centers are six (6) Containment air cooler fans, one (1) Containment polar crane, one (1) 480-7 motor control center and one (1) pressurizer heater bank SCR unit. The SCR unit supplies power to th'ree (3) of the pressurizer heater circuits.

There is a 480-V feeder breaker for each of the above circuits, nine (9) total. The 480-V supply breakers and, on the double 7 ended load centers, the tie breakers, if used, act as backup to the feeder breakers involved.

All feeder, supply and tie breakers incorporate static trip l

devices for protection and will trip only the breakers to which they are connected. All-trip devices are direct acting and I ;

l

independent of the breaker control power; therefore, separate battery sources are not required for the feeder breakers and the supply or tie breakers.

The fan and crane feeders are each protected by (a) a long time delay trip device for overload protection, (b) an instantaneous, trip device for short circuit protection, (c) an undervoltage trip device, and (d) a ground fault detector.

The motor control center feeder mad the 480-V supply and tie breakers are each protected by (a) a long time delay trio device for overload protection, (b) a short time delay trip device for short circuit protection, and (c) a ground fault detector.

The pressurizer heater bank SCR feeder is protected by (a) a long time delay trip device for overload protection, and (b)'a short time delay trip device for short circuit protection.

The three (3) cressurizer heater circuits from the SCR unit into the Contaicment are each protected by a thermal-magnetic breaker for overload and short circuit protection and are located on the SCR unit. In addition, there are current limiting fuses in the SCR unit.

The load center supply and tie breaker protective trip devices

' will be coordinated with the load center feeder breaker cro-tective trip devices to ensure the proper sequence of operation.

All trip devices will be set to operate before the penetration is damaged. except for the pressurizer heater SCR unit feeder breaker. The SCR unit feeder breaker trip devices will be, set to protect the SCR unit which may or may not protect the _ _

pene trations. The pressurizer heater penetrations fed from the SCR unit will be protected as described in the preceding paragraph. In the event that the supply or tie breaker trip devices cannot be coordinated with the penetration damage curve, a second load center feeder breaker with protective trip devices vill be provided between the feeder breaker and the penetration. The protective devices vill be set to ensure that either feeder breaker will protect the penetra '

tion. The need for the additional breaker can only be determined af ter the da= age curve for the penetration is available.

(3) 480-V Motor control Center Systems There are amoroximately 160 circuits into the Containment from the 480-V mo,cor control centers. The motor control center main bus feeder breakers, which are located in the 480-V load centers, are not used for backup protection because of their large rating relative to the individual load breaker racings.

The feeders are individually protected in the motor control i

center by either a thermal-magnetic breaker for overlo'ad i

and short circuit protection or a combination controller unit for short circuit protection and thermal overload relays for overload protection. In addition, the nine pressurizer heater banks have ground fault detectors. Each feeder will also have a thermal-magnetic breaker for over-l load and short circuit protection located between the motor

[

control center and the penetration to supplement the pro-tection in the motor control center. In the event that the circuit. overload, such as a very small valve motor, is less than the minimum setting of the overload ele =ent of the supplemental breaker, the penceration vill be designed D

to withstand the overload without damage..

(4) Lov Voltage Control Systems The majority of low voltage control circuits are self-limiting in that the circuit resistance limits the fault current to a level which does not damage the penetration.

'Jhere, on a case by case basis, a circait' is found not to

~

be self-limiting, primary a'nd backup fuse coordination or an additional feeder breaker will ensure safe operation.

(5) Instrument Systems The energy levels in the instrument systems are sufficiently low such that no damage can occur to the Containment penetration.

The above discussion covers both Class II and non-Class IE penetrations.

QUESTION 18 Page 9.9 describes what is apparently an electrical cooling system for Auxiliary Feedwater Pump room. Diversity was the basis for requiring engine driven Aux. feedvater pumps, yet apparently electrically powered room cooling is necessary to assure the engine-driven function. Please clarify.

Response to Question 18 i

The loss of room cooling does not have any impact on Auxiliary Feedwater System operation and, in fact, does not prevent the engines from running during a loss of all AC power for the design basis two-hour period. During that period, and beyond, the diesel-driven auxiliary feedwater pump room's temperature has been conserva-tively calculated to reach a maxi =ca of 150*F ambient air temperature.

b -

y

-M By locating the temperature-sensitive controls and instrumentation outside the room, the engine is capable of continued operation up to 180*T which is higher than predicted to be achievable.

QUESTION 19 in respect to the Volcanic ash problem:

x a.

Are the diesel-engine air filters designed to prevent dis-abling uptake of ash to the engine during this situation?

b.

What other air intakes have been evaluated to ensure continued, safe operation to shutdown during this condition such as:

Control room ventilation and cooling Diesel generator r#r cooling Aux feedvater engine air cooling Service water motor cooling Any other critical air cooling system Resnonse to Question 19 (1) Combustion Air for Safety-Related Diesel Engines An oil bath filter in conjunction with a settling tank is used to provide combustion air filtration for each standby diesel and for each diesel driver for the auxiliary feedvater pumps. This arrangement is illustrated in Figure 19-1 which shows the system in its normal operating mode. Figure 19-2 shows 'the operating mode during which the settling tank's y

contents can be replaced with clean oil without discontinuing cembustion air supply to the engines.

l~

N During normal operation, conbustion air passes through louvers into an inertial dust separator from which the majority of the ash particles are " bled off" via the Bleed Air Fan.

(The bleed air stream is about 102 of the total outside air entering the system.) The operating principle of these dust separators is such that they

~

have no tendency to " load up" since the ash particles would continuously be rejected to the enviz.as.

The relatively' small quantity of ash that could get carried throneh the dust separator would get caught in the oil bath filter. The volume of oil in the system is the determining factor in its ability to accommodate a J

postulated ash loading. In this connection, the~ effec-tive quantity of oil has been increased by using a settling tank, through which the oil is normally circulated.

In the event that the oil in the settling tank should require replacement, this could be accomplished without shutting' off the combustion air flow, as is illustrated in Figure 19-2.

However, the si:a of the settling tank has been conservatively designed to permit operation throughout the entire period of the design basis volcano without oil replacement.

(2) Ventilation Outside Air Intakes for Safety-Related Systems 7

Safety-related equipment cooling during postulated accidents is generally accomplished using fan coil units which recirculate air within the area, the heat removal medium being the safety-related ESF Chilled Water System.

In this connection, loss of ventilation air to these spaces does not affect the rooms' anbient temperature a

and, thus, the ability of the equipment to operate is not.

4

i i= paired. This general approach has also been used for the staudby diesel engine rooms and the auxiliarv faadwater diesel engine rooms. It should be noted that for the latter rooms, however, loss of the cooling units i

would not prevent the diesel drivers from operating during the two-hour period postulated for the loss of all AC power scenario. (This is discussed further in

.d the response to Question 18.)

Regarding the Control Room Standby Cooling and Filtering System (see Figure 19-3), the operator has the option of putting either of the redundant systems in a full recircu-lation mode. As discussed in PSAR Section 6.4.3.3, based upon 10 persons oce..pying the control room, sufficient air 4

quality exists to remain in this mode for up to 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br />.

The operator iould also have the option of introducing outside air i'ntermittently during this period. It is esti-mated that 12 minutes of operation in the latter mode vould permit the introduction of 12,000 cubic feet of outside air and would be sufficient to replenish the oxygen for an

~

additional four hours of control room occupancy.

The backup service water pump rooms use "all-air" cooling systems rather than fan coil units due to the remoteness of

'the Spray Pond Intake Structure from the ESF Chilled Water System. This scheme is illustrated in Figure 19-4.

Inertial dust separators similar to those discussed above are used at the intakes which would remove the majority of the ash.

However, to accommodate any ash carried over, the equipment will be designed.to be " ash-proof".

Apart from the foregoing, there are no ventilation air intakes critical to safe plant shutdown.

? l

QUESTION 20_

followed by For a main steam line failure inside containment

~

the first random failure being that of the opposite main steam line isolation valve to close, describe how excess flow is prevented through "non-qualified" valve failures such as turbine

~

by-pass valves.

In this connection, clarify the rationale which, in some designs, assumes that the large LOCA is " coincident (!)" with an earthquake assuming no LOCA, the failure of other kinds of, " passive"

but, elements _(such as main steam lines in containment) cannot be tolerated - since subsequent application of the single random failure criterion would destroy critical active services.

1 Response to Question 20 The main steam lines from the steam generators up to and including

\\

the first restraints downstream of the main steam isolation valves are designed to Group B quality standards and Seismic Category I Therefore, a main steam line break inside Containment requirements.

~

An approximate and an SSE are considered independent events.

for a main steam line break following the probabilistic assessment scenario outlined in this question yields a probability less than one in ten million per reactor year (see NUREG-0138, page 1-10).

It has been concluded (NUREG-0138, page 1-10) that a main stream line break of the type envisioned in this question would have a upgligible contribution to the overall risk relative to other possible accident scenarios having a greater or equal likelihood of occurrence.

SSE and main stream line break inside Containment Thus, a concurrent is not considered a credible design basis accident. The design and performace of the non-safety grade components in the main steam

. i I

~ - -.,,, -,

e.

c system are compatible with the accident conditions for which they are called npon to perform as a backup to a single failure in safety-grades components (NUREG-0138, page 1-11).

OUESTION 21 Are the main feedvater isolation valves designed to provide the

- x closing function in a bi-directional flow sense? Is instrumen-tation diversified to assure main feedwater flow interruption when required? Does this include separate d-c or inverter powered systems?

.What prevents suurious closure of main feedwater systems in the light of the critical need to stop such flow when necessary?

What in the esticated frequency of such closures as the original accident?

Response to Question 21 (1) The feedwater isolation valves are designed for isolation in either direction of flow against maximun anticipated differential pressures.

(2) Feedwater isolation signals for Pebble Springs are actuated by the Engineered Safety Features Actuation Sys tem (ESFAS). Diverse input signals to the ESFAS, any of which can cause feedwater isolation, are low steam generator pressure (600 psig).. low reactor coolant pressure (1600 psig), and high Containment pressure (4 psig).

(3) The two independent and redundant ESFAS channels are powered from separate redundant uninterruptable 120-V AC power supplies. During normal operation, or when the standby diesel generators are running,120-V AC power for each ESFAS channel is supplied through a rectifier-inverter h

O s

combination. Upon loss of AC input power to a rectifier, the set of station batteries associated with the power bus supplies power to the inverter for up to two hours, if required. Should an inverter fail, an automatic transfer j

to direct AC power, bypassing the inverter, occurs. With

\\

this arrangement, no interruption of power to the ESFAS j

occurs during a loss of offsite power. Refer to Pebble h

1 Springs PSAR Section 10.4.7.3.7 and Figure 8.3-2 for more i

i information.

(4)- The likelihood of a spurious closure of a feedwater isolation valve for Pebble Springs is minimized through design features of the ESFAS. The ESEAS analog signal initiating circuits employ 2 out of 3 logic to prevent the failure of a single analog element (pressure transmitter, etc) from causing a spurious isolation. Furthermore, the ESFAS digitial logic actuation channels, which process the analog initiating circuit information and generate the isolation signals to the valves,

~

are designed so that loss of electrical power does not result in an actuation signal. The reliability of the ESFAS is enhanced by frequent on-line testing using built-in test features; refer to PSAR Section 7.3.2.

In addition, loss of ESFAS electrical power is annunciated in the control room.

Each feedwater isolation valve is equipped with an independent pneumatic actuation system, as described in PSAR Section 10.4.7.3.7.

The design is such that a single failure in the electrical power supply system vill not cause a spurious isolation valve closure oue to spurious operation of a single air solenoid valve. As in the case of the ESFAS, fr'equent testing of the l

solenoid valves using built-in test features will be used to l

l enhance reliability.

l (5) We are not aware of any detailed studies of the frequency of spurious closure of feedvater isolation valves.

3..

QUESTION 22 The SER indicates that certain cables will be tested for water resistance by submergence..

How often will this be done and what is the probable frequency of exposure to this, condition during operation?

s Is this sort of testing program proposed for the electrical wiring and penetrations within containment. If not, why not?

Response to Question 22 Long-term cable moisture absorption tests are prototype tes ts

~

of single-conductor cables or single-conductor elements of multi-conductor cables as indicated in Pebble Springs PSAR Section 8.3.1.1.10 This test is performed prior to purchasing the cables and is intended to qualify each particular insulating No

- compound for long-term performance while immersed in water.

further moisture absorption tests will be performed. However, past experience with insulation materials capable of passing the

~

tests described in PSAR Section 8.3.1.1.10 indicates that the life expectancy of the finished cable should'be unaffected by continuous water submergence. All cables are magger tested after installation and prior to operation to ensure that insulation integrity was not compromised during the installation and cable pulling process.

It should be noted that 'the moisture absorption tests apply to all cables routed in the plant regardless of the location in the plant or the raceway type.

Only the following cables, however are considered to be subject to flooding:

(1) Cables in manholes and undergr. cud duct banks p

from the main buildings to outside remote buildings, structures or equipment during vet weather. --

(2) Reactor vessel out-of-core detector cables run 1

in conduit which is embedded in concrete in the Containment at the Elevation 740'-0" level during a

.l postulated DBE.

However, all junction boxes will be watertight.

~

e Containment penetration assemblies are tested in accordance with IEEE 317, which includes considerations for the most severe environmental conditions postulated for the penetrations, includ-ing the effects of moisture associated with the Containment Spray System. The Containment penetrations will not be subjected to water immersion during operation.

QUESTION 23 In once-through steam generator designs, the auxiliary feedwater system must respond very promptly after main feedwater is tripped.

Furthermore, the main feedvater system is presumably assured to trip during any significant seismic event.

Against these conditions it appears to be poor practice not to seismically qualify the condensate storage tank as the viable

" passive" source of critical feedwater following a post-earthquake trip and shutdown. The present design does not require this but, instead, depends on the electrically driven (stopped and restarted on diesel power) service water system to provide suction to the Auxiliary Feedvater pumps. For this particular condition, the advantage of the diverse engine driven Aux Feedvater pumps is lost since suction must be provided by the electrically powered service water pu=ps.

Why has the design evolved in this manner?

?

L f -

Resnonse to Question 23 The single failure of one standby diesel generator will not affect safe plant shutdown following a significant seismic event and a loss of main feedvater. Only in the event of an SSE and subsequent failure of both standby diesel generators would the auxiliary feedwater (AFW) supporting systems (eg, backup service water from the Seismic Category I spray pond) fail to provide the source of feedwater necessary to allow the Auxiliary Feedwater System safety, i

function to be performed.

The failure of both standby diesel generators is an extremely unlikely event. The failure of both standby diesel generators concurrent with a significant seismic occurrence is an event of such low probability that it is not a credible design, basis.

Therefore, upgrading the condensate storage tank to Seismic Category I is not considered necessary or justified.

QUESTION 24 From the standpoint of finding the worst credible situation in the context of the maximum race and degree of subcooling of the unbroken primary coolant system, it appears that main steam line failure within containment (which disables pressurizer heaters and provides ECCS trip signals) coupled with failure of main feedwater trip, is probably the worst configuration (It is also presumably intolerable, if persistent, from the standpoint of containment pressurization).

Disc'uss the consequences of this event in respect to:

Degree and rapidity of return of fission power after rod a.

insertion.

'l b.

Thermal gradients in most severely affected parts of reactor vessel and steam generators and subsequent sudden b __

1 i

+

rise of primary coolant pressure to safety valve setpoints after chilling the interior face of the vessel.

Maximum containment pressure as function of time of con-c.

tinued run-on of main and/or auxiliary feedvater flow to the f ailed steam generator.

]

Response to Question 24,

The design of the main and auxiliary feedvater systems is such -

that feedvater is automatically terminated to the steam generator with the broken steam line. One of the design bases for automatic termination is to prevent Conteinment overpressurization for Shutoff of postulated steam line breaks _inside the Containment.

-i main feedwater is accomplished by terminating feedwater pump operation, closing the feedwater control valves, and closing the main feedwater isolation valves. All of these components are in i

Termination of auxiliary feedwater is accomplished by series.

closure of an auxiliary feedwater control valve and an auxiliary feedwater isolation valve. These components are also in series.

NRC Safety grade signals are used to isolate these components.

The regulations require designs be based on single failures.

transient postulated by this event would require multiple failures and, therefore, goes beyond current design rules.

(1) In the steam line break analyses performed, the criterion of no return to criticality has been met. However, as shown in Pebble Springs PSAR Figure 15.14-1, there will be a momentary This return-to-

" return-to-power" at end-of-life conditions.

power (=omentary neutron power increase af ter reactor trip due 10 seconds to suberitical multiplication) takes place about into the accident, produces a power peak of about 25:, and is of a small duration. The system analyses performed that yield this subcritical return-to-power are very conservative and do no't account. fully for the influence of the secondary

system design in reducing the cooldown rate for localized core effects, such as negative reactivity feedback from core voiding.

This feedback would tend to offset flux perturbations in a stuck rod channel.

The NRC Staff has expressed some concern regarding local peaking and thermal conditions in the area of the stuck rod.

In a number of meetings with the NRC in August and September 1977, B&W presented steam line break analyses of the return-to power Laterval that accounted for local voiding in evaluating fuel rod performance in the area of the stuck rod. The approach presented was a simplified two-dimensional study that was accepted by the Staff as giving adequate assurance of fuel pin integrity. However, further justifica-tion of this approach has to be provided, and B&'d has committed to benchmarking this simplified technique against more refined analyses using three-dimensional computer codes such as PDQ or FLAME. These results are expected to be presented by the end of this year.

In su= mary, B&W continues to use and meet the criteria of no return to criticality. Under very conservative assumptions,

however, return-to power during a steam line break is calculated,

-but no DNB or fuel pin fail'ure is predicted.

(2) Thermal stresses in the reactor vessel and steam generator have been evaluated for steam line breaks and other rapid cooling transients. The steam line break analyzed in Pebble

~ 7 Springs PSAR Section 15.14 yielded a low temperature of about 520*F.

Stress analysis results show that there is no problem

, ith tube integrity or crack initiation in the reactor vessel w

due to thermal shock at this temperature range.

(3) Evaluations for a typical Contain=ent have been made by B&W during the development of the autcmatic isolation system.

)

\\

These studies were parametric, and the parameters that were varied were (a) containment free volume, (b) air cooler capacity, (c) spray system capacity, and (d) auxiliary feedwater' flow rate. These parameters were varied within the single failure criterion.

A " worst case". condition postulated was that of all available q

auxiliary feedvater flow directed to the break and then into the Containment with the assumption that the Containment heat removal eapability was reduced by 50%.

The results of this evaluation indicate ' d2at a 58 psig Containment pressure would be reached in about 15 minutes. Conversely, if all Contain=ent heat removal capacity were available, the Containment pressure would be about 29 psig in 15 minutes. While these values are not specifically for the Pebble Springs Containment, they do give an indication of the time-dependent pressures that could be expected following a steam line break without isolation of the auxiliary feedwater flow.

QUESTION 25 In the startup of never design B&W systems, using comparatively large pumps and piping and using a water-solid secondary system, the temperature of the water in the secondary system is raised to 400-500* and subsequently the secondary is drained until normal level is obtained. Has the Staff examined the safety aspects of this system?

Response to Question 25 See response to Question 7.

' l u--

-.m, _ s..

~. -

j, -

o o tipr9 pp ee p;r fhM* t Q3 QUESTION 26 ia j

considering such matters. as (.1) off-site power failure, (2) con-o awk-(see ices 21 preceding) and recent incidents of failures in denser vacuum failure, (3) spurious main feedvater valve closure

-: u "j

ik auxiliary feedvater systems it appears that, single failure r;

%I i criteria notwithstanding, at least short term failures of the

.J]

auxiliary feedvater system must be considered to estimate the f.:

needed reliability of such system.

1 u

f, k' hat, for instance, vould be the peak primary system pressure,

' ',% j.

consequences to primary coolant system safety and relief valves and rate of primary coolant loss following failure of the Auxiliary Vf, Feedvater pumps when needed?

f (:k

.a Response to fNestion 26 D e feedvater sfstems are designed to current NRC regulations.

V Since these regulations include criteria for design and analysis a

assuming one single failure, and tha safety grade Auxiliary r u 'i Teedvater System c'ontains multiple redundant trains (four 50%-size c.

Q'

- capacity pumps are installed with independent power sources), the Pebble Springs design complies with the latest requirements.

[Q ]

Postulation of an event whereby all feedvater is lost requires j'n multiple failures in the main and auxiliary feedvater systems.

ii' Nonetheiess, a preliminary analysis has been made to determine the j [

3=

b event sequence, assuming that all feedvater is lost instantaneously without regard for a realistic mechanism. ne following is an 1:

' r estimate of the sequence of events expected:.

r g

?.

a :

.y r

4 l-f L

y _.._.

c

.2.-

t wit l2 !,

~

~.

N

~

LA

m r

y_-

~..

Event h-Time w u

4.,

HM All feedwater is lost and the RCS begins to w

0 see iM*i increase in pressure. *

&.7?O

+

V4 V:

7dy?

e 7 see Reactor trips on high RCS pressure.

o

-=.

p Pressurizer begins to relieve decay heat via

%..._(i r10 see steam to the RC drain tank at the pressurizer

": 2 s "

x Eh safety valve setpoint of 2500 psig (RCS pressure

>-7

&..c...

about 2740 psig).

mme

n...:-

=

• %;1' e 2 min Reactor coolant expansion causes the pressurizer to become water solid, and water T M,:

' 'h")

relief to the RC drain tank begins (RCS

^C pressure about 2500 psig).

'y Cia *,w v-T$)

9,5.55'

<10 min Containment pressure increases to the ESFAS l.4 S E 7.1 setpoint (4 psig), and high pressure ECCS coolant injection to the core starts automatically.

1 r45 min High pressure ECCS injection flow heat removal rate is about equal to the decay heat generation Prior to this t,ime, boiling has occurred in rate.

the core; and af ter this time, it vill diminish.

A coolable geometry is maintained at all times.

~

I fl '

b, 1.ong' term ECCS high pressure injection vill continue to provide coolant from the borated water storage tank (BWST). When the BWST low-Icvel signal is reachedg the operator can switch the.

L i

ECCS high pressure coolant injection to the

~

l-

.-recirculation mode, if auxiliary or main feedwater has not been restored (see Pebble

-[

Springs Section 6.3.1.4.1 for a discussion on this mode).

r

-52 w

w ~Wo ra p

' ~ ' ? ";?? ~ ~ ' 1 ~ ~-~ ~..~

~ ~ - - ~ ~..

i-

+

a

~

~

'?hY

'-~ _

. L._

.j

..z

((f,

a

g. m

.o o

,y.

ve, -

G (1B0' Et.60W)

~

e c-x

n. e..

etW)2 D'ta. Y,.A..

I F UP.PE..uar tes v: x R

lm:

f

=

p

_e.;

Q.

.M G..

x 50

c. ~;

e' s;

3.1-,'

/.(pzesseg.1EEE t,

E e

s

/

w k.

g 1--

h C(U.E9EAINAD)

'/

/

D(t anGR i

Lp vor tre,)

- - t e m)-

a

/-

g E'

]

h, ji

{.

in ;

B(VOjT VMNG.S)

/..

'i rtess DwucontR)

)L,

~

~

~

s g

a ; f..

g***

A(cogs)

~

m h-..

h(Lo.ugg,.- -

wm-;

l n

r j

~

..n d~.

I(ceixo)

Figure 6-1 B(N Sys:cm as a Sealed Manometer for a Relatively p-;

5 i Small LOCA I.

T?

*NWN 1

s 4~

g_

Ar.

+

Ji_w _

' ^ ^

__r

u.

gn:

N s

m-J, go'Et.60W)

@#~,

(

WE h

f.

3 n-

. o%

m em.

..w..

1 w

n#4 EM

/,F upper W MG)

~

m

/

a.

j

.=

a. =, 2 c1 1

M74 v.,

1 g

g p'EPREsscGE4 (n.

e's e,'

t, /

9=

g.

s s,.

w e

o

, e,

,oe gy 0,d (u m z 1 Y

b) 9, GQs

• • s O(,t.onsa T

.o.-

nor tr.c4

1. s pc5-c-

,'GE97.SATON L

o u

a.-

_. -..g t <

car 2

-jge:c-

,,,Q y't

\\

p, o

6. (ves VM.N B

e e

, V,o

.x

'.Sn.vv9st a9 coaucomsg) o, o

4 a

o0 j

a8 n

f

\\.(core)

' A'

~

svust. :-

L. wny 2

v I(ccs.bLtch h ; _

. ?. -

Figure 6-2 Flow Pattern During RCS Repressurization FolicAng a Relatively Scall LOCA 29 s

m

. q e

u

U

.N OTE S:

VALVE POSITIONS ARE INDICATED DURING NORMAL POWER OPERATION 1.

2...THE GCHEME 15 TYPICAL FOR THE. OTHER STEAH GENERATOR ESFAS ESFAS TO COND. DUMP

I I##

CLOSE VALVEb

.CLOSE

)

)

f 1

H51V

()

70 TURSINE

)

)

X e

TO COND. DRAINS ScX l SCH N

)

)

ESFA5 Ct.CS E

)

g MFIV HEATED FEEDWATER Scr scil Y-(

/

CLEAHuR RECIRC.TO COND.

C

[E

)

. ESFAS P

I f-SE h

i 6CI pCH no 4"

l

(

q' 3**

7.DRAlWS q f

neo t 9 7

'~

=

~

r, n

" SCI I SC'U-ESFAS

.CLOSE DRAIN TO COND.

STEXM GENERATOR h

+,

-).

3>&L r

H

-H

. FROM OTHER *a.G TYP.op}

['

(

(

f-

. REclRC. PUMP S.G. D R AIN S' t-- vcr S C.H 5-I

' VACUUM HEATUP SYSTEh Figure 7-1 l

A

Tests t -l e

815 TINS USES TO ACulEVE Bouflat COLD SticT80WN altereste method Me hed e

Fe sale.

freesweleer elessels sellet Sv etee _

Frede steer opsey er ellow prosaustees to seet

1. . Sedwee BC pressere Ife,g )38)'F) 8 esserelly 8C

.b

• setesel sisselettea ac twee opereales

1. . Clrselate BC for heet reaeval, shoolcel eieles

• Sesse addlelen (see 2

• CRDN lasertfee 845T wie NFl

1. . Sedese teoster powet 3

• Ausillery feedvetor feedvator end stese flew to

1. . Bedece SC temposetore Ife,g >100'F) ease through stese gemeseter 1

SF

• sti f ame swst

1. . Add wates to SC5 Ifer seassastles essesp.F Mabeep flew free mekamp taek 3

sunt sheetcal oddlatee)

Meheep flew free eebeep tank BFI t,e seal lejettlee

3. Best injestles les BC peep operealen ebeve 130'F mene Leteews flew to fittese

3. turitisettee fes water slosamp la pseparealee and deelteraliset few satselleg gene

• Deser best flew through

1. . Reduce SC temposatese (Te,g (303'F) sessy best seelese 4

Dda Mil te oestllssy Desey best flew to

1. . seemse BC peessere lie,g (308't. BC peeps evelliery spesy eps ey 2

idle l,

• dFl/LF8 with berle Add bests as!d free CA483 esid free SWST

1. . Cheelsel ekwtdeve to makeup eyetse 4

CA&BE Letdown flow thsough degeellier Furge ashaup test

3. Segeseles Les pressettser f!Il Add altseges to alth ester

1. Callepelag pressetteer stees bubble

,pressettser NS pese Add altsegen sq eshaup task

3. Furgleg makeup teak Close cose fleeding testation pese I. Feevent sese fleeding eposettes estwes Cr ne.e I

IC-,e.e.t so il.g este,

<!. econ e..eter to C.....C,. e.d este,s.

.i letdeve seelers, desey best peepe med CC.

,,lo:de

..e,l.

L,es,e.l..e,l.

seelste

l...te,..e.C,ese..... s..co.t..tle.

Itene

,ee,,le, Freeensleer vest to

3. Begeseles pressesleer lateswa lies I

.p Vest hydrogen free eskeep teak pose t

8. tweglag asheep tomb posse seneup tonk vent Desselfler

f. Degeseleg

/

teeenslet for gelag to sold sheldown.

Safety SeeJo.

testione ese selety grade.

Sesevall C&488 - Cheelsel SC = seester Ceelsegg 5F = secondary Fleagg WUSP = Maheep and f.elticelleet Dus - Bessy West Coeltag Water.

AJJillee and asten Secoveryl N3 = p!ssegen Supplyg CF = Cese fleedlag! and CCW = Ceepeneet 8

e e

5 em e

a I

TRMM ' A'

- - -.- s.

3 OUTSIDE A:R Y (SEISMIC CAT.I DUCT s (INLETASV. DOORI

/

STRUCTURE

['

,5, i2 l

^

^

g

\\m

.HISSILE l

BARRIE RS g

N i

. r.

. -c -..

OUTSIDE ]

j FL.EL.0E?.'-G l

A'"

I g

.T.a M N,'B'

.i.-

.I! /

'l g

OUTSIDE g.-.l7=.

A A

,j; d

i g

AIR

}

.; g l

5

'f d'

PART PLAN Coh!TP.0L BL OG g

.uvj e

,e OUTSIDE AIR 9 _ y..

INLET DUCT A

1 l

OUTSIDE. AIR EL.8P'/-6,-y l

I.. INLET r

i i

i SECT. A-K PEBBLE SPRINGS NUCLE AR PLANT MAIN CONTROL ROOM OUTSIDE AIR INLET OPENINGS -

FIGURE 16-1

O a

COHGUSTION AIR _

b

• TO DIESEL El'GINF.

OUT SIDE L

AIR'lilLET L 1

DUST ~.SEPA RATOR."r Hj;

_ OIL BATH

1. 7;EGU All2.lM6 BLE ED. AIR FAN ' d

[w LINE ZFILTER I

4:

w e

i w

.f;.. -

/

FLOW. '

e j f.',, '.

)

d%_-_.o__.

30 El'OW 5

s

^.

VENT.~ ~~ '~ /. Di<AIN COCK.

..,A...

Y COCl(._.

.. L. _ 3 - WA...

r

}-

'.(STRAIGHT.: :PO.5LTION Hjl h

g 8

-i:

~'

GAG E ~ ::.

~

stAgg y. g SETTLING.. p

.y.[; ' ;f' h.TA,

/., i TAN K I.C' 7,..

gi,..

g,., -

_)T q'

8*

', s - E,

's

'*.s - ;

.eip l

..r.,.

l

?..,..

n'.-

'?

DRAIN VALVE'.

,,~DCEEAN.OUJ

• \\*

5 -

> 0ll BATH;,AND SETTLING.. TANK

' ' O I L PU H RT_.

. ev.

,. e l

ij.,,ig ):

. ;3: ' ' 1 N hlOR M^ L. O PE RATIO N ~ ~

FIGURE 19 -I

~

,_7

-3

@ymI,'ii B.

@ !}i. % -

. l yt.'.u

- e @. V;..

. n. 7 I

_ = - - _

A n

v a

l cOMeusTION AIR V

.? -

=-

OUT SIDE i TO DIESEE ENGlHE AIR INLE~f d

k._

8 OU5T. SEPARATOR'--+

Hj:

.OlL BATH.

• - = EQUALIZING.1 BLEED AIR EAMi.

llH

': FILTER

- CIN E i

1

.w

- ELOW

/.

I h.

i

,3

l+ = =.

a NO FLOW

~

H VENT

-s

~._.._2.~._B..WAY'. C QC K' _.

DRAIN COCK l

(. BIG 117. ANGLE._P_OS1. TJON) r,'.

ilt l

--lll k-

__gji I

._GA6E

./.t SETTLING / M

.g i'

',,iGLASS 9 ","g.. TANK.J.

,, Q ";

7

'l.

..v lll e'.- '

"I

'CLE A N i OUI-~~~.~'

DRAlH VANE ~

..a CLE4[_ SETTLING iThN K'.. WIT H

. Zoll _punp:

. n,.

' FiduRE 19 - 2 '

_..It' _

j.. i... _ _,__

lei.i.".. 01 L. BAT 11 FlLTERT_ OPE RATING,..

4:hN'

.i.N, fi, y' / ::]if,

.. i f. y. ~

,. i: 4 (j ' ( k( 6.s,(p) :.rt -

p q O

A

y-

~

1 l

. :.' ISOLATION. DAMPER

~

'CillLLED WATER Coll j

(CLOSE FOR..RECIRC.ULATION MODE)

ELECTRIC 11EATillG DUST.. SEPARATOR ANDJ.FAH--

FILTER HOUSIN6 i

COIL

(

7 8

C' '

8 i

E H

li

(

_q, E

f f>)

O)

OUT SIDE Xo E

11

-i y'

i f.

O

}

P A

P C

> P A

2 R D A c

c AIR SUPPLY AIR FWN

.. BCEE D'JAIR '

MRECIRCUL~ATION. DUCT a

)

.~... CONTROC.RO~OMZ5 TAN DBYi' _..

.a 00 L I N G}T EI LT E R I N G ~ W ST E IC.~..~..'.1 ~ '~

.IRAIN ' A' CONTROL f

goo n;___

t l'.

.si.

n v

4

j;..

.c,.

b J'..

ISOLATION DANPER....

.L.,;

Dust. =Eiw2 ATOP _..^urrFANW,

FILTER.IiOUSING --

ClHEEE.9.. WATER" COIL -

.~..

.. ~ '9

• T _'.;

f,?

cp E

- f g'

I t

it E

__OUTSIDE n

e r-'a r

3 m

4 o

3 w

d.

.A F

- --.g g g ' " ~

,tp 1

,,. :.t.'s.

.__.5.W c

D R

A c

c

.. :.?: ~.

.v:

ELECTRIC. HEATING COIL

/ 9.' 7 T.

SUFiELY_ AIR _EAN~

0

- B E E.. E. D_~~A,I R-

c b

a _ +._._.

s 2 ' Q; 7..

-..w.T R A'\\N,.,. B,

... Vl.9,.;.r..

ElGURE' 19 -L.

j*. :-

-.=

. ' " ". +

i'

>f

..e 4... g;1, g.

a 1

9.

a

~

9 SHUT O.cF~ DAM PE R']

ZOUST SEPARATOR b

f---EXHAUET AIR

/

LOUVER WITil

- VUTSIDE ~_g B'ACK DRAFT 1

1 N

DAM PE R s

AIR g"

+

-- l s

~

g N

N SUPPLY FAN

/

BLEED AIR I(9 L TRAlbT.%t. PUMP. ROOM.. '.

~

T OL EE D FAN __ - ~

Y _~SiluTOEF_.37 ASPER ~/

a; DU5T1.SEFA RATOR'...

E%HAUST AIR

./ "

___ [_

. / ": -

)

COUVER WITH r _:pOUTSIDE.----

5 s

7. -

BKCK DRAFT r

l OAM PER c

AIg

.I

.: f

[

' -i

?

s O:

d A

3gppiLY. FAN

[f":BCEED AIRT!

I. '.

[-$,f.,.

. 'Ji/ -

_. BLEED FAH_T" ~

'TRAI.N__tPdi PUMR_ ROOM ~ ~_

.';i.4 :.;

e-

- n;a -

n.. <.e

z...-

..%. y.

.e.*

e*

u,..-

1.,'y I BNCKUP.' SERVICE WATER:YTAKESTRUCTURE '._

}-}-(j)'(

VENTil ATION 5YSIEEi-

~-

- ' A

. FIGURE 19-4 ~

i

. '9J5 !. -- " ~.~..

,yI

[%p;i.

,qer uY..:..

fc yy.r:

p..

A A

__ __.