Text

Rept on Saint Lucie 1 Natural Circulation Cooldown on 800611. Certificate of Svc Encl
	ML19340E903
Person / Time
Site:	Crane, Saint Lucie
Issue date:	01/14/1981
From:	Imbro E; NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To:	;
Shared Package
ML19340E901	List: ML19340E900; ML19340E903;
References
	NUDOCS 8101160175
	Download: ML19340E903 (28)
	v • d • e

~:

~

h REPORT ON THE SAINT LUCIE I NATURAL CIRCULATION COOLDOWN ON JUNE 11,1983 by the OFF [ FOR ANALYSIS AND EVALUATION OF OPERATIONAL DATA Prepared by:

E. V. Imbro

a NOTE: This report documents results of studies conpleted to date by the Office for Analysis and Evaluation of Operational Data with regard to a particular operating event. The find #ngs and recorriendations contained in this report aree pro.ided in support of other ongoing NRC activities concern-ing this event. Since the studies are ongoing, the report is not necessarily final, and the findings and recoernendations do not represent the position or requirements of the responsible '

program office of the Nuclear Regulatory Commission.

B101150 \\ 7(

...o-

..m g

.-c-

=*

TABLE OF CONTENTS Page Table of Contents..........................

1 Preface...............................

11 1.

Event Description......................

1 2.

Event Sequence Analysis.................,.

6 3.

Findings..........................

9 4.

Recommendations.......................

18 l

5.

Co ncl u s i o ns..........,..............

22 List of Figures Figure 1.

Pressurizer Level vs Time..................

23 2.

Shutdown Cooling Flow Diagram..........

24

3. T and T vs Time......................

25 H

c i

=

w.

e he g

m.

,y s

PREFACE The findings, recomendations, and conclusions contained in this report are based on information gathered through informal channels between Florida Power & Light, Combustion Engineering, and the U.S.

h'uclear Regulatory Comission Headquarters and Regional Offices. To the extent possible, the information used in this report has been verified by cross checking with other P.,urces. The findings contained in this report relate mostly to Saint Lucie I and other Combustion Engineering pressurized wa ter reactors.

However, similarities among pressurizer water reactors leads us to be'lieve that many of the findings and re:c=endations may be broadiy and generally applicable to all pressu'rized water reactors. To this end, we recomend that a plant by-plant review not possible in this investigation, be undertaken by others, to assess the applicability of these findings and recommend-ations to other PW'R's and to analyze and evaluate plant unique design features not addressed in this investigation. The scope of this report, although primarily limited to Saint Lucie I, does address some of the design features unique to Westinghouse plants with Upper Head Injection and Babcock & Wilcox plants as they relate to a cooldown by natural circulation.

l

-./2-

= = = = = = = --;

r--

SAINT LUCIE NATURAL CIPCULATION COOLDOWN FOLLOWING SIMULTANEOUS LOSS OF COMP'ONENT COOLING WATER TO ALL REACTOR COOLANT PUMPS 1.

Event Description On June 11, 1980 while operating at full poner one of the two containment isolation valves in the component cooling water (CCW) return line from the reactor coolant pumps (RCP's) failed closed causing a simultaneous loss of CCW to all RCP's. CCW is supplied to all RCP's and Control Rod Drive (CRD) air coolers through a single line penetrating the containment that is connected to the non-esential CCW beader, designated as " Header N".

Like' wise the CCW return from the RCP's and the CRD air coolers is via a single line penetrating containment. The containment isolation function for both the supply and return lines is provided by two air operated valves in series in each line.

Both containment isolation valves in each line are located outside the containmer.t.

j The failure cf the CCW return valve was caused by a short circuit of a terminal board that controls the solenoid valve' in the air' supply line to CCW valve. The short circuit of the terminal board, due to the effects of moisture produced by steam leakage from a flanged connection in a blowdown line, caused the valve to fail closed at 0226 hours0.00262 days 0.0628 hours 3.736772e-4 weeks 8.5993e-5 months . After unsucces.sfully attempting to reopen the CCW return valve, the reactor was nanually tripped l

at 0233 hours0.0027 days 0.0647 hours 3.852513e-4 weeks 8.86565e-5 months . RCP IBl was tripped at 0234 hears and RCP's l Al, lA2, and 1B2 i

were tripped at 0235 hours0.00272 days 0.0653 hours 3.885582e-4 weeks 8.94175e-5 months . The RCP's were running approximately 8 to 9

__,_.J.

_e..

minutes without CCW prior to being tripped. This is within the allowable 10 minute time interval specified in the plant procedures.

At 0238 hours0.00275 days 0.0661 hours 3.935185e-4 weeks 9.0559e-5 months the operator jogged RC'P 1B1 to aid in establishing natural circulation.

The plant cooldown on natural circulation was commenced at 0300 hours0.00347 days 0.0833 hours 4.960317e-4 weeks 1.1415e-4 months by dumping steam via the atmospheric dump valves. This is the normal means of cooling the plant. CCW was reestablished to the RCP's at 0350 hours0.00405 days 0.0972 hours 5.787037e-4 weeks 1.33175e-4 months by junpering an air supply to the CCW return valve te reopen it.

The RCP's were not restarted since R,CP lower seal cavity temperature had exceeded the 250*F Ifmit specified by the pump manufacturer (Byron-Jackson).

Th,e manufacturer recommends that the RCP shaft seals be renoved for inspection if the lower seal cavity temperature goes above 250'F.

Since the RCP lower seal cavity and temperature had exceeded 250*F, the licensee continued cooling down b'y natural circulation in an expeditious manner but w1 thin t.ie cooldown rate permitted by Plant Technical Specifi-cations. During the time interval from 0601 to 0630 hours0.00729 days 0.175 hours 0.00104 weeks 2.39715e-4 months the Reactor Coolant System (RCS) pressure was lowered from 1140 psig to 690 psig.

As can be seen from Figure 1, at 0613 hours0.00709 days 0.17 hours 0.00101 weeks 2.332465e-4 months pressurizrr level oscillations l

l were observed when the charging pumps were aligned to the auxiliary spray connection in the pressurizer. As the charging pump flow was alternated between the auxiliary spray connection and the normal charging connection in the cold leg, the pressurizer level was observed to increase at a rate approximately tr.n times as rapidly as could be acCNnted for by the charging flow rate when the charging pumps were in the spray mode and 2

e l

3 ---

__r____'______

- mM '

~

rapidly dec'rease when in the normal charging mode. This behavior is i

indicative of a void som:where'in the RCS other than the pressurizer.

Sanples of reactor coelant taken ind'icated that there were not enough dissolved gases in the coolant to account for the magnitude of the level oscillations observed. This lead to the conclusion that the void was steam rather than noncondensable gases.

It ap, ears that the steam bubble was formed in the reactor vessel head since during natural circulation there is essentially no flow through this region of the reactor vessel to affect its cooldown. Although the subcooling margin (calculated using either T or the core exit H

temperature) indicated about 200*F subcooling, it was not representative of'the subcooling rargin of the fluid in the reactor vessel head.

Since the steam bubble in the reactor vessel head was not affecting the natural circulation flow, the cooldown was continued and at 1051 hours0.0122 days 0.292 hours 0.00174 weeks 3.999055e-4 months Shutdown Cooling (SDC) Loop B, as shown in Figure 2, was put into service.

During the time interval of 0613 hours0.00709 days 0.17 hours 0.00101 weeks 2.332465e-4 months when the hubble was first fomed until 1227 hours0.0142 days 0.341 hours 0.00203 weeks 4.668735e-4 months when the Low Pressure Safety Injection (LPSI) Punp 1A was started to take the RCS and pressurizer solid and to raise the RCS pressure; pressurizer level was maintained by alternately charging through the auxil-iary spray and normal charging connections and by starting and stopping 1

letdown flow from the RCS. At approximately 1237 hours0.0143 days 0.344 hours 0.00205 weeks 4.706785e-4 months the pressurizer level indicator went off scale high indicating that the pressurizer was solid. LPSI Pump 1A was left dead headed against the RCS in the injection mode until 1347 hours0.0156 days 0.374 hours 0.00223 weeks 5.125335e-4 months when it was secured and the mini-flow line was shut by closing the motor operated valves (MOV's) in the common line to the Refueling Water Tank (RN).

3

_ _ y - f_ ---l -

- ~.

r

After the pressurizer was taken' water solid, the operators tried unsuccess-fully to raise the RCS pressure above 200psig (produced by running LPSI Pump 1 A dead headed against the RCS) by using charging pumps, pressurizer heaters, and securing letdown. The absence of a pressure rise indicated that there was a leakage path from the RCS. During the 90 minutes the LPSI Pump 1A was running in the injection mode the RWT 1evel was observed to increase by 4 inches which corresponds to 4500 gallons or a flow of 50 GPM to the RWT.

Upon securing of LPSI Pump 1 A and closing the MOV's in the mini-flow line RCS pressure showed a slight increase. Sinultaneously a recheck of valve alignment revealed that the manualli operated mini ' low valve on LPSI Pump 1B was 1/2 turn open and it was subsequently closed. Since the securing of,LPSI Pump 1 A, the closing of the MOV's is.he mini-flow line, and the tightening of the manually operated valve occurred at essentially the same time, the actual leakage. path from the RCS to the RWT could not be deter-mined. The potential leakage paths as shown in Figure 2 are as follows:

a.

Through the LPSI Punp 18 manually operated mini-flow valve to the RWT or, b.

Backflow through the LPSI Pump 1 A discharge check valve and through the mini-flow line to the RWT.

Once the leakage path to the RWT was isolated, the RCS pressure responded in a normal manner. At 1432 hours0.0166 days 0.398 hours 0.00237 weeks 5.44876e-4 months letdown flow was ' established greater than the charging flow and a bubble was drawn in the pressurizer. The RCS 4

O eO-

="

A-

I behaved normally indicating that the steam bubble in the reactor vessel head was no longer present. The coolddwn was continued until the RCS war, in Mode 5 so that the RCP shaft seals could be replaced.

~

O f

e l

5

~

_a.

l 2.

Event Sequen:e Analysis o

As shown in Figure 3, following the reactor trip and the subsequent RCP trip both T and T initially decreased as expected. T, which is H

C C

controlled by the secondary side saturation conditions, stabilized at about 532*F the zero load temperature. T, after initially decreasing, H

l began to climb steadily at a rate of approximately 5'T per minute.

This increase in T caused the operator to betone concerned that natural H

circulation was not being established. Subsequently, RCP 1B1 was started momentarily at 0238 hours0.00275 days 0.0661 hours 3.935185e-4 weeks 9.0559e-5 months and run for. about 1 minute, in which time T H

dropped to a value equal to T. After RCP 1B1 was stopped, T increased C

H again at approximately 5'T per minute until it peaked at 555'F.

It then be'gan to decrease indicating that natural cir',alation had been established.

Approximately 25 minutes af ter CCW flow had been lost to the P.CP seals, the lower seal cavity temperature on all four pumos exceeded the 250*F temperature limit. Upon exceeding 250*F, Bryon Jackson recommends that the RCP shaft seals be removed and inspected for degradation.

Cooldown was commenced at 0300 hours0.00347 days 0.0833 hours 4.960317e-4 weeks 1.1415e-4 months using the atmospheric dump valves i

and RCS pressure was reduced by spraying into the pressurizer via the auxiliary spray connection. Since the RCS pressure strip chart recorder l

l l

l 6

l

~

.~

f~~~

~.c J J ' ' ~ ~' ~ ~

~~

goes off scale below 1500 psig it cannot accurately be determined at what pressure the steam bubble was formed.

It appears from looking at the strip chart of pressurizer level that the steam bubble (estimated to be about 750 cubic feet) was formed about 0613 hours0.00709 days 0.17 hours 0.00101 weeks 2.332465e-4 months at which time the RCS pressure was somewhere between 1140 and 693psig. This would bra:ket the temperature of the fluid in the upper head between 563*F and 504*F. Assuming a linear decrease in pressure between 0600 and 0630 hours0.00729 days 0.175 hours 0.00104 weeks 2.39715e-4 months would infer a temperature of the fluid in the reactor vessel head of about 536'F. The full power T is 585'F which means that over the H

four hours after the reactor trip this region of the reactor vessel cooled down only about 50'F.

This represents a cooldown rate of about 12:5'F per hour in the upper head as compared with approximately a 60'F per hour cocidown rate of the bulk RCS inventory over the same interval.

During the cooldown, erior to taking the pressurizer water solid, the pressurizer level was being controlled by alternatively spraying into the pressurizer and charging through the normal charging connections in the cold legs. The auxiliary spray was periodically initiated to keep the pressurizer level above the heater cut out point. The continued controlled variation of pressurizer level was beneficial in that the expansion and contraction of the steam bubble in the reactor vessel caused cool fluid to be flushed in and out of the vessel head area thus enhancing the cooldown of this region of the vessel. The centinued insurge and outsurge of the pressurizer, however, exceeded the I

capability of the pressurizer heaters to maintain RCS pressure. The con.

tinuing decay of RCS pressure necessitated the use of the LPSI Punp 1 A 1

7 l

9

in the injection mode simultane'ously with the LPSI 1B in the shutdown cooling node in order to maintain an adequate subcooling margin. At the time when LPSI Pump 1A was started in the injection mode, the pressurizer pressure and the subcooling margin had dropped to approxi-mately 110psig and 50*F respectively.

O e

m 8

~

[ ~

~

~~

'~~

+

3.

Findings The rapid depressurization of the RCS resulted in a plant condition a.

that was not anticipated by the plant operators. Although the actual safety significance of drawing a steam bubble in the reactor vessel head during the natural circulation cooldown appears to be small, the plant response did initially puzzle the plant operators. Thiscob have resulted in the plant operators taking actions that were inco,ry.ect~

Although this3s notjege in th,is instance, it does indicate that.

operator guidance needs to be developed in this area.

b.

The jogging of the RCP to aid in the establishment of natural circula-

' tion appears to have been unnecessary. The plant operators, apparently concerned over the increasing T, decided to jog RCP 1B1, the first H

pump tripped.

RCP 1B1 had been run 8 minutes following the loss of CCW prior to its being tripped, 2 minutes less than the 10 minutes allewable time specified in plant procedures.

Prior to jogging RCP 181, core T was approaching the normal full power T.

Energency Operating Procedures at the plant indicate that one of the criteria to ensure that natural circulation has been established is that core T is less then the nornal full power T.

As seen in Figure 3, CE plants exhibit a charac-teristic increase in T during the establishnent of natural circulation.

H Considering that T increased again at the same rate and stabilized at H

approximately the same temperature after RCP IB1 was stopped would tend to indicate that jogging was unnecessary in establishing natural circula-tion.

During the incipient stages of establishing natural circulation, 1

9 1

- -. ~ ~.

..s 4

[*'

~~~ ~

E E ~

4-

.. ^~

b

Cperators need to be made arare that T will initially decrease H

then rise and peak suddenly. While jogging the pump caused no prob-lem it did increase the potential for seal failure. Operator guid-ance in how to recognize natural circulation needs to be expanded.

The continued sloshing of the pressurizer eventually led to a c.

condition that resulted in the sinultaneous use of the LPSI pumps in the SDC and injection modes to naintain an adequate subcooling margin.

When aligned in this manner, the check valve on the dis-charge side of the LPSI pump in the injection mr,de becomes the only barri.* betweer. the reactor coolant fluid and the Refueling Water

, Tank (RWT). Since the RWT vents to atmosphere, a leaky check valve in this system alignment creates an unnonitored leakage path for primary coolant activity. The leak tightness of the check valves on the discharge side of the LPSI pumps needs to be periodically

' verified.

d.

The formation of the steam bubble in the reactor vessel did not inhibit natural circulation flow; although, it is estimated by the licensee that the size of the bubble was about 750 cubic feet.

Information provided by the licensee indicates that the bubble extended down about 10 inches below the reactor vessel closure fl ange. This left a 36 inch margin above the top of the hot leg-which corresponds to approximately 300 cubic fest of reactor vessel O

10 l

} ~ v v }[*

~

O

l vol ume. The steam bubble size would have had to be 1050 cubic feet before it would have reached the top of the hot legs.

Intuitively, it would appear that if the RCS pressure is slowly

(

decreased (causing a correspondingly slow expansion of the bubble) it is not likely that a bubble of this size (1050 cubic feet) would be achieved, since as the size of the bubble is increased the vapor liquid interface moves out of the upper head region to a progres-sively cooler region of the reactor vessel. This would tend to condense the steam. Also, the liquid temperature approaches the measured T as the surface moves'-from a stagnat flow area to one H

that is in the natural circulation flow path. This cooling effect also tends to inhibit further formation of steam.

On the other hand, a very rapid decrease in RCS pressure will result in a rapid rise of pressurizer level and may result in the expan ion of the bubble in the reactor vessel head into the hot legs.

In this case, the dynamics of the situation may not permit sufficient time for condensation of the steam bubble in the reactor vessel. However, this nay not be a problem in that the vapor should condense either in hot legs or in the steam generator tubes.

In any case, it may be desirable to maintain the pressurizer level between specified bounds during the level oscillations to assure 'that the vapor remains in the reactor vessel.

11 emoes 9

p.e..

,w.

+

..en.

ow * * -

7

Q ' *.

A rapid depressurization couid be a prob 1r.m for B8W plants, partic-e.

ularly if they are cooling down on natural circulation on one steam generator, as a steam bubble might form in the " candy-cane" region of the inactive hot leg. Once a bubble forms in this inactive hot leg either due to flashing in the " candy-cane" or due to vapor expanding out of the reactor vessel natural circulation could be precluded in the inactive loop.

It may be difficult to totally condense a steam bubble once forced simply by repressurization of the RCS. If the liquid surface is quiescent the liquid acts as a piston and the increase in the pressure causes the bubble temperature to increase.

The increase in RCS pressure causes a corresponding increase in the saturation temperature of the bubble and if the process of repressur-ization is adiabatic, it is thermodynanically impossible to condense the vapor.

The stean bubble an only be condennd by cooling of the bubble which may be a relatively slow process because of the hot walls of the RCS piping.

f.

During a natural circulation cooldown with an idle steam generator, it may be possible to form steam bubbles in the upper regions of the "U" tubes.

Although these bubbles can be condensed by cooling down the secondary side of the steam generator, the concern is that a rapid collapse of these bubbles could cause a large decrease in pressurizer level and possibly drain the pressurizer. Ope rators,-

should be alerted to this possibility when starting an idle steam

~

generator during a natural circulation cooldown.

/

12 l

l i

f

g.

The RCS ' pressure boundary is an extremely reliable passive safety feature in that uith the exception of the RCP seals, it requires no dependence'on auxiliary systems to' perform its function of contain-ing the reactor coolant. The integrity of the reactor coolant pump seals depends on cooling water. This cooling water is provided either by seal injection or CCW to the thermal barrier heat exchanger or inte-gral seal cooler. Saint Lucie, as well as all other operating Combustion Engineering plants, does not have seal injection as a backup to CCW for seal cooling.

In addition, the CCW is supplied to the RCP in such a manner that a single failure can stop cooling flow to all RCP seals.

Since the loss of CCW to the RCP' seals may cause degradation of the RCS pressure boundary even if the RCP's are stopped, the CCW supply to the RCP's should be highly reliable even though it may not be from a safety grade source. Considsration should be given to upgrading the reliability of the system supplying cooling water to the RCP sea'.s.

i h.

Saint Lucie was rapidly depressurized due to a valid concern over the capabi ity of the RCP seals to maintain their integrity.

Following an extended less of CCW to the seals, the lower seal cavity temperature limit will be exceeded in about 25 minutes with.the reactor at hot stand-by. At this point, the pump manufacturer recommends seal removal. and l

inspection. This guidance, however, is not reflected in plant operating procedures. The manufacturer also recommends, as indicated in plant

'\\

procedures, that the pumps not be run more than 10 minutes with 0CW. In c

g a similar loss of CCW in 1977, the RCP's were run for 12 minutes and

\\

c.

this resulted in failure of the lower stage of a seal in one RCP. The June 11,1983 incident showed no damage to the two RCP'seais inspected.

These RCP seals were run for 8 to 9 minutes without CCW.

Based on a 13

-. c..T -.. :.

. = :.. ~ L..

.:=.-

~

~ ~;- -

== +3 _,

~"~

- ~ - -

single data point, the 10 mi,nute criterion appears to be a good one; but considering the seal stage failure in 1977 after 12 min-utes, which clearly could have been influenced by other variables, the question of whether 10 minutes is close to a threshold is also raised. There appears to be a lack of data on the "off-design" performance of RCP seals.

i. The consequences of forming a steam bubble in the reactor vessel head on plants equipped with Upper Head Injection (UHI) were studied to determine possible adverse effects of UHI actuation to condense the steam bubble.

UHI is a passive injection system provided for core cooling during a LOCA. The injection pressure of the UHI was selected such that during a large LOCA the vessel head would be filled with water curing injection. The UHI flows are high in this case due to the rapid depressurization of the RCS.

UHI during a small break LOCA would be into a two phase region but the flows would be low due to the slowly decreasing RCS pressure.

In either of these cases the ootential for water hammer is ninimized.

The UHI system is initially isolated during the cooldown and depres-suri:ation process as required by normal operating procedures. gow-ever, during natural circulation, if a steam bubble were formed in the vessel head, actuation of UHI would be one method of collapsing the bubble available to the operators.

In this mode of operation, the 14

/ -

____;---3.-----------__

s

.L

UHI could be rapidly injecting into steae filled piping which might produce wat'er hammer by rapid steam' condensation. This same situation could possibly occur over some spectrum of intermediate size breaks.

In some plants equipped with UHI, the nozzles were added to the vessel head after the final heat treatment. Since the failure of a UHI line on the RCS side of the check valves results in c. LOCA, it is inportant that the UHI piping will withstand any water hammer loads imposed by injection into the steam filled lines. Consequently, it shou'Id be verified that the UHI piping will withstand potential water hammer loads associated with the use of UHI when injecting into steam filled lines either to collapse a steam bubble in the reactor vessel head or during a LOCA.

j. One design feature that was provided on Saint Lucie was a valve closure on Hi T for the CCW return valves provided for each RCP. The func-tion of this automatic closure is to prevent reactor coolant from going into CCW system in the event of a tube rupture in the RCP Seal Coole r.

Therefore, if the CCW outlet temperature from the RCP Seal Cooler exceeds the inlet temperature by more than '203*F the air operated outlet valve will close.

l l

The April 1977 natural circulation cooldown was caused by failure of a containment air compressor which caused these return valves to fail

(

closed.

After the air supply was restored, the operators were unable to l

open these valves since the Hi T closure logic prevented the valves from 15 i

~

...__ _., o _

=

being quickly reopened so t'at CCW could be reestablished to the h

seal s.

Following Gis incident,' Saint Lucie has provided a reset feature on these valves with a 10 second ' time delay incorporated.

This permits the valves to be reopened after closure on Hi T

but the valves will automatically close if the differential temper-ature does not go below 203*F within 10. seconds.

Incorporation of a reset feature on other plants that have an automatic isolation of CCW on failure of the RCP seal cooler would

^

permit a more rapid restoration of CCW.

k.

As the event progressed at Saint Lucie, the control room became ir.creasingly occupied as plant personnel responded to the off nomal plant conditica. Although the severity of the event did r.ot warrant activiation of the Technical Support Center (TSC), it appears that the TSC could have been used to assist the operators in the evaluation of the event. This would have gotten some plant personnel out of the control room and pemitted the operators to function in a quieter environment.

i t

l Tge TSC is officially activated in accordance with the Site Emergency Plan only during more serious events.

Putting the Emergency Plan into effect implies notification of the State, activation of the Offsite Support Center and other functions geared toward an event of greater severity. Clearly, there exist many events or situations where '

16

++-#

= = = -== -

.==-y

- =

* * + - - - - ~ - -

it trould be beneficial to activate the TSC function without requir-ing the Sit'e Emergency Plan to b' invoked..

e Criteria should be established to allow the use of the TSC function by itself, apart from the Site Emergency Plan during events of less severity which progress over a relatively longer period of time.

1.

The RCP manufacturer, Bryon-Jackson, recomends that the RCP seals be renoved and inspected if the lower seal cavity temperature exceeds 25 0* F.

At the Saint Lucie plant the lower seal cavity temperature for each RCP is indicated on the main control board and is high alarmed at 170'F.

During normal operation this temperature is about 100-110*F. Although the lower seal cavity temperature is logged twice a shift, a situation could arise where this temperature could exceed the recomended 250*F limit for a period of time and return to a temperature below 250'F without being noticed by the plant operators.

This situation could be avoided if in addition to the high temperature alare at 170*F a high-high temperature alarm was also installed to actuate at 250*F.

In reviewing the Saint Lucie procedure for Reactor l

l l

Coolant Pump Off Normal Operation, no mention was made of the manufac-l l

turers recomendation for RCP seal inspection if the lower seal cavity tenperature exceeds 250*F.

d.

Recommendations During the incipient stages of establishing natural circulation, oper-a.

ators should be made aware that T will initially decrease then rise H

17

~~-

r:

.-J - --

=T

4 fairly rapidly and peak suddenly..This guidance would preclude unnecessary concern or starting of RCP's.

b.

Consideration should be given to alarming the lower seal cavity temperature if it exceeds the recommended limit of 250'F.

Cooldown procedures during natural circulation should be expanded c.

t' specify a non-mandatory rate of depressurization which if adhered to would avoid formation of a bubble in the reactor vessel head.

d.

Procedures should be developed to guide the operators in responding to a bubble formed in the reactor vessel head. These procedures should include some definite limits on the controlled oscillations

' of pressurizer level, if this procedure is recon, ended to aid in cooling the head. Enphasis should also be placed on the fact that it nay not be possible to condense a steam bubble by repressurization without cooling.

Operator training should be expanded to allow operators to quickly e.

recognize the symptoms of formatio 5 of a void'in the RCS other than the pressurizer.

/

18 n.

- *: ~ L. : f. - [ ~ ". ' - -

=-

luu

Plant procedures should be f'rnalated addressing the sinultaneous f.

o Use of the i. PSI pumps in both th'e injection and shutdown cooling mode.

Particular attention should be directed to any potential leakage paths from the RCS to the P.WT.

g.

leak tightness of the check valves in the LPSI pump discharge lines needs to be pt-fodically verified. These valves should be included in the Inservice Test Program.

h.

Analytical models used in accident and transient analysis should be

~

exa.ined to verify that they properly acccunt for the observed thereal and hydraulic decoupling of the reactor vessel head region from the remainder of the reactor vessel.

i. Consideration should be given to the potential for the formation or accumulation of vapor in the " candy-cane" of the B&W reactors, particularly in the inactive 'szop when natural circulation cocidown is being accomplished with a stgia steam generator.

j. Plants not using seal injection and having a High T closure feature on the CCW discharge valves from the RCP should consider installing a time delay reset that would permit temporary override of the closure feature.

19

~.T

~.

- _ _:n_ -.- -.

...3 i -

k.

Loss of Instrument Air Procedures should be reviewed and revised as necessary to address the potential effects of extended loss of instrument air on RCP operation.

1.

Consideration should be given to providing a supply of cooling water to the RCPs that will not be totally disabled by a single

~

failure.

Consideration should be given to providing a means to measure m.

temperature in the reactor vessel'. head, Definitive data based on operating experience or testing should n.

be obtained from RCP vendors for pumps not provided with seal injection:

(1) The time interval pump seals can survive on an idle pump without CCh' at normal operating RCS temperature and pressure.

l (2) The time interval in which RCPs should be stopped 'following,

the loss of CCW to preclude seal failure.

The potential for water hammer due to steam condensation in Upper' o.

Head Injection (UHI) lines should be evaluated.

\\

t 20'

1

. g

. ~ _. -.

p.

Graduated c,riteria should be developed to allow partial implemen-tation of the Site Emergency Plan for incidents of less severity that progress over a significant period of time.

i q.

Revise plant procedures as necessary to include RCP manufacturers recom.endation on seal inspection if lower seal cavity temperature exceeds 250'F.

i e

e 4

4 i

o e

4 21

~~.m

...m..

_~ **

~~T-

'~

^

~

m

5.

Conclusions Tne primary sig'nificance of the June 11, 1980 natural ciec;1ation cooldown is that the formation of the steam bubble in the reactor head was unexpec.ted

- ~........ _.... -..

by t'he plant operators and was not immediately recognized by them. This_

L could have led to the operators taking improper corrective action,_

although this was not the case. However, operator training needs to be expanded so that the fortnation of a steam bubble in the reactor head can be promptly recognized by the operators during a rapid depressurization while undergoing natural circulation cooldown. Recognizing that the reactor vessel head is not in good thermal and hydrauic comunication with the remainder of the reactor vessel and that the formation of a steam bubble in the head is possible; procedures should be developed to guide the operator in plant depressurization to avoid bubble formation.

Under certain conditions, rapid depressurization is necessary or desirable; therefore, procedures should also be developed to guide the operator in cooling down the plant by natural circulation with a stean bubble in the reactor vessel head.

l l

Although further investigation, as enumerated in the recor*r.dations, is nect cary, the voiding of the reactor vessel head does not represent an imediate safety concern. Clearly, it is a plant condition that should be avoided, if possible. However, formation of a steam void 'in the reactor vessel head did not in any way impede natural circulation. Other than the problem caused by the leakage of reactor coolant to the Refueling k'ater Tank, apparently due to a partially open valve ~, the reactor was brought to a cold shutdown condition in an' orderly manner, considering the new situation that confronted the plant operators.

22 l

l

g N

7 O

s.

O-N

.g.....

. 4..

...._.............O o

f

.a*

g...

g

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

s s-e....

)"..

..e4 2.

~.

,O

. 9......

l e

g..

...g

. 4....

I..

g C

g

...a..

5

-t.

e.........

3.-

..a

.l

......... I a

I i..

..',e m=.;

+

e

..l...

t

,.g......,. t.. e..

y 4,

.g.

i...............

3 3_

....e....

g

...e

(...

s e.....

e 4

.e e.

e

....l

.l.

l

.e

.J.

9 j:..

....l..

l J

s..........

.I..

I'..........i g..

g......

. e...

.......... j.

8

.l.

-......a.,..

.........-..i.

a.

...e

........l

.....j...

O.. -

l.

l

..e.

.~g l " *.'*

. ;....J.......

t.J*.

.C...

g.

4,-

g...

e.

')...

t g...

. {...

g...

e l

4...

e........

4 e,.=*

..e...

,..........n.. J.

g.

q

...e........*.

.....l.

= =.

2 8 e........

G

+N

.......o

.l r

g"

. e e

2 i

i i

..]

' 't c

[ ' ' f. ~-

f 23 -.

m O

. pq we m6 e

m.m W MM M~*

-.. w q-w-

ei-r

OL AL AU JT B

'O iO 1O O

I I

'C PC PH PH c

P "C

S OS OS OS O

OC OC O. C O

'R tR LR R

L t

4 M

y TN E

EM D

I N

x S IN 4

N M

A y

N T I

N O

C I

T N

EE DMN IS I T A U T lI s

N N

O N M

O C

A M

4 li G

A

)

I D

f WO-L F

G hL

.N

/

I

2. L VA O

E O e

n C

g i

R L

N n

V W io O

l t

RD G

v le e

S la ue N R r i T

c n IL E V

iL U

c O G p )r e g H

S O N P

to o R in S

C A M

S a t

wlo l

A B

Nl U

d e l

C r

o o C

P e p F

WX aO n

t OE N

mw r

D O

e n u o T

p o I" imd I

T E

A T

Otis lu N

t N-C n u E

E ir P 8 h

Sil I

J A

N iMS i

L

(

N N

i I

l IO l'

T Y

l M (V I

l T j A

B g i T

l 1

A i

1 r

E N

L F

U A

C I

T l

S e

R

~

v M!

la e

I C

V v)

E D

l r

a o p

R o -

Vt

[' D a

I I l ' i i iI

,l 1,'

t p

r D O S

e o p 4

MI d~

T E t

N e

SO ESDN R e

a d n t

G DEE U

e e r

e e e

t g

IN LJP v la p

a a C L O T R

l l

aV r r K T L N V

O eh L E 4

YI F

p p ET N W L

d r

n e o

Oi a A

E E UA T (R L

A e s t

- D L B E FW A

B H.

p o o

r l

i(

E E

OC M

A R

M V E E R LV N 3>4 O A AO t>

NVH YMd i

2. gc %g,

'K M 5

b 15 g

)!

.i l

[

I

' lS. -

1 t

l i

a 3

. *...o.

l

........... s e....

g

.,9...

g..

. I.....

9....

8

..l...

-i

.1 ll 4

g..

..... -.. I -

.t

(

....,,...,....3.......

..g g

g

..g..

...... -..g

..g..

t..

.J,......

e....

.a...

.5...

m. e

.....-......... '...e....

.a

.i l

4.,.

.I.

.l.

....l....

'....e

.8...

.4

.g m.

..e.

=....

....=....

.e.....

7....suii,

.T..

.....y.....

....,....g...

/

..e..

.I

. =....

.....a

.... +..

..........g.....

. m....

.....mmi. nm.

9

= = =...

.mm

....... l 4...... i

=..

...................J..

...d

...e...

'e

=

.....i

. m....p

.3 it 3..-..

l l..

.e.

.......i........

.mm.

...jg

. = =.,

t

.e..

9.

-..~.......

_...e........

.......1..................

3....

. f..

y

.......... =..

....._...i.

ma....

.g.

m..

g

+..-,

j

.=

..........6...

.........m_.

........ e.

.s.m g.

..g

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

..e

.g.-..

...y

..-................3.-.l......

.a

..g....

....g

..-.m..

w

==

.um.....

..8.*..

.....I 6

i...

a-

-.8...

..L._..,

........g-.

..-J.

.J........

.e

..J.

..~..F....._...f..

e. r..J i.

-.a.._.

.I..

g.

t.......

._9....."1.-.......

.....L.r.... f _......

..J..

._....,L.

._-..e..

...~l........... -... _... _...

_.-4-.,.

~.. _.

.... y. - --

.......l...

.a

. = -..

.3

.........._l.......

..I.

. _ _..'T...

4.

._1.....

..._f~.~.'"*....

.....T".

.6..L.._.#..*.,._...._...

.._4

.......~ t.

.._...............3.

~ -. _.

~.

....l

.7.

.c.

~._.e.

....J.....

...f....,....a.1

.8...

_..*1....

.l..

.... 1

. m.

....e...,...... _...

L.

1..

8

. ~..,.

s

.!.*. ~.._...._

~.

a....

.. 1....

e t.

n a...

~... ~... _

.4--...a L...._...

........"f"..._.3.~~.~""...'

l..

..e...

..M......_..,

-........_...f..... -l..

........J 8._.

w.

s..

........b..

........ _ _.. I I _.

_.. ~ _.. _

...-.l

...I e

.-..../,..................f.

,.,.. i.

. i

. m..

g...

.1 g

..._....,...-1_-,..

8

_..-_._....,__J,.._*.J....._

....,,.......I'.....

a...

1 L..,-...

.... - -.....i....,

....3_._.--....

. T".

..1.....

. *.....=.

. ~....

.........-"".1..1,...

.I

..1.4

......3

......T.... -

/.... _.... -....__......

e

,,8 g..

,, _..e_-

i;._

,,,L,.

i.

-........'").

-e.

7

,g

-y,=

t~,,

--I-..

..m,.....

.I..,...

~-.

... _3..

.J 9..

....... j....

_. _ _.,..""3....a._....

.3

.3,

.i.-....

3. -

._a....,_

... = -

..-......T

.~_-

.e..

-~

"~

p-

."'t.

,,. =

..1.

.',.9-..__4__

.,"3..,,.,.

g. -=

- ~-

... i.

.e

~

. I.. -

p;.

.........._.........._...3...

t.-.

-..,3.

...1.........

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

..~.g-..

. ~.,

. = - -...

..g

...........g-...........

~-

.,,*.--r

~,....

".,I_,

,I........{..._.

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

==-t.-...,.

+.

+'

..g.~.. -

..g

=

_ 3,.

........I........

,...1..

.... -.,,,...~+.. ~.~3_.

.....1..

........3....,

..1..

.s -

....... w....

~.......

J.

. :-.....,. ein..

.i......

... _..~

.L._-

..l...

..3-

.. = =.

=. -.

... =.

.~

........ i

/..

-. =....-......

=

=--.

.. _....... s.

...L............,.

e

."1.

i...

...L..

I

..-e..

'"1.._,

8 4

I s

I e

I a

l y *.

. e

...-l

......1 8.

g

..~

.l..-....

l ~. e.

w...

..,1

._t i

3.

._. g e.-

n e....

.I.

.h..

.....i e

25 er q r, g, n \\

7,

..: gl.l 0

23 g

s g

g wi

--....... ~-.

.. % +

-__>m,,..-..-w

--*...b

... ~.

_m

- e_

- L.-

umm h

.w-

k UNITED STATES OF AMERICAN NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING P0APD In the Matter of

)

METROPOLITAN EDIS0N COMPANY, ET AT.

)

Docket No. 50-289

)

(Restart)

(Three Mile Island Nuclear Station,

)

Unit 1)

)

AFFIDAVIT OF LAURENCE E. PHILLIPS STATE OF MARYLAND

)

COUNTY OF MONTG0MERY )

I, Laurence E. Phillips, being duly sworn, depose and state:

1.

I am a Section Leader of the Thermal-Hydraulics Section in the Core Performance Branch, Division of Systems Integration, Office of Nuclear Reactor Regulation, U.S. Nuclear Regulatory Comission, Vashington, DC 20555.

I have supervisory responsibility for the review of the reactor core thermal-hydraulic design and behavior including the review of functional requirements for core monitoring systems to provide capat'ility for detection and response to inadequate core cooling conditions.

2.

I have submitted a statement of Professional Qualifications in my testimony responding to UCS Contention 7, Sholly Contention (6)b and ANGRY Contention 5 (b).

3.

I have answered the ten MET-ED December 23, 1980 interrogatories and the answers given are true and correct to the best of my knowledge.

.A b

7

) y *'

'b;~c, (.

- l1. &

Subscribed and Sworn to Laurence E. Phillips

/

bef me this /d th day i

of 1981 bshk ktY _ )

3(ry PubMc

~

Not f/

MY COMMfS$loN EWPIRES JULY 1,1562 l

(

UNITED STATES OF AMERICA NUCLEAR REGULATORY COMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of

)

METROPOLITAN EDISON COMPANY, ET AL.

)

Docket No. 50-289

)

(Restart)

(Three Mile Island Nuclear Station,

)

lnit 1)

)

CERTIFICATE OF SERVICE I hereby certify that copies of "NRC STAFF'S RESPONSES TO LICENSEE"S TEN INTERROGATORIES ON REACTOR VESSEL ETER LEVEL OF DECEMBER 23,1990" in the above-captioned proceeding have been served on the following by deposit in the United States mai!, first class, or, as indicated by an asterisk, through deposit in the Nuclear Regulatory Comission's internal mail system, this 14th day of January,1931:

Ivan W. Smith, Esq.*

Mr. Steven C. Sholly Atomic Safety and Licensing Board 304 Soeth Market Street U.S. Nuclear Regulatory Commission Mechanicsburg, PA 17055 Washington, DC 20555 Mr. Thomas Gerusky Dr. Walter H. Jordan Bureau of Radiation Protection 881 W. Outer Drive Dep3rtment of Environmental Oak Ridge, TN 37830 Resources P.O. Box 2063 Dr. Linda W. Little Harrisburg, PA 17120 5000 Hermitage Drive Raleigh, NC 27512 Mr. Marvin I. Lewis 6504 Bradford Terrace George F. Trowbridge, Esq.

Philadelphia, PA 19149 Shaw, Pittman, Potts & Trowbridge 1800 M Street, N.W.

Metropolitan Edison Company Washington, DC 20006 ATTN:

J.G. Herbein, Vice President Karin W. Carter, Esq.

P.O. Box 542 505 Executive House Reading, PA 19603 P.O. Box 2357 Harrisburg, PA 17120 Ms. Jane Lee R.D. #3, Box 3521 Honorable Mark Cohen Etters, PA 17319 512 E-3 Main Capital Building Harrisburg, PA 17120 Senator Allen R. Carter, Chairman Joint Legislative Committee on Walter W. Cohen, Consumer Advocate Energy Department of Justice Post Office Box 142 Strawberry Square,14th Floor Suite 513 Senate Gressette Bldg.

Harrisburg, PA 17127 Columbia, SC 29202

Ms. Gail P. Bradford Jordan D. Cunningham, Esq.

ANGRY Fox, Farr and Cunningham 245 West Philadelphia Street 2320 North 2nd Street York, PA 17401 Harrisburg, PA 17110 John E. Minnich, Chairman Louise Bradford Dauphin Co. Board of Commissioners T'il ALERT Dauphin County Courthouse 315 Peffer Street Front and Market Streets Harrisburg, PA 17102 Harrisburg, PA 17101 Ms. Ellyn R. Weiss Robert Q. Pollard Sheldon, Harmon & Weiss 609 Montpelier Street 1725 I Street, N.W.

Baltimore, MD 21218 Suite 506 Washington, DC 20006 Chauncey Kepford Judith H. Johnsrud Thomas J. Germine Environmental Coalition on Deputy Attorney General Nuclear Power Division of Law - Room 316 433 Orlando Avenue 1100 Raymond Boulevard State College, PA 16801 Newark, NJ 07102 Ms. Frieda Berryhill, Chairman Atomic Safety and Licensing Board Coalition for Nuclear Power Plant Panel

Postponement U.S. Nuclear Regulatory Conmission 2610 Grendon Drive Washington, DC 20555 Wilmington, DE 19808 Atomic Safety and Licensing Appeal Ms. Marjorie M. Aamodt Panel (5)*

R.D. #5 U.S. Nuclear Regulatory Commission Coatesville, PA 19320 Washington, DC 20555 John Levin, Esq.

Docketing and Service Section (7)*

PA Public Utilities Commission Office of the Secretary Box 3265 U.S. Nuclear Regulatory Commission Harrisburg, PA 17120 Washington, DC 20555 N

N J sep R. Gray

/

uns 1 for NRCJtaff

.s

,

ML19340E903

Text

Navigation menu

Search