ML17309A190

Use of Lake Ontario Water in Steam Generators During Hot Shutdown.

ML17309A190
Person / Time
Site:	Ginna
Issue date:	02/28/1981
From:	Copley S, Leibovitz J, Pearl W NWT CORP.
To:
Shared Package
ML17309A169	List: ML17309A168 ML17309A190
References
NWT-167, NUDOCS 8106300308
Download: ML17309A190 (150)

Text

I 1

" h. 704K il iceà uelzv ROCHESTER GAS AND ELECTRIC CORPORATION o 89 EAST AVENUE, ROCHESTER, N.Y. 14649 JOHN E. MA IER VICE PRESIDENT June 23, 1981

'/4 TEt.EPHONE ARE* COOE 7IO

=6'>

0I<l i54g 2700 l~, j I

Director of Nuclear Reactor Regulation Attention: Mr. Dennis M. Crutchfield, Chief Operating Reactors Branch 55 U.S. Nuclear Regulatory Commission Washington, D.C. 20555

Subject:

SEP Topics V-10.B, V-11.A, V-11.B, VI-7.C.1, VII-3, and VIII-2, R.E. Ginna Nuclear Power Plant Docket No. 50-244

References:

(1) Letter from Dennis M. Crutchfield, NRC, to John E. Maier, RGE, SEP Topics, V-10.B, V-ll.B, and VII-3 (Safe Shutdown Systems Report), May 13, 1981.

(2) Letter from Dennis M. Crutchfield, NRC, to John E. Maier, RGE, SEP Topics V-11.A, V-11.B, and VI-7.C.1, dated April 24, 1981.

(3) Letter from Dennis M. Crutchfield, NRC, to John E. Maier, RGE, SEP Topics VII-3 and VIII-2, dated April 2, 1981.

Dear Mr. Crutchfield:

This letter is in response to the SEP topic assessments provided in the three above-referenced letters. Due to the intimate relationship of the "Safe Shutdown" topics V-10.B, V-ll.A, V-11.B and VII-3 addressed in these three letters, all of our comments are provided concurrently in the three attached responses.

This should aid the inclusion of our comments into the NRC's "SEP Integrated Assessment".

Very truly yours, +~a Attachments ohn E. Maier ili

RGKE responses to NRC Assessment of SEP Topics V-10.B, RHR System Reliability, V-ll.B, RHR Interlock Requirements, and VII-3, Systems Required for Safe Shutdown (Safe Shutdown Systems report), May 13, 1981.

In RG6E's January 13, 1981 response to the NRC's November 14, 1980 "Safe Shutdown Systems" assessment, a number of comments were made which have not been incorporated into Revision 2 of this assessment, transmitted by letter dated May 13, 1981.

We feel these comments were valid, and should be incorporated.

For continuity, these comments will be listed below (with their original comment numbers):

On page 5, Pi in S stem Passive Failures, the NRC assumes piping system passive failures"...beyond those normally postulated by the staff, e.g., the catastrophic failure of moderate energy systems...". Although shown that safe shutdown following such an event could it is be achieved, it is not considered that such an evaluation should even be made. As noted by the staff, clearly beyond a reasonable that this design basis. Itit is is thus recommended paragraph be deleted from the evaluation. Subsequent evaluations to this "criterion",

such as those related to the CCW system on page 22 and 23, should also be deleted.

In paragraph g on page 66, it is noted that, when applying the power diversity requirements of BTP ASB 10-1 in event of an SSE, no means to supply feed to the steam generators exists. It was determined that this was acceptable, based on low likelihood of occurrence.

This conclusion is correct; however, since BPT ASB 10-1 does not consider an SSE in conjunction with the loss of all A.C. power, there is no need to even make the evaluation. The comparisons in the SEP program should be to current criteria, rather than to arguable extrapo-lations. Reference to loss of all A.C. power in conjuncton with an SSE should thus be deleted from this paragraph.

12. On page A-4, it is noted that additional systems are required to achieve cold shutdown for a PWR than for a BWR because of a difference in the definition of cold shutdown. This does not appear to be a reasonable basis. System requirements should be based on specific safety reasons. The NRC should be consistent in its requirements for cold shutdown, or provide a technical basis for any differences."

~ f 7

Staff position 1 states that "the licensee must develop plant operating/emergency procedures for conducting a plant shutdown and cooldown using only the systems and equipment identified in Section 3.1 of the SEP Safe Shutdown Systems Report." RGsE disagrees with the need for these procedures.

We reiterate the comments provided in our January 13, 1981 response that the operator should perform a cooldown with the best equipment available to him at the time. If a piece of non-safety equipment is available, and would be the most beneficial for performing a required function, that this piece of equipment would be used. Ifititis expected is not available, the operator could fall back on the use of safety-grade equipment. But does not intend to commit plant personnel RGGE to use only safety-related equipment, is available and more effective. We feel that if non-safety equipment it would be impossible to determine when a "safety-grade-only" cooldown procedure would ever be implemented. As long as the safety-grade equipment is available (and the safe shutdown assessment concudes that it is), RGSE considers that the necessary safety requirements are met.

RGSE also notes that no regulatory basis for this requirement is provided. It is admitted in Section 4.5 of the Safe Shutdown report that "the need for procedures for these evaluations is not, identified in Regulatory Guide 1.33...".

Section 4.5 then goes on to say that the basis is found in BTP RSB 5-1 and SEP Topic VII-3. But BTP RSB 5-1 merely references RG 1.33, and this is the assessment of SEP Topic VII-3.

Therefore, since no basis for this "requirement" exists, and we do not feel that it would even be beneficial, and since the Safe Shutdown report did conclude that the capability for attaining cold shutdown using only safety-related equipment exists, RG&E concludes that this staff position should be deleted from consideration.

Staff position 3 does not appear to take into account the information provided in our March 27, 1981 submittal regarding SEP Topic V-11.A. Enclosure 3 to that submittal provides the valve equipment specification, noting that the 700, 701, 720 and 721 MOV's are designed such that they physically are unable to open against a differential pressure of greater than 500 psi. This ensures that an intersystem LOCA caused by the opening of the outboard valves, plus leakage of the inboard valves, cannot occur, since the outboard valves cannot open.

Even how without this provision, it is difficult to comprehend the Ginna arrangement could result in an "Event V". By administrative procedure, the RHR valves are key-locked closed, with power removed. Further, interlocks are provided for the inboard RHR valves. Thus, for an "Event V" to occur would require the:

1) failure of the administrative procedure requiring power lock-out (at the breaker),

2) failure of the administrative procedure governing operation of the valve at power,

3) failure of the inboard isolation valve,

4) failure of the relief valve (RV 203) which has a capacity of 70,000 lb/hr at its 600 psig setpoint, to relieve the leakage past the inboard RHR valve.

This set of failures is considered very remote. When coupled with the fact that the RHR valve design prevents opening of the valves against a greater than 500 psi differential pressure, it is RGEE's conclusion that the possibility of an intersystem LOCA should not be a credible design basis. No additional modifications, such as diverse interlocks for the outboard valves, are warranted.

Staff position 5 states that "the operating procedures for the Ginna plant should be modified to direct the operator to cooldown and depressurize to RHR initiation .parameters within 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> whenever the Service Water System is used for steam generator feedwater..." This position is based on the reference BNL-NUREG-28147, "Impure Water in Steam Generators and Isolation Condensers." We have had this report reviewed by NWT Corporation. NWT-167, "Use of Lake Ontario Water in Steam Generator During Hot Shutdown" (attached) concludes that, "although not recommended from the standpoint of maximizing component life, and operation for periods up to several days is not expected to result in any significant cracking or in deterioration of steam generator integrity."

RG&E therefore concludes that a specific directive to cool down and depressurize to RHR initiation conditions is not warranted, and should not be included in a procedure. The capability to do this does exist, however, and could be used if determined to be necessary at the time.

l

\

J" v

1 v ~

Jl

RG&E responses to NRC letter of April 24, 1981 regarding SEP Topics V-11.A, "Isolation of High and Low Pressure Systems", V-11.B, "RHR Interlock Requirements", and VI-7.C.1, "Independence of Redundant Onsite Power Systems".

1. The Safety Evaluation for SEP Topic V-11.A, "Requirements for Isolation of High and Low Pressure Systems", specifies that the outboard RHR valves should have diverse interlocks to prevent opening .when the RCS pressure is greater than RHR system design pressure.

RGGE rationale for not providing these additional interlocks is provided in comment 3 of Attachment 1 of this transmittal.

2. The safety evaluation also required that interlocks be installed on the CVCS suction valves (200A, 200B, 202), to prevent a possible overpressurization of the CVCS letdown line outside containment. RGEE has noted in our March 27, 1981 letter on this SEP Topic that a relief valve (RV 203), with a capacity greater than the combined capacity of the three orifices, would relieve the pressure buildup caused by closure of the contain'ment isolation valve'71. No overpressurization of the CVCS would thus be expected.

RGGE has also evaluated the potential consequences of such an overpressurization event, with a subsequent small LOCA outside containment, and determined that no unacceptable consequences would result. This break would be a small LOCA outside containment (maximum flow of 140 gpm), and would be terminated by closure of valves 200A, 200B, and 202 either by operator action or automatically by low pressurizer level. Radiological consequences would be minimal, since no fuel damage would result. This event is specifically evalu-ated by SEP Topic XV-16, "Radiological Consequences of Failure of Small Lines Carrying Primary Coolant Outside Containment." RG&E has provided information concerning this topic by letter dated June 18, 1980 from L. D. White Jr. to Mr. Dennis M. Crutchfield.

The RG&E conclusion is that, based on the availability of RV 203 to prevent overpressurization, together with the lack of unacceptable consequences due to an overpressurization, no interlocks or other modifications are required for the CVCS suction valves.

3. The safety evaluation further states that position indication is required on the CVCS discharge check valves. As stated in our March 27, 1981 letter on SEP Topic V-11.A, we do not believe that this line should be classified as a low pressure

I system connected to the RCS, since the piping is 2500-1b piping throughout its length (to the positive displacement charging pump). RG&E has had no experience with'ailures of the positive displacement charging pump pistons to hold primary system pressure, nor would any failures be anticipated.

Our contention that the charging line is not a line of concern is borne out by a memo from Edson G. Case to Raymond F. Fraley, "Isolation of Low Pressure Systems from Reactor Coolant System", dated July 11, 1977. That letter transmitted an NRC study of this subject to the ACRS, and evaluated all potential lines of concern. The charging line was not included.

To verify that the charging line was not a valid "Event V" concern, RG&E calculated the PWR Check Valve Event Tree (Section 4.4 of NASH-1400), using the charging line con-figuration (two in-series check valves and a charging pump piston). Very conservatively assuming that both check valves were undetected open, and that the probability of the charging pump~iston failure was equal to a check valve failure, the Q calculated for this configuration was determined to SeUM1.4 x 10 /year. This is a low enough value to obviously be of no concern.

RG&E therefore considers that check valve position indi-cation is not needed on the chargingline check valves.

With respect to the SEP Topic Assessment V-11.B, no comments are necessary, since the resolution of outstanding issues is addressed in the topic assessment for SEP Topic V-11.A.

The additional information requested for SEP Topic VI-7.C.1 is presently being developed. It is anticipated that this information can be furnished to the NRC by July 15, 1981.

l L

J

RGEE responses to NRC letter of April 2, 1981, concerning SEP Topics VII-', "Electrical, Instru-mentation, and Control Feature of Systems Required for Safe Shutdown", and VIII-2, "Diesel Generators".

It appears that all comments provided by RG&E in our January 1981 and January 30, 1981 letters concerning these topics 23, have been properly incorporated.

Based on the resolution of all open items, and the removal of diesel generator testing from SEP Topic VIII-2, RGEE concludes that both of these topics are complete, with no outstanding issues to be carried into the Integrated Assessment.

0 I

H i

tL N

IP I

NWT 167 February 1981 USE OF LAKE ONTARIO WATER IN STEAM GENERATORS DURING HOT SHUTDOWN W. L. Pearl S. E. Copley J. Leibovitz Prepared for Rochester Gas & Electric Company Corporation 7015 REALM DRIVE, SAN JOSE, CALIFORNIA 95119

'I 0

I 1

h

'!'

~, 1

'

zf This document was prepared for the Rochester Gas & Electric Company.

Neither the NWT Corporation nor any person acting on its behalf assumes any responsibility for liability or damage which may result from the use of any information disclosed in this document. )

)

INTRODUCTION The possibility of using Lake Ontario water as an emergency PWR feedwater supply for more than 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> during which the plant would be brought to cold shutdown is being considered. The maximum steaming rate during such a period would be 100,000 pounds/h (200 gpm) at a temperature of 350'F. As a'consequence of steaming, impurities of the untreated Lake Ontario water will concentrate in the steam generator. Of major concern is the possible risk of stress corrosion cracking (SCC) of steam generator materials in contact with the concentrated solution thus formed. To address this concern, the chemistry variation in the liquid phase as steaming proceeds at 350'F was estimated with emphasis on pH.

Then, the possible potential for SCC was assessed on the basis of these estimates and available SCC data.

pH VARIATION AT 350'F 'UPON STEAMING LAKE ONTARIO WATER A. Computer Model ing The composition of Lake Ontario water as determined by RGE is given in Table l.

TABLE 1 LAKE ONTARIO WATER ANALYSIS ppm Cal cium "35 Nitrate 2.5 Magnesium 8 Phosphate 0.3 Sodium 13 Fluoride 0.15 Potassium 3.6 Silica (as Si02) 0.25 Aluminum 0.1 Dissolved Oxygen 9.5 Chloride 32 Ammonia (as Nitrogen) 0.24 Sulfate 35 Estimates of the water chemistry variation upon steaming were developed using the following assumptions:

1. Since aluminum and silica are in s'toichiometric proportion in Lake Ontario water (Table 1), they are assumed to precipitate as aluminum silicate (clay) upon concentrating 'and therefore are removed from solution.

2. Since calcium occurs in the water (Table 1) in large excess over phosphate, it is assumed to precipitate all the phosphate as calcium hydroxy apatite (Ca5(P04)30H) and remove it from the solution. The calcium in solution is decreased by the corresponding amount.

3. Fluoride and nitrite are assumed to behave as chloride. Potassium is assumed to behave as sodium.

4 ~ Sodium and chloride in solution are assumed to remain completely dissociated.

5. Calcium carbonate precipitation is neglected. Degasification of CO~

by steaming is assumed to occur.

r ~

l~

6. The concentration of sodium and calcium chlorides is assumed limited by a solubility of 5 molal.

7. Chemical equilibrium expressions of references 2 and 3 apply.

On this basis, the liquid solution pH variation upon steaming at 350'F was estimated as a function of concentration factor defined as the mass ratio of total water (steam + liquid) to liquid water residual. The results are presented graphically in Figure 1. It is important to note that the definition of pH used here is that followed by Mesmer4 in the determination of the dissociation constant of water at high temperatures, viz, the negative of the logarithm of the hydrogen ion concentration (not of its activity). Similarly, neutral pH is defined as that where the hydrogen and hydroxyl ion concentrations are e'qual.

This neutral pH is a function of ionic strength. Therefore, the pH variation of the concentrated solutions must be considered in relation to that of neutral pH, also plotted in Figure 1. For basic solutions as is the case considered here, it is important to bear in mind that the hydroxyl ion concentration is expressed in terms of pH as'follows:

10pH-2NpH OH (where NpH is the neutral pH v'alue) and that when the neutral pH varies together with the ionic strength as the liquid solution is being concentrated upon steaming, the basicity of the solution may not be appreciated from the solution pH alone. The equivalent NaOH concentration is more suitable for this purpose and is plotted also in Figure 1.

B. Discussion Steam Generator Bulk Water Based on-'a maximum feed rate of 200 gpm to the steam generator and a total steam generator liquid volume of approximately 12,000 gallons, a maximum of one steam generator volume is steamed away each hour. Therefore, under maximum steaming conditions, the concentration factor achieved in the bulk steam generator water is t + 1 where t is the number of hours of steaming.

10 Solution pH

.400 O

300 o Equivalent NaOH Concentration Neutral pH 200 8

lot 0

10 100 1000 10000 Time, h Figure l. Variation of Steam Generator pH with Steaming at 350'F (feeding Lake Ontario water at 200 gpm)

l-

!

t I,

The variation with time of the equivalent sodium hydroxide concentration in the steam generator with steaming of emergency Lake Ontario feedwater then can be followed on Figure l. It is seen that a maximum equivalent NaOH concentration 'f about 300 ppm will be reached in the steam generator bulk water when 15 to 20 steam generator volumes will have been converted to, steam, i.e., in approximately twenty hours. Further boiling should then decrease the equivalent NaOH concen-tration as magnesium and/or calcium hydroxides and/or calcium sulfate precipitate with increased concentrating. The decrease reaches a limit (at about 20,000 steam generator volumes converted to steam, i.e., in 20,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />) when sodium and calcium chlorides start to precipitate also. This limit is estimated at about 100 ppm equivalent NaOH for Lake Ontario water composition as specified in Table 1 and with the assumptions already stated. The assumptions seem reasonable and, at any rate, can be tested experimentally with a small autoclave from which known amounts of Lake Ontario water would be boiled away at 350'F at constant liquid level in the autoclave.

Crevices The estimated equivalent NaOH solution concentration in steam generator crevices will depend upon the relative degree of crevice solution concentration above the bulk water. In tube to tube support plate crevices, there may be a distribution of relative concentration factors of unity and hi gher.

The chemistry in a crevice would lead that of the bulk in the sense that the chemistry of a specific crevice would travel the same curve (Figure 1) as the bulk but would be at a point on the curve somewhat ahead of the bulk. Since the causticity of Lake Ontario water is not a strong function of concentration, this does not pose a problem.'ndeed it is expected in this case that after a short period of steaming, the crevice chemistry will be less basic than that of the bulk.

Coolin Water Com osition The NWT chemistry modeling work discussed herein is based on the chemical composition of Lake Ontario water summarized in Table 1 as supplied by RGE.

~ ~ '

It is possible that seasonal changes in -the characteristics'f the lake water may,result from the interrelation between source river flowrates, industrial pollution and/or acid rain. NWT has no relevant data to assess such effects.

It may be desirable that analyses made of Lake Ontario water during different seasons and under various conditions be fed into the NWT chemistry model. In this manner the safety of feeding Lake Ontario water, over the range of likely chemical compositions, can be verified.

POTENTIAL FOR SCC A. Corrosion The most aggressive solution expected based on the modeling work is 300 ppm NaOH, with sl0 ppm 02 (see below) at 350'F. Although laboratory data regarding these exact conditions are not available, data are available which can be extrapolated to assess the maximum corrosion rates expected for a given range of conditions.

van Rooyen and Kendig'ite Westinghouse data indicating that U-bends of Alloy 600 in deaerated lOX NaOH crack after several months of exposure.

Figures 2 and 3 summarize Westinghouse tests'hich show that at least 100 days of exposure to deaerated 10% NaOH at 600'F is required to produce a detectable crack in stressed Alloy 600.

Figure 4 shows data gathered by Berge and Donati.~ These curves are for yield stressed C rings at 660'F. Extrapolating the curve for mill annealed Alloy 600 to 300 ppm NaOH yields a minimum time of 3500-4000 hours to induce a 0.5 millimeter crack.

The data presented above are for deaerated systems and are consistent with van Rooyen's conclusion that Alloy 600 in 105 NaOH would not crack for several months. In the presence of oxygen, the susceptibility of Alloy 600 to SCC may be increased. Figure 5 shows stress corrosion behavior in 600'F high purity water containing varying amounts of oxygen in the gas phase above the water and adjusted to pH 10 at startup with ammonia. As the oxygen content of the gas phase increased, the percent of the specimens attacked and extent of the attack increased. As noted in Figure 5 the average life in the 18-week test varied from no cracking with 1X oxygen in the gasphase (<2 ppm oxygen in the water) to 7 weeks with 100Ã oxygen in the gas phase (<200 ppm oxygen in the water).

<<Ilree and Hichels and later Sedriks, et al.," reported less than 20K cracking after 27 days for Alloy 600 (2 common heat treatments) in aerated, 50/. NaOH at 570'F.

~ ~

( ~

wats Cress Stress N ttt There((alt/

I sss(s/es0 Ahneetd Treats(s ZT5/al tta 50//50 ss tla/IO 4 g .C00 V0/V.4 ~ e ea st t

~ tsta NXI t(e(ees(re Tle( ~ I hest Figure 2. Crack Depth as a Function of Time, Stress Level and Material Condition for ID Pressurized Capsules Exposed to Deaerated 10$ NaOH at 600'F, Heat 5C M.A. T.T.

ROT .011 O.e 8901 .OV ~

SSM .03t ~

0.3 T3Q .04$

.07 ss 105 HaOH 5 04 310'C 6

Cwlrs05 1-U05 MIH Antsealtd yte505 0.3 8'01

~ J ea 0.2 0.1 ~ I&

~

e

~ Therthally Treated t000 Dlssstsre Tits(e 1 hrsl Figure 3. Crack Depth as a Function of Exposure Time for Mill Annealed and Thermally Treated Inconel Alloy 600 Exposed to Deaerated 10K NaOH at 600'F'

,4000 C Rings Stressed to a=Ys According to ASTM STP 425

~o 3000 0V S S

~

a)

E I

r lA Alloy 600 HT 16h 1300 F o 2000 E

r 0)

I I K0 1000 Al 1 oy 600 MA 0

1,000, 10,000 '00,000 NaOH Concentration, ppm Figure 4. Resistance to Stress Corrosion Cracking of Alloy 600 Mill-Annealed-or Heat Treated at 1300 F as a Function of Deaerated Sodium Hydroxide Concentration at 600'F'

l

~ ~

X NO.OF SPECIMENS, MILL ANN. 8 l2 l2 l2 t2 W PERCENT CRACKED 0 0 0 58 92 IOO W

AV UFE IN 18 WEEK TESTS >8 I8 ie l7 l5 7 I20 AV,OF MAX. CRACK DEPTHS,MIL 0 0 0 2I 50 52 CI t o~ IOO O

Z o 80

~C PI: 60 16 12 8 8 8 8

W~

cL> 40 18 18

.O I 12

ÃQ W W

~~ 2O 18 18 il 18 I8 18 18 18 18 18 1818181818 18 IR ALLOY 600 HEAT N 27 85 4 2 478256 4 78256 478 256 478256 OXYGEN IN GAS NIL . NIL 2I 100 AT START, (%)

BALANCE OF GAS, HYDROGEN ARGON'ITROGEN NITROGEN AIR Figure 5. Stress Corrosion Behavior in Crevice Areas in Hill Annealed Inconel 600 Double U-bend Specimens in 600'F High Purity Water Adjusted to pH 10 with Ammonia at Star tup' 10

e 1

Laboratory studies show that there is a significant temperature dependence of caustic stress corrosion cracking as illustrated in Figures 6 and 7.

These results are. for pressurized capsules exposed to 10K and 50% NaOH at varying stresses at temperatures ranging from 650 to 550'F. As can be seen, reducing the temperature below,600'F .significantly extends the time for SCC to occur. This temperature dependence is further illustrated in Figure 8 where temperature is plotted versus rate constant for both 105 and 505 NaOH.

8. Oxygen The lake water fed to the generators probably would be air saturated (approximately 10 ppm 02). However, at 350'F the Kp (the equilibrium ratio between steam phase and liquid phase) for oxygen is slightly greater than 5000. Even though the dynamic distribution in practice may not reach true equilibrium conditions, the net effect of the high KD value is that recriculated steam generator coolant will contain oxygen concentrations lower than 10 ppm. This recirculated coolant will dilute the oxygen concentration of incoming feedwater with a net oxygen level in the downcomer of ~1 to 10 ppm, depending on the recirculation ratio under the contingency conditions.

C. Conclusion With the significantly lower concentrations of sodium hydroxide (max 300 ppm),

oxygen concentration <10 ppm and the lower temperature (350'F) involved, the contingency of feeding Lake Ontario water to the Ginna steam generators shouId result in no measureable damage to steam generator internals. A'jthough not recommended from the standpoint of maximizing component life, such operatio~

for periods up to several days is not expected to resu1t in any significant cracking or in a deterioration of steam generator integrity.

~ ~

NaOH 50% 10% 575 F II 0 .5 to 5 mils 50 30 A

0)

>10 mils N

Ql L) flJ 5 to 10 mils 0.5 to 5 10 mils 0

0 1000 2000 3000 4000 5000 6000 7000 Exposure Time, hrs Figure 6. Caustic Cracking of Mill Annealed Alloy 600 at 575'F (Lines depict zones of crack depth from 105 NaOH at 600'F) 12

NaOH 50% 10% 550 F 7 0 <0.5 mils Cl 0 5 to 5 mils g 0.5 to 5 mils 50% NaOH 650'F 50 40

~

~ 30 A

~I 10 mils 07 20 5 to 10 mils 10 0.5 to 5 mils 0

0 1000 2000 3000 4000 5000 6000 Exposure Time, hrs Figure 7. Caustic Cracking of Mi11 Annealed A11oy 600 at 5 and 650'F )Lines depict zones of crack depth <<om NaOH at 600'F) 13

g' s

"a

'll

200 MA I-600 150 f5 I I V O IJ 4J lQ N C c5 D

a%4 0

100 50 0

530 550 570 590 610 630 Temperature, F FigUre 8. Indicated Variation in Rate of SCC with Temperat 14

'a ."S~~

'

'PP<~p~~"':~ " .

COMMENTS ON VAN ROOyEN ANP KE!i01G S REPORT'he referenced repo'res basica)]y i5 a broad summary covering a large volume of data applicable in part to stainless steels and in part to Alloy 600. We

'gg" 4"~

~

are generally in agreement with heir nine summary conclusions, but find

"~ is@'"

it to apply'their broad-brush treatment to the specifics of a PWR hot 8,>>'ifficult

~4'~ g p-,'--

shutdown with lake'water'dded to the steam generators 5t 350'F. Their document is misleading for s'uch an application in two respects:

'w C>" ..;~

0',$ 4 t fkkcf

1. Caustic Concentration Their statement that ..."For the nurposes of SCC predictions, it has to be assumed that the time to form dangerous levels of NaOH, once impurities have been introduced, is short, i.e., one day or less" does not fully

,(,- -

s:i recognize the specific concentration chemistry of the cooling water involved nor the low heat flux available and the cutback in steaming rate a period of hot shutdown. \'uring

!n the case of the Lake Ontario water, for example, the maximum NaOH concentration reached is 300 ppm (after steaming ~20 steam generator vo>umes) with a decrease in concentration thereafter. I, ra S',

~ tl

2. Temperature All of the test work referenced in the referenced reports was performed in the temperature range of 550 to 630'F. With the significant temperature dependence of caustic SCC as shown above, the concern at 350 F is many tires less than is indicated from the data quoted by the authors.'ased on the above three considerations. it is our assessment that the generalized time limit of 36 ho~rs in :he report~ is not directly applicable to the Ginna steam generators steam'ng " 350'F while fed by Lake Ontario water.

15

~ 1

~ V 1

~v

.REFERENCES

1. Harhay,

.1981.

A., Rochester Gas 8 Electric, Personal Communication, February r

ll,

2. Leibovitz, J., and Sawochka, S. G., "Modeling the Effects of Condenser Inleakage on PWR Chemistry", presented at 41st Annual International Water Conference, Pittsburgh, Pennsylvania, October 1980.

3. Leibovi;tz, J. and Sawochka, S. G., "Modeling of Cooling Water Inleakage Effects in PWR Steam Generators, Topical Report, Research Project 404-1",

Electric Power Research Institute, May 1980, to be published.

4. Mesmer, R. E., Baes, C. F., Jr., and Sweeton, F. M., "Boric Acid Equilibria and pH in PWR Coolants", Proceedings 'of the 32nd International Water Con-ference, Pittsburgh, P'ennsylvania, November 1971, pp. 55-65.

van Rooyen, D., and Kendig, M. W., "Impure Water in Steam Generators and Isolation Generators", Brookhaven National Laboratory, June 1980 (Draft-BNL-NUREG-2814?).

6. Airey, G. P., "Effect of Processing Variables on the Caustic Stress Corrosion Resistance of Inconel Alloy 600", presented at NACE Meetingi March 1979 (Paper Number 101).

7. Berge, Ph. and Donati, J. R., "Materials Requirements for Steam Generator Tubing", presented at International Conference on Materials Performance in Nuclear Steam Generators, St. Petersburg, Florida, October 1980.

8. Copson, H. R. and Economy, G., "Effect of Some Environmental Conditions on Stress Corrosion Behavior of Ni-Cr-Fe Alloys in Pressurized Wate~".

Corrosion, 24, No. 3, pp. 55-65 (March 1968).

9. McIlree, A. R. and Michels, H. T., "Stress Corrosion Behavior of Fe-<<->>

and Other Alloys in High Temperature Caustic Solutions", Corrosion. 33.

No. 2, pp. 60-67 (February 1977).

10. Sedriks, A. J., et al., "Inconel Alloy 690-A New Corrosion Resistant Material", -Corrosion Engineering (Japan), 28, No. 2, pp. 82-95 (1979).

Burstein, S., WEPCO, ltr to H. R. Denton, NRC, dtd November 23 with attachments.

16

st<

'0 Vj ~

f,'

J I

,r 4

HWT 167 February 1981 USE OF LAKE ONTARIO WATER IH STEAN GENERATORS OURIHG HOT SHUTDOWN W. L. Pearl S. E. Copley J. Leibovitz Prepared For'ochester Gas & Electric Company Co%oration 7015 REALM ORIVE, SAN JOSE, CALIFORNIA 95119

This document was prepared for the Rochester Gas & Electric Company.

Neither the %/T Corporation nor any person acting on its behalf assumes any responsibility for liability or damage which may result from the use of any information disclosed in this document.

INTRODUCTION The possibility of using Lake Ontario water as an emergency PWR feedwater supply for more than 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> during which the plant would be brought to cold shutdown is being considered. The maximum steaming rate during such a period would be 100,000 pounds/h (200 gpm) at a temperature of 350'F. As a consequence of steaming, impurities of the untreated Lake Ontario water will concentrate in the steam generator. Of major concern is the possible risk of stress corrosion cracking (SCC) of steam generator materials in contact with the concentrated solution thus formed. To address this concern, the chemistry variation in the liquid phase as steaming proceeds at 350'F was estimated with emphasis on pH.

Then, the possible potential for SCC was assessed on the basis of these estimates and available SCC data.

6. The concentration of sodium and calcium chlorides is assumed limited by a solubility of 5 molal.

7. Chemical equilibrium expressions of references 2 and 3 apply.

On this basis, the liquid solution pH variation upon steaming at 350'F was estimated as a function of concentration factor defined as the mass ratio of total water (steam + liquid) to liquid water residual. The results are presented graphically in Figure l. It is important to note that the definition of pH used here is that followed by Mesmer" in the determination of the dissociation constant of water at high temperatures, viz, the negative of the logarithm of the hydrogen ion concentration (not of its activity). Similarly, neutral pH is defined as that where the hydrogen and hydroxyl ion concentrations are equal.

This neutral pH is a function of ionic strength. Therefore, the pH variation of the concentrated solutions must be considered in relation to that of neut. al pH, also plotted in Figure 1. For basic solutions as is the case considered nere, it is important to bear in mind that the hydroxyl ion concentration is expressed in terms o pH as follows:

10pH-2NpH OH (where NpH is the neutral pH value) and that when the neutral pH varies together with the ionic strength as the liquid solution is being concentrated upon steaming, the basicity of the solution may not be appreciated from the solution pH alone The equivalent NaOH concentration is. mor e suitable for this purpose and is plotted also in Figure 1'.

B. Discussion Steam Generator Bulk Water Based. on a maximum f'eed rate of'00 gpm to the steam generator and a total s.earn generator liquid volume of. approximately 12,000 gallons, a maximum of one steam generator volume is steamed away each hour. Therefore, under maximum steaming

<<nditions, the concentration factor achieved. in the bulk steam .generator water.

,

is t + 1 ~he~e t is the number of hours of steaming.

The variation with time of the equivalent sodium hydroxide concentration in the steam generator with steaming of emergency Lake Ontario feedwater then can be followed on Figure 1. It is seen that amaximum equivalent NaOH concentration of about 300 ppm will be reached in the steam generator bulk water when 15 to 20 steam generator volumes will have been converted to steam, i.e., in approximately twenty hours. Further boiling should then decrease the equivalent NaOH concen-tration as magnesium and/or calcium hydroxides and/or calcium sulfate precipitate with increased concentrating. -The decrease reaches a limit (at about 20,000 steam generator volumes converted to steam, i.e., in 20,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />) when sodium and calcium chlorides start to precipitate also. This limit is estimated at about 100 ppm equivalent NaOH for Lake Ontario water composition as specified in Table I and with the assumptions already stated. The assumptions seem reasonable and, at any rate, can be tested experimentally with a small autoclave from which known amounts of Lake Ontario water would be boiled away at 350'F at constant liquid level in the autoclave.

Crev',ces The es imated equivalent NaOH solution concentration in steam generator crevices wi 11 depend upon the relative degree of crevice solution concentration above the bulk water. In tube to tube support plate crevices, there may be a distribution of relative concentration factors of unity and higher.

The chemistry in a crevice would lead that of the bulk in the sense that the chemistry of a specific crevice would travel the same curve (Figure I) as the bulk but wouId be: at. a- point on the; curve. somewhat ahead of'he bulk. Since

'E the causticity of Lake Ontario water is not a. strong function of concentration, this does not pose a problem. Indeed it is expected in this case that after a short period of steaming, the crevice chemistry will be less basic than that of the bulk.

Coolin Mater Comaosition The NWT chemistry modeling work discussed herein is based on the chemical composition of Lake Ontario water summarized. in Table I as supplied by RGE.

POTENTIAL FOR SCC A. Corrosion The most aggressive solution expected based on the modeling work is 300 ppm NaOH, with s'10 ppm 02 (see below) at 350'F. Although laboratory data regarding these exact conditions are not available, data are available which can be extrapolated to assess the maximum corrosion rates expected for a given range of conditions.

van Rooyen and Kendig'ite Westinghouse data indicating that U-bends of Alloy 600 in deaerated 10~ NaOH crack after several months of exposure.

Figures 2 and 3 summarize Mestinghouse tests'hich show that at least 100 days of exposure to deaerated 10~ NaOH at 600'F is required to produce a detectable crack in stressed Alloy 600.

igure 4 shows data gathered by 8erge and Donati.'hese curves are for yield stressed C rings at 660'F. Extrapolating the curve for mill annealed Alloy 600 to 300 ppm NaOH yields a minimum time of 3500-4000 hours to induce a 0.5 millimeter crack.

The data presented above are for deaerated systems and are consistent with van Rooyen's 'onclusion that Alloy 600 in 10% NaOH would not crack for several months. In the presence of oxygen, the susceptibility of Alloy 600 to SCC may be increased. Figure 5 shows stress corrosion behavior in 600'F high purity water containing varying-amounts of'.oxygen in the- gas phase above the water and adjusted to pH 10't startup with ammonia.'s the oxygen content of the gas phase increased, the percent of the specimens attacked and extent of the attack increased. As noted in Figure 5 the average life in the 18-week test varied from no cracking with 1~ oxygen in the gas phase (<2 ppm oxygen in the water) to 7 weeks, with 100~ oxygen in the gas phase (<200 ppm oxygen in the water).

~cIlree and Nichels and later Sedriks, et al., 'eported less than 20~

~~~cking after 27 days for Alloy 600 (2 common heat treatments) in aerated, 50K NaOH at 570 F.

7

40QO C Rings Stressed to cr=Ys According to ASTM STP 425 3000 O 4P CJ A11oy 600

~ ~ WQ HT 16h 13QO'F

>000 1000 A11oy 600 MA 1,000 10.,000 100,000 NaOH Concentration, ppm Figure 4. Resistance to Stress Corrosion Cracking of A11oy 60Q Mi 11-Annea1ed or Heat Treated at 1300'F as a Function of Oeaerated Sodium Hydroxide Concentration at 600'F~

9

~~

Laboratory studies show that there is a significant temperature dependence of caustic stress corrosion cracking as illustrated in Figures 6 and 7.

These resul ts are for pressuri zed capsul es exposed to 10>> and 50'~ NaOH at, varying stresses at temperatures ranging from 650 to 550'F. As can be seen, reducing the temperature below 600'F significantly extends the time for SCC to occur. This temperature dependence is further illustrated in Figure 8 where temperature is plotted versus rate constant for both 10% and 50~ HaOH.

B. Oxygen The lake water fed to the generators probably would be air saturated (approximately

.'0 ppm 02). However-, at 350'F the K0 (the equilibrium ratio between steam phase and liquid phase) for oxygen is slightly greater than 5000. Even though the dynamic distribUtion in practice may not reach true equilibrium conditions, the net effect of the high K0 value is that recriculated st am generator coolant

~ ~ill contain oxygen concentrations lower than 10 ppm. This recirculated coolan=

'~>>l'ilute the oxygen concentration of incoming feedwater with a net oxygen

evel in the downcomer of ~l to 10 ppm, depending on the recirculation ratio under the contingency conditions.

CD Conclusion the significantly lower concentrations of sodium hydroxide (max 300 ppm),

oxygen concentration s10 ppm and the lower temperature (350'F) involved, the contingency of feeding Lake Ontario water.to the Ginna steam generators should.

result in no measureable damage to steam- generator internaTs. Although:

from the standpoint of maximizing component life, such operation not'ecommended for periods up to several days is not expected to resu1t in any significant cracking or in a deterioration of steam generator integrity.

11

NaOH 50Z 10X 550'P Q <0.5 mils 0.5 to 5 mils g 0.5 to 5 mils 50X NaOH 650'F 50 N II 40 30 10 mils 0 5 ro 10 mils 10 0.5 to 5 mils 0

0 1000 2000 300Q 40QO 5000 Exposure Time, hrs Figure 7'. Caustic Cracking of Mill Annealed-Alloy 600 at 550'F 55 and 650'F (Lines depict zones of crack depth <<om NaOH at 600'F) 13

~,smA~

'4%44+Av w COMMENTS ON YAH ROUEN ANP yE!;g..-".'5 ."-.EPORT

,;v~p+W<.-: " '-

The referenced reports bas<ca]]v :-... broad summary covering a large volume of data applicable ')n part to ...z::.i .ss steels and in part to Alloy 600. Me are generally in agreement wi:h :".eir nine summary conclusions, but find

~I,&g it difficult to apply'the)r broad-br's~ treatment to the specifics of a PMR hot shutdown with lake water added

o '.'.".= s earn generators at 350'F. Their document

, Ifg m'$

is misleading for such an applicat on in two respects:

tt<<. ~

~ I

1. Caustic Concentration Their statement that ..."For ..-. -;rposes of SCC predictions, it has to be assumed that the time to c; "..:vgerous levels of NaOH, once impurities have been introduced, is shor.. :.=., one day or less" does not fully recognize the specific conce"'.r'= .:;; chemistry of the cooling water involved nor the low heat .'! x ;:;a;'able and the cutback in s.earning ra .e during a period of hot shu.co<< .':: -.he case of .he Lake Ontario water,

=or example, the maxiaam .'4C:":; r':,",-ration reached s 300 ppm (after szeam', ng ~20 s.earn genera:or .: .. -

'.; "'. th a decrease in concentra ti on thereafter.

2. Temperature All of the test work referenced '"";he referenced report was performed in ficant tempel dependence of caustic SCC as shown .above, the concern at 350'F is many ti" es less than is indicated. from <~<<'a quoted'y the authors.

Based on the above three conside<< -ions. it is our assessment, that the generalized time limit of 36 ho""s '" = e report's not directly app>>c to the Ginna steam generators stea "'n"= "': 350'F while fed by Lake Ontario water.

15

Cy (4~p,R Rangy 0 UNITEO STATES NUCL'EAR REGULATORY COMMISSION WASHINGTON, O. C. 20555

/c &~

/y April 24, 1981

+w*w~

'ocket,'No. 50%44 LS05 04-035 Mr. John E, Maier Vice, President Electric and Steam Production Rochester Gas & Electric Corporation 89 East Avenue Rochester, New York 14649 Dear Mr. Maier;

SUBJECT:

SEP TOPICS V-ll.A, ISOLATION OF HIGH AND LOW PRESSURE SYSTEMS, V-ll,B, RHR INTERLOCK REQUIREMENTS AND VI-7.C.l, INDEPENDENCE OF REDUNDANT Q5SITE POWER SYSTEMS -. R, E, GINNA NUCLEAR POWER PLANT We have reviewed your letter of March 27, 1981 and agree with resolving open items during topic evaluations rather than deferring a decision to the Integrated Assessment, To this end, we are enclosing a revised safety evaluation of Topic V-ll.A.

We have also reviewed your comments on the draft Technical Evaluation Report (TER) SEP Topic V-ll.B dated January 8, 1981, Your comments on Topic V-'1.B are covered by Sections 3.1 and 3.2 of our safety evalua-

'EP tion on SEP Topic V-ll,A. We are enclosing a revised Technical Evaluation Report on Topic V-ll.B which incorporates a reference to Section 3,1 and 3.2 of our safety evaluation report on Topic V-ll.A.

We are enclosing a request for additional information on SEP Topic VI-7,C,l where we do not have sufficient information to reach an independent safety assessment.

Sincerely,'

Dennis M, Crutchfield, ief Operating Reactors Branch No, 5 Division of Licensing

Enclosure:

Wa for V-ll.A SER guestjons f'rTopic SEP SEP Topic VI-.7.C,1 cc w/enclosure'ee next page

SAFETY EVALUATION TOPIC: V-ll.A Requirements for Isolation of High and Low Pressure Systems Several systems that have a relatively low design pressure are connected to the reactor coolant pressure boundary', The valves that form the interface between the high and low pressure systems must have sufficient redundancy and interlocks to assure that the low pressure systems are not subjected to coolant pressures that exceed design limits. The problem is complicated since under certain operating modes (e.g,, shutdown cooling and ECCS injection) these valves must open to assure adequate reactor safety, As noted in EG&G Report 1285 (Appendix A), Ginna has three systems with a lower design pressure rating than the RCS, that are directly connected to the RCS.

The RHR, SIS, and CVCS system do not meet current licensing requirements for isolation of high and low pressure systems as specified below.

(1) The RHR system is not in compliance with the current licensing require-ments of BTP RSB 5-1 since none of the isolation valves will automatically close if RCS pressure exceeds RHR design pressure. Also, the outboard isolation valves have no interlocks to prevent RHR overpressurization, and the inboard valve interlocks are neither diverse nor independent, (2) The SIS is not in compliance with the current licensing requirements of SRP 6.3 since the MOVs in the low pressure injection lines have no interlocks to prevent opening where the RCS pressure and the single check v'alve in each line is not tested.

(3) The CVCS is not in compliance with current licensing requirements for isolation of high and low pressure systems contained in BTP EICSB-3 since the suction and discharge line solenoid-operated valves have no interlocks to prevent system overpressurization, and the discharge line check valves have no position indication available in the control room.

Because of the severe consequences of a LOCA outside of containment and the lack of assurance that these isolation valves could be closed against signifi-cant flow under the resulting environmental conditions, the RHR isolation valves and the CVCS suction valves should be modified to satisfy the functional require-ments of BTP RSB 5-1 and BTP EICSB-3. The modifications of this equipment should meet current criteria for seismic and environmental qualification. The schedule for installing these modifications will be determined during the integrated assessment portion of our review.

The basis for requiring diverse interlocks in the RHR outboard isolation valves is that, if an operator opens the outboard valve and the inboard valve leaks, an uncontrolled LOCA outside of containment (Event V) could possibly occur.

.. The basis for not requiring interlocks in. the low pressure injection system is that, since the contractor's report was published, a check valve test program has been established, and thus, the system now satisfies the single .

failure criterion.

The basis for requiring the interlocks on the suction and position indica tion on the discharge check valves in the CVCS system is that a failure of the relief valve to function when required .may'ead to overpressurization and a subsequent event V.

The basis for not requiring interlocks on the CVCS discharge valves is that the check valves in series with the positive displacement pumps satisfy the single failure criterion as long as check valve position is known and the pump capacity is verified periodically.

noted, in our systems it previously safety evaluation of SEP Topic V-lO.B, As is not necessary to close-the RHR valves automatically on increasing reactor coolant system pressure during startup because of the overpressurization pro~

tection system. (See also Topic V-3.)

. hcGIj

( C~

C eW UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D. C. 20555 r~r "

a G21 la~i

+s*e+

Docket No. 50-244 LS05-81-02-060 Mr. John E. Maier Vice President

. Electric and Steam Production Rochester Gas 8 Electric Corporation 89 East Avenue Rochester, New York 14649

Dear Mr. Maier:

RE: SEP TOPICS V-II.A, ISOLATION OF HIGH AND LOW PRESSURE SYSTEMS, AND VI-7.C.1, INDEPENDENCE OF REDUNDANT ONSITE POWER SYSTEMS-R.E. GINNA NUCLEAR POWER PLANT Enclosed are final evaluations of SEP Topics V-II.A and VI-7.C.l for R.E. Ginna Nuclear Power Plant. These assessments compare your facility, as described in Docket No. 50-244, with the criteria currently used by the regulatory staff for licensing new facilities. These reports have been revised to reflect the factual comments provided by your January 8, 1981 letter.

Your observations with regard to the acceptability of alternative designs and the use of administrative controls will be considered during our preparation of the integrated safety assessment for your plant. However, it must be pointed out that the currently approved version of Regulatory Guide 1.139 is Revision 0. Revision 0 requires diverse interlocks.

These evaluations will be basic inputs to the integrated safety assess-ment for your facility. As previously stated, these assessments may be revised in the future if your facility design is changed or if NRC criteria relating to this subject are modified before the integrated assessment is completed.

Sincerely,,

Dennis M. Crutchfield, ief Operating Reactors Branch 85 Division of Licensing

Enclosure:

Draft SEP Topics V-II.A and VI-7.C.1 cc w/enclosure:

See next page

Mr. John E. Maier R. E. GINNA NUCLEAR POWER PLANT 50-244

. i DOCKET NO.

cc w/enclosure:

Harry H. Voigt, Esquire Director, Technical Assessment LeBoeuf, Lamb, Leiby and MacRae Division 1333 New Hampshire Avenue, N. M. Office of Radiation Programs Suite 1100 (AW-459)

Washington, D. C. 20036 U. S. Environmental Protection Agency Mr. Michael Slade Crystal HalI f2 12 Trailwood Circle Arlington, Virginia 20460 Rochester, New York 14618 Rochester Committee for U. S- Environmental Protection

-

Sci entif ic Inf ormation Agency Robert E. Lee, Ph.D. Region II Office P. 0. Box 5236 River Campus ATTN: EIS COORDINATOR Station 26 Federal Plaza Rochester, New York 14627 New York, New York 10007 Jeffrey Cohen Herbert Grossman, Esq., Chairman New York State Energy Office Atomic Safety and Licensing Board Swan Street Building U. S. Nuclear Regulatory Comnission Core 1, Second Floor Washington, D. C. 20555 Eopire State Plaza Albany, New York '2223 Dr. Richard F. Cole Atomic Safety and Licensing Board Director, Technical Development U. S. Nuclear Regulatory Comission Programs Mashington, D. C. 20555 State of New York Energy Office Agency Building 2 Esquire State Plaza Dr. Emneth A. Luebke Albany, New York 12223 Atomic Safety and Licensing Board U- S. Nuclear Regulatory Comnission Rochester Public Library Mashington, D. C. 20555 115 South Avenue Rochester, New York 14604 Hr. Thomas B. Cochran

~

Natural Resources Defense Council, Inc.

Supervisor of the Town 1725 I Street, N. M.

of Ontario Suite 600 107 Ridge Road Mest Washington, .D. C. 20006 Ontario, New York 14519 Resident Inspector Ezra I. Bialik Assistant Attorney General R. E. Ginna Plant Environmental Protection Bureau

.c/o U. S. NRC New York State Department of Law 1503 Lake Road 2 World Trade Center Ontario, New York 14519 New York, New York 10047 Richard E. Schaffstall, Executive Director for SEP Owners Group 1747 Pennsylvania Avenue, NW Washington, D.C. 20006

0130J SEP TECHNICAL EVALUATION TOPIC V-11.A ELECTRICAl, INSTRUMENTATION, AND CONTROL FEATURES FOR ISOLATION OF HIGH Ai%) LOW PRESSURE SYSTEMS FINAL DRAFT R. E. GINNA NUCLEAR STATION Docket No. 50-244 January 1981 S. E. Mays 1-26-81

CONTENTS

1.0 INTRODUCTION

2~0 CRITERIA ~ ~ ~ ~ o ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

2.1 Residual Heat Removal (RHR) System . 1 2.2 Emergency Core Cooling System ~ 2 2.3 Other Systems ~ 2 3.0 DISCUSSION AND EVALUATION ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 3 3.1 Residual Heat Removal (RHR) System . ~ ~ ~ ~ ~ "~ ~ ~ ~ ~ ~ 3 3.2 Safety Injection System 3 3.3 Chemical and Volume Control System . 4 4o0

SUMMARY

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 5

>F0 REFERENCES ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ o ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 6

SEP TECHNICAL EVALUATION TOPIC V-ll.A ELECTRICAL, INSTRUMENTATION, AND CONTROL FEATURES FOR ISOLATION OF HIGH AND LOW PRESSURE SYSTEMS FINAL DRAFT R. E. GINNA NUCLEAR STATION

1.0 INTRODUCTION

The purpose of this review is to determine if the electrical, instrumentation, and control (EI&C) features used to isolate systems with a lower pressure rating than the reactor coolant primary system are in compliance with current licensing requirements as outlined in SEP Topic V-llA. Cuzrent guidance for isolation of high and low pres-sure systems is contained in Bxanch Technical Position (BTP) EICSB-3, BTP RSB-5-1, and the Standard Review Plant (SRP), Section 6.3.

2. 0 CRITERIA 2.1 Residual Heat Removal (RHR) S stems.'solation requirements for RHR systems contained in BTP RSB"5-1 axe:

1. The suction side must be provided with the following isolation features:

a. Two power-operated valves in series with posi-tion indicated in the control room.

b. The valves must have independent and diverse interlocks to prevent opening if'he reactor coolant system-(RCS) pressuxe is above the design pressure of the RHR system.

C ~ The valves must have independent and diverse interlocks to ensure at least one valve closes upon an increase in RCS pressure above the design pressure of the RHR system.

2. The discharge side must be provided with one of the following features:

a. The valves, position indicators, and interlocks desczioed in (1)(a)'hxough (1)(c) above.

bo One or more check valves in series with a normally-closed power-operated valve whicn has its position indicated in the control room.

If this valve is used for an Emergency Core Cooling System (ECCS) function, the valve must open upon receipt of a safety injection signal (SIS) when RCS pressure has decreased below RHR system design pressure.

c. Three check valves in series.

d. Two check valves in series, provided that both may be periodically checked for leak tightness----

and are checked at least annually.

2.2 Emer enc Core Coolin S stem. Isolation requirements for ECCS are contained in SRP 6.3. Isola'tion of ECCS to prevent overpres-surization must meet one of the following features:

l. One or more check valves in series with a normally<<

closed motor>>operated valve (MOV) which is to be opened upon receipt of a SIS when RCS pressure is less than the ECCS design pressure

2. Three check valves in series

3. . Two check valves in series, provided that both may be periodically checked for leak tightness and are checked at least annually.

J

• with the RCS must meet tne following isolation requirements from BTP EICSB-3:

At least two valves in series must be provided to isolate the system when RCS pressure is above the system design pressure and valve position should be provided in the control room

2. For systems with two NOVs, each NOV should have independent and diverse interlocks to prevent opening until RCS pressure is below the system design pressure and should automatically close when RCS pressure increases above system design pressure

3. For systems witn one check valve and a MOV, the MOV should oe interlocked to prevent opening if RCS

pressure is above system design pressure and should automatically close whenever RCS pressure exceeds system design pressure.

3.0 DISCUSSION AND EVALUATION There are three systems at R. E. Ginna Nuclear Station which have a direct interface with the RCS pressure boundary and have a design pressure rating of all or part of the system which is less than that of the RCS. These systems are the Chemical and Volume Control System (CVCS), the Safety Injection System (SIS), and the Residual Heat Removal (RHR) sys tern.

3.1 Residual Heat Removal S stem. The RHR system takes a suction on the RCS loop A hot leg, circulates the water through the RHR system heat exchanger, and discharges to the RCS loop B cold leg. Two motor-operated valves in series provide isolation capabilities in both the suction and discnarge lines. Each of these NOVs has position indica-tion in the control room. The inboard (closest to the RCS) valves are interlocked to prevent opening if RCS pressure is above RHR system design pressure. However, both valves use the same pressure switch and relay to provide this interlock. The outboard valves have no pressure interlocks. None of the valves will automatically close if RCS pres-suze increases above RHR system design pressure during RHR system operation.

The RHR system is not in compliance with the current licensing requirements of BTP RSB-5"1 since none of the isolation valves will automatically close if RCS pressure exceeds RHR design pressure. Also, the outboard isolation valves nave no interlocks to prevent RHR overpressurization, and the inboard valve interlocks are neither diverse nor independent.

3.2 Sazet In'ection S stem. One SIS subsystem consists of two accumulators pressurized with nitrogen with each accumulator isolated rzom the RCS by a pair of check valves. There are connections upstream oz eacn cneck valve that can allow them to be tested. A normally-open

motor-operated isolation valve upstream of the check valves for each accumulator has position indication in the control room. Each MOV is opened automatically, i.f closed, upon receipt of a safety injection signal.

The second SIS subsystem consists of two loops, each supplied by a safety injection pump. Each pump discharges to the hot and cold legs of one RCS loop. Isolation is provided by two check valves in series for each branch of the safety injection loop. The cold leg check valves are testable. The check valves in the lines supplying the RCS hot leg for each SIS loop are not testable. However, the MOV in each hot leg is locked shut with power removed and is not required for accident mitigation. A motor-operated isolation valve with position indication in the control room is provided in each branch of the cold leg.discharge lines. These valves open upon receipt of a safety injection signal, but have no interlocks preventing opening when RCS pressure is above SIS design pressure.

The third SIS subsystem uses the RHR system to provide low pres-sure water from the refueling water storage tank to the reactor vessel head (core deluge). Isolation is provided by a MOV in series with a check valve in each of two branches. The MOVs open upon receipt of a safety injection signal but have no interlocks to prevent opening when RCS pressure is above SIS design pressure.

The SIS is not in compliance with the current licensing require-ments of SRP 6.3 since the MOVs for the low pressure injection lines have no interlocks to prevent opening when RCS pressure exceeds SIS design pressure.

3.3 Chemical and Volume Control S stem. The CVCS takes water from the RCS and passes it through a regenerative heat exchanger, an orifice to reduce its pressure, and a nonregenerative heat exchanger before reducing its pressure 'further by the use of a pressure control valve. After filteri.ng and cleanup, the water may be returned to the RCS by the use of the cnarging pumps, which increase .the water pressure

and pass it througn the regenerative heat exchanger to eitner the hot or cold legs of the RCS or to the pressurizer auxiliary spray line.

The CVCS suction line isolation is provided by a manually-operated solenoid valve in series with three parallel solenoid-operated valves.

Each of these valves is operated from the control room and has valve position indicated. None of the valves have interlocks to prevent opening or to automatically close if the pressure exceeds the design rating of the low pressure portions of the system.

The CVCS discharge line isolation is provided by a common dis-charge line check valve and a branch check valve in each of the three branches downstream, of the common check valve. Drain fittings on the discharge line upstream of each check valve can allow the valves to be tested. There is no position indication available in the control room for the check valves. There are solenoid isolation valves in each discnarge line branch which have position indication in the control room, but these valves have no interlocks to prevent system overpressurization.

The CVCS is not in compliance with current licensing requirements for isolation of high and low pressure systems contained in BTP EZCSB-3 since the suction line solenoid-operated valves have no interlocks to prevent system overpressurization, and the discharge line check valves have no position indication available in the control room.

4. 0 SUaiRY The R.'. Ginna Nuclear Station has three systems with a lower design pressure ratin'g than the RCS, which are directly connected to the RCS. The CVCS, SIS, and RHR system do not meet current licensing requirements for isolation of high and low pressure systems as speci-fied below.

~ I

~ ~

1. The CVCS solenoid-operated valves have no pressure-related interlocks, and the discharge line check valves have no position indication available in the control room as required by BTP EICSB-3

2. The MOVs in the low pressure SIS lines have no pressure-related interlocks required by SRP 6.3

3. None of the RHR system isolation valves automati-cally close if RCS pressure increases above RHR system design pressure during RHR system operation, and the outboard isolation valves have no pressure-related interlocks as required by BTP RSB-5-1. The interlocks for the inboard isolation valves are neither diverse nor independent.

5 0 REFERENCES

1. NUREG-075/087, Branch Technical Positions EICSB-3, RSB-5-1; Standard Review Plan o.3.

2. Updated Final Facility Description and Safety Analysis Report, Ginna Nuclear Power Plant, Unit No. 1.

3. RG&E drawings 33013-422, -424'257 426'27'28'32'33'434,

-435, and -436.

4. RGGE drawings 10905-280, -285, -287, -295, -296, -300, and -301.

TOPIC Y-11,8(SYSTH'8)

SEE TOPIC V-'lO.A

Cy

'A c<

+~

gy,R

+**++

Rfgy

~o

~O UNITED STATES NUCL'EAR REGULATORY COMMISSION WASHINGTON, D. C. 20555 April 24, 1981 gj /

Docket'No. 50-244 LS05 04-035 Mr. John .E. Mater Vice President Electric and Steam Production Rochester Gas 8 Electric Corporation 89 East Avenue Rochester, New York 14649 Dear Mr. Maier;

SUBJECT:

SEP TOPICS V-ll.A, ISOLATION OF HIGH AND LOW PRESSURE SYSTEMS, V-ll,B, RHR INTERLOCK REQUIREMENTS AND VI-7.C.l, INDEPENDENCE OF REDUNDANT VISITE POWER SYSTEMS - R. E. GINNA NUCLEAR POWER PLANT We have reviewed your letter of March 27, 1981 and agree with resolving open items during topic evaluations rather than deferring a decision to the Integrated Assessment, To this end, we are enclosing a revised safety evaluation of Topic V-ll.A.

We have also reviewed your comments- on the draft Technical Evaluation Report (TER) SEP Topic V-ll.B dated January 8, 1981, Your comments on SEP Topic V-ll.B are covered by Sections 3.1 and 3.2 of our safety evalua-tion on SEP Topic V-ll,A. We are enclosing a revised Technical Evaluation Report on Topic V-ll.B which incorporates a reference to Section 3,1 and 3.2 of our safety evaluation report on Topic V-ll.A.

We are enclosing a request for additional information on SEP Topic VI-7,C,1 where we do not have sufficient information to reach an independent safety assessment.

Sincerely,'nclosure:

Dennis M, Crutchfield, ief Operating Reactors Branch No, 5 Division of Licensing SER for SEP Topic V-ll.A Questions for SEP Topic VI-.7.C.l cc w/enclosure:

See next page

Hr. John E. Maier CC Hariy H. Voigt, Esquire Director, Criteria and Standards LeBoeuf, Lamb, Lei by and NacRae Division 1333 Ne~( Hampshire Avenue, N. W. Office of Radiation Programs Suite 1100 (ANR-460}

Washington, D. C. 20036 U. S. Environmental Protection Agency t<r. Michael Slade Washington, D. C. 20460 12 Trailwood Circle Rochester, New York 14618 U. S. E nvi ronmenta1 .P rotect i on Agency Ezra Bi al ik Region II Office Assistant Attorney General ATTN' IS COORDINATOR Environmental Protection Bureau 26 Federal, Plaza New York State Department of Law New York, New York 10007 2 World Trade Center New York, New York 10047 Herbert Grossman, Esq., Chairman Atomic Safety and Licensing Board Jeffrey Cohen U. S. Nuclear Regulatory Commission New York State Energy Office Washington, D. C. 20555 Swan Street Building, Core 1, Second Floor Dr. Richard F. Cole Empire State Plaza Atomic Safety and Licensing Board Albany, New York 12223 U. S. Nuclear Regulatory Comnission Mashington, D. C. 20555 Director, Technical Development Programs Dr. Emeth A. Luebke State of New York Energy Office Atomi c Saf ety and Licensing Board Agency Building 2 U. S. Nuclear Regulatory Cowrission Empire State Plaza Mashington, D. C. 20555 Albany, New York 12223 Hr. Thomas B. Cochran Rochester Public Library Natural Resources Defense Council, Inc.

115 South Avenue 1725 I Street, N. M.

Rochester, New York 14604 Suite 600 Washington, D. C. 20006 Supervisor of the Town of Ontario Ezra I. Bi alik 107 Ridge Road West Assistant Attorney General Ontario, New York 14519 Environmental Protecti on Bureau New York State Department of Law Resident Inspector 2 World Trade Center R. E. Ginna Plant New York, New York 10047 c/o U. S. NRC 1503 Lake Road Ontario, New York 14519

4 ~< ~

w-' ~ -.

~ m EGG 1183-4154 21 April 1981 Energy Measurements Group, SYSTFMAT/C EVALUATION PROGRAM RFVIEVf QF NRC SAFETY TOPIC V-II.B ASSOCIATED WITH THE ELECTRICAL, IHSii-UiYiKNTATIQN,Ai~'D CONTROL PQRTIOF~S QF THE RESIDUAL HEAT RED"<OVAL.SYSTEM FOR THE GINNA NUCLEAR PQVZER PI ART SAN RAMON OPERATIONS 29C". GLG <<r CW C~NYC4 PC 9c N 84MCQ, CAglsGANIA 9(<9 9

EGG 1183-4]S4 SYSTEMATIC EVALUATION PROGRAM REYIDV QF NRC SAFETY TOPIC V-31.B ASSOCIATED VflTH THE ELECTRICAL, INSTRUMENTATION'ND CONTROL PORTIONS OF THE RESIDUAL HEAT REMOVAL SYSTEM FOR THE GINNA NUCLEAR POWER PLANT

ABSTRACT This report documents the technical evaluation'nd review of NRC safety topic Y-,li.B, associated with the elec.rical, instrumentation, and control por.ions of the residual heat removal (RHR) system for the Ginna nuclear power plant. Current licensing criteria are used to evaluate. the overpressure protection and incependence of the RHR sys.em.

FQREMCRO Thi s repor. i s supp 1 i ed as part of the Systematic Ev al uati on Program being conducted for the U.S. Nuclear Regul atory Commi ssion by Lawrence Livermore National Laboratory. The work was performed by. EGEG, Energy Measurements Group, San Ramon Operati ons for Lawrence Livermore National Laboratory under U.S. Oepartment of Energy contract number OE-AC08-75NV01183.

TABLE OF CONTENTS Page

1. INTRODUCTION............ 1

2. C'URREHT L I CEHS I HG CRITERIA 'o 3

~....

~ ~ o e ~ ~ ~ ~

3 ~ REY IFh GUIDE'NES - ~ -.. ' ~ 5

4. SYST"-t1 DESCRIPTiON........... 7

5. EVALUATION AND CONCLUSIONS....... ~ . 9

6. SUM~V RY ~ ~ ~ ~ ll R=".=ERENCES ~ ~ ~ o A HRC SAFETY TOPICS RELATED TO THiS REPORT... 13'PPENDIX A-1

SYSTEMATIC EVALUATION PROGRAM REVIEM OF NRC SAFETY TOPIC V-11.B ASSOCIATED WITH THE ELECTRICAL, INSTRUMENTATION, AND CONTROL PORTIONS .

OF THE RES'IDUAL HEAT REMOVAL SYSTEM FOR THE GINNA NUCLEAR POMER PLANT

1. INTROOUCTIQN A number of plants have residual heat removal (RHR) sys ems in 0

which the design pressure rating is lower than the reactor coolant system (RCS) pressure, boundary to which the system is connected. The RHR system normally is located outside of primary containment and has motor-opera ed valves (MOVs),which isolate it from the RCS. There is, therefore, a po-tenti al that these systems woul d be sub'cted to pressure stresses in excess of their design rating if the isolation MOVs were opened inadvert-ently while the RCS was above the RHR system design pressure rating. This could result in a LOCA ou-side contairment and a loss of ref lood capability since 'he coolant inventory could . be lost. Generally, interlocks are provided to prevent isolation MOVs frcm opening under high RCS pressure conditions.

It is importan to incorporate features into the system design

. which will pr event overpressuri zing the low pressure-rated RHR systems which interface with the reactor coolant pressure boundary. The current licensing cri teria requires redundant, diverse interlocks to prevent opening of the isolation MOVs when RCS pressure exceeds RHR pressure design

limits. The current licensing criteria also requires automatic closure of the isolation NOYs when RCS exceeds RHR pressure design limits.

The objective of this review is to ensure that the plant has adequate measures to protect a low pressure-rated RHR system that inter-faces with .he RCS from failures due to excessive pressure and that such protection is suitably redundant and diverse.

This review applies to the interlocks associa.ed with the isola-tion NOYs of the RHR system. Other protection schemes such as doubl e-testable che k valves are discussed'n reports on other NRC Safety Topics.'

2. CURRENT L'ICEN SING CRiTERIA Branch Technical Position ICSB-3 [Ref. 13, entitled "Isolation of Low Pressure Systems from the High Pressure RCS," states that:

The isolation MOYs should have independent and diverse interlocks to prevent opening unless the primary system pressure is below the subsystem design pressure. Also, the isola.ion NOY operators should receive a signal to close the valves automatically when the primary sys em pressure ex-ceeds the subsystem design pressure.

Branch Technical Position RSB S-I iRef. 2j, entitled "Oesign Re-qu'.rements for the Residual '~eat Removal System," states that:

I sol ati on shal 1 be provi ded by at 1 e st two power-operated valves .in series, and the val ves snail have independent diverse interlocks to prevent the valves from being opened unless the RCS pressure is below the RHR system design pres-sure. The valves shall have independent, diverse interlocks to protect against one or both valves being open during an increase above RHR system design 'pressure. If the RHR system discharge line is used for an emergency core cooling sys em (ECCS) function, the power-operated valve is to be opened upon receipt of a safety injec ion signal once the reactor coolant pressure has decreased below the ECCS design pressure.

3. REVIEW GUIDELINES The NRC guidelines used in this review are as follows: t (1) Identify the valves which isolate the RHR system fran the reactor coolant pressure boundry: (Refer to NRC memorandum from B. L. Siegel, RSB, to P.

A. Oi Benedetto, SEP; which is enclosure 3 of a letter from Crutchfield NRC, SEPB, to Dittmore, LLNL, dated 6'-10-80 [Ref. 3]).

(2) Evaluate (he design features which provide protec-tion =-gainst the overpressurization cf tne RHR system.

(3) Identify the related topic reviews in an appendix u this report.

(4) Compile a list of the major EISC systems that are necessary for OBE and for safe shutdown of the plant. Submit the compilation of necessary items for safe shutdown as an appendix to NRC Safety Topic YII-3, entitled "Systems Required for Safe Shutdown."

(5) If power is locked-out to the RHR isolation MOYs, revie~ to determine if any functions of the inter-locks or perm..'ssives are adversely affec .ed. (The report on NRC Safety Topic Yi-7.C, among others, sta .es which values have power locked out).

4. SYSTEM DESCRIPTION The RHR loop consis .s of two pumps, two heat exchanoers, and the necessary valves, piping, and instrumen;ation. During plant cooldown, coolant flows frcm the RCS to the RHR pumps, through the tube side of the RHR hea exchangers and back to the RCS. The single inlet line to the RHR loop commences at he hot leg of r actor coolant loop A, through two re-dundan . pumps .'nd thei,r associated heat exchangers, and back to the cold leg of reactor coolant loop 8 via a single header.

r The RHR pumps and hea. exchangers serve dual funct',cns. Al:hough the normal du".y of .he RHR pumps and heat exchangers is performed during periods of reac.or shutdown, this equipment is aligned during the injection phase after a loss-of-coolant-accident (LOCA) to perform the low-head safety injection '(LPSI) function. In addition, during the recircul eton phase of a LOCA the capability may be divided between the. core-cooling function and the cor tainment-cooling function as a part of the containment spray system.

5. EVALUATION AND CONC'SIONS The suction line of the RHR system is isolated from the loop A hot leg of the RCS by MOV-700 and MOV-701 in series. The discharge line of the RHR system is isolated frcm the loop B cold leg of the RCS by MOV-720 and MOV-721 in series. I:Ref. 4, drawing 33013-436-A].

All permissive interlocks associated wi .h the RHR sys em isola-tion MOYs are designed to open the valves; there are no permissive int r-locks associated with isola .ion MOY closure.

Sec .ion 4.1 of.. the SEP review o Safe Shu down Syste.,s [Ref; 5j states that the permissive interlocks requi red to open the four RHR sys .em isola.ion valves are as listed below:

MOV-700....RCS pressure must be less than 410 psig.

RHR suction valves MOV-850A and MOY-850B fran the containment sump must be closed.

MOY-701... . The val ve .is operated by a key swi tch.

RHR sue .ion valves MOV-850A and MOY-8508 from .he containment sump must be closed.

MOV-720....No in;erlocks exist; valve operated by key swl tch ~ 1 MOV-721....RCS pressure must be less than 410 psig.

The RHR system discharge line is not used for an ECCS function that would require MOY-720 or MOY-721 .o open; however, a branch of .he RHR discharge line provides low pressure sa ety injecton (LPSI) to the reac.or vessel via paral 1 el lines . isolation between he RHR system and LP S injec ion into .he reactor ve'ssel is provided by two separate paths from the RHR discharge line,~ith each path containing an MOY and check valve, MOY-852A and check valve 853A provide isolation in one path, while MOY-852B and check valve 8538 provide isola ion in the other path I:Ref. 4, drawing

33013-436-A; Ref. 6, drawing 33013-432-A]. The LPSI isolation MOYs open on a SI signal regardless of RCS pressure; .here are no interlocks associated

~ith closure of the 'LPS'I isolation MOYs, although key swi.ch closure cap-ability is provided.

Section -".1 of the SEP review of Safe Shutdown System [Ref. 53 states in part that:

A branch of the RHR discharge line provides low pressure safety injection (LPSI) to the reactor vessel via parallel lines with one normally closed mo or-operated valve (NOV) and one check valve in each line. The MOY position indi-cation is provided in the cortrol room anc these valves

. eceive an open signal coincident, with the safety injection (S;) signal. The MOVs in the LPSI lines open ,on an SI signal before RCS pressure drops be'ow RHR design pressure.

The plant ccmplies to all EIEC aspec .s of the "RHR Interlock Requi rements" review criteria listed in Section 2 of this report except for the fol 1 owiag:

The plant RHR system does not satisfy BTP ICSB 3 [Ref.

13 and BTP RSB 5-1 [Ref. 23 because the RHR discharge and sucti on i sol ati on MOYs do not have independent diverse interlocks to prevent opening the valves until RCS pressure is below 4'0 psig. Only the inboard valves MGY-7GO and MGY-721 have this irterlock. The outboard valves MOY-701 and MOY-720 are manually controlled with key-locked swit"hes. By procedure, MOV-701 and MOY-720 are not opened un.il RCS pressure is less than 410 psig.

(2) The plan RHR system does not satisfy BTP ICSB 3 [Ref.

1j and BTP RSB 5-1 [Ref. 23 because all RHR isolation MOYs lack an in.erlock feature to close them when RCS pressure increases above the RHR design pressure.

(3) The plant RHR sys.em does not satis y BTP ICSB 3 [Ref.

lj and in the BTP RSB 5-1 I.Ref. 2] because the isolation MOVs LPSI lines (MOY-852A and MOY-852B) ooen .on an SI signal before RCS pressure drops below RHR design pressure.

6.

SUMMARY

The plant RHR interlock system fails to satisfy current. licensing criteria for the following reasons:

The RHR suction and discharge isolation MOVs do not have independent diverse interlocks to prevent opening the isolation MOYs until RCS pressure is below 410 pslgo (2) All RHR isolation MOYs lack an interlock feature to close them when RCS pressure increases above RHR design pressure.

(3) The isolation MOYs in the LPSI lines open on an SI signal regardless of RCS pressure.

The resolution of items 1, 2 and 3 are presented in Sections 3.1 and 3.2 of SEP Topic V-ll.A.

REFERENCES

1. U.S. Nuclear Regulatory Commission, Branch Technical Position ICSB 3, "Isolation of Low Pressure Systems from the High Pressure Reactor Coolant System."

2. U.S. Nuclear Regulatory Commission, Brancn Technical. Position RSB 5-1, "Design Requirements of the Residual Heat Removal System."

3. NRC (0. M. Crutchfield) letter to LLNL (M. H. Di tmore), dated June 10, 1980. '

Ginna drawing, 33013-436-A, "Auxiliarv Coolant System". ~ ~

SEP Review of Safe Shu.down Systems for the R.E. Ginna Nuclear Power Plant, Revision 1, qnda~ed.

6. Ginna drawing, 33013-432-A, "Safety injection System."

APPENDIX A NRC SAFETY TOPICS RELATED TO THIS R PORT 1 III-1, "Classifica.ion of Structures, Systems and Components."

2. III-10.A "Thermal Overload of MOVs."

3. Y-10-.B, "RHR System Rel iabil i ty."

V-11.A, "Requirements for Isolation of High and Low Pressure Systems."'.

Y1-?.C "ECCS Single Failure Criterion and Requiremen.s for Locking Out Power to Valves Including Independence of Interlocks on ECCS Valves.."

6. YIV.-3, "Svstems Required for Sa,e Shutdown "

?. XVI, "Technical Specifications".

TOPIC V-12,A SEE TOPIC II-4. E

TOPIC V-0 SEE TOPIC I I-2. 8

' ~ ,A .HNENT

X ~ loll

=, ~fg+j e~~P'/gg ROCHESTER CPS PHD ELECTRIC CORPORPTIOH ~ S9 EAST AVEHUE, ROCHESTER, N.Y. I46<9 c'~+oic V<~'RiSI "EN7 i ici ooc -io 6 '6 2 OO June 23, 1981 Director of Nuclear Reactor Regulation Attention: Mr. Dennis M. Crutchfield, Chief Operating Reactors Branch 55 U.S. Nuclear Regulatory Commission ~~l washington, D.C. 20555 SEP Topics V-10.B, V-ll.A, V-11.B, VZ-7.C.1, )1'ubject:

VZZ-3, and VIIZ-2, R.E.p Ginna Nuclear Power Plant Docket No. 50-244

References:

(1) Letter from Dennis M. Crutchfield, NRC, to John E. Maier, RGE, SEP Topics, V-10.B, V-ll.B, and VZZ-3 (Safe Shutdown Systems Report), May 13, 1981.

(2) Letter from Dennis M. Crutchfield, NRC, to John E. Maier, RGE, SEP o ics V-11.A, V-11.B, and VZ-7.C.1, dated April 24, 1981.

(3) Letter from Dennis M. Crutchfield, NRC, to John E. Maier, RGE, SEP Topics VIZ-3 and VIII-2, dated April 2, 1981.

Dear Mr. Crutchfield:

This letter is in response to the SEP topic assessments provided in the three above-referenced letters. Due to the intimate relationship of the "Safe Shutdown" topics V-10.B, V-11.A, V-11.B and VZI-3 addressed in these three letters, all of our comments are provided concurrently in the three attached'esponses.

This should aid the inclusion of our comments into the NRC's "SEP Integrated Assessment". r Very truly yours, ohn E. Maier Attachments r

RG&E responses to NRC Assessment, of SEP Topics V-10.B, RHR System Reliability, V-ll.B, RHR Interlock Requirements, and VII-3, Systems Required for Safe Shutdown (Safe Shutdown Systems report), May 13, 1981.

1. In RG&E's January 13, 1981 response to the NRC's November 14, 1980 "Safe Shutdown Systems" assessment, a number of comments were made which have not been incorporated. into Revision 2 of this assessment, transmitted by letter dated May 13, 1981.

We feel these comments were valid, and should be incorporated.

For continuity, these comments will be listed below (with their original comment numbezs):

assumes piping system passive failures"...beyond those normally postulated by the staff, e.g., the catastrophic failure of moderate energy systems...". Although shown that safe shutdown following such an event could it is be achieved, it is not considered that such an evaluation should even be made. As noted by the staff,.it is clearly beyond a reasonable design basis. It is thus recommended that this paragraph be deleted from the evaluation. Subsequent evaluations to this "criterion",

such as those related to the CCW system on page 22 and 23, should also be deleted.

ll. In paragraph g on page 66, it is noted that, wnen

a. plying the power diversity requirements of BT? ASB 10-1 ' event o= an SSE, no means to su p}y feed to the steam generators exists. It was cetezmined that this was acceptable, based on low likelihood of occurrence.

This conclusion is correct; however, since BPT ASB 10-1 does not consider an SSE in conjunction with the loss of all A.C. power, there is no need to even make the evaluation. The comparisons in the SEP program should be to current criteria, rather than to arguable extrapo-lations. Reference to loss of all A.C. power in conjuncton with an SSE should thus be deleted from this paragraph.

12. On page A-4, it is noted that additional systems are required to achieve cold shutdown for a PWR than for a BWR because of a difference in the definition of cold shutdown. This does not appear to be a reasonable basis. System requirements should be based on specific safety reasons. The NRC should be consistent in .its requirements for cold shutdown, oz provide a technical basis for any differences."

Staff position 1 states that "the licensee must develop plant operating/emergency procedures for conducting a plant shutdown and cooldown using only the systems and equipment identified in Section 3.1 of the SEP Safe Shutdown Systems Report." RG&E disagrees with the need for these procedures.

We reiterate the comments provided in our January 13, 1981 response that the operator should perform a cooldown with the best equipment available to him at the time. If a piece of non-safety equipment is available, and would be the most beneficial for performing a required function, that this piece of equipment would be used. If ititis expected is not available, the operator could fall back on the use of safety-grade equipment. But RG&E does not intend to commit plant personnel to use only safety-related equipment, if is available and more effective. We feel that non-safety equipment it would be impossible to determine when a "safety-grade-only" cooldown procedure would ever be implemented. As long as the safety-grade equipment is available (and the safe shutdown assessment concudes that it is), RG&E considers that the necessary safety requirements are met.

RG&E also notes that no regulatory basis for this requirement is provided. It is admitted in Section 4.5 of the Safe Shutdown report that "the need for procedures for these evaluations is not identified in Regulatory Guide 1.33..." ~

Section 4.5 then goes on to say that the basis is found in BTP RSB 5-1 and SEP Topic VII-3. But BTP RSB 5-1 merely references RG 1.33, and this is the assessment of SEP Top'c VII-3.

Therefore, since no basis for this "requirement" exists, and we do not feel that it would even be beneficial, and since the Safe Shutdown report did conclude that the capability for attaining cold shutdown using only safety-related equipment exists, RG&E concludes that this staff position should be deleted from consideration.

Staff position 3 does not appear to take into account the information provided in our March 27, 1981 submittal regarding SEP Topic V-ll.A. Enclosure 3 to that submittal provides

'he valve equipment specification, noting that the 700, 701, 720 and 721 NOV's are designed such that they 'physically are unable to open against a differential pressure of greater than 500 psi. This ensures that an intersystem LOCA caused by the opening of the outboard valves, plus leakage of the inboard valves, cannot occur, since the outboard valves cannot open.

3 Even how without this provision, it is difficult to comprehend the Ginna arrangement could result in an "Event V". By administrative procedure, the RHR valves are key-locked closed, with power removed. Further, interlocks are provided for the inboard RHR valves. Thus, for an "Event V" to occur would require the:

1) failure of the administrative procedure requiring power lock-out (at the breaker),

2) failure of the administrative procedure governing operation of the valve at power,

3) failure of the inboard isolation valve,

4) failure of the relief valve (RV 203) which has a capacity of 70,000 lb/hr at its 600 psig setpoint, to relieve the leakage past the inboard RHR valve.

This set of failures is considered very remote. When coupled with the fact that the RHR valve design prevents opening of the valves against a greater than 500 psi differential pressure, it is RG6E's conclusion that the possibility of an intersystem LOCA should not be a credible design basis. No additional modifications, such as diverse interlocks 'for the outboard valves, are warranted.

Staff position 5 states that "the operating procedures for the Ginna plant shoula be modified to direct the operator to cooldown and depressurize to RHR initiation parameters within 36. hours whenever the Service Water System is used for steam generator feedwater..." This position is based on the reference BNL-NUREG-28147, "Impure Water in Steam Generators and Isolation Condensers." We have had this report reviewed by NWT Corporation. NWT-167, "Use of Lake Ontario Water in Steam Generator During Hot Shutdown" (attached) concludes that, "although not recommended from the-standpoint of maximizing component life, and operation for periods up to several days is not expected to result in any significant cracking or in deterioration of steam generator, integrity."

t RG&E therefore concludes that a specific directive to cool down and depressurize to RHR initiation conditions is not,.

warranted, and should not be included in a procedure. The capability to do this does exist, however, and could be used if determined to be necessary at the time.

RG&E responses to NRC letter of April 24, 1981 regarding SEP Topics V-11.A, "Isolation of High and Low Pressure Systems", V-11.B, "RHR Interlock Requirements", and VI-7.C.1, "Independence of Redundant Onsite Power Systems".

The Safety Evaluation for SEP Topic V-11.A, "Requirements

,for Isolation of High and Low Pressure Systems", specifies that the outboard RHR valves should have diverse interlocks to prevent opening when the RCS pressure is greater than RHR system design pressure.

RG&E rationale for not providing these additional interlocks is provided in comment 3 of Attachment 1 of this transmittal.

2.. The safety evaluation also required that interlocks be installed on the CVCS suction valves (200A, 200B, 202), to prevent a possible overpressurization of the CVCS letdown line outside containment. RG&E has noted in our March 27, 1981 letter on this SEP Topic that a relief valve (RV 203), with a capacity greater than the combined capacity of the three orifices, would relieve the pressure buildup caused by closure of the containment isolation valve 371. No overpressurization of the CVCS would thus be expected.

RG&E has also evaluated the potential consequences of such an overpressurization event, with a subsequent small LOCA outside containment, and determined tha" no unacceptable consequences would result. This break. would be a small LCCA outside containment (maximum flow of, 140 gpm), and would be terminated by closure of valves 200A, 20GB, and 202 either by operator action or automatically by low pressurizer level. Radiological consequences would be minimal, since no fuel damage would result. This event is specifically evalu-ated by SEP Topic XV-16, "Radiological Consequences of Failure of Small Lines Carrying Primary Coolant Outside Containment." RG&E has provided information concerning this topic by letter dated June 18, 1980 from L. D. White Jr. to Mr. Dennis M. Crutchfield.

The RG&E conclusion is that, based on the..availability of RV 203 to prevent overpressurization, together with the lack of unacceptable consequences due to an overpressurization, no interlocks or other modifications are required for the CVCS suction valves.

3. The safety evaluation further states that position indication is required on the CVCS discharge check valves. As stated in our March 27,, 1981 letter on SEP Topic V-ll.A, we do not believe that this line should be classified as a low pressure

system connected to the RCS, since the piping is 2500-1b piping throughout its length (to the positive displacement charging pump). RG&E has had no experience with failures of the positive displacement charging pump pistons to hold primary system pressure, nor would any failures be anticipated.

Our contention that the charging line is not a line of concern is borne out by a memo from Edson G. Case to Raymond .F. Fraley, "Isolation of Low Pressure Systems from Reactor Coolant System", dated July 11, 1977. That letter transmitted an NRC study of this subject to the ACRS, and evaluated all potential lines of concern. The charging line was not included.

To verify that the charging line was not a valid "Event V" concern, RG&E calculated the PNR Cneck Valve Event Tree (Section 4.4 of WASH-1400), using the charging line con-figuration (two in-series check valves and a charging pump piston). Very conservatively assuming that both check valves were undet'ected open, and that the probability of the charging pump~iston failure was equal to a check valve failure, the Q calculated for this configuration was determined to 5eU> 1.4 x 10 /year. This is a low enough value to obviously be of no concern.

RG&E therefore considers that check valve position indi-cation is not needed on the charging line check valves.

i ith respect to the S=P .opic Assessment V-11.3, no comments are necessary, since the resolution of outstandi..g issues is addressed in the topic assessmen" "or SEP 'pic V-3.3..A.

The additional information requested for SEP Topic VI-7.C.1 is presently being developed. It is anticipated that this information can be furnished to the NRC by July 15, 1981.

RG&E responses to NRC letter of April 2, 1981, concerning SEP Topics VZI-3, "Electrical, Instru-mentation, and Control Feature of Systems Required for Safe Shutdown", and VIII-2, "Diesel Generators".

Zt appears that all comments provided by RG&E in our January 23, 1981 and January 30, 1981 letters concerning these topics have been properly incorporated.

Based on the resolution of all open items, and the removal of diesel generator testing from SEP Topic VIII-2, RG&E concludes that both of these topics are complete, with no outstanding issues to be carried into the Integrated Assessment.

NWT 167 February 1981 USE OF LAKE ONTARIO WATER IN STEAN GENERATORS DURING HOT SHUTDOWN W. L. Pearl S. E. Copi ey J ~ LelboVl KZ Prepared for'ochester Gas & E1ectric Company Corporation 7015 REALM DRIVE, SAN JOSE, CALIFORNIA 95119 810680 0~

This document was prepared for the Rochester Gas 6 Electric Company.

Neither the NWT Corporation nor any person acting on its behalf assumes any responsibility for liability or damage which may re'suit from the use of any information disclosed in this document.

INTRODUCTION The possibility of using Lake Ontario water as an emergency PWR feedwater supply for more than 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> during which the plant would be brought to cold shutdown is being considered. The maximum steaming rate during such a period would be 100,000 pounds/h (200 gpm) at a temperature of 350'F. As a consequence of steaming, impurities of the untreated Lake Ontario water will concentrate in the steam generator. Of major concern is the possible risk of stress corrosion cracking (SCC) of steam generator materials in contact with the concentrated solution thus formed. To address this concern, the chemistry variation in the liquid phase as steaming proceeds at 350'F was estimated with emphasis on pH.

Then, the possible potential for SCC was assessed on the basis of these estimates and available SCC data.

pH VARIATION AT 350'F 'UPON STEAMING LAKE ONTARIO WATER A. Computer Modeling The composition of Lake Ontario water as determined by RGE is given in Table 1.'ABLE 1

LAKE ONTARIO WATER ANALYSIS ppm Calcium 35 Nitrate 2.5 tlagnesium 8 Phosphate 0.3 Sodium 13 Fluoride 0.15 Potassium 3.6 Si i ca (as Si 02) 1 0.25 Aluminum 0 ~ 1 Dissolved Oxygen 9.5 Chloride 32 Ammonia (as Nitrogen) 0.24 Sul fate 35 Estimates of the water chemistry variation upon steaming were developed using the following assumptions:

1. Since aluminum and silica are in stoichiometric proportion in Lake Ontario water (Table 1), they are assumed to precipitate as aluminum silicate (clay) upon concentrating 'and therefore are removed from solution.

2. Since calcium occurs in the water (Table 1) in large excess over phosphate, it is assumed to precipitate all the phosphate as calcium hydroxy apatite (Ca5(P04)30H) and remove it from the solution. The calcium in solution is decreased by the corresponding amount.

3. Fluoride and nitrite are assumed to behave as chloride. Potassium is assumed to behave as sodium.

4. Sodium and chloride in solution are assumed to remain completely .

dissociated.

5= Calcium carbonate precipitation is neglected. Degasification of CO~

by steaming is assumed to occur.

6. The concentration of sodium and calcium chlorides is assumed limited by a solubility of 5 molal.

7. Chemical equilibrium expressions of references 2 and 3 apply.

On this basis, the liquid solution pH variation upon steaming at 350'F was estimated as a function of concentration factor defined as the mass ratio of total water (steam + liquid) to liquid water residual. The results are presented graphically in Figure 1. It is important to note'that the definition of pH used here is that followed by Mesmer" in the determination of the dissociation constant of water at high temperatures, viz, the negative of the logarithm of the hydrogen ion concentration (not of its activity). Similarly, neutral pH is defined as that where the hydrogen and hydroxyl ion concentrations are equal.

This neutral pH is a function of ionic strength. Therefore, the pH variation of the concentrated solutions must be considered in relation to that of neutral pH, also plotted in Figure 1. For basic solutions as is the case considered here, it is important to bear in mind that .he hydroxyl ion concentration is expressed in terms o pH as follows:

= 10pH 2NpH OH (where NpH is the neutral pH value) and that when the neutral pH varies together with the ionic strength as the liquid solution is being concentrated upon steaming, the basicity of the solution may not be appreciated from the solution pH alone. The equivalent NaOH concentration is more suitable for this purpose and is plotted also in Figure l.

B. Discussion Steam Generator Bulk Water Based. on a maximum feed rate of 200 gpm to the steam generator and a total steam generator liquid 'nlume of approximately 12,000 gallons, a maximum of one steam generator volume is steamed away each hour. Therefore, under maximum steaming-

~onditions, the concentration factor achieved in the bulk steam generator water is t + 1 where t is the number of hours of steaming.

Solution pH

~

400 o)

O zoo 0 5

0 Pt 0

Equivalent NaOH Concentration Neutral ft I pH 0

200 I 8 100 10 100 1000 10000 Time ~ ll 3 Figure 1. Variation of Steam Generator pH witH Steaming at 350'F (feeding Lake Ontario water at 200 gpm)

The variation with time of the equivalent sodium hydroxide concentration in the steam generator with steaming of emergency Lake Ontario feedwater then can be followed on Figure 1. It is seen that a maximum equivalent NaOH concentration of about 300 ppm will be reached in the steam generator bulk water when 15 to 20 steam generator volumes will have been converted to steam, i.e., in approximately twenty hours. Further boiling should then decrease the equivalent NaOH concen-tration as magnesium and/or calcium hydroxides and/or calcium sulfate precipitate with increased concentrating. The decrease reaches a limit (at about 20,000 steam generator volumes converted to steam, i.e., in 20,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />) when sodium and calcium chlorides start to precipitate also. This limit is estimated at about 100 ppm equivalent NaOH for Lake)Ontario water composition as specified in Table 1 and with the assumptions already stated. The assumptions seem reasonable and, at any rate, can be tested experimentally with a small autoclave from which known amounts of Lake Ontario water would be boiled away at 350'F at constant liquid level in thy autoclave.

Crevices The estimated equivalent NaOH solution concentration in steam generator crevices will depend upon the relative degree of crevice solution concentration above the bulk water. In tube to tube support plate crevices, there may be a distribution of relative concentration factors of unity and higher.

The chemistry in a crevice would lead that of the bulk in the sense that the chemistry of a specific crevice would travel the same curve (Figure 1) as the bulk but would be at a point on the. curve somewhat ahead of the bulk. Since 1

the causticity of Lake Ontario water is not a strong function of concentration, this does not pose a problem. Indeed it is expected in this case that after a short period of steaming, the crevice chemistry will be'ess basic than that of the bulk.

Coolin Water Com osition The NWT chemistry modeling work discussed herein is based on the chemical composition of Lake Ontario water summarized in Table 1 as supplied by RGE.

It is possible that seasonal changes in the characteristics'f the lake water may result from the interrelation between source river flowrates, industrial pollution and/or acid rain. NWT has no relevant data to assess such effects.

It may be desirable that analyses made of Lake Ontario water during different seasons and under various conditions be fed into the NWT chemistry model. In this manner the safety of feeding Lake Ontario water, over the range of likely chemical compositions, can be verified.

6

POTENTIAL FOR SCC A. Corrosion The most aggressive solution expected based on the modeling work is 300 ppm NaOH, with slO ppm 02 (see below) at 350'F. Although laboratory data regarding these exact conditions are not available, data are available which can be extrapolated to assess the maximum corrosion rates expected for a given range of conditions.

van Rooyen and Kendig~ cite Westinghouse data indicating that U-bends of Alloy 600 in deaerated 10~ NaOH crack after several months of exposure.

Figures 2 and '3 summarize Westinghouse tests'hich show that at least 100 days of exposure to deaerated 10% NaOH at 600'F is required to produce a detectable crack in stressed Alloy 600.

Figure 4 shows data gathered bv Berce and Oonat',.'hese curves are for yield stressed C rings at 660'F. extrapolating the .curve for mill annealed Alloy 600 to 300 ppm NaOH yields a minimum time of 3500-4000 hours to induce 5 millimeter crack.

The data presented above are for deaerated systems and are consistent with van Rooyen's 'onclusion that Alloy 600 in 10% NaOH would not crack for several months. In the presence of oxygen, the susceptibility of Alloy 600 to SCC may be increased. Figure 5 shows stress corrosion behavior in 600'F high purity water containing varying amounts of oxygen in the gas'phase above the water and adjusted to pH 10 at startup with ammonia. As'the oxygen content of the gas phase increased, the percent of the specimens attacked and extent of'he attack increased. As noted in Figure 5 the average life in the 18-week test varied from no cracking with 15 oxygen in the gas phase (<2 ppm oxygen in the water) to 7 weeks with 100~ oxygen in the gas phase (<200 ppm oxygen in the water).

McIlree and Michels and later Sedriks, et al., 'eported less than.20K cracking after 27 days for Alloy 600 (2 common heat treatments) in aerated, 50< NaOH at 570'F.

7 ~

J aal Theeeenr Wrsl asn anneeieg lraaaea m<a itg C Tgrl XI ~ ~ 'R g .aa UIIII 0 n IXIIV. A se

. I00 01 XCO 41SI cKn KCO Csoesure Tlee I hrss Figure 2. Crack Depth as a Function of Time, Stress Level and Material Condition for ID Pressurized, Capsules Exposed to Deaerated 10~ NaOH at 600'F Heat '%C MA TT 9COT .011 1901 .OTT 150 .03l L5 7343 .OQ

.01 6 LA 310 C ae C&nTs 1 1500 MIIIAnnealed 0'e ea yleMI 0.3 g

.01 aa LT O. I ~eee s TTIarinally Treetssl tCOT Tsponsre Tlaie 1 nrsl

'Figure 3. Crack Oepth as a Function of gxlIosure Time for Mill Annealed and Thermally Treated Inconel Alloy 600 Exposed to Deaerated 10K NaOH at 600'F'

4000 C Rings Stressed to a=Ys According to ASTM STP 425 3000 Al 1 oy 600 HT 16h 1300'F 2000 1000 Al 1oy 600 MA 0

1,000 10,000 100,000 NaOH Concentration, ppm Figure 4. Resistance to-Stress Corrosion Cracking of Alloy 600 Mi 11-Annealed or Heat Treated at 1300'F as a Function of Deaerated Sodium Hydroxide Concentration at 600'F~

NO. OF SPECIMENS, MILL ANN. 8 12 l2 l2 I

I'20 PERCENT CRACKEO AV LIFE IN IB WEEK AV OF MAX. CRACK TESTS DEPTHS,MIL 0 0

IB '0'2 0

IB 0

18 0

58 17 2l 9?

I5 50 IOO

,7 52 o

o~ IOO O

'Z 80 8 I

I-i: 60 is i2 Zg IR r

+)

'"

i8 o

40 ie

. CI I

'l2 18 i8 20 j

ie ie is is ie ie ie ie is ie is is 8 I lO lO lA I IP IA IAIR t IA A 0

ALLOY 600 HEAT N 2 7 B5 4 2 4 7B25 6 4 7B256 478 256 476256 OXYGEN IN GAS NIL NIL 21 IOO AT START, (%)

BALANCE OF GAS, HYOROGEN ARGOhk NITROGEN NITROGEN AIR Figure 5. Stress Corrosion Behavior in Crevice Areas in Mill Annealed Enconel 600 Oouble U-bend Specimens in 600'F High Purity Water Adjusted to pH 10 with Ammonia at Startup'0

Laboratory studies show that there is a significant temperature dependence of caustic stress corrosion cracking as illustrated in Figures 6 and 7.

These results are for pressurized capsules exposed to 105 and 50% NaOH at varying stresses at temperatures ranging from 650 to 550'F. As can be seen, reducing the temperature below 600'F significantly extends the time for SCC to occur. This temperature dependence is further illustrated in Figure 8 where temperature is plotted versus rate constant for both 10K and 505 NaOH.

8. Oxygen The lake water fed to the generators probably would be air saturated (approximateiy

'.0 opm 0>). However-, at 350'F the KO (the equilibrium ratio between steam phase and liquid phase) for oxygen is slightly greater than 5000. Even though the dynamic distribution in practice may not reach true equilibrium conditions, the net effect of the high KO value is that recriculated steam generator coolant

~ Iill contain oxygen concentrations lower than 10 ppm. This recirculated coolant wil': dilute the oxygen concentration of incoming feedwat r with a net oxygen level in the downcomer of ~1 to 10 ppm, depending on tl e recirculation ratio under the contingency conditions.

C. Conclusion

~ With the significantly lower concentrations of sodium hydroxide (max 30o ppm),

oxygen concentration <10 ppm and the lower temperature (350'F) involved, tIie

~o~tingency of feeding Lake Ontario water to the Ginna steam generators should result in no measureable damage to steam generator internals. Although'ot recommended from the standpoint of maximizing component life, such operat~o~

for periods up to several days is not expected to resu1t in any significant.

cracking or in a deterioration of steam generator integrity.

11

NaOH 50X 10X 575'F g 0 .5 to 5 mils 50 40 30

>10 mils

')0 5 to 10 mils 0.5 to 5 10 mils 0 1000 2000 3000 4000 5000 6000 7000 Exposure Time, hrs Figure 6-. Caustic Cracking of Hill Annealed Alloy 600 at 575'F (Lines depict zones of crack depth from lOX NaOH at 600 F) 12

NaOH 50Z 10X 550'7 0 <0.5 mils 0.5 to 5 mils g 0.5 to 5 mils 50X NaOH 650 p 50 40 30

>10 mils 0 n

m 20 5 to 10 mils 0

10 0.5 to 5 mils 0 1000 2000 3000 4000 5000 6000 Exposure Time, hrs Figure 7. Caustic Cracking of Mill Annealed Alloy 600 at 55 and 650'F ]Lines depict zones of crack depth <<om NaOH at 600'F) 13

200

~

I MA I-600 150 C )C t0 K Zs I I V U 4J tC 0)

CI 100 50 0

530 550 570 590 610 630 Temperature, F Figure 8. Indicated Variation in Rate of SCC with Temperature 14

~~ ~

i'r.:.

L. I E E I

COMMENTS ON YAN ROQ'fEN ANO yE!;0:O'5 REPORT' The referenced reports basica11v:-. i broad summary covering a large volume of data applicable, in part to ..~:rl.ss steels and in part to Alloy 600. ge t>$

are gene~ally in agreement with .'.e'r nine summary conclusions, but find it E

difficult to apply their broad-br>qn treatment to the specifics of a pMR hot shutdown with lake water added to ..he steam generators at 350'F. Their document EEE~~f is misleading for such an application in two respects:

L E ~

~ \

E

1. Caustic Concentration Their statement that ..."For .".,-;rposes of SCC predictions, it has to be assumed that the time to .'c. "..:ngerous levels of NaOH, once impurities have been introduced, is shor:. :.=., one day or less" does not fully recognize the specific concen .r'.:.n chemistry of the cooling water involved nor the low heat '.lux::;'.i'able and the cutback in steaming rate during a period of hot shutcown.'" the case of the Lake Ontario water for example, the maximum NaCH ...""-=.".-raw'.on reached is 300 ppm (after steaming ~20 steam generator -" -'; with a decrease in concentration thereafter.

2. Tempera ture E All of the test work referenced in -he referenced report~ was performed ir.

the temperature range of 550 to =-:G=F. With the significant temperature dependence of caustic SCC as shown above, the concern at 350 F is many ti~s less than is indicated fnm the, data quoted by the authors.

Based on the above three considerations. it is our assessment that the generalized time limit of 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> i, =.'"e report~ is not directly applicable to the Ginna steam generators steamin. at 350'F while fed by Lake Ontario water.

15

~ rt E

.REFERENCES Harhay, A., Rochester Gas 8 Electric, personal Communication, February ll,

. 1981.

2. Leibovitz, J., and Sawochka, S. G., "Modeling the Effects of Condenser Inleakage on PWR Chemistry", presented at 41st Annual International Mater Conference, Pittsburgh, Pennsylvania, October 1980.

3. Leibovi;tz, J. and Sawochka, S. G., "Modeling of Cooling Water Inleakage Effects in PWR Steam Generators, Topical Report, Research Project 404-1",

Elec'tric Power Research Institute, May 1980, to be published.

Mesmer, R. E., Baes, C. F., Jr., and Sweeton, F. M., "Boric Acid Equilibria and pH in PWR Coolants', Proceedings of the 32nd International Mater Con-ference, Pittsburgh, P~nnsylvania, November 1971, pp. 55-65.

5. van Rooyen, D., and Kesdig, M. W., "Impure Water in Steam Generators and Isolation Generators", Brookhaven National Laboratory, June 1980 (Draft-BNL-NUREG-28147).

Airey, G. P., "Effect of Processing Variables on the Caustic Stress Corrosion Resistance of Inconel Alloy 600", presented at NACE Meeting~

March 1979 (Paper Number 101).

7. Berge, Ph. and Donati, J. R., "Materials Requirements for Steam Generator Tubing", presented at International Conference on Materials performance in Nuclear Steam Generators, St. Peter sbur g, Fl ori da, October 1980.

8. Copson, H. R. and Economy, G., "Effect of Some Environmental Conditions on Stress Corrosion Behavior of Ni-Cr-Fe Alloys in Pressurized Water' Corrosion, 24, No. 3, pp. 55-65 (March 1968).

9. McIlree, A. R. and Michels, H. T., "Stress Corrosion Behavior of Fe"<<->>

and Other Alloys in High Temperature Caustic Solutions", .Corr'osion. 33.

No; 2, pp. 60-67 (February 1977).

10. Sedriks, A. J., et al., "Inconel Alloy 690-A New Corrosion Resistant Material", 'Corrosion Engineering (Japan), 28, No. 2, pp. 82-95 (1979).

4 Burstein,. S., WEPCO, ltr to H. R. Denton, NRC, dtd Novembe~ 23~

with attachments.

.,ACHMENT

'

P

%rt '.Z ROCHESTER GAS AND ELECTRIC CORPORATION ~ 89 EAST AVENUE, ROCHESTER, N.Y. 14649 C  !>>O'IC i ~ Ni OaC -io,5-'9-2 OO-

~

June 23, 1981 Director of Nuclear Reactor Regulation Attention: Mr. Dennis.M. Crutchfield, Chief Operating Reactors Branch 55 U.S. Nuclear Regulatory Commission 1(p Nashington, D.C. 20555

Subject:

SEP Topics V-10.B, V-ll.A, V-ll.B, VI-7.C.lg VZX-3, and VIII-2, R.E. Ginna Nuclear Power Plant Docket No. 50-244

References:

(1) Letter from Dennis hi. Crutchfield, NRC, to John E. Maier, RGE, SEP Topics, V-10.B, V-ll.B, and VII-3 (Safe Shutdown Systems Report), May 13, 1981.

(2) Lette" from Dennis M. Crutchfield, NRC, to John E. Maier, RG"g SEP opics V-ll.A, V-ll.B, anc VX-7.C.lf dated April 24, 1981.

(3) Letter from Dennis M. Crutchfield, NRC, to John E. Maier, RGE, SEP Topics VII-3 and VIII-2, dated April 2, 1981.

Dear Mr. Crutchfield:

This letter is in response to the SEP topic assessments provided in the three above-referenced letters. Due to the intimate relationship of the "Safe Shutdown" topics V-10.B, V-ll.A, V-ll.B and VII-3 addressed in these three letters, all of our comments are provided concurrently in the three attached'espon'ses.

This should aid the inclusion of our comments into the NRC's "SEP Xntegrated Assessment".

Very truly yours, +~s ohn E. Maier Attachments I/i 810630 0'3M

RGS E responses to NRC Assessment of SEP Topics V-10.B, RHR System Reliability, V-11.B, RHR Interlock Requirements, and VII-3, Systems Required for Safe Shutdown (Safe Shutdown Systems report), May 13, 1981.

1. In RG&E's January 13, 1981 response to the NRC's November 14, 1980 "Safe Shutdown Systems" assessment, a number of comments were made which have not been incorporated into Revision 2 of this assessment, transmitted by letter dated May 13, 1981.

We feel these comments were valid, and should be incorporated.

For continuity, these comments will be listed below (with their original comment numbers):

On page 5, Pi in S stem Passive Failures, the NRC assumes piping system passive failures"...beyond those normally postulated by the staff, e.g., the catastrophic failure of moderate energy systems...". Although shown that safe shutdown following such an event could it is be achieved, it is not considered that such an evaluation Itit should even be made. As noted by the staff, is clearly beyond a reasonable design basis. is thus recommended that this paragraph be deleted from the evaluation. Subsequent evaluations to this "criterion",

such as those related to the CCW system on page 22 and 23, should also be deleted.

1n paragraph g on page 66, it is noted that, wnen a plving the power diversity requirements of BT? ASB 10-1 in event of an SSE, no means to su ply feed to the steam generators was exists. It was determined that acceptable, based on low likelihood of occurrence.

this This conclusion is correct; however, since BPT ASB 10-1 does not consider an SSE in conjunction with the loss of all A.C. power, there is no need to even make the evaluation. The comparisons in the SEP program should be to current criteria, rather than to arguable extrapo-lations. Reference to loss of all A.C. power in conjuncton with an SSE should thus be deleted from this paragraph.

12'. On page A-4, it is noted that additional systems are required to achieve cold shutdown for a PWR than for a BWR because of a difference in the definition of cold shutdown. This does not appear to be a reasonable basis. System requirements should be based on specific safety reasons. The NRC should be consistent in .its requirements for cold shutdown, or provide a technical basis for any differences."

Staff position 1 states that "the licensee must develop plant operating/emergency procedures for conducting a plant shutdown and cooldown using only the systems and equipment identified in Section 3.1 of the SEP Safe Shutdown Systems Report." RG&E disagrees with the need for these procedures.

We reiterate the comments provided in our January 13, 1981 response that the operator should perform a cooldown with the best equipment available to him at the time. If a piece of non-safety equipment is available, and would be the most beneficial for performing a required function, it is expected that this piece of equipment would be used. If it is not available, the operator could fall back on the use of safety-grade equipment. But RG&E does not intend to commit plant personnel to use only safety-related equipment, if non-safety equipment is available and more effective. We feel that it would be impossible to determine when a "safety-grade-only" cooldown procedure would ever be implemented. As long as the safety-grade equipment is available (and the safe shutdown assessment concudes that it is), RG&E considers that the necessary safety requirements are met.

RG&E also notes that no regulatory basis for this requirement is provided. It is admitted in Section 4.5 of the Safe Shutdown report that "the need for procedures for these evaluations is not identified in Regulatory Guide 1.33...".

Section 4.5 then goes on to say that the basis is found in BTP RSB 5-1 and SEP Topic VII-3. But BTP RSB 5-1 merely references RG 1.33, and this is the assessment of SEP op'c VII-3.

Therefore, since no basis for this "requirement" exists, and we do not feel that it would even be beneficial, and since the Safe Shutdown report did conclude that the capability for attaining cold shutdown using only safety-related equipment exists, RG&E concludes that this staff position should be deleted from consideration.

Staff position 3 does not appear to take into account the information provided in our March 27, 1981 submittal regarding SEP Topic V-11.A. 3 to that submittal provides

'he valve equipment Enclosure specification, noting that the 700, 701, 720 and 721 MOV's are designed such that they:physically are unable to open against a differential pressure of greater than 500 psi. This ensures that an intersystem LOCA caused by the opening of the outboard valves, plus leakage of the inboard valves, cannot occur, since the outboard valves cannot, open.

~ ~

Even how without this provision, it is difficult to comprehend the Ginna arrangement could result in an "Event V". By administrative procedure, the RHR valves are key-locked closed, with power removed. Further, interlocks are provided for the inboard RHR valves. Thus, for an "Event V" to occur would require the:

1) failure of the administrative procedure requiring power lock-out (at the breaker),

2) failure of the administrative procedure governing operation of the valve at power,

3) failure of the inboard isolation valve,

4) failure of the relief valve (RV 203) which has a capacity of 70,000 lb/hr at its 600 psig setpoint, to relieve the leakage past the inboard RHR valve.

This set of failures is considered very remote. When coupled with the fact that the RHR valve design prevents opening of the valves against a greater than 500 psi differential pressure, it is RG&E's conclusion that the possibility of an intersystem LOCA should not be a credible design basis. No additional modifications, such as diverse interlocks for the outboard valves, are warranted.

Staff position 5 states that "the operating procedures for tne Ginna plant should be modified to direct the operator to cooldown and depressurize to RHR initiation parameters within 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> whenever the Service Water System is used for steam generator feedwater..." This position is based on the reference BNL-NUREG-28147, "Impure Water in Steam Generators and Isolation Condensers." We have had this report reviewed by NWT Corporation. NWT-167, "Use of Lake Ontario Water in Steam Generator During Hot Shutdown" (attached) concludes that, "although not recommended from the standpoint of maximizing component life, and operation for periods up to several days is not expected to result in any significant cracking or in deterioration of steam generator. integrity."

RG&E therefore concludes that a specific directive to cool down and depressurize to RHR initiation conditions is not,.

warranted, and should not be included in a procedure. The capability to do this does exist, however, and could be used if determined to be necessary at the time.

RG&E responses to NRC letter of April 24, 1981 regarding SEP Topics V-ll.A, "Isolation of High and Low Pressure Systems", V-11.B, "RHR Interlock Requirements", and VI-7.C.1, "Independence of Redundant Onsite Power Systems".

The Safety Evaluation for SEP Topic V-11.A, "Requirements for Isolation of High and Low Pressure Systems", specifies that the outboard RHR valves should have diverse interlocks to prevent opening when the RCS pressure is greater than RHR system design pressure.

RG&E rationale for not providing these additional interlocks is provided in comment 3 of Attachment 1 of this transmittal.

2. The safety evaluation also required that interlocks be installed on the CVCS suction valves (200A, 200B, 202), to prevent a possible overpressurization of the CVCS letdown line outside containment. RG&E has noted in our March 27, 1981 letter on this SEP Topic that a relief valve (RV 203), with a capacity greater than the combined capacity of the three orifices, would relieve the pressure buildup caused by closure of the containment isolation valve 371. No overpressurization of the CVCS would thus be expected.

RG&E has also evaluated the potential consequences of such an overpressurization event, with a subsequent small LOCA outside cpntainment, and determined tha" no unacceptable consequences would result. This break would be a small LCCA ou side conta'ment (maximum low of, 140 gpm), and would be terminated by closure of valves 200A, 20GB, and 202 either by operator action or automatically by low pressurizer level. Radiological consequences would be minimal, since no fuel damage would result. This event is specifically evalu-ated by SEP Topic XV-16, "Radiological Consequences of Failure of Small Lines Carrying Primary Coolant Outside Containment." RG&E has provided information concerning this topic by letter dated June 18, 1980 from L. D. white Jr. to Mr. Dennis M. Crutchfield.

The RG&E conclusion is that, based on the..availability of RV 203 to prevent overpressurization, together with the lack of unacceptable consequences due to an overpressurization, no interlocks or other modifications are required for the CVCS suction valves.

3. The safety evaluation further states that position indication is required on the CVCS discharge check valves. As stated in our March 27, 1981 letter on SEP Topic V-11.A, we do not believe that this line should be classified as a low pressure

system connected to the RCS, since the piping is 2500-1b piping throughout its length (to the positive displacement charging pump). RG&E has had no experience with failures of the positive displacement charging pump pistons to hold primary system pressure, nor would any failures be anticipated.

Our contention that the charging line is not a line of concern is borne out by a memo from Edson G. Case to Raymond F. Fraley, "Isolation of Low Pressure Systems from Reactor Coolant System", dated July 11, 1977. That letter transmitted an NRC study of this subject to'he ACRS, and evaluated"all potential lines of concern. The charging line was not included.

To verify that the charging line was not a valid "Event V" concern, RG&E calculated the PWR Check Valve Event Tree (Section 4.4 of WASH-1400), using the charging line con-figuration (two in-series check valves and a charging pump piston). Very conservatively assuming that both check valves were undetected open, and that the probability of the charging pump~iston failure was equal to a check valve failure, the Q calculated for this configuration was determined to 5eU> 1.4 x 10 /year. This is a low enough, value to obviously be oz no concern.

RG&E therefore considers that check valve position indi-cation is not needed on the charging line check valves.

~;ith respect to tne ="=P .opic Assessment V-11.3, nc commen"s are necessary, since the resolution of outstanci..g issues is addressed in ti e topic assessr..en" or SEP epic V-ll.A.

The additional information requested for SEP Topic VI-7.C.1 is presently being developed. It is anticipated that this information can be furnished to the NRC by July 15, 1981.

RG&E responses to NRC letter of April 2, 1981, concerning SEP Topics VII-3, "Electrical, Instru-mentation, and Control Feature of Systems Required for Safe Shutdown", and VIII-2, "Diesel Generators".

It appears that all comments provided by RGSE in our January 1981 and January 30, 1981 letters concerning these topics 23, have been properly incorporated.

Based on the resolution of all open items, and the removal of diesel generator testing from SEP Topic VIII-2, RG6E concludes that both of these topics are complete, with no outstanding issues to be carried into the Integrated Assessment.

NWT 167 February 1981 t

USE OF LAKE ONTARIO WATER IN STEAM GENERATORS DURING HOT SHUTDOWN W. L. Pearl S. E. Copley J. Leibovitz Prepared foi Rochester Gas & Electric Company Corporation 7015 REALM ORIVE. SAN JOSE, CALIFORNIA 95119

This document was prepared for the Rochester Gas & Electric Company.

Neither the %/T Corporation nor any person acting on its behalf assumes any responsibility for liability or damage which may result from the use of any information disclosed in this document.

INTRODUCTION The possibility of using Lake Ontario water as an emergency PWR feedwater supply for more than 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> during which the plant would be brought to cold shutdown is being considered. The maximum steaming rate during such a period would be 100,000 pounds/h (200 gpm) at a temperature of 350'F. As a consequence of steaming, impurities of the untreated Lake Ontario water will concentrate in the steam generator. Of major concern is the possible risk of stress corrosion cracking (SCC) of steam generator materials in contact with the concentrated solution thus formed. To address this concern, the chemistry variation i'n the liquid phase as steaming proceeds at 3k'F was estimated with emphasis on pH.

Then, the possible potential for SCC was assessed on the basis of these estimates and available SCC data.

.pH VARIATION AT 350'F UPON STEAMING LAKE ONTARIO WATER

,A. Computer Model ing The composition of Lake Ontario water as determined by RGE is given in Table l.'ABLE 1

LAKE ONTARIO WATER ANALYSIS ppm

~ I III'al cium 35 Nitrate 2.5 Ma gnes i um Phosphate 0.3 Sodium 13 Fluoride 0.15 Potassium 3.6 Silica (as Si02) 0.25 Aluminum 0.1 Dissolved Oxygen 9.5 Chloride 32 Ammonia (as Nitrogen) 0.24 Sul fate 35 Estimates of the water chemistry variation upon steaming were developed using the following assumptions:

1. Since aluminum and silica are in stoichiometric proportion in Lake Ontario water (Table 1), they are assumed to precipitate as aluminum silicate (clay) upon concentrating 'and therefore are removed from solution.

2. Since calcium occurs in the water (Table 1) in large excess over phosphate, it is assumed to precipitate all the phosphate as calcium hydroxy apatite (Ca5(P04)30H) and remove it from the solution. The calcium in solution is decreased by the corresponding amount.

3. Fluoride and nitrite are assumed to behave as chloride. Potassium is assumed to behave as sodium.

4'. Sodium and chloride in solution are assumed to remain completely .

dissociated.

5= Calcium carbonate precipitation is neglected. Degasification of CO~

by steaming is assumed to occur.

6. The concentration of sodium and calcium chlorides is assumed limited by a solubility of 5 molal.

7. Chemical equilibrium expressions of references 2 and 3 apply.

On this basis, the liquid solution pH variation upon steaming at 350'F was estimated as a function of concentration factor defined as the mass ratio of total water (steam + liquid) to liquid water residual. The results are presented graphically in Figure l. It is important to note that the definition of pH used here is that followed by Mesmer" in the determination of the dissociation constant of water at high temperatures, viz, the negative of the logarithm of the hydrogen ion concentration (not of its activity). Similarly, neutral pH is defined as that where the hydrogen and hydroxyl ion concentrations are equal.

This neutral pH is a function of ionic strength. Therefore, the pH variation of the concentrated solutions must be considered in relation to that of neutral pH, also plotted in Figure 1. For basic solutions as is the case considered here, it is important tc bear in mind that .he hydroxyl ion concentration is expressed in terms o pH as follows:

10pH-2NpH OH (where NpH is the neutral pH value) and that when the neutral pH varies together with the ionic strength as the liquid solution is being concentrated upon steaming, the basicity of the solution may not be appreciated from the solution pH alone. The equivalent NaOH concentration is more suitable for this purpose and is plotted also in Figure l.

B. Discussion Steam Generator Bulk Water Based on a maximum feed rate of 200 gpm to the steam generator and a total steam generator liquid Mlume of approximately 12,000 gallons, a maximum of one steam generator volume is steamed away each hour. Therefore, under maximum steaming conditions, the concentration factor achieved in the bulk steam generator water is t + 1 where t is the number of hours of steaming.

3 Solution pH

~

400 p) rt 300 n

0 O

Pl Equivalent NaOH Concentration Neutral pH 6 200 3

I I

100 0

10 100 1000 10000

'l'9 me, h Figure 1. Variation of Steam Generator pH wltli Steaming at 350'F (feeding Lake Ontario water at 200 gpm)

The variation with time of the equivalent sodium hydroxide concentration in the steam generator with steaming of emergency Lake Ontario feedwater then can be followed on Figure 1. It is seen that a maximum equivalent NaOH concentration of about 300 ppm will be reached in the steam generator bulk water when 15 to 20 steam generator volumes will have been converted to steam, i.e., in approximately twenty hours. Further boiling should then decrease the equivalent NaOH concen-tration as magnesium and/or calcium hydroxides and/or calcium sulfate precipitate with increased concentrating. The decrease reaches a limit (at about 20,000 steam generator volumes converted to steam, i.e., in 20,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />) when sodium and calcium chlorides start to precipi tate also. This limit is estimated at about 100 ppm equivalent NaOH for Lake Ontario water composition as specified in Table 1 and with the assumptions already stated. The assumptions seem reasonable and; at any rate, can be tested experimentally with a small autoclave from which known amounts of Lake Ontario water would be boiled away at 350'F at constant liquid level in the autoclave.

Crevices The estimated equivalent NaOH solution concentration in steam generator crevices will depend upon the relative degree of crevice solution concentration above the bulk water. In tube to tube support plate crevices, there may be a distribution of relative concentration factors of unity and higher.

The chemistry in a crevice would lead that of the bulk in the sense that the chemistry of a specific crevice would travel the same curve (Figure 1) as the bulk but would be at a point on the curve somewhat ahead of'he bulk. Since s

the causticity, of Lake Ontario water is not a strong function of concentration, this does not pose a problem. Indeed it is expected in this case that after a short period of steaming, the crevice chemistry will be'ess basic than that of the bulk.

Coolin Water Com osition The NWT chemistry modeling work discussed herein is based on the chemical composition of Lake Ontario water suamarized in Table 1 as supplied by RGE.

It is possible that seasonal changes in the characteristics'f the lake water may result from the interrelation between source river flowrates, industrial pollution and/or acid rain. NMT has no relevant data to assess such effects-It may be desirable that analyses made of Lake Ontario water during different seasons and under various conditions be fed into the NMT chemistry model. In this manner the safety of feeding Lake Ontario water, over the range of likely chemical compositions, can be verified.

POTENTIAL FOR SCC

~

A. Corrosion The most aggressive solution expected based on the modeling work is 300 ppm NaOH, with slO ppm 02 (see. below) at 350'F. Although laboratory data regarding these exact conditions are not available, data are available which can be extrapolated to assess the maximum corrosion rates expected for a given range of conditions.

van Rooyen and Kendig~ cite Westinghouse data indicating that U-bends of Alloy 600 in deaerated 10 NaOH crack after several months of exposure.

Figures,2 and '3 summarize Westinghouse tests'hich show that at least 100 days of exposure to deaerated 10% NaOH at 600'F is required to produce a detectable crack in stressed Alloy 600.

Figure 4 shows data gathered bv Serge and Oonati.'hese curves are for yield stressed C rings at 660'F. Extrapolating the curve for mill annealed Alloy 600 to 300 ppm NaOH yields a minimum time of 3500-4000 hours to induce a 0.5 millimeter crack.

The data presented above are for deaerated systems and are consistent with van Rooyen's conclusion that Alloy 600 in 10% NaOH would not crack for several months. In the presence of oxygen, the susceptibility of Alloy 600 to SCC may be increased. Figure 5 shows stress corrosion behavior in 600'F high purity water containing varying. amounts of oxygen in the gas'hase above the water and adjusted to pH 10 at startup with ammonia. As the oxygen content of the gas phase increased, the percentof the specimens attacked and extent of'the attack increased. As noted in Figure 5 the average life in the 18-week test varied from no cracking with 1% oxygen in the gas phase (<2 ppm oxygen in the water) to 7 weeks with 1005 oxygen- in the gas phase (<200 ppm oxygen in the water).

<<Ilree and Michels and later Sedriks, et al., 'eported less than.20Ã cracking after 27 days for Alloy 600 (2 common heat treatments) in aerated, 50K NaOH at 570'F.

7

Inseuus <<sn Qress Ssiil Theres III I htpsI SsO Anneaieg Trasseg TI$ 14l

%11ÃI 4 4 Ullgg 0 4 9 .OOT UOI TI.O sa

. OIO XCO 41O lOXI OXO Tspasuse lies lhrsi Figure 2. Crack Depth as a Function of Time, Stress Level and Material Condition for ID Pressurized Capsules Exposed to Deaerated 10< NaOH at 600'F Heel  % C M.A. T.T.

gC03 .Oll STOT .OZT 35$ .03l L5 T34Z .04O

.OZ le RsOTI B LO 31VC C&isgg T-UOS B

ea ylel41 MillAnnealed CI L3 M

.01 sa LZ e

~ emO

~ He

~

I~ Thefinslly Treated OCOO fspossifo TTsie 1hfgl

'Figure 3. Crack Oepth as a Function of Exposure Time for Mill Annealed and Thermally Treated Inconel Alloy 600 Exposed to Deaerated lOX NaOH at 600 F'

4000 C Rings Stressed to a=Ys According to ASTM STP 425 3000 S- S-O 47 Alloy 600

'r 4i HT 16h 1300 F 2000

'r

<J 1000 Alloy 600 MA 0

1,000 10,000 100,000 NaOH Concentration, ppm Figure 4. Resistance to Stress Corrosion Cracking of Alloy 600 Mill-Annealed or Heat Treated at 1300'F as a Function of Deaerated Sodium Hydroxide Concentration at 600'F~

9

(Vl NO. OF SPECIMENS, MII.L ANN. B I2 l2 f2 V

,III IIIJ

'.Z I20 PERCENT CRACKED AV I.IFE IN IB WEEK AV.OF MAX. CRACK TESTS DEPTHS,MII 0

IB 0

"0'2 0

IB 0

IB 0

5B IT 2l 92 I5 50 IOO T

52

~ 1O o IOO Z

80 8 I

I-I:

4I 60 l2 8 Zg 1 4 4- ') I8

. ~,OX'<<

40 Ie 4

'1~ ZO Ie Ie 1I 18 I8 IS IS IS IS IS IS IS Ie I8 IS te 5 '4 I4 I4

0. I 14 14 14 14 14 ~ 14 14 14 14 I ALLOY 600 HEAT Nq 2785 4 2 4 78256 I 4 78256 478256 478256 OXYGEN IN GAS NIL NIL 2I IOO AT START. ('/4)

BALANCE OF GAS, HYDROGEN ARGON; NITROGEN NITROGEN AIR Figure 5. Stress Corrosion Behavior in Crevice Areas in Mill Annealed lnconel 600 Double U-bend Specimens in 600'F High Purity Water Adjusted to pM 10 with Ammonia at Startup'0

Laboratory studies show that there is a significant temperature dependence of caustic stress corrosion cracking as illustrated in Figures 6 and 7.

These results are for pressurized capsules exposed to 105 and 50K NaOH at varying stresses at temperatures ranging from 650 to 550'F. As can be seen, reducing the temperature below 600'F significantly extends the time for SCC to occur. This temperature dependence is further illustrated in Figure 8 where temperature is plotted versus rate constant for both 10$ and 50> NaOH.

B. Oxygen The lake water fed to the generators probably would be air saturated (approximately

'0 ppm 02). However-, at 350'F the KO (the equilibrium ratio between steam phase and liquid phase) for oxygen is slightly greater than 5000. Even though the dynamic distribution in practice may not reach true equilibrium conditions, the net effect of the high KO value is that recriculated steam generator coolant

'iill contain oxygen concentrations lower than 10 ppm. This recirculated coolant will dilute the oxygen concentration of incoming feedwater with a net oxygen

'evel in the downcomer of ~l to 10 ppm, depending on the recirculation ratio under the contingency conditions.

C. Conclusion With the significantly lower concentrations of sodium hydroxide (max 30o ppm),

oxygen concentration <10 ppm and the lower temperature (350'F) involved, the contingency of feeding Lake Ontario water to the Ginna steam generators should result in no measureable damage to steam generator internals. Although n<<

recommended from the standpoint of maximizing component life, such operatio~

for periods up to several days is not expected to resu1 t in any significant cracking or in a deterioration of steam generator integrity.

NaOH 50X 10X 575 F g Q .5 to 5 mils 50 40 30

>10 mils 10 5 to lp mils 0.5 to 5 10 mils 0

0 1000 2000 3000 400p 5000 6000 7000 Exposure Time, hrs Figure 6-. Caustic Cracking of Ni11 Annea1ed A1loy 600 at 575 F (Lines depict zones of crack depth from lOX NaOH at 600'F) 12

~ ~

~V NaOH 50X 10X 550'F 1't ':

0 <0.5 mils 0.5 to 5 mils g 0.5 to 5 mils 50X NaOH 650 F 50 40 30 A >10 mils

'h lj La (Q 20 5 to 10 mils 10

0. 5 to 5 mils 0

0 1000 2000 3000 4000 5000 6000 Exposure Time, hrs Figure 7. Caustic Cracking of Mill Annealed A1]oy 600 at 550 F and 650 F (Lines depict zones of crack depth from at 10'aOH 600 F) 13

200 MA I-600 150

)C c5 X

I ZI U U N

100 50 0

530 550 570 590 610 630 Temperature, 'F "igure 8. Indicated Variation in Rate of SCC with Temperature 14

i I ~ ~ ~~ ~ ~~ I ~ ~ ~ ~ ~

~

~ ~

.r i',

i f'p~

<!

.c.%

i I'

i'4$ P.

COMMENTS ON YAN Rpp>EN ND KE!iD!O'5 "EppRT The referenced reports bas<callv ;- broad summary covering a large volume it >N4 of data applicablein part to .ta:: l;ss steels and in part to Alloy 6pp. Me ic>b:

are generally in agreement with :hr r nine summary conclusions, but find it dif<<cult to apply their broad-bri~n treatment to the specifics of a pMR hot 444 c H~,~

shutdown with lake water added to he steam generators at 350'F. Their document g ~ <<f ~'C 1a is misleading for such an application in two respects:

~ ~

~*

r ra 'I

1. Caustic Concentration

v. <+e Their statement that ..."For .-;;";rposes of SCC predictions, it has to be assumed that the time to .'c.  : ngerous levels of NapH, once impurities have been introduced, is shor-.. :.=., one day or less" does not fully recognize the specific concen . '.: n chemistry of the cooling water invo1ved nor the low heat E'  : .ible and the cutback in steaming rate during a period of hot shutco'~> n :.he case of the Lake Ontario water, for example, the maximum .laCH =," =.ntration reached is 300 ppm (after steaming ~20 steam genera or u ..'s; ".<ith a decrease in concentration thereafter.

2- Temperature All of the test work referenced in -;he referenced report'as performed in the temperature range of 550 to =.G=F- With the significant temperature dependence of caustic- SCC as shown above, the concern at 350'F is many times less than is indicated lcm the data quoted by the authors.

r Hased on the above three considerations, it is our assessment that the generalized time limit of 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> in =he report's not directly applicable to the Ginna steam generators steami"qt 350'F while fed by Lake pntario water.

15

.REFERENCES

1. Har hay, A., Rochester Gas In Electric, Personal Communication, February ll,

.1981.

2. Leibovitz, J., and Sawochka, S. G., "Modeling the Effects of Condenser Inleakage on PWR Chemistry", presented at 41st Annual International Mater Conference, Pittsburgh, Pennsylvania, October 1980.

3. Leibovitz, J. and Sawochka, S. G., "Modeling of Cooling Water Inleakage Effects in PWR Steam Generators, Topical Report, Research Project 404-1",

Electric Power Research Institute, May 1980, to be published.

4. Mesmer, R. E., Baes, C. F., Jr., and Swekton, F. M., "Boric Acid Equilibria and pH in PWR Coolarrts', Proceedings of'the 32nd International Water Con-ference, Pittsburgh, Pt.nnsylvania, November 1971, pp. 55-65.

5. van Rooyen, D., and Keadig, M. W., "Impure Water in Steam Generators and Isolation Generators", Brookhaven National Laboratory, June 1980 (Draft-BNL-NUREG-28147).

6. Airey, G. P., "Effect of Processing 'lariables on the Caustic Stress Corrosion Resistance of Inconel Alloy 600", presented at NACE Meeting%

March 1979 (Paper Number 101).

7. Berge, Ph. and Donati, J. R., "Materials Requirements for Steam Generator Tubing",=presented at International Conference on Materials Performance in Nuclear Steam Generators, St. Petersburg, Florida, October 1980.

Copson, H. R. and Economy, G., "Effect of Some Environmental Conditions on Stress Corrosion Behavior of Ni-Cr-Fe Alloys in Pressurized Water' Corrosion, 24, No. 3, pp. 55-65 (March 1968).

9. McIlree, A. R. and Michels, H. T., "Stress Corrosion Behavior of'Fe-Cr>>

and Other Alloys in High Temperature Caustic Solutions", .Corrosion. 33 No. 2, pp. 60-67 (February 1977).

10. Sedriks, A. J., et al., "Inconel Alloy 690-A New Corrosion Resistant Material", "Corrosion Engineering (Japan), 28, No. 2, pp. 82-95 (1979}.

4 Bursteln, S , WEPCO, ltr to H. R. Denton, NRC, dtd November 23, 1979, with attachments.

16

"'Age If I

~ I g