ML20040A994
| ML20040A994 | |
| Person / Time | |
|---|---|
| Site: | Point Beach  |
| Issue date: | 01/15/1982 |
| From: | Fay C WISCONSIN ELECTRIC POWER CO. |
| To: | Harold Denton Office of Nuclear Reactor Regulation |
| References | |
| RTR-NUREG-0737, RTR-NUREG-737, TASK-2.K.2.13, TASK-TM TAC-46897, TAC-46898, NUDOCS 8201230043 | |
| Download: ML20040A994 (14) | |
Text
-_
i i
i
\\%sconsin Electnc eom coumr I
231 W. MICHIGAN, P.O. BOX 2046. MILWAUKEE, WI 53201 i
anua 5, 1982 j
qu '
c),
U l
?
3 s il m
1 Mr. Harold R. Denton, Director p f ' 1, Y ?
r
)
Office of Nuclear Reactor Regtd ation
-t W i7 b82a
~
J U.
S. NUCLEAR REGULATORY COMMISSION C.
g#
]
Washington, D.
C.
20555
- C i
/\\
%v..[i(\\ ' '
Gentlemen:
I DOCKET NOS. 50-266 AND 50-301 REACTOR VESSEL RESPONSE TO THERMAL SHOCK j
POINT BEACH NUCLEAR PLANT, UNITS 1 AND 2 t
i One of the post-TMI documents issued by the Nuclear Regulatory Commission was NUREG-6737, entitled " Clarification j
of TMI Action Plan Requirements."
Contained in this do::ument l
is item II.K.2.13 which required an analysis of the thermal-mechanical conditions in the reactor vessel during postulated i
small break transients.
This analysis was required of all 1
pressurized water reactor (PWR) licensees and a submittal was i
requested by January 1, 1982.
With agreement of the NRC Staff, it was decided that this issue would be treated generically by the several reactor vendor owners groups.
The intent of tht j
generic analysis was to provide a conservative assessment of j
reactor vessel integrity to postulated thermal transients and
]
to demonstrate the minimum amount of time for licensees to develop logical and orderly plans for addressing these transients i
in the event that potential problems were identified.
It was
(
realized that this generic treatment which utilized conservative material property categories would introduce broad analysis j
conservatisms that might be overly restrictive when applied to any specific reactor vessel.
4 The results of the Westinghouse Owners Group generic j
analysis, in the form of WCAP-10019, which satisfies the NUREG-0737 l
item II.K.2.13 requirement, were recently transmitted to you by a
letter dated December 30, 1981 from Mr. O. D. Kingsley, Jr.,
l Chairman of the Westinghouse Owners Group.
Included in the generic report is Table III.2.1 which indicates the minimum acceptable times before additional analyses or remedial actions i
must be taken for each Westinghouse designed reactor vessel.
i This table indicates that the Point Beach reactor vessels are
)
acceptableforaminimumoffouryears(toaboutOctober1985)[O before further action is required. Furthermore, the generic j
analysis contains many conservatisms and we believe future
./I 4
8201230043 820115
?
PDR ADOCK 05000266 P
r o
Mr. Harold R. Denton January 15, 1982 plant-specific analyses will substantially extend these time intervals. to this letter presents a discussion of some of the differences between the generic analysis performed by Westinghouse and the actual material properties, system configurations and previous analyses appropriate to the Point Beach Nuclear Plant.
For example, Point Beach Units 1 and 2 have been operated for the last two years with cores designed for low neutron leakage.
While a specific analysis has yet to be performed to determine the exact reduction in neutron fluence at the reactor vessel inner wall, section 5 of WCAP-10019 discusses this approach and states that this could reduce the neutron flux (or increase the remaining time before additional analyses need to be completed) by a factor of 2.
Also, the Point Beach Unit 1 reactor vessel materials surveillance program has demonstrated significantly lower temperature shifts as a function of irradiation than either the Westinghouse prediction curve or the method of Regulatory Guide 1.99 which was used for the small break transient analyses.
This lower temperature shift would result in significantly longer minimum times before additional analyses need to be completed for the Point Beach reactor vessels.
While we believe that further plant-specific reviews are required, and we intend to initiate those reviews in the near future, it is our conclusion that the generic results in WCAP-10019 are quite conservative and clearly demonstrate that the Point Beach reactor vessels are safe for continued operation.
Very truly yours, l'v/
/
C. W.
Fay Assistant Vice President Enclosure Copies to Mr.
C. F. Riederer (PSCW)
Mr. Peter Anderson (WED)
NRC Resident Inspector
ATTACHMENT 1 POINT BEACH NUCLEAR PLANT REACTOR VESSEL INTEGRITY BY WISCONSIN ELECTRIC POWER COMPANY A.
INTRODUCTION The Westinghouse Owners Group report on reactor vessel integrity (WCAP-10019) describes the generic approach to reactor vessel analysis and provides results of that analysis in general terms. The report was prepared and submitted to NRC in response to NUREG-0737, item II.K.2.13.
This attachment comments upon the results of the generic analysis, as well as the results of the previous fracture toughness analysis performed for the Point Beach Nuclear Plant reactor vessels.
This attachment also identifies some specific generic analysis conservatisms and discusses potential future actions which may be taken to preclude adverse effects on the reactor vessel from postulated thermal transients.
Plant-specific fracture mechanics evaluations for normal operation and accident conditions have previously been performed for the reactor vessels of both units of the Point Beach Nuclear Plant (PBNP).
The accident conditions evaluated in these previous analyses were the large Loss of Coolant Accident (LOCA) and the large Steam Line Break (SLB).
The results of these analyses were provided to the NRC by a Wisconsin Electric letter to Mr. B. Rusche, dated March 4, 1977.
This transmittal provided nine Westinghouse WCAP reports (-8735 through -8743) pertaining to heatup and cooldown curves, fatigue crack growth evaluations, ASME III Appendix G analyses, large accident fracture mechanics evaluations, and the material testing results from Unit 1 Capsule S.
B.
CONSERVATISMS Section IV.1 of WCAP-10019 discusses conservatisms which are inherent in a generic assessment of reactor vessel thermal shock.
Certain of these conservatisms are specifically applicable to the Point Beach reactor vesse2s as discussed in the following sections:
1.
Actual Materials Data:
The material surveillance program for each PBNP reactor vessel consists of five specimen capsules (plus one standby capsule).
To date, three capsules have been removed from each Page 1 of 12
vessel and the specimens tested.
Figure 1 is extracted from the last Unit 1 capsule test report (WCAP-9357, transmitted to NRC by Wisconsin Electric letter of November 8, 1978) and is a summary plot of the Unit 1 capsule test data versus the transition temperature shift curves. The data points are all below the appropriate Cu weld curve; the copper content of the weld metal specimens is 0.24% (see WCAP-8743).
- Thus, the Westinghouse prediction curve is conservative with respect to PBNP Unit 1.
These curves are also conservative with respect to PBNP Unit 2.
2.
RT Prediction Curves:
The generic fracture mechanics anNSfsesforthesmallLOCA, small SLB, and Rancho Seco Transient were performed utilizing the reference temperature adjustment method of Regulatory Guide 1.99.
The Regulatory Guide data are based upon high flux irradiation data in the high fluence regime and appear to overpredict the temperature shift for power reactors (lower flux irradiation) as the fluence approaches the end-of-life conditions. In a March 1979 EPRI Special Report (NP-1103-SR; Pressure Boundary Technology Program:
Progress 1974 Through 1978; Section 5), this is documented along with projects being initiated to attempt to resolve the various technical parameters.
As the Point Beach Unit 1 surveillance program has shown that the material is conservatively bounded by the Westinghouse temperature increase prediction curve, this curve has been the basis for predicting conservative, but more realistic, irradiation effects for the Unit 1 vessel.
The following brief comparison demonstrates the magnitude of the conservatisms in the two prediction curves.
Actual Capsule R, 30 ft-lb temperature shift 165*F Applicable Westinghouse curve predicted shift 240*F Regulatory Guide 1.99 predicted shift 330*F The radiation fluence on the Unit 1 reactor vessel as a function of time is presented in Figure 2-2 of WCAP-8743.
The fluence on the veggel wa}l will reach the Capsule R exposure (2.22 x 10 n/cm ) at about 18.5 effective full power years (EFPY) (about 23 calendar years at 80% capac periodof10EFPY(1.2x10{gyfactor). From the tigg),
) to 32 EFPY (3.8 x 10 the vessel wall gccumu}ates fluence at a rate of y
about 0.12 x 10 n/cm per EFPY.
Page 2 of 12
l The use of temperature shift prediction curves in the generic analysis is briefly discussed in section 3.b of WCAP-10019.
However, the use of the Regulatory Guide curve is overly conservative.
The above-noted prediction curve difference (90'F) can be worked backwards into operation for the full design life of l
the reactor vessel starting with the 7.78 EFPY noted in Table III.2.1 of the generic analysis report.
This is done by determining the vessel wall fluence for 7.78 EFPY, determining the predicted temperature shift, adding the 90*F prediction curve difference, determining the fluence necessary to obtain this temperature shift, and finally observing that the resulting fluence is greater than the estimated Point l
Beach end-of-life vessel wall fluence.
Thus, in the fluence range of interest, the use of the Regulatory I
Guide prediction curve can make a difference of about
{
l 24 EFPY (or 30 calendar years at 80% capacity).
I While the foregoing comparison is not directly applic-I able to the accident analyses as a stress evaluation I
is not included, it does demonstrate the impact of the conservatism imposed by Regulatory Guide 1.99.
Similar conservatisms apply to PBNP Unit 2.
3.
Low Neutron Leakage Fuel Cycles:
For Point Beach Unit 1, Cycles 1 through 7 (1970 through 1979), new fuel was located on the core periphery as was contem-plated in the original design.
Beginning with Cycle 8 (1980), core loading patterns employed a Low Leakage Loading Pattern (LLLP) design and assemblies with several previous cycles of burnup were positioned at certain locations on the core periphery.
The nearest fuel assembly locations with respect to the Unit 1 reactor vessel longitudinal welds are referred to as positions H-13 and F-1.
Table 1 presents a simplified burnup history of the Unit 1 fuel assemblies located in these two positions.
Thus, the neutron exposure of the Unit 1 longitudinal welds for the last two years has been reduced below the fluence levels which have been predicted.
The LLLP was also fully imple-mented for PBNP Unit 2 for Cycle 7 (1980).
4.
Three Dimensional Modeling:
As discussed in section VI.2 of WCAP-10019, three dimensional thermal hydraulic models have not yet been developed for analysis purposes.
Thus the analysis of temperatures, times, and flow directions does not presently account for the location of the longitudinal welds with respect to the reactor vessel nozzles. Figure 2 is a schematic representation of the inside circumference of the PBNP Unit i reactor vessel.
Figure 2 was developed from the drawings contained in the vessel technical manual and shows the relationship of the inlet and outlet nozzles with respect to the vessel beltline Page 3 of 12
For the large and small LOCA thermal-hydraulic evaluations a hot leg break is assumed with safety injection flow into the cold legs.
As illustrated by the figure, the weld centerline is radially offset from the inside radius of the cold leg nozzle by about 35 inches.
The PBNP Unit 2 reactor vessel has no longitudinal shell welds.
5.
Weld Location, RT and Chemistry: The Point Beach reactor vessels wE9I constructed to the applicable codes in effect when the vessels were ordered.
At that time the material properties testing requirements were not oriented to the fracture toughness properties currently required.
However, as part of the fracture mechanics evaluations performed and submitted to NRC in March 1977, conservative estimates of RT for the Point Beach reactor vessel materials ann beldments were developed by the methods of Section 5.3.2,
" Pressure-Temperature Limits" of the NRC Standard Review Plan.
These values are documented in WCAPs l
-8743 and -8737 for Units 1 and 2, respectively.
The l
weldment properties are reproduced in Table 2 attached hereto.
For Unit 1, the limiting material is the longitudinal weld in the lower shell, and for Unit 2, it is the circumferential weld between the inter-mediate and lower shells (Unit 2 does not have longitudinal welds).
It should be noted that there is a substantial differ-ence in the chemistries of the two longitudinal welds in Unit 1.
The weld closest to an inlet nozzle (intermediate shell), where the cold water enters the reactor vessel, is a lower copper content weld and will have higher fracture toughness values as a function of time than the lower shell longitudinal weld.
This fact, in combination with three dimensional thermal hydraulic analyses, should be beneficial to the results of thermal shock analyses for Point Beach, Unit 1.
6.
Assumed Analytical Flaw:
As discussed in the fracture mechanics analysis section of the generic report, a surf ace flaw is assumed to exist.
However, as part of the previous overall PBNP reactor vessel program, an inservice inspection of the beltline welds was performed.
The inspections revealed some indications and ASME Code allowable reflectors, but none indicated any measurable through-wall dimension.
Thus, it has been verified that no significant flaw exists in the weldments of either Point Beach reactor vessel.
The Unit I results were reported to the NRC by a Wisconsin Electric letter to Mr. E. Case dated August 18, 1977.
Page 4 of 12
C.
ANALYSIS RESULTS Results of the generic analysis are summarized in Table III.2.1, page 122, of WCAP-10019 for each Westinghouse plant by listing the remaining minimum years of additional operation before further analyses or remedial actions need be completed with respect to thermal shock transients.
The most limiting case using the generic analytical model is the small SLB transient for each of the Point Beach units.
The generic analysis shows acceptable results for at least four additional calendar years or through at least October 1985.
It is to be noted that the generic analysis large LOCA result for Point Beach is presented as six years without the benefit of the warm prestressing phenomenon.
This is essentially the same number as presented in WCAP-8742 trans-i mitted to NRC in March 1977.
Appendix A of that report addresaed l
warm prestressing and, on the basis of the analysis at that time, a preliminary evaluation indicated that warm prestressing would enable the Unit 1 reactor vessel to demonstrate acceptable results during the large LOCA for the design life of the plant.
A more detailed recent evaluation specifically performed for Point Beach and utilizing the fracture mechanics techniques of WCAP-10019 and warm pre-stressing has shown the Unit 1 vessel as acceptable for the full design life.
Warm prestressing is discussed in section 3.b.ix of the generic report (WCAP-10019) and is not repeated here.
D.
PBNP OPERATOR AWARENESS OF THERMAL SHOCK The operators at Point Beach are trained to be aware of reactor vessel thermal shock during normal and accident operating conditions.
The normal and emergency operating procedures have steps that directly address, and deal with, the integrity of the reactor vessel.
The operators receive training in these procedures and the bases for each step are explained.
Some examples of where the reactor vessel integrity concern is addressed in plant procedures are as follows:
1.
The heatup, cooldown and normal operating procedures require the plant to be maintained within the limits of pressure and temperature as defined by curves in the plant Technical Specifications.
These curves ensure proper margins of safety for reactor vessel integrity.
The operators are trained in the use and meaning of these curves.
Should a transient occur that would take the plant into the unacceptable area of this curve (e.g.,
temperature too low or pressure too high for the other parameter), the operator is trained to terminate the initiating event and return the plant to acceptable operating conditions.
Page 5 of 12
2.
During the cooldown procedure and when the plant is in the ccoled down condition, the operating procedures require a low temperature overpressure protection system to be activated.
This system provides vessel integrity protection against pressure transients when in the cooled condition.
The operators are trained in the use and basis for this system.
3.
The steamline break emergency operating procedure provides an additional set of primary system conditions under which safety injection can be terminated and thus minimize, or eliminate, the pressure stress during this thermal transient.
Should a steamline break transient result in a significant cooldown of the plant (to less than 350*F), safety injection can be terminated when primary pressure goes above 700 psig.
This prevents the low temperature and high pressure condition from occurring simultaneously and provides protection to the vessel. The operators are trained in the use and basis for this special safety injection termination criteria.
E.
FUTURE ACTIONS Wisconsin Electric has under consideration a number of future actions that may be taken to demonstrate that reactor vessel thermal shock is not a limiting factor for Point Beach Units 1 and 2 for the expected life of the plant.
These considera-tions include both physical modifications that can mitigate the effect of postulated transients and refined analyses to more accurately predict the integrity of the Point Beach reactor vessels.
These actions are summarized as follows:
1.
Safety Injecticn System Configuration: The Point Beach units presently utilize a safety injection system which injects water into both cold legs and directly into the reactor vessel in the upper plenum region.
The upper plenum is isolated from the reactor vessel wall by the vessel barrel which forms an annular ring separating the cold leg inlet flow from the hot leg outlet flow.
Low pressure safety injection is piped to the upper plenum and will inject water when the primary system pressure drops celow 150 psig.
The accumulators are piped to tne cold legs and inject water when reactor coolant system pressure drops below 700 psig. High pressure safety injection is normally piped to the cold legs. However, in the Point Beach units, it is also piped to the upper plenum region. The high pressure safety injection system injects water when the reactor coolant system pressure drops below 1500 psig.
The normal injection path for the high pressure safety injection system is l
into the cold legs. However, by changing the valve lineup, the water can be injected into the upper l
Page 6 of 12
plenum rather than into the cold legs.
This would allow the water entering the upper plenum to mix with the warm water in and above the reactor core, thus eliminating the possibility of cold water causing vessel wall thermal shock.
Moving the injection location from the cold leg to the upper plenum would be effective in preventing reactor vessel thermal shock for any transient where the primary system pressure remains above 700 psig.
Since the primary system pressure remains above 700 psig for the small break LOCA, the Rancho Seco transient and the small SLB, a change in injection location would resolve any vessel thermal shock concern for these transients.
2.
Continuation Of Low Leakage Loading Pattern: The low leakage loading pattern as described previously was instituted primarily for fuel cycle economic reasons.
However, since it does result in a lower neutron flux at the vessel wall, the Point Beach units will continue to be operated with this type of loading pattern.
3.
Refueling Water Storage Tank Heating: Heating of the refueling water storage tank to increase the temperature of the major source of injected water to a level that reduces the potential for thermal shock is feasible, although the exact manner by which this can be accomplished i
remains to be investigated, as does the benefit which could be derived by this change. The generic report indicates that substantial benefit can be achieved by heating the refueling water storage tank.
4.
Other Considerations: The generic report described additional modifications that may reduce the potential for reactor vessel thermal shock.
In general, these may be applicable to the Point Beach Nuclear Plant, but need to be investigated further.
F.
SUMMARY
The discussions presented in this attachment and in the generic reactor vessel integrity report (WCAP-10019) show that the Point Beach Nuclear Plant reactor vessels do not have an immediate thermal shock concern.
We have described herein the additional evaluations of plant-specific features which can be accomplished for Point Beach and which, in our judgment, will demonstrate that the reactor vessels will maintain their integrity under postulated accident conditions for a significant amount of time beyond that shown in the generic analyses.
Also 1
described is a three-dimensional analysis that could be performed specifically for the Point Beach reactor vessels to demonstrate further the ability of the reactor vessels to withstand thermal shock.
We also discussed certain changes to the plant that could further reduce the potential for reactor vessel thermal shock.
We intend to perform the additional reviews described Page 7 of 12
l and we will take the appropriate actions in the future to assure the continued safe operation of Point Beach Nuclear Plant for its designed lifetime.
I Page 8 of 12
1 TABLE 1 POINT BEACH NUCLEAR PLANT, UNIT 1 NEUTRON FLUENCE NEAR REACTOR VESSEL LONGITUDINAL WELDS PERIPHERAL ASSEMBLY HISTORY CYCLES OF PREVIOUS BURN AT BEGINNING OF INDICATED CYCLE Location Cycle 15*
195*
Number I-12
- H-13 E-2
- F-1 1-7 0
0 0
0 8 (1980) 0 3
0 3
9 (1981) 0 3
0 3
10 (1982) 0 4
0 4
- Major contributor to fluence at RV longitudinal weld.
Note:
Point Beach Unit 2 reactor vessel has no longitudinal weld seam.
Page 9 of 12 l
{
TABLE 2 POINT BEACH NUCLEAR PLANT REACTOR VESSEL WELDMENT UNIRRADIATED bT AND CHEMISTRY The following data are extracted from Tables 3-1 in WCAPs
-8743 and -8738 except for the nickel content which is from WCAPs
-7513 and -7712 for Units 1 and 2, respectively.
The Unit 1 nickel content is reported simply as weld metal and, thus, the value is assigned to all welds.
Estimated Chemistry (Wt %)
Unirradiated Item Cu P
Ni RTNDT ( F)
A.
Unit 1 l
1.
Longitudinal Weld 0.12 0.017 0.57 30 Intermed. Shell 0.19 0.024 0.57 30 2.
Circumferential 0.21 0.021 0.57 10 Weld Inter. to Lower l
Shell 3.
Longitudinal Weld 0.20 0.012 0.57 30 Lower Shell B.
Unit 2 j
1.
Circumferential Weld 0.25 0.018 0.59 30 Inter. To Lower Shell i
Page 10 of 12
WCAP-9357 ANALYSIS OF CAPSUr.E R FROM THE WISCONSIN ELECTRIC POWER COMPANY POINT BEACH NUCLEAR PLANT, UNIT NO. 1 REACTOR VESSEL RADIATION SURVEILLANCE PROGRAM 1-o~
-e m
.=f ts C
i r
e J
.O l.$
a o
i 55 O
d m*
~
s_
c m
.o LL. e O
3 E ni o
o
=e
^
.m 3
-2 m
sr
~
o
-o.5 s* 3:
,v
_ 5S a.
3_
e o w
O.c h
O Wz5
. g.: r a
e-O Qu i
- O 9
w
.. ~ -
o 6o g
m &,a
- m. a
=
< =
o G
V O
i EE y
es R1R 5:
a 6 66
- : =~z i
y *,
- 1 e on
==o r
e o
_.ase.s a
t m,
m
.r
$<OO>O i
4.
i 1
j
'iI i
I i
1 5_
8 8
-8 E
S 9
R 9
- r n
(do) 3SY383Ni 3bO1783dH31 N0lilSNVH1
.g -
-i t
,i>*
FIGURE 1
'P e
Pac e 11 o' ? '
~
~
. S H Ef f.
CF I'
CALCULATION SHEET
~
FILE No.
SUBJECT PBNP Unif 1 R e#c+e v-
\\/esstI mADE sy b_t.D4 Y2Me/
DATE CHKD.BY b
!/ 3 $ Z DATE e
VesneI/H<s.A.
F t o.9e I
w x
y E
270*
O' to"
/8 0*
270*
l l
1 1
I a
I l
l 1
l 1
73 '/y 31.5 W l
- 31. 5
- h 28.5* '
H 28./
8 g
3 ist L
- st L
-+
_ Lg q. _._ _ 4
_4 g
4
_4. _
sr I
s:
Inlet o.t let I n tet o wlet A
e i
t j
~
8 I
I
~ 31 tye i
i I
8 t
T g
^
i i
i
-*l /[ ~
l j
.3 fl',
g l
3 e
94 ks In+. e.,
s h. //
w I
I I
+ 4.n l '
long;18a i
l to =.
I
,r I
I I
E I
I I
A i
t l
---w f5*
w 9L f.
t i
L.o w e I
3 S he ll l
i i
i i
i i
e l
l y
l 4 / 'f. 69 "
l F IG U R E.
2 Page 12 of 12