B14327, Application for Amend to License DPR-61,revising TS Re Steam Generator Repair Criteria,To Reduce Unnecessary Plugging of SG Tubes in Tubesheet Expansion Zone Roll Transition Area
| ML20128E541 | |
| Person / Time | |
|---|---|
| Site: | Haddam Neck File:Connecticut Yankee Atomic Power Co icon.png |
| Issue date: | 01/29/1993 |
| From: | Opeka J CONNECTICUT YANKEE ATOMIC POWER CO., NORTHEAST UTILITIES |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| Shared Package | |
| ML20128E546 | List: |
| References | |
| B14327, NUDOCS 9302100506 | |
| Download: ML20128E541 (30) | |
Text
_- __ __ - __ _ ._ _
NORTHEAST UTILITIES o.nor.i Ome.. . s.io.n su..t o.rm Conn.cucu
[OErORD. CONNECTICUT ($141-027 L ' J $; ";%ll"1%"" (203) 66H000 January 29, 1993 Docket No. 50-211 B14327 Re: 10CFR50.90 U.S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, DC 20555 Gentlemen:
Haddam Neck Plant Proposed Revision to Technical Specifications Steam Generator Repair Criteria Pursuant to 10CFR50.90, Connecticut Yankee Atomic Power Company (CYAPCO) hereby proposes to amend Facility Operating License DPR-61 by incorporating the changes identified in Attachment 2 into the Technical Specifications of the Haddam Neck Pl ant. The proposed amendment to the Facility Operating License is similar to CYAPCO's request of July 31, 1992,'" with two exceptions. The first exception is the permissable postaccident leakage allowed has been revised from 100 gpm to 1.26 gpm per steam-generator. This-change significantly reduces the projected off-site dose consequences for postulated steam-line breaks. The second exception is a revision to the assym4d design basis differential-pressure across the steam generator tubes used for leakage calculations. This change supports the r< vised Icakage limits.
It[IM)UCTION The reason for the' proposed change is to re' duce unnecessary plugging of the Haddam Neck Plant steam generator tubes in the tubesheet expansion zone (EZ) roll transition area. This change will have the..'aneficial effect .of minimixing further reductions in thermal hydraulic operating margins and plant efficiency.
In addition, radiation ex)osure, cost, and outage time associated with the-plugging of these tubes will be reduced. - The proposed change would also allow .
the unplugging and returning to service of tubes which were previously plugged .
due-to cracks in the expansion region,-with its associated improvement in the ,
thermal hydraulic operating margins and. plant efficiency.
(1) J. F. Opeka letter to the U.S. Nuclear Regulatory Commission, "Haddam Neck Plant, Proposed Revision to Technical Specifications-Steam Generator' ?
Repair Criteria," dated July 31, 1992.
ossof ncy.44s : 93021 5 N0129 I PDR ADOCK 05000213- +--
, ? P.- . P DR :
. w
U.S. Nuclear Regulatory Commission B14327/Page 2 January 29, 1993 M WiRQUED Primary water stress corrosion cracking (PWSCC) of alloy 600 steam generator tubes has been diagnosed in the tuberheet EZ roll transition area of steam generator tubes in many pressurized water reactor plants. When existing tube repair (plugging / sleeving) limits based on crack depth are applied to axial cracks, many tubes may require repair which is unnecessary from either a safety or reliability standpoint. Allowing tubes with PWSCC to remain in service can be justified based on a combination of enhanced in service inspection, a repair limit based on crack length rather than crack depth, and a limit on the number of tubes with characterized cracks retained in service.
The combination of augmented inspections and an acceptable crack length-based repair limit provides an alternative to the current 50 percent depth-based repair criterion while maintaining safety and acceptable reliability for the Haddan Neck Plant steam generators which have experienced PWSCC. The criteria used in the proposed technical specification changes are in addition to the degradation specific evaluation criteria in the existing technical specifications which have provided significant operational benefit while maintaining adequate safety margins.
The length-based repair limit for PWSCC induced axial cracks described herein was ado)ted from the generic criteria that was developed for steam generator tubes wit 1 full or partial depth roll EZs. For tubes with partial depth roll EZs, as in the Haddam Neck Plant steam generators, confinement of the cracked section of the tube within the tubesheet precludes tube burst at the flaw location, in this case, the repair limit may be based on maintaining acceptable leakage limits at faulted load for the distribution of cracked tubes which are retained in service.
This proposed change applies to axial or inclined axial cracks where the axial length of the inclined crack is gret.ter than the circumferential length. Tubes with identified distinct circumferential cracks will be repaired. In addition, multiple axial cracks will be limited such that the length of possible undetected circumferential cracks remains acceptable.
The elements of t.he proposed approach include:
. At each scheduled inspection outage, performance of eddy-current inspections of 100 percent of the tube roll EZs in regions of the steam generator where the tube roll expansion zones are susceptible to PWSCC.
i .
U.S. thclear Regulatory Commission B14327/Page 3 January 29, 1993
. Repair of tubes with axial cracks longer than a conservatively established repair limit which includes the following elements:
-- Use of a tube rupture curve bounded by tube burst test correlations for cracks located above the top of the tubesheet,
-- Application of Regulatory Guide 1.121 safety factors,
-- Use of lower bound tube material properties,
-- Correction for tubesheet constraint,
-- Allowance for crack growth between inspections, Allowance for eddy-current crack length measurement uncerttinty, and
-- Restriction on spacing between adjacent axial cracks.
. Calculation of the potential theoretical leak rate expected during postulated accident loads from the cracked tubes that remain in service.
. Calculation of maximum allowable site-specific leak rate at postulated faulted load to ensure that a small fraction of 10CFR100 dose limits are not exceeded.
Implementation of these elements constitutes a defense-in-depth approach that was developed to ensure adequate levels of safety. 1he inspection scope and procedures, crack length and distribution limits, and the leak rate limits developed for tubes with PWSCC ensure adequate margins against failure and ensure calculated 10CFR100 doses within acceptable levels.
The bonnfits of this change are reduction of occupational radiation exposure due to tube plugging, minimizing further reductions in thermal margins due to plugging of steam generator tubes, minimizing the reduction in plant efficiency due to tube plugging, and reduced cost and outage durations. The Advisory Comittee on Reactor Safeguards (ACRS) has previously reviewed the conceptual industryapgroachthatjustifieskeepingtubeswithcertainthroughwallcracks in service. 3 They concluded that: "The continued use of the 40 percent death limit as a repair limit results in a large effort by the licensees and a significant exposure _to workers, and leads to the repair of many tubes that have (2) Meeting among ACRS, NRC, EPRI, if, Labnrela, and Alabama Power Co.,
Bethesda, November 6, 1991.
......s.
..- - " m.- - -.
U.S. Nuclear Regulatory Commission B14327/Page 4 January 29, 1.993 a negligible risk of failure." This industry approach was subsequently documented in detail,',' and is the basis for the approach provided in this request.
OlLCUhkl9B A. Existina Technical Specifications The existing technical specifications prescribe plugging limits. One portion of the plugging limits permits imperfections below the uppermost-one inch of sound roll. Imperfections above the uppermost one inch of sound roll within the tubesheet have a plugging limit criterion equal to 50 percent of the thickness nominal tube wall. The technical specifications also set forth tube sample selection criteria for inspections.
B. Description of Proposed Chance The proposed change modifies the requirements for repairing or removing frnm service defective steam generator tubes. The proposed changes would allow tubes with certain well characterized axial cracks to remain in service. The allowable cracks are limited in length and circumferential extent, but are conservatively assumed to be through the tube wall. Tubes with cracks characterized in this fashion have been shown to be structurally sound and exhibit acceptable levels of leakage with the required safety margins. .The number of tubes with cracks to be retained in service is limited by the primary-to secondary leakage limits. During normal operation. leakage from all steam generator tubes, including any tubas with cracks retained in service, must remain below the specified operational limits. The chiculated theoretical leakage from these cracks under postulated accident conditions is limited such that the dose contribution from the aggregate leakage will remain a small fraction of the 10CFR100 dose limits.
SAFETY ASSESSMD[I CYAPC0 has reviewed the implementation of this revised steam generator tube:
plugging limit to assess the impact on the accidents evaluated as part of the design basis, the potential for creation of a new unanalyzed event, and the impact on the margin of safety, CYAPC0 has determined that the proposed design-would introduce a malfunction of a different type than previously evaluated.
(3) Ward, David (Chairman),," Steam Generator Tube Repair Limits," Letter to Selin, Ivan, November 15, 1991.
(4) EPRI, "PWR SG Tube Repair Limits for PWSC, in Roll Transitions," Rey, 1, NP6864-L, December, 1991. .
U.S. Nuclear Regulatory Commission B14327/Page 5 January 29, 1993 tiowever, CYAPC0 has determined the proposed design change to be acceptable and ;
safe. CYAPC0 has reviewed the proposed technical specification changes _in accordance with 10CFR50.92 and has determined that the changes do not involve a significant hazards consideration. Along with this amendment request, supporting ,
documentation is provided as follows:
. Attachment 1 provides the safety assessment for the proposed I design changes.
. Attachment 2 forwards the revised pages of the technical specifications.
. Attachment 3 provides the significant hazards consideration determination.
The safety assessment concludes that a plant-specific application of generic alternate repair criteria provides the same margin of safety for steam generator tube structural design with respect to bursts. The plant-specific a) plication of the generic alternate repatr criteria provides a calculated tieoretical leakage rate for steam generator tubes having cracks that are permitted to remain in-service. The accident which is affected by this proposed change is a msin steam line break (MSLB). When the calculated theoretical leakage rate is used to evaluate the dose consequences of a MSLB, dose consequences are significantly less than 10CFR100 limits. it should be noted that projected-off-site doses would be less than 10CFR100 limits, even if calculations are performed using the maximum permissable leakage cate and. Standard Review Plan 15.1.5, Appendix. A methodology. Further details are provided in Attachment 1.
H a ce, these proposed technical specification changes are considered safe.
Summary For the reasons identified above, CYAPC0 has concluded that operation of the facility in accordance with the proposed amendment would not involve a significant hazards consideration.
Moreover, the Commission has novided guidance concerning -the application of standards in 10CFR50.92 by providing certain examples (March 6,1986,51FR7751)-
of amendments that are considered not likely to involve a significant hazards consideration. Based on the above and Attachment 3 CYAPC0 believes that example (vi) encompasses our change. Example _ (vi) addresses a change .which either may result in some increase to. the probability or consequences of a previously analyzed accident or may reduce in some way a safety margin, but where the results of the change are clearly within all acceptable criteria with respect to the system or component trecified in the Standard Review Plan, e.g.; a change resulting from the- application of' a small refinement of a previously- used calculational model or design method.
U.S. Nuclear Regulatory Commission B14327/Page6 January 29, 1993 CYAPC0 has reviewed the proposed license amendment against the criteria-of- ;
10CFR51.22 for environmental considerations. The proposed changes do not involve a significant hazards consideration, nor increase the types and amounts of ;
effluents that may be released offsite, nor significantly increase-individual or l cumulative occupational radiation exposures. Based on the foregoing, CYAPC0 j concludes that the proposed changes meet the criteria delineated in >
10CFR51.22(c) environmental (9) for a categorical exclusion from the requirements of an )
impact statement. !
Revision bars are provided in the right-hand margin to indicate a revision to text. No revision bars are utilized when the page is changed solely - to !
accommodate the shifting of text due to additions or deletions.
The CYAPC0 Nuclear Review Board and the Plant Operations Review Committee have reviewed and approved the proposed license amendment and have concurred with the above determination.
Regarding our schedule for this proposed amendment, CYAPC0 intends to fully implement the license amendment within 30 days of its issuance by the NRC Staff.
CYAPC0 desires to implement these changes during the Spring 1993 refueling outage. Hence, NRC action on this proposal is requested by April 1, 1993. This proposed amendmut does not supercede our previous proposal of July 31, 1992.
It is CYAPCO's intention to pursue NRC Staff review and approval of our. previous request in addition to this interim proposal.
In accordance with 10CFR50.91(b), we are providing the State of Connecticut with a copy of this proposed amendment.
Very truly yours, CONNECTICUT YANKl.J ATOMIC POWER COMPANY ,
%FRA J. F. Opeka O Executive Vice President Attachment cc: T. T. Martin, Region 1 Administrator A. B. Wang, NRC Project Manager, Haddam Neck Plant W. J. Raymond, Senior Resident Inspector, Haddam Neck Plant Mr. Kevin McCarthy Director, Radiation Control Unit Department of Environmental Protection Hartford, CT 06116 L
l
U.S. Nuclear Regulatory Connission B14327/Page 7 January 29, 1993 Subscribed and sworn to before me this J'i day of j$ AtJw Ask t , 1993 bwh Ilotary 9
hNah Public Date Connission F,xpires: I'l 31k41 I
l Docket No. 59_-111 E!4327 .
l l
I Attachment 1 Haddam Neck Plant Safety Assessment t
ti January 1993 x
-U.S. Nuclear Regulatory Commission Attachment 1/014327/Page 1 January 29, 1993 Proposed Modifications to Steam Generator Tube Repair Criteria 1his Safety Assessment documents the impact of allowing cracked steam generator tubes to remain in service. The cracks would be limited in length and the total number of tubes with cracks retained in service are limited such that the dose contributions from the theoretically calculated aggregate tube leakage will be limited to a small fraction of 10CFR100 dose limits in the event of a steam line break. A maximum theoretically calculated post steam line break primary-to-secondary leakage of 1.26 gpm per steam generator is used in this safety assessment.
I. Description of the Chanae The proposed design change modifies section 3/4.4.5 of the Technical Specifications. Details of the change are contained in Attachment 2.
Basically, the change would allow some steam generator tubes that may leak following a postulated steam line break (SLB) to remain in service. The number of tubes allowed to remain in service would be limited such that the dose contribution from the aggregate tube leakage will be limited to ,
a small fraction of 10CFR100 dose guideline values in the event of a steam '
line break.
II. Discussion A. Backaround The proposed alternate repair criteria would allow tubes with postulated throughwall cracks in the Haddam Neck steam generators to remain in service without repair. The cracks are located in the roll expansion region, which is defined as the bottom 5 inches of the tube, approximately 18 inches from the secondary side of the tubesheet. The criteria define an allowable axial crack length such that the structural integrity of the tube is assured during normal operation and- postulated accident conditions as required by Regulatory Guide 1.121, " Bases for Plugging Degraded PWR Steam Generator Tubes" August 1976._ The basis for the proposed allowable length is the Haddam Neck Plant-specific application of a generic alternate repair criteria developed by a committee of industry -
experts under the cognizance of EPRI (Reference [1]).
The generic alternate repair criteria were initially develo)ed-for steam generator tubes which are fully expanded through-the tu)esheet with cracks located at or above the top of the tubesheet in the expansion transition region. The length limitations of the generic-criteria are to preclude burst of the cracked tubes with the required safety margins. The length limitation includes allowances for inspection measurement uncertainty and potential crack growth
l l
U.S. Nuclear Regulatory Commission Attachment 1/014327/Page 2 January 29, 1993 based on a 95% confidence level of the composite industry data base.
As indicated in Reference (1), application of the same length limitations to the Haddam Neck steam generator tubes, which are only partially expanded through the tubesheet, is extremely conservativs.
The tubesheet completely surr; nds the tube in the region where the cracks are located. Depending o.. the length of axial cracks assumed to be present, the tube can either withstand a pressure load of three times the normal operating pressure on its own or will expand and come in contact with the tubesheet prior to reaching the pressure which would cause burst if the crack were in the free span of the tube. Thus, burst of the tube within the tube sheet is precluded regardless of the length of the axial crack assumed to be present in the region where the alternate repair criteria are applicable.
B. Structural Consideration Primary water stress corrosion cracks (PWSCC) have been identified in steam generator tubes at the Haddam Neck Plant. The cracks are located in the roll expansion region of the tube. The roll expansion region consists of the expanded portion of the tube plus the transition zone between the expanded and unexpanded portions of the tube, Figure 1. Since the tubes in the Haddam Neck steam generators are expanded only a small portion of the length within the tubesheet, any cracks which develop in the roll expansion region are totally contained within the tubesheet. For the purpose of this analysis, the bottom 5 inches of the tube will be considered the roll expansion region and cracks outside of this area will not be addressed.
Steam generator tubes with cracks which are retained in service must remain structurally sound over the next operating cycle under normal operating and postulated accident conditions. Structurally sound means that the tube will not burst or separate and pull out of the tubesheet with the required safety margins. Use of the generic repair criteria ensure that crack tubes retained in service will remain structurally sound.
(1) iltgedgral limits Basad on Generi,S Criteria Generic alternate repair criteria have been developed to address cracks in the expansion region of steam generator tubes, Heference [1]. The structural basis of the generic criteria is an experimentally determined burst curve for axial cracks of various lengths. The generic criteria assume that the tubre are fully exp'anded through the tubesheet and the tracks are located in the expansion transition region at or above the top of the tubesheet. Cracks are limited in axial length and circumferential extent under these conditions in
U.S. Nuclear Regulatory Commission Attachment 1/014327/Page 3 January 29, 1993 order to preclude burst of the tube with the required safety margins. Tube plugging limits developed using these criteria are extremely conservative for tubes which are only partially expanded through the tubesheet as in the case of tie Haddam Neck Plant, since tube burst is prevented by the presence of the tubesheet. The axial limiting crack length is determined for the Haddam Neck Plant using the following equation from Reference (1):
A-a +a,, -a, aa, where: A - allowable axial crack. length a - tube rupture equation reference crack length a,, - correction for tube sheet constraint a, - allowance for crack growth i a - allowance for NDE uncertainty i
a For the Haddam Neck Plant, the relevant input parametsrs are listed in Table 1 along with the tabulation of the results.
For the Haddam Neck Plant, the limiting crack length is 0.45 inches.
Hence, tubes with axial- cracks less than or equal to 0.45 inches in length in the expansion region meet the applicable regulatory requirements and are safe to be maintained in operation.
(2) Other Structural Considerations Additional )lant specific testing and analysis was performed to ensure tie structural acceptability of retaining cracked tubes in service in the Haddam Neck Plant. Table 2 shows the tube loads used in this testing and analysis. Burst of _a tube with axially oriented cracks of any length in the defined roll expansion region cannot occur due to the physical presence _ of the tubesheet which surrounds the tube. An ~ increase in internal pressure' would cause the tube to bulge outward and would be restrained by the tubesheet. The maximum ex)ansion of the tube is limited to the clearance' between the tu)esheet hole' and the tubes or 0.014- inch nominal clearance 'on the-diameter. A worst case stack-up of tolerances would add 0.010 inch to' the nominal clearance for' a maximum of 0.024 inch clearance on the diameter or 0.012 inch 'on the radius. Since the thickness of the tube wall is 0.055 .0055 inch, a flap caused by cracks in the tubes could be postulated to expand radially to' the tubesheet contact point while still overlapping the thickness of the tube by 0.043 inch. I.eakage through this opening would be greatly inhibited and would be a
i ,
U.S. Nuclear Regulatory Commission Attachment 1/814327/Page4 January 29, 1993 orders of magnitude less than the classic fishmouth type ,
rupture which could occur if the tubesheet wcs not present.
Since tube burst is precluded by' the presence of the tubesheet, structural integrity of the tube is ensured provided the cracks cannot result in physical separation of the tube and pull out-from the tubesheet hole.
A series of mechanical pull tests specifically for the Haddam Neck Plant (Reference (2)) were performed to determine the structural strength of SG tubes with cracks present. Test specimens were manufactured using prototypical 0.75 00,-0.055 inch wall, Inconel Alloy 600 tubing rolled into carbon steel collars. Slots were electric discharge machined (EDM) through the tube wall at various distances below the roll expansion transition to simulate throughwall stress corrosion cracks.
Specimens were manufactured with either 7 or 15 slots inclined at an angle of 15, 30, or 45 degrees to the longitudinal axis of the tube. In addition, speciment were manufactured with 36 slots inclined at a 30' angle to the longitudinal axis.
Thirty-six was the maximum number of slots which could be -
machined using the available equipment.
The length of the EDM slots was 0.5 inch, which was considered-larger than the average length of eddy current test (ECT)_
indications present in the SG tube roll expansions. The width of the machined slots was 7 mils. This width is much wider than the width of stress corrosion cracks and while structural tests would produce a prototypical to conservative result, leakage tests would be expected to be overly conservative.
Complete specimens were heated to 1100'F for four hourse to simulate the effects of the highest possible temperature the tubes could have experienced during SG assembly or subsequent operation. The mechanical pull tests were then performed at room temperature. The results were adjusted to account for the decrease in material strength at operating temperature and the minimum material properties of actual Haddam Neck Plant tubing based en a 90/95 lower tolerance limit of the tubing certification values. :Some specimens were tested without collars in order to determine the effect of cracks present in the free span of the tube.
The test fixture which held the specimens was - specifically -
constructed to allow the tube to rotate when an axial load was applied to the tube. -This decision was made after an analysis was performed which determined that- allowing the tube to.
rotate would result in a lower, and hence more conservative, failure load. The tube has a tendency to rotate when a_ load.
f U.S. Nuclear Regulatory Commission Attachment 1/014327/Page 5 January 29, 1993 is applied due to the moment generated by the slots machined at an angle to the longitudinal axis. This moment acts to
- straighten" the array of slots and in the process causes plastir. deformation at the tip of the slot. If rotation is prevented, the portion of the axial load causing the rotation must be absorbed by the test fixture. Since the test fixture would r.ow be absorbing a portion of the load placed on the specimen, the observed specimen failure load would be higher and nonconservative. in the actual steam generator, friction between the tube and the tubesheet would absorb a portion of the axial and rotational loads placed on the tube. However, this effect was neglected for conservatism.
The results of the mechanical pull tests indicated that the load bearing capability of degraded tubes was strongly dependent on the angle of the cracks from the longitt'ainal axis. The number of cracks arosent and the location of the cracks within the roll expans'on had little effect on the load bearing capability of tie tube. As expected noncollared specimens failed at a slightly lower load than collared specimens, since the collar would tend to inhibit deformation of the tube, in all cases, the test failure load was significantly above the limiting case load of three times normal operation.
Following the mechanical testing, an analytical model was constructed to allow the formulation of design curves for various crack lengths and orientations. The model bounds crack lengths up to 1.5 inches and crack angles up to 60' from the longitudinal axis. The model conservatively predicts the results of the mechanical testing. The design curves constructed from the analytical model are shown in Figure 2.
The conclusions of this analysis and testing are that tubes with arrays of cracks less than 1.5 inches in length and 60' in angle from the longitudinal axis met all the requirements of Regulatory Guide 1.121, and are safe to be maintained in operation.
Additional experimental. tests were performed with tubes with circumferential cracks. Experhental data (Reference (3])
indicate that the extent of circumferential cracking needed to produce axial separation of the tube is very large. (i.e.,
approximately a through wall crack extending 250' around the circumference, or a part-through wall crack 80% of; the wall thickness and 360' around the circumference). These crack sizes are readily detectable by standard NDE methods for tube inspection. By limiting the maximum allowable circumferential crack length, which theoretically could be present between 1
b
l U.S. Nuclear Regulatory Commission Attachment 1/B14327/Page 6 January 29, 1993 axial cracks and remain undetected, to less than 1 inch, structural integrity of the tube is ensured.
C. Leakaae Considerations The cracks in the e.vpansion region of the tube to be retained in service are assumed to be through the tube wall and as such could potentially leak primary water to the secondary side of the SG. The icvel of leakage under norma) operating conditions is limited to the current Technical Specification requirement of 150 gallons per day (gpd). Under postulated accident conditions, cracks could potentially open up and leak at an increased rate. The number of tubes with cracks retained in service will be limited such that the calculated theoretical leak rates will remain acceptable. A main steam line brnk is the limiting accident considered in this case since it res' ts in the highest possible pressure differential between the pimary and secondary sides and provides a leak path outside of containment. A differential pressure of 2050 psi is used in the leakage calculations. This pressure is equivalent to the normal primary side operating pressure and a zero secondary side pressure. This is considered reasonable since many MSLB scenarios would actually result in decreasing primary-side pressure. If, in fact, operating charging pumps repressurize the primary system, the o)erators would take action to limit primary-side pressure to less t1an or equal to normal operations pressure, 2050 psi. This value is considered conservative based on the steam itne break pressure curve presented in Figure 3.
(1) 11addam Neck plant Specific Abolication of Generic Methodoloav int _Qgterminina Leakaag Theoretical leakage under accident conditions through cracks of a given size are calculated using the methodology provided in Reference [1] adapted to plant specific conditions. Cold leg conditions are assumed since these result in slightly higher leakage rates.
Table 3 provides a list of the parameters used. As )er Reference [1], the leakage rate from an axial crack in a taale with internal pressure is given by the following formula.
q
I U.S. Nuclear Regulatory Commission Attachment 1/014327/Page 7 January 29, 1993 OnKb/2P/ p where Q is leak rate (in*/sec)
K is the discharge coefficient (dimensionless) 6 is leakage area (in')
P is effective differential pressure (psi) p is liquid density (lb/in')
The leakage area is obtained by the summation of an elastic and plastic component of the crack opening area during steam line break conditions. The discharge coefficient will increase with crack size. Experimental data were used to detcrrr.ine the discharge coefficient as a function of crack '
size in Reference [1]. For the Haddam Neck Plant, the discharge coefficient has been benchmarked against test data as described below. For the Haddam Neck Plant, Table 4 provides a summary of the crack width and the resulting leak rate for crack lengths from 0.05 to 0.5 inches.
(2) Oualification of Methodploav for Determinina leakaae As an additional check of the Reference [1] methodology for determining leakage, leakage levels were determined for ten tubes removed from the McGuire 1 steam generators and compared with leakage determined in laboratory tests (Reference (4)).
The test leakage levels for the ten tubes were a factor of 40 greater than the leakage measured.in the entire SG prior to tube removal, indicating some crack opening occurred during the tube removal process. :Using average tube material properties, a good correlation is obtained between the test values and the predicted results. The- leakage values calculated for the Haddam Neck Plant specific case employed the same methodology with the exception that 95 1ercent -
confidence lower linit material properties were usec - The Haddam Neck Plant calculated values are expected to.be quite cons (rvative when compared to actual steam generator conditions .
(3) leakaae Through Tube to Tubesheet Annulus The maximum theoretical leakage under steam line break conditions is calculated assuming .that the tube _ is severed.
U.S. Nuclear Regulatory Commission ,
Attachment 1/814327/Page 8 January 29, 1993 l circumferential1y 18 inches below the top of the tubesheet and the only restriction to leakage is the tube to tubesheet annulus. The maximum tolerance stack up and cold leg conditions are used. A value of 5.3 gal / min was calculated for a single tube.
D. frJtek Growth Rate n comparison of the crack length determined by ECT is made for tubes with cracks present during consecutive inspections. .The data consist of 72 cracks retained in service between 1987 and 1989 as part of the F* criteria. The 1989 length is subtracted from the 1987 length to obtain_ a delta length. On average the delta length is zero, indicating no growth of the cracks.
Although zero crack growth rate has been found, as a conservatism a crack growth rate was calculated using the Reference [1]
methodology. For the Haddam Neck Plant the value used is 0.029 inches /EFPY or 0.037 inches per operating cycle.
E. Tube Material Properties The material test certifications for the tubes used in the ,
construction of the Haddam - Neck. Plant steam generators, were reviewed and the mean, standard deviation and 95 percent confidence lower limit values determined. The 95 percent lower limit represents the value for which 95 percent of the tubes have an equal-or greater value. Material property data was not available for 500 of the approximately 15,000 tubes in the Haddam. Neck steam generators. Lack of data for these small number of tubes- is not l expected to significantly affect these values,
- f. Other Considerations Tubes with identified cracks whose circumferential extent exceeds L the axial extent do not meet the proposed criteria and would be repaired or removed 'from service. Circumferential cracks theoretically could be present between identified axial cracks and.
remain undetected due to nondestructive test limitations. The current state of the art rotating pancake coil- (RPC) eddy, current
-testing can reliably identify circumferential cracks in the presence of axial cracks spaced greater than approximately 0.4 inches? A throughwall circumferential crack conservatively will be_ assumed-to be present between axial cracks spaced closer than 0.'4 inches. The extent of axial cracking retained in service is limited 'such that the_ theoretically undetected circumferential crack extent does not -
exceed 1 inch of the-tube circumference. The 1-inch-limit on the circumferentir.1 extent assumed to be present- is a conservative number based ' on burst tests of circumferential flaws, Refer-l i
0.5. Nuclear Regulatory Commission Attachment 1/814327/Page 9 January 29, 1993 ence [3). Tubes with throughwall circumferential flaws up to 1.4 inches in circumferential extent were tested and found to meet Regulatory Guide 1.121 requirements for margin to bursts. The 1-inch limit assures that axial separation of a tube with characterized cracks retained in service will not occur under normal operation or postulated accident conditions.
In addition to remaining structurally sounc, tubes with cracks retained in service must not exhibit unacceptable levels of leakage under normal operation or postulated accident ccnditions. Leakage under normal operation is limited to the current Technical Specification requirement of 150 gallons per day (gpd) per steam-generator. Because of the location of the cracks approximately 18 inches into the tubesheet, the likelihood that the tube to the tubesheet annulus is packed with sludge, and the fact that not all cracks would be expected to be through the tube wall, the contribution to the caseline operating leakage by these tubes is expected to be small. Examination of the steam generator leakage history of the Haddam Neck Plant steam generators shows no change in the baseline leakage levels following the plugging of 200 tubes with cracks in the expansion region. A postulated increase in the operating baseline leakage does not decrease margins of safoty since the required shutdown limits remain the same and any developing condition which increases the rate of leakage would require plant-shutdown sooner than the condition where the baseline leakage was Zero.
Leakage from undetected circumferential cracks which theoretically could be present is assumed to be negligible, since the actual through wall length would be very small, the increase-in load in the axial direction due to an increase in pressure is one half of the increase in load in the hoop direction, and the presence of the tubesheet restricts leakage. The calculated maximum leakage under accident conditions from the 181 tubes which were previously plugged, but meet the proposed criteria, was determined to be 0.92 gpm.
Cracks which are through the tube wall in the roll expansion region could potentially allow primary coolant in contact with- the tubesheet. The effect of primary coolant on the tubesheet was considered. Corrosion of the tubesheet is not a concern due to the low level of oxygen present in the primary coolant and the tendency of any corrosion products which do form to expand and block access of the coolant to the tubesheet base metal. The limited amount of tubesheet corrosion possible would tend to restrict the tube to tubesheet annulus, further limiting leakage.
I U.S. Nuclear Regulatory Comission Attachment 1/B14327/Page 10 January 29, 1993 111. Impjtc1 of Channe on Consiquences of an Accident A. Lmpact on Previous 1v Evaluated Accidenti The only a*cident potentially impacted by the change is the steam line break accident. However, since the change is related to steam generator tubes, the impact on a postulated steam generator tube rupture (SGTR)isalsoevaluated.
The change cannot impact the probability of occurrence of a steam line break (SLB). As discussed above, the impact of the change on structural integrity of the SG tubes is such that the tubes that would be allowed to remain in service would be structurally sound.
Therefore, there is no increase in the probability of occurrence of an SGTR.
Since the tubes will remain structurally sound during normal operation, the change does not increase challenges to safety systems. There may be the potential for increasirg : a likelihood of plant shutdown if the leak rate increase,: during normal operation. However, as discussed above, operatilg experiences at t1e Haddam Neck Plant indicates no additional Ir akage is expected.
Thus, this potential is expected to be small. The change does not affect any malfunctions evaluated for the SLB or SGTR accidents.
The change does not increase the postulated primary to secondary leakage following a steam generator tube rupture since no significant increase in DP across the unaffected tubes could occur.
Therefore, it does not affect the consequences of an SGTR.
The change does not modify any failure modes. Therefore, the change cannot impact the consequences of a previously evaluated malfunction.
The postulated primary to secondary leakage following an SLB could increase as a result of the change. This would increase the calculated dose consequences of a steam-line break but would result in doses that are significantly below the limits of 10CFR100.
B. Potential for a New Unanalyzed Aqcident The change could increase the primary to secondary leakage following a postulated SLB. This leakage would be limited such that the dose contribution from the aggregate tube leakage will be limited to a fraction of 10CFR100 dose guidelines in the event of a steam line break. Based on current ant. lysis, this value will be limited to 1.26 gpm per steam generator. This leak rate is larger than currently assumed but is not large enough to compromise the mitigation of an SLB. Therefore, it is concluded that the
U.S. Nuclear Regulatory Comission Attachment 1/814327/Page 11 January 29, 1993 possibility of an accident of a different type than previously evaluated is not created by the change.
1he change would allow steam generator tubes that could potentially leak following an SLB to remain in service. In the design basis analyses, it is assumed that there is no increase (over the existing level specified secondary in thefollowing leakage Technical Specification) a steam line break.in the primary The existingto technical spectfication does not allow any tubes to remain in service that are known to have the potential for increased leakage following an SLB. Since the change would allow a postulated postaccident leak rate, this is considered a malfunction of a different type than previously evaluated, i.e., increased primary to secondary leakage post-SLB.
The proposed change will maintain the required margin of safety of the steam generator tubes. Also, the change does not compromise accident mitigation following an SLB. The increased leakage is judged to negligibly impc;t core or containment predicted postaccident parameters. Therefore, the margin of safety with respect to mechanical issues is not negatively impacted.
Radiological consequences are discussed in Section Ill.C below.
The proposed change does not introduce any new malfunctions. The only malfunction evaluated is failure of steam generator tubes following an SLB.
C. Radi910sjtal conseauences For this propcsed change the only accident which needs to be considered from a radiological viewpoint is the main steam line break. This is the only accident which would result in a larger 4p across the tubes than that which exists during normal operations and hen e, may increase the leakage rate through the cracks. For the steam generator tube rupture accident and the control rod ejection accident, the primary to secondary leakage is not expected to increase as a result of the accident and hence there would be no change in accident assumptions.
The radiological consequences for a Haddam Neck Plant main steam line brelk were calculated by the NRC Staff in 1982 during SEP and are presented in Reference [5]. The calculation was done in accordance with Standard Review Plan 15.1.5, Reference (6).
The assumption used for primary to secondary leakage in the Standard Review Plan and by the NRC Staff is that the postbreak primary to secondary leakage is equal to the Technical Specification primary to secondary leak rate limit. This is a nonmechanistic simplifying assumption that may, under certain cunditions, be nonconservative. ,
1
U.S. Nuclear Regulatory Commission Attachment 1/814327/Page 12 January 29, 1993 Prior to the accident (steam line break), operation with through wall defects in the tubes is permitted. Depending on the type of defect and location, leakage could increase as a result of the break and depressurization of the secondary side. The theoretical calculated leakage values discussed are calculated assuming the ,
cracks are above the top of the tubesheet and takes no credit for.
the fact that the cracks are located in tubes in an area entirely within the tubesheet. This tube /tubesheet crevice would further restrict primary / secondary flow even for the case where the annulus area was assumed to be clean. The annulus area may contain corrosion products that would additionally restrict flow. Hence, the source of leakage for the radiological assessment is itself a conservative number. In addition to source term conservatism, the-following dose assessment assumption conservatisms exist.
- 1. 95% worst-case meteorology--approximately 10100 times higher than typical meteorology, especially for a release such as this where thermal plume rise can be expected.-
1
- 2. No iodine DF in the steam generators- although the affected generator will be dry, there would still be significant plateout on generator surfaces such as the steam dryers and in ;
our specific case here, in the roll region. A minimum DF of 10 has been used on other dockets (See Table 5.)
- 3. lodine spiking factor of 500--For the Haddam Neck Plant 'a spiking factor of approximately 25 is typical for a reactor trip and depressurization. (Reference [7]).
- 4. The plant is assumed to operate -at its Tech. Spec primary coolant limit of 1.0 pCi/gm DEQ I-131. Typical operation is 10-1000 times less than 111s.
- 5. The plant is assumed to operate at its secondary side Tech. .
~
Spec. limit of 0.1 C/gm-I-131. This is over a 1000 factor conservative and should not even be considered in the dose calculation, as the contribution will be small compared to the primary activity released. In Reference'[5), initial secondary side activity accounts for 60% of the doso, in the NRC's SEP analysis of the Haddam Neck MSLB.
- 6. The plant is assumed to operato at:its primary to secondary Tech. Spec leak rate limit. Typically, the plant, operates at 10-1000 times less than this limit,
- 7. The plant is assumed to operate with the maximum number of degraded tubes allowed by the 1.26 gpm limit and all of these tubes leak at their cinservatively calculated leak rate.
Expected actual leakage would be the normal operating leak i
i U.S. Nuclear Regulatory Commission Attachment 1/B14327/Page 13 January 29, 1993 rate limit of 0.1 gpm or less since primary side pressure would actually be expected to decrease during the HSLB transient.
' Hence, the conservatisms in the steam line break dose calculation I more than offset an increased leak rate. !
CYAPC0 would also like to note that in order to approach 10CFR100 .
limits, all of the SRP assumptions must exist simultaneously. To i Incur a MSLB simultaneous with 95% meteorology, high coolant i activity, maximum iodine spiking, minimum iodine DF and maximum !
primary to secondary leakage has a combined probability.of less than 10*/ year. Hence, the SRP presented accident is really not a credible event.
Since mechanistically leakage -will increase through the- cracks following a steam line break and with the above ~ conservatism in mind, the SRP guidelines can be modified and an assumed increased leakage used to determine if the proposed change is safe. The pro)osed Technical Specification correlates theoretical crack-leacage determined in accordance with the Haddam Neck Plant specific ,
application of Reference (1) with the dose-calculation-methodology described below.
A main steam line break event theoretical leakage is calcu'.ated.
This leakage rate is then used in the dose calculation methodology.
A postaccident leak rate of 1.26 gpm is used. The potential.' leak rate is 10 times. the current Technical Specification operational limit. Even at this leak rate the' proposed change is safe.
Per Reference [5), it can be determined that the limiting dose,
~
1.e., the one closest to its. limit, is-the Exclusion Area Boundary (EAB) thyroid dose for the accident initiated spike. - W ale-body:
doses are relatively insignificant without any fuel fai_ lures.
The NRC~ Staff calculated EAB thyroid dose during SEP Reference [5]
was 10.8 REM. Of this, only 4 REM was due' to- the primary to'-
secondary leakage. The 7 REM due to preexisting secondary side activity can realistically be neglected.~ The NRC Staff assumed a primary to secondary leakage of 0.4 gpm in the affected oenerator.
Thus,usinga1.26gpmleak.agerateassumption,theresultingdose is approximately 12.6 REM.
l 5
1-+-r-- .,,& e s - - - - - _ _ - . . - - - . - - _ - _ _ . _ - > - . . - s.- - . m -
1 I
U.S. Nuclaar Regulatory Commission Attachmont 1/014327/Page 14 Janua;y 29, 1993 The resulting offsite dose using the 1.26 gpm postaccident leakage is significantly within 10CFR100 limits (i.e. less than the 30 rem j guideline of the SRP) and not a safety concern given the low dose and low probability of the initiating event.
IV. SAFETY DETERMINAT10B 1
The evaluation above concluded that the proposed change would introduce a ,
malfunction of a different type than previously evaluated. Therefore, the '
sections below assess the safety aspects of the proposed change.
The proposed change could increase the primary to secondary leakage ,
following an SLB. This postulated increased leakage would be limited such 4
that the dose contribution from the aggregate tube leakage will be limited >
to a small fraction of 10r,FR100 dose limits-in the event of a steam line break. The Emergency Operating Procedures contain appropriate instructions to mitigate the consequences of a postulated SLB concurrent ,
with an SGTR. Therefore, the procedures provide appro)riate operator guidance for mitigating this postulated event. Basec. on this, the probability that the operators will fail to mitigate an ' SLB is not increased.
The proposed change would allow higher primary to secondary . leakage
, M1owing a postulated SLB. This increased leakage would be limited such l that it would not compromise the ability of the high-pressure. safety injection (HPSI) system to mitigate an SLB. The HPSI capacity exceeds the ,
anticipated leak rate. Sufficient water exists in the refueling. water tank to compensate for such leakage until the primary side can be cooled down and put-on-RHR. Therefore, the proposed change does not affect the probability of failure of mitigation equipment.
The postulated consequences following an SLB would increase as a result of .
the change. However, as discussed above in Section III.2 Radiological Consequences, the consequences would be a small fraction of 10CFR100 limits. Using realistic assumptions results-in the increased _ public risk' ,
being negligible.
The change does not increase the probability of any accident. Also, the postulated increased consequences following an SLB will negligibly impact public risk. Therefore, the proposed change is judged to be safe, ,
5
)
U.S. Nuclear Regulatory Commission :
814327/ References /Page 1 January 29, 1993 ,
REFERENCES (1) J. A. Maccoun, Electric Power Research Institute, letter to NRC Attention Emmett L. Murphey dated March 26, 1992 transmitting EPRI NP-6864 L, PWR Steam Generator Tube Repair Limits: Technical Su Expansion Zone PWSCC in Roll Transitions (Revision , December 1)pport 1991.Document for (2) WCAP 11280, " Development of Tubesheet Roll Region Plugging Criteria for Steam Generator Tubes in the Connecticut Yankee Power Plant," Revision 1, dated January 1987.
(3) WCAP 12916 " Northeast Utilities Tube Specimen Burst Tests," dated May 1991.
(4) C. R. Faye and D. T. Martin, ' Leak and Burst Tests of the Expansion
-Transmitted -and Support Areas of Tubes Exam McGuire Unit 1 Steam Generators, Volume I, Steam Generator Tube Test Data." B&W Report RDD: 90-5160 01:01, dated October 1990.
(5) Dennis Crutchfield letter to W. G. Counsil, dated November 9, 1982, SEP-Topic XV-2.
(6) NRC Standard Review Plan 15.1.5, Appendix A, " Radiological Consequences of Main Steam Line failures Outside Containment of a PWR."
(7) J. P. Adams and C. L. . Atwood, "The Iodine Spike Release Rate During a Steam Generator Tube Rupture," June 1991 EGG Idaho, Nuclear Technology Vol. 94.
i e e figure 1 Expanded Roll Area Diagram a -
.7500 + DIA t: .0550 d.0055 _ _-
.640 DIA (NDH.) ,
WLL TKk.
i.
1
/ %
\ \ nt -
l /
-mtsnta 0.25
/ ,, TOP OF ROLL
< , BOTTOM OF ROLL
, !/ -
l l / .
/ /
/ / 4.25 LUlGTH OF ROLL i / l s
/ /
/ / I '
.25; , CLADD1 4G Tmt tso 0A pI hl I 1 1.031 * .005 (PITCH)
.22 4.00
. 03 =
.6595 DIA (NDM.)
.7640 4 .0025 DIA
. .o ,
Figure 2 .
Design Curves i
. ULTIM ATE PULL LOAD (LOS)
12000 10000- )
8000- ,
6000 - ,
4000-x _
.2000 .
0000 i i i i . . . i i i i O 5 10 15'.20 25 30 35 40 45 50 55 60 MULTIPLE CRACK SLANT ANGLE (DEGREES) 0,5 inch Cracks 1.5 inch Cracks-e SAFETY FACTOR = 3 4
-r, m Figure 3 Steam Line Break from HFP 4-Loop Operation .;
=
I I i i --
i ,
g - .
e E8- - _
.m m- __
fmH -
W~ ._
'e
-N .
1 .
~~
.h , _
I t i : L i ,
d SD- 150 290- 320- ,'400' 450 -550. 690f
-TIME (SECOND) b 1
- U.S._ Nuclear Regulatory Commission Table 1/B14327/Page 1
--January 29,-1993-
=
Tabla 1 Haddam Neck Parameters for Determining Allowable Crack Size IDDut Parameters Tube wall thickness - 0.055 inches: ,
Average tube radius = 0.348 inches-Normai-Primary Pressure - Psig - 2050 Psig Normal Secondury Pressure - Psig - 675 Psig-Tube Yield Strength'(psi)-G 600 F - 36,960
- 95% lower confidence lim't Tube Ultimate Strength (psi) 0 600 F - 91,675
- 95% lower confidence limit Resul ts -
Reference crack length a - 0.528; inches Correction for tubesheet constraint. a,, =
_0.060 inches Allowances for Crack Growth - a ,. - 0.037 inches.
NDE uncertainty; allowance la,w. .
0.102 inches -
Al?owable crack length A -- .528 + .0601- .037 - 0.102-
- 0.45 l
l p
l
'c I
La w
U.S. Nuclear Regulatory Conninsion
' Table 2/B14327/Fage 1 January 29, 1995 Table 2 TUBE LOADS USED IN ANALYSIS DIFFERENTIAL AXIAL ANALYSIS PRESSURE FORCE SAFETY -VALUE fosi) Llhil FACTOR Llb.3L Normal Operation 1375 633 3 1899 Steam Line Break 2650 1220 1.43 1744
(
' 4 '[
)7 t
4U.S.t Nuclear!Regul Atory Commission -
- Table'3/B14327/Page 1; ,
January 29, 1993 Table 3 i Haddam Neck: Plant.
Parameters for Determining Leakage for Crack L,qn.thi_p205 q to 0.5 inches ,
Tube wall thickness - inches - 0.055 inches-
-Average tube radius. - 0.348 inches Poisson's. Ratio- - 0.33-1 8
Elastic modulus 0 600 F = 29.5 x 10 psi
- Maximum' Accident' Pressure =- 2050 psi Density 0:539 F and 2050 psi =
0.027442 lb/ inches * ,
' Saturation Pressure 0 539'F = 954.9 psi
-Secondary Accident' Pressure - O psi ;
Tube Ultimate Strength 9 600 F - 91,675 psi-
- 95%-lower confidence limit Tube Yield Strength 9 600 F - 36;960 psi
- 95% lower confidence limit
.i g t
ai g I,.
I U.S. Nuclear Regulatory Commission L Table 4/B14327/Page 1 January 29, 1993 Table 4 Haddam Neck Plant Calculated Crack Opening and Associated Leakage Crack Leakage Width Rate
[ rack Size finches) igal/ min) 0.05 0.000033 0.0001 0.10 0.00007 0.0002 0.15 0.000117 0.0006 0.20 0.000175 0.0012 0.25 0.00025 0.0021 0.30 0.000342 0.0036 0.35 0.000459 -0.0058 0.40 0.000657 0.01 0.45 0.000927 0.017 0.50 0.001279 0.0282 I
l
_ _- ______ __ -