ML20207N461

Notice of Denial of Util 851106 Application for Amend to License DPR-50,revising Tech Specs Re Steam Generator Tube Plugging Limitations

ML20207N461
Person / Time
Site:	Three Mile Island
Issue date:	12/23/1986
From:	Stolz J Office of Nuclear Reactor Regulation
To:
Shared Package
ML20207N126	List: ML20207N124 ML20207N461
References
NUDOCS 8701140233
Download: ML20207N461 (4)

Text

y w  : , . ;,- m,,

w ,

m ,-

. _,, s- , . .

~ >

^

7590-01 c y <

UNITED STATES NUCLEAR REGULATORY COMISSION~

METROPOLITAN EDISON COMPANY JERSEY CENTRAL POWER AND LIGHT COMPANY PENNSYLVANIA ELECTRIC COMPANY-GPU NUCLEAR CORPORATION ,

THREE MILE ISLAND NUCLEAR STATION, UNIT NO. 1 DOCKET NO. 50-289 NOTICE OF DENIAL OF AMENDMENT TO FACILITY OPERATING LICEt '.

AND OPPORTUNITY FOR HEARING The U.S. Nuclear Regulatory Commission (the Commission) has denied a request by GPU Nuclear Corporation, et al. -(the licensees) for an amendment to Facility Operating License No. DPR-50 issued to GPU NuclearcCorporation for operation of the Three Mile. Island Nuclear Station, Unit No. 1 (TMI-1) located in Dauphin County, Pennsylvania. Notice of consideration of issuance of this amendment and opportunity for prior hearing was published in the FEDEPAL' REGISTER on January 6,1986(51FR459).

The amendment would revise the provisions in the Technical Specifications relating to the steam generator tube plugging limitations in accordance with the licensees' application for amendment dated November 6,1985. Basically, the present Technical Specifications require repairing or removing from service c . a steam generator tube when a defect exceeds 40% of the tube wall thickness.

The proposed amendment would maintain the 40% throughwall limit on the secondary side of the tube but replaces the limit on the primary side of the tube with a sliding s' ale c which goes from 40% to 70% throughwall depending on the size of the defect.

8701140233 DR 861223 ADOCK 05000289 PDR

~

This proposed amendment is the subject of litigation and discovery for hearing scheduled to start.in March 1987. However. the hearing is limited to .

issues raised by the contentions which relate to the adequacy of eddy current testing (ECT) and to concerns about the environmental effects of operational plant chemistry. There are many additional technical issues associated with the request which are not within the scope of the hearing issues. These matters are within the authority of the Commission's staff.

Specifically, independent of hearing contentions, the Commission conducted a detailed safety evaluation of the fracture mechanics methodology used by the licensees to justify a 70% Once Through Steam Generator (OTSG) tube plugging limit. The Commission has, in a Safety Evaluation (SE) dated December 23, 1986 , concluded that the licensees' analyses are not technically acceptable. Thus, without regard to those issues under litigation, the Commission has decided that a 70% tube plugging limit is not acceptable, and i

the proposed amendment is denied.

The licensees were notified of the Commission's denial of the proposed l

Technical Specification changes by letter dated By , the licensees may demand a hearing with respect to the

denial described above and any person whose interest may be affected by this l proceeding may file a written petition for leave to intervene.

A request for hearing or petition for leave to intervene must be filed with the. Secretary of the Commission, U.S. Nuclear Regulatory Commission, Washington, D.C. 20555, Attention: Docketing and Service Branch, or may be delivered to the Commission's Public Document Room, 1717 H Street, N.W.,

Washington, D.C., by the above date, i

[ 14 copy of any petitions should also be sent to the Office of the General Counsel-Bethesda, U.S. Nuclear Regulatory Commission, Washington, D.C. 20555, and to Bruce W. Churchill, Shaw, Pittman, Potts and Trowbridge, 2300 N Street, N.W., Washington, D.C. 20037, attorney for the licensees.

For further details with respect to this action, see (1) the application for amendment and supplement, dated November 6, 1985 and October 3, 1986-(respectively), and (2) the Commission's letter and SE to GPU Nuclear Corporation dated December 23, 1986, which are available for public inspection at the Commission's Public Document Room, 1717 H Street, N.W., Washington, D.C. and at the Government Publications Section State Library of Pennsylvania, Education Building, Commonwealth and Walnut Streets, Harrisburg, Pennsylvania 17126. A copy of the Commission's letter and SE may be obtained upon request addressed to the U.S. Nuclear Regulatory Commission, Washington, D.C. 20555, Attention:

Director, Division of PWR Licensing-B.

Dated at Bethesda, Maryland, this 23rd day of December 1986.

FOR THE NUCLEAR REGULATORY COMMISSION

/S/

John F. Stolz, Director PWR Proiect Directorate #6 Division of PWR Licensing-B l

l l

• See previous white for concurrences

[ Distribution: Docket File FR File.Ringram TRoss JThoma RWeller PWR#6fA PWR#6 PWR#6 PWR#6 PWR#6 OGC-Bethesda l RIngram TRoss;eh JThoma RWeller k4sdSt'olz (dd kli i>/\/86 11/19/86 1724/86 10/28/86 ISs/GL/86 ; 2, /86 5

l-

,.c

~

~

1. A copy of any petitions-should also be sent to the Office of the General Counsel-Bethesda, U.S. Nuclear Regulatory Commission, Washington, D.C. 20555,

.and to Bruce W. Churchill, Shaw, Pittman, Potts and Trowbridge,'2300 N Street, N.W., Washington, D.C. 20037, attorney for the licensees.

For further details with respect to this' action, see.(1) the application for amendment and supplement, dated November 6, 1985 and October 3, 1986 (respectively), and (2) the Comission's letter and SER to GPUN Corporation -

dated ,=which are available for public inspection at the Comission's Public Document Room,1717 H Street, N.W. , Washington, D.C.. and a 'at the Government Publications Section, State Library of Pennsylvania, Education Building, Comonwealth and Walnut Streets, Harrisburg, Pennsylvania 17126. A copy;of the 'Comission's letter may be. obtained upon request addressed to the U.S.

' Nuclear Regulatory Comission, Washington, D.C. 20555, Attention: Director, Division-of'PWR Licensing-B.

Dated at Bethesda, Maryland, this FOR THE NUCLEAR REGULATORY COMMISSION.

John F. Stolz, Director PWR Project Directorate #6

, Division of PWR Licensing-B Distribution: Docket File FR File RIngram TRoss JThoma RWeller QW PWR# PWR# 4 WR#6 PWR#6 OGC-Bethesda RIngrab TRoss;7 PWR#6-I e JThom[A RWeller JStolz

/4 /vf/86 # /g /86 h /zy/86 /* /4T/86 / /86 / /86 8hl*Ylfi Q0

-sa hIl '1l0

f* "% UNITED STATES f , ,%,,

y g NUCLEAR REGULATORY COMMISSION c WASHINGTON, D. C, 20555 n

, p SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATED TO NRC REVIEW OF THE USE OF FRACTURE MECHANICS METHODS TO JUSTIFY OTSG TUBE PLUGGING CRITERIA METROPOLITAN EDISON COMPANY JERSEY CENTRAL POWER AND LIGHT COMPANY PENNSYLVANIA ELECTRIC COMPANY GPU NUCLEAR CORPORATION THREE MILE ISLAND NUCLEAR STATION, UNIT N0. I m

DOCKET NO. 50-289 1.0 BACKGROUN_D By letter dated November 6,1985 (Reference 1), GPU Nuclear Corporation (GPUN or the licensee) requested a change in the Technical Specifications for Three Mile Island, Unit 1 (TMI-1) which would rasult in a revision to the plugging limits for Once Through Steam Generator (OTSG) tubes. GPUN, in the past, has plugged steam generator tubes based on a general 40% throughwall plugging limit.

This generic steam generator tube plugging limit defines as acceptable a tube with a uniform wall loss of 360* in circumferential extent, unlimited axial length and depth up to 40% of the tube wall thickness. TMI-1 was teraporarily authorized to operate with a primary side tube defect of up to 50% throughwall penetration and 0.55 inches in length (note other restrictions also apply).

This limit was only effective until the Cycle 6 refueling outage which began November 1, 1986. The proposed TSCR 148 criteria would further revise the depth and length of allowed defects located on the primary side of the tubes. It specifies a plugging limit of up to 70% throughwall depth for small length defects, decreasing to 40% throughwall depth for longer defects.

The proposed criteria apply to primary side (inner surface of the tube wall) defects only. Upper and lower tube sheet secondary faces and tube support entry and exit. locations in the primary side, which are areas of reduced eddy current sensitivity, are excluded. The plugging limit for defects in these areas as well as those on the outer surface of the tubes (secondary side) remains 40% of the nominal tube wall thickness.

By letter dated October 3, 1986 (Reference 2), the licensee concluded that because of design and performance differences of the B&W OTSG, Regulatory Guide 1.121 methodology fnr determining allowable minimum wall thickness in steam generator tubes does not apply. The NRC staff has reviewed this letter, and the arguments contained therein, but maintains that the guidance in the Regulatory Guide is applicable to OTSGs.

2.0 DISCUSSION The licensee has performed analyses in an attempt to demonstrate that a margin equivalent to the existing 40% throughwall (and unlimited length) plugging criteria can be provided by a tube with a defect greater than 40% of the tube wall thickness but with a finite length. These analyses used methodologies A-, , / , .m D / Y)x- e' /

e'ncompassed by elastic-plastic fracture mechanics and net section collapse. .

This safety evaluation will only address the use of fracture mechanics and net sectinn collapse. Other subjects of review including eddy current testing (ECT) and potential chemistry effects will be presented when the NRC staff's final Safety Evaluation Report (SER) is issued addressing the entire amendment request.

3.0 NRC STAFF REVIEW AND EVALUATION OF STRUCTURAL INTEGRITY In regard to the applicability of Regulatory Guide 1.121 to B&W OTSGs, it is the staff's position that the basic engineering guidance contained in Regulatory Guide 1.121 is applicable to all steam generators regardless of individual design and performance induced differences. The Regulatory Guide contains sufficient instructions and flexibility to allow its use in evaluation of tubes in all domestic steam generator designs. Regulatory Guide 1.121 describes the derivation of minimum allowable tube wall thickness by considering all appropriate tube loads developed under normal and postulated accident conditions.

It requires that infonnation be developed to provide a basis for ensuring tube integrity will be maintained during postulated design basis accidents. Design criteria used to establish structural integrity of the tubing should result in defining the minimum tube wall thickness that can sustain, with adequate margins, pressure and themal loads under nonnal operating and accident conditions.

Regulatory Guide 1.121 does not explicitly require that primary and secondary stresses be combined to meet the allowable limits of ASME Section III under accident conditions. However, it is incumbent upon the licensee to demonstrate that adequate margins exist in comparing the predicted to the required tube capacity. A licensee is free to use approaches or methods other than those described in Regulatory Guide 1.121, if the licensee can technically justify such approaches or methods. In this instance GPUN has failed to technically justify its proposed approach.

The licensee has used elastic-plastic fracture mechanics (tearing instability or J-T analyses) and net section collapse criterion (NSCC) analyses (also known as limit load analyses) in an attempt to justify the proposed tube plugging criteria. These analyses (Reference 3) purport to demonstrate that the explicit and implied safety margins to tube rupture, as recommended in Regulatory Guide 1.121, are met by these criteria. However, based on the analysis submitted, the NRC staff cannot conclude that GPUN's application and use of fracture mechanics is appropriate.

The NRC staff concludes that the GPUN safety analysis contains significant errors and questionable assumptions in the areas of load development, material properties, and analytical conclusions. The NRC staff's position on specific aspects of these analyses is provided as follows:

A. Load Development As input to the fracture mechanics and NSCC analyses, the licensee used axial tube loads under normal operating and faulted conditions as obtained from Reference 4. Although submitted as topical reports, neither Reference 4 nor its successor, Reference 5. has been reviewed or approved by the NRC staff; as such, the magnitude of the axial tube loads, a critical element in these analyses, is a source of uncertainty and remains to be verified.

_- - __ __ ________ __J

The licensee has previously used these same axial tube loads as p rt of a justification for the kinetic expansion repair process used on TVI-1 steam genera tors. However, this was a unique plant-specific repair pr6 cess. The uncertainty associated with axial tube loads was not as crucial in this repair process as it would be to establish tube plugging limits Fracture mechanics methodology is a central cornerstone of TSCR 148. Therefore, the magnitude of axial tube loads is a source of uncertainty that should be verified.

Regulatory Guide 1.121 provides explicit and implied srfety margins for tube loading. Regulatory Guide 1.121 explicity states that tube loading should have a safety factor of 3.0 under nomal operating conditior.s. The Regulatory Guide further states that the margin of safety against tube failure under postulated accident conditions, such as a loss of coolant accident (LOCA),

steam line break, or feedwater line break concurrent with the safe shutdown earthquake (SSE) should be consistent with the margin of safety determined by the stress limits specificied in NB-3225 of Section III of the ASME Boiler and Pressure Yessel Code. In accordance with this criteria, under accident conditions a tube should not be loaded more than 0.7 times its ultimate load strength. The inverse of 0.7 is 1.47.8 and this is an implied safety factor under accident conditions.

GPUN extracted axial tube loads from Reference 4 and multiplied them by 3.0 and 1.428 respectively to obtain limiting load conditions as input to their a nal.' s i s . The resultant loads and associated stresses were derived on a linear elastic basis. Use of these derived linear elastic loads and strasses in NSCC analyses for circumferential defects of up to 70% throughwall peaetration results in predicted failure of the tube under faulted conditions (a] plied stress exceeds material flow stress) and demonstrates that the plegging criteria proposed by the licensee have less than a safety factor of 3.0 for nomal operation. However, since these values are beyond the proportional limit for the tube material, the licensee calculated " actual" tube loads in the non-linear regime and used these calculated " actual" tube loads as inputs to their analysis. Since these calculated " actual" tube loads i are smaller than the derived linear elastic loads, the licensee concluded that

! the predicted safety margins are satisfactory. While the licensee is correct in the assertion that non-linearly calculated tube loads and stresses are lower in magnitude'than associated pseudo-elastic values, it is technically unjustifiable to input non-linearly derived loads and stresses into NSCC analyses (a linear elastic calculational procedure).

In GPUN's evaluation of the faulted condition during a Main Steam Line Break l (MSLB) accident, pressure and maximum applied axial loads have been considered, but bending loads due to flow-induced vibration during MSLB have not. These I loads could be very significant for circumferentially oriented flaws and should have been addressed.

l l

I'n sumary: (1) the axial tube loads under normal operating and faulted conditions have not been verified by the NRC, (2) GPUN incorrectly inserted non-linearly derived loads and stresses into a linear elastic calculational procedure, and (3) bending loads due to flow-induced vibration during a MSLB have not been addressed.

B. Material Properties The tube material (Inconel 600) true stress-true strain curve needed to evaluate stability of defects by J-T analyses was obtained by the licensee using an estimation procedure. This estimation procedure has not been validated by GPUN. Also, since the licensee did not obtain true stress-true strain data for the tube material at 600 F, an important source of uncertainty was introduced into the J-T analyses perfonned.

Another significant source of uncertainty in the J-T analyses stems from the licensee's attempt to approximate the J-R curve for the tube material (Inconel 600) by employing a lower bound curve for 304 stainless steel with the power law fit. Without some' J-R data representative of Inconel 600 tube material, the licensee has not demonstrated that 304 stainless steel data are conservative with respect to the actual tube material. Furthermore, employment of data extrapolations based on a puver law fit could result in non-conservative J-T analyses. GPUN has not demonstrated that use of the power law fit is conservative in this application. Additionally, in order to be valid, J-R curves must be representative af the geometry to be analyzed (i.e., a part throughwall flaw and a thin ligament). However, the derived J-R curve has not been proven to be representative of the tube and flaw geometry.

The stress at which net section collapse of a flawed tube is predicted to occur is called flow stress. The value of flow stress is a critical factor in NSCC analysis. Reference 7 recomends using a value of 35m for flow stress when other data are not available. Sm is the allowable stress intensity at temperature from ASME Code Section III, Appendix I. The licensee used an estimated value of 35m for flow stress or specifically 69.9 ksi for Inconel 600 tubes at the temperatures of concern. The staff has concerns for this specific value of flow stress for two reasons:

1. First,'an estimate of flow stress is inappropriate when proven material yield and ultimate strengths at temperature are available, as they are in this instance. Taking an average of these values,

! the flow stress is detennined to be 63.5 ksi; this is approximately 10% lower than the licensee's estimate of 69.9 ksi. On the basis of this input parameter alone, the licensee's evaluation of axial and circumferential defects in tubes by NSCC analyses is nonconservative by values of at least 10%. As a consequence, the safety factors claimed to be inherent in the proposed plugging limit criteria do not exist. ,

l l

l

]

J.

l

2. Secondly, the use of 35m may not be a conservative value as an estima'te of flow stress. In Reference 8, it has been shown that using 35m as an estimate of flow stress is nonconservative for axial flaw orientations. Recently, in Reference 6, it was reported that 3Sm was also nonconservative:for circumferentially flawed pipe.

Admittedly, the conclusion in Reference 6 will. require further review,' but it does raise questions.on the validity.of using 3Sm as a conservative estimate of flow stress. Additionally, References 6,'7, and 8 were principally concerned with pipes. The licensee.is using these results for tubes, which have thinner walls and different -

characteristics.

1In sumary: (1) the tube material .true stress-true strain curve . estimation -

procedure was not validated, (2) GPUN did not demonstrate that the use of data from 304 stainless steel with a power law fit is conservative.for the actual tube material (Inconel 600), (3) GPUN did not demonstrate the derived J-R curve is representative of the tube and flaw geometry, and -(4) GPUN's

. value for flow stress, a critical factor in the analysis, is nonconservative

'when compared to proven material yield and ultimate strengths at temperature.

C. Analytical Results The' licensee's. sitated purpose for performing J-T instability analyses based upon elastic-plastic fracture mechanics and material toughness .

properties was to assure unstable tearing does not occur and net section collapse is the governing failure mechanism for a cracked tube. Since J-T analyses were applied to a configuration' (i.e., thin ligaments,-.large plastic zone size) not consistent with the . theory and principles of elastic-plastic fracture mechanics and because the material properties input to the analyses are suspect, the staff believes that GPUN has not j . achieved the purpose stated above.

Nonetheless, even if the NRC staff were to assume without adequate proof that

_ net section collapse is the governing failure mechanism for a cracked tube, i the licensee's analytical results do not support the' proposed tube repair

l. limit criteria ;in TSCR 148. Utilizing the licensee's own analytical results l with no changes for questions raised in Sections 3.A and B of this Safety Evaluation, the results indicate that the recomended safety factors of Regulatory Guide 1.121 are not met for axial defects greater than 60%

$,athroughwallpenetration. Specifically, for axial defects greater than 60%

• throughwall penetration, the licensee calculates a safety margin of

S2.?7 under normal operatina conditions (vice the recomended 3.0) and 1.188 l 'under accident conditions (vice the recomended 1.428). GPUN attempts to o ,lustify their conclusions using the followino two approaches.

.v l

3

- first, GPUN states that tube burst test data show significant margins over predictions. This is equivalent to saying that calculated safety factors of 2.27 and 1.148 are "close enough" to desired safety factors because the calculations are over-conservative. The generic tube burst test data referenced by the licensee are not sufficient to show that their calculations are conservative with respect to the THI-1 tubes. Tube burst data for Alloy 600 tubes are very limited and a majority of these tests were performed on tubes with thicker walls (0.05 inch) than those found at THI-1 (0.034 inch).

These tests did not include circumferential cracks. They were done on relatively new tubes with at most two and generally only one artificially induced defect on the outer tube surface. There are iraufficient tube burst data to conclude that desired safety factors are met wi1.h the thinner walled THI-1 OTSG tubes with various size defects. Additionally, the-NRC staff does not concur that GPUN calculations are over-conservative and believes they should be corrected for reasons discussed in Sections 3.A and 8 of this Safety Evaluation. These concerns cast serious doubts on the implied margin of safety in the licensee's submittal.

Second, by letter dated October 3,1986, GPUN attempts to justify their calculated safety margins by casting doubt on the validity of Reculatory Guide 1.121 for OTSGs. As discussed at the beginning of Section 3.0 of this Safety Evaluation, the NRC staff policy continues to be that Regulatory Guide 1.121 is applicable to all types of steam generators includino OTSGs. However, licensees are free to use approaches or methods other than those described in Regulatory Guides if they are technically justified. In their October 3, 1986 letter, GPUN offers suggestions for an alternative approach to calculating safety factors for OTSGs. However, GPUN does not provide a detailed technical justification for an alternative approach. In absence of any technical justification of an alternative approach, the NRC staff will continue using the guidance of Regulatory Guide 1.121 as our acceptance criteria.

4.0 CONCLUSION

S The elastic-plastic fracture mechanics and net section collapse analyses performed by the licensee in support of proposed steam generator tube plugging limit criteria are not adequate to demonstrate that such criteria will meet the factors of safety recommended by Regulatory Guide 1.121.

Neither does the licensee offer an acceptable alternative to the safety factors of Regulatory Guide 1.1?1. Further, the licensee's analyses contain significant errors and questionable assumptions in the areas of load development, materials properties and analytical conclusions. The NRC staff, therefore, rejects the proposed 70% tube plugging limit.

References

. 1. Letter from H. D. Hukill (GPUN) to J. F. Stolz (NRC), TMI-1, " Technical Specification Change Request No. 148", dated November 6, 1985.

2. Letter from H. D. Hukill (GPUN) to J. F. Stolz (NPC), TMI-1, " Regulatory Guide 1.121 Extent of Applicability to OTSG's", dated October 3,1986.

3. GPUN Report TOR-690, " Comparison of GPUN Proposed 0TSG Tube Plugging Criteria to Regulatory Guide 1.121", October 1985.

4. Babcock & Wilcox Report 8AW-1588, " Determination of Minimum Required Tube Wall Thickness for 177-FA Once-Through Steam Generators", April 1980.

5. Babcock & Wilcox Report BAW-10146, " Determination of Minimum Required Tube Wall Thickness for 177-FA Once-Through Steam Generators", October 1980.

6. Memorandum from Michael Mayfield, "0TNSROC Technical Note: Comments nn Stainless Steel Pipe Flaw Evaluation Using IWB-3640", May 13,1986.

7. Ranganath, S. and Norris, D. M., " Evaluation Procedure and Acceptance Criteria for Flaws in Austenitic Steel Piping", Draft No. 10, Sponsored by Subcommittee on Piping, Pumps, and Valves of PVRC of the Welding Research Council, July 1983.

8. EPRI Report, " Evaluation of Flaws in Austenitic Steel Pipe", April 11, 1986.

l l