ML20052C179

Updates Status of Work on Plant Steam Generator Failure Analysis,Repair Plans & Safety Evaluations.Confirmation Requested of Util Position That Hearings Are Not Needed. Marked-up Viewgraph TMI-1 OTSG Status Review, Encl

ML20052C179
Person / Time
Site:	Three Mile Island
Issue date:	04/30/1982
From:	Hukill H GENERAL PUBLIC UTILITIES CORP.
To:	Eisenhut D Office of Nuclear Reactor Regulation
References
5211-82-104, NUDOCS 8205040468
Download: ML20052C179 (110)

Text

_ _ _ - - .

l l

GPU Nuclear NUCIMr P.O. Box 480 Middletown, Pennsylvania 170S7 717-944-7621 Writer's Direct Dial Number:

April 30, 1982 5211-82-10 O!

n Office of Nuclear Reactor Regulations " q.

Attention: D. G. Eisenhut, Director Division of Licensing ~

h

~t -

United States Nuclear Regulatory Commission f " @y $

Washington, D. C. 20555 -

3 []l} -

F2

" J ;f 'y

Dear Sir:

"6 Three Mile Island Nuclear Station, Unit 1 (TMI-1 N o, Operating License No. DPR-50 Docket No. 50-289 The purposes of this letter are:

o To provide an updated status of our work on the TMI-1 Steam Generator Failure Analysis.

To outline the proposed repair and our plans to complete needed safety evaluations and fulfill regulatory requirements.

o To request early NRC action to confirr that you agree with our approacn including the fact that hearings are not needed and will not be required by the NRC.

We recognize that the possibility of an unreviewed safety question must be considereo in accomplishing the repair program itself, as well as in the return of the steam generators to service. We will, of course, carry out the required safety evaluation and submit it to the NRC whether or not an unreviewed safety question is identifieo.

The material enclosed is, in large part, the same as that presented to the NRC Staff and consultants, in the publicly noticed meeting on April 7, 1982.

While some confirmatory work remains to be completed, we are confident that the repair program will fully resolve safety concerns associated with the damaged tubes and result in return of the steam generators to a condition meeting all design and safety performance criteria.

00l

//

8205040468 820430 PDR ADOCK 05000289 P PDR GPU Nuclear is a part of the General Public Utihties System

Mr. D. G. Eisenhut' 5211-82-104 As identified in the course of our presentation and discussed with you and your Staff late that day, we believe that return of the TMI-1 steam generators to service following repair can be accomplished without any degradation in their ability to meet all design requirements, thusly no significant hazard exists. The basis for these conclusions is included in the attachments and can be summarized as follows:

The failure mechanism is now known.

The attack is highly localized within the upper tubesheet. It occurred when the plant was cold.

This location allows for repair in such a way as to return the Unit to within design conditions.

From a safety standpoint, the location of attack and the fact that it occurred when the plant was cold do not pose the significant increased risk of large abrupt increases in primary to secondary leakage typical of other steam generator tube problems.

In addition to our normal safety reviews and evaluations of this situation, we have established an independent safety review group of highly qualified experts for this matter. The group will provide a final written assessment which will be made available to the NRC. Additional information on the scope and composition of the group is provided in the attachments.

On April 7, 1982 we also discussed with the Staff whether a public hearing on the steam generator repairs would be required or recommended prior to resumption of operations. We noted that, even if, contrary to our expectations, a license amendment is necessary, we believe that we and the Staff would conclude that no significant hazards considerations were involved.

Return of TMI-l to service is of great importance to the Company and its customers and to proceeding with the cleanup of TMI-2. If it were now known that a public hearing were to be held or offered, we would, of course, be urging prompt issuance of an appropriate notice in order that at least the hearing preliminaries could be accomplished in parallel with the Staff's review. However, we see no safety considerations which would warrant the time and expense of a public hearing; and, a hearing, even if noticed at this tiine, would, in all probability, delay acceptance of the program many months beyond completion of the Staff review.

., j

I o .

Mr. D. G. Eisenhut 5211-82-104 t

In view of the information above, we request written advice as to whether the Staff would be prepared to take the position (1) that a public  ;

hearing, whether or not requested by others, would not be legally required to '

be held prior to issuance of a license amendment, if needed, and resumption of operation, and (2) that the staff would not on its own initiative propose such a hearing. '

~

We recognize that the so-called Sholly Amendment to the NRC Authorization Bill may be enacted. Under the pending legislation, we understand the Commission would be required to issue implementing regulations concerning the criteria for determining whether an amendment involves no '

significant hazards consideration and other procedural matters. Therefore we '

would urge that the Staff reviews of this matter be made with the possible need for meeting such criteria in mind.

An early reply to this letter including the basis for any NRC disagreements with our position is requested.

Very truly yours, f

fM>

H. D. Hu <ill Director, TMI-1  ;

i e

HDH/k1k i

Attachments cc

R. Haynes J. Stolz R. Jacobs T. Novak i

i i

l l

l l

l i

UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION In the Matter of )

)

METROPOLITAN EDISON COMPANY ) Docket No. 50-289

) (Restart)

(Three Mile Island Nuclear )

Station, Unit No. 1) )

SERVICE LIST Chairman Nunzio J. Palladino Dr. Reginald L. Gotchy U.S. Nuclear Regulatory Commission Atomic Safety and Licensing Appeal Washington, D.C. 20555 Board Panel U.S. Nuclear Regulatory Commission Commissioner Victor Gilinsky Washington, D.C. 20555 U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Ivan W. Smith, Esquire Chairman Commissioner John F. Ahearne Atomic Safety and Licensing Board U.S. Nuclear Regulatory Commission U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Washington, D.C. 20555 Commissioner Thomas M. Roberts Dr. Walter H. Jordan U.S. Nuclear Regulatory Commission Atomic Safety and Licensing Board Washington, D.C. 20555 Panel 881 West Outer Drive Docketing and Service Section Oak Ridge, Tennessee 37830 Office of the Secretary U.S. Nucicar Regulatory Commission Dr. Linda W. Little Washington, D.C. 20555 Atomic Safety and Licensing Board Panel Gary J. Edles, Esquire 5000 Hermitage Drive Chairman Raleigh, North Carolina 27612 Atomic Safety and Licensing Appeal Board James M. Cutchin, IV, Esquire U.S. Nuclear Regulatory Commission Office of the Executive Legal Director Washington, D.C. 20555 U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Dr. John H. Buck Atomic Safety and Licensing Appeal John A. Levin, Esquire Board Panel Assistant Counsel U.S. Nuclear Regulatory Commission Pennsylvania Public Utility Commission Washington, D.C. 20555 P.O. Box 3265 Harrisburg, Pennsylvania 17120 Christine N. Kohl, Esquire Atomic Safety and Licensing Appeal Robert Adler, Esquire Board Panel Assistant Attorney General U.S. Nuclear Regulatory Commission 505 Executive House Washington, D.C. 20555 P.O. Box 2357 Harrisburg, Pennsylvania 17120

1

' John E. Minnich William S. Jordon, III, Esquire Chairman, Dauphin County Board Harman & Weiss of Commissioners '1725 Eye Street, N.W., Suite 506 Dauphin County Courthouse Washington, D.C. 20006 Front and Market Streets Harrisburg, Pennsylvania 17101 Robert Q. Pollard 609 Montpelier Street Walter W. Cohen, Esquire Baltimore, Maryland 21218 Office of Consumer Advocate 1425 Strawberry Square Chauncey Kepford Harrisburg, Pennsylvania 17127 Judith H. Johnsrud Environmental Coalition on Nuclear Power Jordan D. Cunningham, Esquire 433 Orlando Avenue 2320 North Second Street State College, Pennsylvania 16801 Ilarrisburg, Pennsylvania 17110 ,

Marvin I. Lewis Ms. Louise Bradford 6504 Bradford Terrace TMI ALERT Philadelphia, Pennsylvania 19149 1011 Green Street ,

liarrisburg, Pennsylvania.17120 Marjorie M. Aamodt 1

R. D. 5 Ellyn R. Weiss, Esquire Coatesville, Pennsylvania 19320 Harmon & Weiss 1725 Eye Street, N.W., Suite 506 Thomas J.-Germine, Esquire Washington, D.C. 20006 Deputy Attorney General Division of Law - Room 316 Steven C. Sholly 1100 Raymond Boulevard Union of Concerned Scientists Newark, New Jersey 07102 1346 Connecticut Avenue, N.W., Suite 1101 l Washington, D.C. 20036 j

i Cail B. Phelps .

ANGRY 245 West Philadelphia Street York, Pennsylvania 17404 4

i 9

5

.-- , . . - , . w - --

. - April 28, 1982 ATTACHMENT I OTSG Reoair Technical Summary The following information provides a summary of the failure analysis program, repair plan / program and licensing aspects of the TMI-1 OTSG tube leak problem. The discussion refers to the various slides which were used on April 7,1982, to present a status review to the NRC. These slides have been bound and are attached hereto.

I. Failure Analysis Procram A. What Hannened?

Three Mile Island Unit I has remained in a shutdown condition for approximately 3 years as ordered by the Nuclear Regulatory Commission on July 2, 1979. Slide 4 shows the primary system pressure and temperature during this shutdown period. As shown, the primary coolant system was heated up and pressurized in late August /early September,1981 for hot functional testing (HFT).

This HFT was conducted to test equipment and plant modifications in anticipation of completion of the TMI-1 restart hearings.

Subsequent to this heatup, the plant was again placed in a cold shutdown condition. In late November the primary system pressure was increased to 45 psi for further equipment testing. At this point in time, both activity measurements and baron contamination on the steam generator secondary side indicated the presence of primary to secondary leakage. The primary coolant system was depressurized and drained down to a level just above the steam generator upper tubesheets. The secondary side was pressurized with nitrogen to approximately 15 pounds and a visual inspection was conducted in order to identify the leaking steam generator tube or tubes. During this testing in early December, a total of over 130 tubes were identified as leaking in the two steam generators and a non-destructive examination of the OTSG tubes using eddy current techniques was commenced. The initial eddy current examinations indicated that there were probable defects in thousar.ds of tubes and GPUN established an internal task force to evaluate the status of the TMI-1 steam generators and reactor coolant system.

B. What Resources Are Assistino GPUN?

As soon as GPUN realized the extent of damage to the steam generators, an extensive failure analysis and industry experience review was commenced. GPUN requested assistance from numerous experts in the nuclear power industry and research laboratories as detailed on Slides 2 and 24. To date, 19 failed tubes have been removed from the TMI-l steam generators with independent failure analysis being conducted by B&W, Battelle Columbus and Westinghouse. There are plans to pull an additional six to ten tube samples to be used to confirm the status of steam generator tubes that have been accepted by eddy current evaluation and for corrosion testing.

, April 28, 1982 ATTACHMENT I (cont'd)

C. What Do We Know?

1. Eddy Current Results Eddy current examination of the steam generator tubes proceeded during December and early January utilizing the standard differential probe and multiple frequencies. This examination has proven to be an acceptable examination technique in the areas of the steam generator tubes between the upper and lower roll transitions. The adequacy of the standard differential probe has been demonstrated by both 100% correlation with defects in the removed tube samples and good correlation between the standard circumferential coil and pancake coil and the standard differential probe and absolute probes. By an empirical process, the eddy current inspection contractor (Con Am) identified a frequency mix using 400 kilohertz and 800 kilohertz signals which minimized the roll transition

• signal and improved the sensitivity to detect defects at this t

location. We later confirmed by the inspection of tube samples that defects existed at the roll transition. However, it was soon recognized that the standard differential probe could not adequately define the status of the roll transition, tube exit, and heat af fected tube area because of the signal interference from the transition and exit. Therefore, GPUN in conjunction wiv Con Am and Zetech developed an absolute probe with four coils in order to inspect these areas at the very top and bottom of the tubes. In addition, the eddy current manipulator was modified to allow rotation of the new 4x1 probe resulting in 360 coverage. Mockup testing has confirmed the adequacy of the 4x1 probe to detect defects 40% through wall or-greater. The 100% correlation with metallurgical samples also confirms the adequacy of the 4x1 probe to detect the type of defects that exist in the TMI-1 OTSG.

Although eddy current examination of the steam generators is continuing, GPUN has projected the probable defect population as a function of both radial and axial position within the steam generator. These projections are shown on Slides 6, 7 and 8 and result in an estimate of between 8,000 and 10,000 tubes which will require repair due to defects at or near the roll transition. These projections specifically exclude the location in the tube immediately below the seal weld to the upper tubesheet. The new 4x1 probe is qualified to detect defects at this location but both the number of tubes that are defective at this location and the implications of defects at thi location have not yet been finally determined. However, it is expected that a majority of all the tubes will be shown to have defects in this location. The existence of cracks in the tubes at this location was first noticed by metallurgical examinations of the pulled tube sartples.

'See Slide 22 for location of the roll transition.

, . April 28, 1982 ATTACHMENT I (cont'd)

2. Failure Analysis The failure analysis activities have centered around the ,

metallurgical analysis of tube samples, the chemical analysis of TMI-1 water samples, operating history review, OTSG fabrication history review and tube stress analysis. These activities have been aimed at identifying the combination of material condition, tube stress and aggressive environment that caused damage to the OTSG tubes at TMI-1. The failure mechanism should, based on evaluating the combination of these conditions, be able to explain the timing of cracking, the material failure mode, the contaminant source and the axial / radial crack distribution as identified by eddy current.

With regard to material conditions, both the fabrication history and the metallurgical analysis of the tube samples indicate that the tube material is sensitized. Based on the tube metallurgical analysis, the cracking has been caused by intergranular stress corrosion initiated from the inside diameter. In order to explain the circumferential orientation of the cracking, it is necessary to determine when the axial stress in the tubes exceeded the hoop stress. The fabri:ation of the steam generator involves installing the tubes with an axial tensile preload in order to minimize the potential for 4

tube buckling during operation. In addition, cooldown transients apply axial tensile loads to the steam generator tubes. During periods of ncrmal operating pressure and temperature, the hoop stress is tensile and the axial stress is compressive in the tubes. Thus, it has been concluded that the circumferential cracks identified in the TMI-l OTSG tubes must have occurred during a period of cooldown or cold shutdown.

With regard to the environment, initial analysis of the contaminants on the cracked surface identified the presence of sulfur. An extensive evaluation of potential sulfur contamination sources and the existence of sulfur contamination in TMI-l systems was conducted. The results of this evaluation are summarized on Slides 32 and 33. Although there are several potential sources for sulfur contamination, an evaluation of their concentrations, volumes and dilution in the reactor coolant system has led to the conclusion that the most probable source of sulfur contamination in sufficient quantities to result in aggressive concentrations in the reactor coolant system is the sodium thiosulfate. The scenario for the injection of sodium thiosulfate involves: 1) leakage from the sodium thiosulfate tank to the suction of the building spray pumps over the long period of shutdown, 2) surveillance testing l

of the building spray pumps in June, August and September of 1981 in anticipation of TMI-1 restart (resulting in addition.of sodium thiosulfate contamination to the BWST), 3) subsequent l

, April 28, 1982 ATTACFNENT I (cont'd) injection of BWST water into the reactor coolant system prior to and during the August / September,1981, hot functional testing (HFT) and 4) reduced (aggressive) sulfur species formed i during HFT.

The overall summary of the failure analysis status as of April 7, 1982, is detailed on Slide 26. It indicates that the TMI-1 OTSG tube cracking is caused by intergranular stress corrosion, initiated on the ID surface, and sulfur is identified as the contaminant which most probably caused the cracking. The cracking probably occurred subsequent to the RCS heatup and cooldown that was conducted in late August /early September, 1981, and was then detected the next time the reactor coolant system was pressurized in late November 1981. As noted on Slide 31, the primary system water level subsequent to the August / September heatup varied from about the 13th tube support plate to the upper tubesheet during the month following the heatuo. It is believed that this water level is important in that it provided an opportunity for concentration of the contaminant on the tube surface in sufficient quantity to initiate and rapidly propagate the tube cracking.

3. Corrosion Testino GPUN has initiated numerous corrosion tests in order to duplicate and, therefore, confirm the failure scenario, identify whether or not the water currently in the TMI-1 reactor coolant system is aggressive, and confirm that subsequent to the planned repairs the TMI-1 OTSG tubing will be acceptable for continued operation. In addition, a testing program is being implemented to identify what reactor coolant 1

system clean up of sulfur contamination if any is required and how this clean up should be accomplished. This program is summarized on Slide 83 with hydrogen peroxide identified at l this time as a likely candidate additive for RCS cleanup. The preliminary corrosion test results are summarized on Slide 35.

4. Reactor Internal Insoection In order to confirm that other mGerials within the reactor coolant system have not been damaged by the environment which caused the steam generator tubes to crack, GPUN has initiated a comprehensive review and inspection of reactor coolant system components and materials. This program involves removal of the reactor vessel head and the inspection or test of approximately 1000 items. The objectives, program plan and inspection plan are summarized on Slides 44 through 47. The program involves classifying items by material condition, environmental exposure, applied stress, and safety significance. Candidates have been selected for inspection and testing that are April 28, 1982 ATTACH 4ENT I (cont'd) representative of the worst metallurgical, chemical or stress conditions expected with an additional objective of minimizing man-rem exposure as much as practical. This inspection program is well underway and has shown no evidence of attack on any reactor internals.to date.

D. What Future Work Is To Be Done?

Slides 9 and 88 provide a summary of the future work and/or decisions that still remain. This remaining work entails: 1) completion of the eddy current examination in the heat affected area and other work to complete determination of the status of OTSG tubes, 2) completion and evaluation nf the inspection of other RCS components and materials, 3) establishing the need for and the approach to be used in cleaning up sulfur in the reactor coolant system, 4) selecting, removing and evaluating the final tube samples, 5) completion and evaluation of the corrosion test program, 6) the development, qualification testing, and implementation of a tube expansion process in order to repair the OTSG's.

II. Reoair Plan /Procram A. What Is The Reoair Criteria?

The repair criteria is summarized on Slides 48, 49 and 50. The criteria require adequate pullout load capability, primary to secondary leakage as low as reasonably achievable and within the reactor coolant leakage and radioactive effluent limits of the technical specifications, and confirmation that the repair will maintain the thermal and hydraulic performance within the acceptance criteria for both normal operating and design basis accident conditions.

8. How Will The OTSG Be Reoaired?

The failure analysis information available to date continues to indicate that the large majority of the defects in the steam generator tubes are located within the first few inches below the upper surface of the upper tube sheet. The tube material below this area is essentially like new except in limited areas where surface intergranular attack and small pits several grains deep exist (as found by metallurgical examination) and where defects have been identified by eddy current. The local areas are considered typical of S.G. tubes after several years of service.

In accordance with the technical specification requirements, GPUN plans to plug and remove from service all tubes with indications of defects outside the tube sheet area. For those tubes tt.,.t are defective within the tube sheet, the repair plan is to expand the tube within the tube sheet below the defect indications so that the repaired tube has adequate strength and leak tightness.

. April 28, 1982 ATTACHMENT I (cont'd)

Preliminary plans call for 1) cleaning the tube ID in the area of the repair, 2) heating the upper tubesheet tube crevice to drive moisture out of the crevice, 3) expanding the tube for a distance of approximately 8-10 inches from the top of the upper tube sheet in order to close the crevice, 4) cleaning the tube ID to remove residue from the expansion process (if required), 5) conducting a second expansion (if required for leak tightness or load carrying capability) approximately 1 inch long using a mechanical roll at a location near the bottom of the expansion and below any defect indications. Depending on the total number of tubes requiring repair GPUN may decide to conduct the expansion on all tubes. As shown on Slide 77, a 10 inch expansion leaves approximately 376 tubes with defects in the tube sheet that may either be plugged or

expanded depending on current tooling and future sleeving l considerations.

The repair program outlined above is possible because of the condition of the remaining tubing and the fact that the defects are located within the tube sheet area. Given these facts, it is possible to establish a new mechanical and essentially leak tight joint below the defect in a manner which is similar to the original configuration of the steam generator. The original configuration involved a 1 inch minimum roll at the top of the existing tubes.

An extensive qualification program involving pullout load tests, thermal cycling tests and leak tests to be conducted over the next several months is expected to provide confirmation of the acceptability of this repair. Slide 60 and 61 provide a summary of the points to be addressed by the qualification program. In addition, GPUN's review of the industrial experience in both steam generator and other reactor components indicates that mechanical joints have been used successfully in numerous applications in order to form an essentially leak tight joint. GPUN intends that the repair process and program as currently identified with subsequent leak testing and leakage monitoring, will provide assurance that the probability of abnormal primary to secondary leakage during operation is very low and the generator will conform to its original design basis.

Slide 84 details the testing plans to be used to confirm the adequacy of the OTSG repair. These testing plans involve drip and bubble tests with a differential pressure of 150 psi and a leak test with a differential pressure of 1500 psi. The sensitivity during the bubble test is expected to be approximately .1 gal per day / tube and the sensitivity during the 1500 psi differential pressure test is about 10 gal, per day / steam generator. These sensitivities give us confidence that we can confirm adequate leak tightness of the steam generator prior to criticality, thus continuing to assure minimum risk to the health and safety of the public. Once the reactor is critical, leak rate will be monitored continuously with a sensitivity of approximately 10/GPD/OTSG.

Corrective action will be taken to plug or repair steam generator tubes based on excessive total leak rate and/or the rate of change of leak rate.

ATTACHENT 2 License / Technical Saecification Conformance Summary The TMI-1 Technical Specifications identify the key parameters that set the envelope within which the margins of safety originally reviewed and approved by the NRC will not be exceeded. The Code of Federal Regulation Title 10 Part 50.59 established conditions that if met will allow a licensed facility to be modified without obtaining prior NRC approval. In that section specific tests are outlined to guide licensees in properly making that determination. The tests are twofold. First the change must not require a change to the unit technical specifications and second, the change must not involve an unreviewed safety question.

An unreviewed safety question is defined in that part to exist "(1) if the probability of occurrence cr the consequences of an accident or malfunction of equipment important to safety previously evaluated in the safety analysis report may be increased; or (ii) if a possibility for an accident or malfunction of a different type than any evaluated previously in the safety analysis report may be created; or (iii) ir the margins of safety as defined in the basis for any technical specification is reduced."

At present we are evaluating the above criteria regarding the extent of damage, the cause of failure, and the repair method. This evaluation will be the subject of a future report to the staff. The completion of the repair activities as discussed in the Technical Summary and the attached slides is not expected to increase the possibilities or probabilities of events or consequences of events as compared to the current licensing basis.*

With regard to the TMI-1 Technical Specification, the key functions of the steam generator are retention of reactor coolant pressure boundary integrity, and the implied maintenance of adequate heat transfer capability and reactor coolant flow area. The following discussion highlight the portions of the Technical Specifications that focus on the maintenance of steam generator tubes as a portion of the reactor coolant pressure boundary. As noted in the attached presentation the material safety functions related to heat transfer and adequate RCS flow area should not be significantly impacted by the repair processes.

• See Northern States Power Company (Monticello Nuclear Generating Plant, Unit 1), 00-79-5, 9 N.R.C. 588, 591 (1979) for an example of the Staff's application of the definitions of "unreviewed safety question" found in 10 CFR 50.59. Note particularly the emphasis placed on the consequences of an accident or malfunction, rather than on the mechanisms of causation.

ATTACHMENT 2 (cont'd) i The TMI-1 Technical Specifications require in Section 3.1.1.2 that both steam generators be operable for temperatures above 2500F. The basis of this specification is to insure system integrity against leakage under normal and transient conditions. One requirement in declaring the generators operable is execution of T.S. 4.19, which defines a periodic test program intended to demonstrate that the structural integrity of the tube

' portion of the OTSGs is maintained (Bases, T.S. 4.19).

T.S. 4.19 addresses specifically the tube area between the tubesheets as requiring inspection (T.S. 4.19.4.8) and defines accceptance criteria for that inspection area (T.S. 4.19.4). Further, the specification exempts from consideration leakage in the area of the tube-to-tubesheet joint (T.S.

4.19.3.C.1).

Not addressed is the area within the tubesheets ex:luding the tube-to-tubesheet joints, although clearly some portions of this area would be expected to affect the tube's structural integrity. This portion of the-tube coincides with the location of the majority of cracks identified in the OTSG tubes. Therefore, GPUN has examined the regulatory and design bases for the OTSG tubes to determine for what portion of the tube area in the tubesheet the tube inspection acceptance criteria (T.S. 4.19.4) should be applied, and for what portion of the tube area in the tubesheet the

acceptance criteria seem to be inappropriate. '

This question was considered looking at a repaired steam generator tube, expanded to some point below our identified cracking. T.S. 4.19 and the Regulatory Guides 1.121 and 1.83, refer to the structural integrity, requirements applicability as the " tube portion of the reactor coolant

, pressure boundary" and the " heat transfer surface." The GDC in 10CFR5] use 4

the same words in establishing design bases. It is the purpose of the expansion qualification program to demonstrate thst the new tube to tubesheet joirt will carry all normal and transient loads and remain essentially leaktight. The tube area above that joint would no longer be a part of the heat transfer surface or the reactor coolant pressure boundary.

As such, it would be inappropriate to apply the acceptance criteria for the freestanding tube (T.S. 4.19.4) to the remainder of the tube, however, GPUN would apply the acceptance criteria of the balance of T.S. 4.19 and plug tubes, as required. After repair, the cracked areas are no longer part of

- the RCS pressure boundary or heat transfc. surface.

4 1

, ATTACIGENT 3 THIRD PARTY REVIEW OF THI-1 OTSG REPAIR PROGRAM CHARTER I. PURPOSE It is the intended purpose of this " Third Party Review" (TPR) to provide e time!y, independent, obj ect ive , safety evaluation of all activities detined in this charter for conformance to: 1) the NRC rules &

regulations governing the operation of TMI-1; and 2) the adequacy of the steam generator repair program that will allow safe operation of the nuclear unit.

II. SCOPE The s cope of this review is generally limited to activities associated with the identification of failure mechanisms and repairs of the TMI-1 Once Through Steam Generators (OTSG's). The specific task areas to be reviewed are described in more detail in Section IV of this charter. It is the intent of CPUN Management to fully develop and implement repairs to the TMI-l OTSG's within the provisions of 10CFR50.59. It is expected that the TPR will promptly notify GPUN Management of any circumstance not already identif Aed by GPUN, that fails to meet these standard s.

III. MEMBERSHIP The membership of the TPR body shall include individuals with expertise in the following specialty areas:

A. Steam Generator Design and Performance - 1 member B. Chemistry - 1 member C. Materials - 1 member D. Stress Analysis - 1 member E. Safety Analysis - 1 member F. Plant Operations - 1 member C. Non-Destructive Examination - 1 member The TPR shall have a GPUN individual assigned as Secretary for the review. He will be the general interf ace for the TPR membership and CPUN. A Chairman will be elected and report the results of the review directly to the Vice President - Technical Functions in this assignment.

The Secretary will arrange for all review material, meetings and rec ord keeping for the team. Any specific information requests within the scope of the TPR should be to the Secretary. 'Ihe Secretary will be o non-voting member in any matters of the TPR seeking concensus cpinions.

l I

1 Rev. 3, 4/27/82

ATTACHMENT 3 (cont.)

M:mbars cf tha .TPR w'ill eithar be from cutside the GPUN org'cnizcticn or from tha psrtion of GPUN net rssp:nrible for tha stee:2 gsnarctor repair er TMI-1 operations .

IV. SPECIFIC REVIEW AREAS A. Failure Analysis Program This program is intended to identify the cause of tube cracking and means to arrest it. The program will also include an evaluation of other portions of the reactor coolant system to determine if corrosion mechanism extended out of the steam generator boundaries.

B. Eddy Current Examination Program This program is to develop and implement an eddy current examination method to identify the extent of the tube cracking problem.

C. OTSG Performance Evaluation It is the object of this ef fort to evaluate the impect of the repair procedures on the performance of the steam generators, especially in the area of safety analysis .

D. Repair Criteria This program is intended to provide guidance concerning the type of repair to be done on the damaged tubes.

E. OTSG Repair Program This program covers the actual repair of the. steam generators.

V. MEETING FREQUENCY The TPR shall initially meet to receive a presentation by GPUN and its consultants on the current status and direction of the OTSG Repair Program. At that meeting, they will be presented with initial reference c:aterial that will enable them to assess products then available.

A secot, meeting will be scheduled by CPUN to make a final presentation to the TPR prior to initiation of the final production repairs.

Other meetings may be held at the discretion of the TPR.

VI. RECORDS The Secretary shall maintain records of all meetings of the TPR. The Secretary shall also maintain a record of all documents reviewed by the nembership. Portions of the material may be proprietary in nature.

Appropriate arrangements shall be made to protect proprietary information when it is used.

Rev.3,4/27/82

ATTACHMENT 3 (cont.) ,

Trcnscripts may be taken of the final review meeting and used as part of tha documer.e stion package available to the NRC in support of the OTSG R: pair Program.

1 A final report will be prepared by the review team which sununarizes its findings and conclusions - regarding the safety adequacy of the repair program. The report should make an explicit finding that the approach proposed by CPUN is adequate if that is the conclusion of the review, a

i Rev. 3,4/27/82 4

. l

. . i l'

ATTACHMENT 4 MEMBERS, TMI-l Third Party Review e STEAM CENERATOR DESIGN AND PERFORMANCE ED J. WAGNER - Director, Engineering and Design, Breeder Reactor Division, Burns and Roe, Inc.

e CHEMISTRY DAVID J. MORGAN - Nuclear Analyst, Plant Systems Analysis Section, P.P. & L.

e MATERIALS DR. R. W. WEEKS - Associate Director Materials Science Division, Argonne National Laboratory e STRESS ANALYSIS DR. ARTURS KALNINS - Professor of Mechanics, Lehigh University

e SAFETY ANALYSIS WILLIAM LAYMAN - Department Manager, Generic Safety Analysis Electric Power Research Institute e PLANT OPERATIONS STAN HOLLAND - System Production Engineer - Duke Power Co.

e NON-DESTRUCTIVE EXAMINATION STEPHEN BROWN - Principal Engineer - Inspection, Electric Power Research Institute e SECRETARY ED WALLACE - Manager, PWR Licensing, CPU Nuclear s

i ENuclear TMI-1 OTSG Status Review ,

1 .

l l l April 7,1982 i l

l l

/

l l

l l

. .. _. 1._ .

TMI-1 OTSG STATUS REVIEW I. INTRODUCTION -

R. F. WILSON II. FAILURE ANALYSIS -

DR. R. L. LONG 1

l t

III. STEAM GENERATOR REPAIR -

J. PEARSON IV, OTSG REPAIR PROGRAM OVERVIEW - D. G. SLEAR V. CONCLUSIONS /

SUMMARY

R. F. WILSON l

1 1

. BETHESDA, MARYLAND 4/7/82 Q - -,--e--

2 Organizations Actively Working with GPUN on Steam Generator Program B&W, Lynchburg and Alliance Research Labs EPRI and consultants Battelle Laboratories MiT 4

ORNL (CONEWl

,,a,p< .< uw - n 4/7/82

3 TMI-1 Steam Generator i l

ELEVATION CROSS SECTION

! W 1

t PRIMARY S10E D ra -v >

(INSIDE TUBES) J L g? ~T, ',' 2lg. . -!_ . " _

, UPPER fE5- - u i TUBESHEET (UTS) [MyWirs-52[

AUXlLIARY F EDWA Yh yg7g ,, , _ , _ } }g ,

gqjj';;

(AFW) mn -

we

@ $M 23Q . h.a Q+v. - - _ .

! 'l l' nm -

b j.,:il

'f:'T -

T l 3 ..fii,'

a6_. . 2': *D[$$  ? ' ~ "%

m,.

STEAM OUTLET ,

ela3dbh,bN{N-g LANE .-

l F.1AIN FEEDWATER l yha ni'rtr m.

OTUBE N0ZZLES (MFW) y

- -l s .n,.u* i, ..

i

[ e L

(i, g-I SECONDARY SIDE __ (

• dEXTERNAL TO TUBES)

(;dm .[ _

l k I .Y

( w 1* " -

1 6

m-Weight, operating. . . . . . . . . . . . . . 637 tons

,,. H eig ht. . . . . . . . . . . . . . . . . . . . . . . 73 feet

"'7"" * - Primary flow. . . . . . . . . . . . . . . . . . 6 9X106 #/hr.

l SUP =- Steam flow . . . . . . . . . . . . . . . . . . . 6.1 X 106 #/hr.

PLATES "i

=- Number tubes . . . . . . . . . . . . . . . . . 15531 1,,o , Tube size, material . . . . . . . . . . . . . 0.625" od, .034 wall l LOWER > inconel 600 SECONDARY j y" ' <l i.

Manufacture date . . . . . . . . . . . . . . 5/69 to 11/70 HANDHOLD i 1 I

1 4/7/82

i i

4 Reactor Coolant System 4

550-

-2585 500-l -2350 450-( ll

-2115 l 400- 1 ll l

II q

-1880 y l350- lll l lt , -1045 E a 300- ll I m. l i

I

=

l -1410

$2 3 250- lli l

H E -1175 ll l 1 1 200- i; 3

, l -900  !

v '

150- 1 llll 1

-905 100- i

-470 50 p:..................c........:.:..,

. ql -235 AMJJASONOJFMAMJJASONDJFMAMJJASOND 1979 1980 l

l l

I l

l l.

I l 4/7/82 l

_. = -

5 TMI-1 STEAM GENERATORS EXTENT OF ATTACK e STEAM GENERATOR e NUMBER OF LEAKING TUBES 200 TO 500 e NUMBER OF TUBES WITH INDICATION 8000 TO 10,000 0F SIGNIFICANT DEFECTS l

e UNRESOLVED AREA AT SURFACE OF UNKNOWN UPPER TUBESHEET e ELSEWHERE e MATERIALS POTENTIALLY SUBJECT TO SIMILAR ATTACK USED ELSEWHERE IN REACTOR

- EXAMINATION PROGRAM GETTING UNDERWAY

- ATTACK, IF ANY, REQUIRES RIGHT COMBINATION OF MATERIAL CONDITION, STRESS, LOCAL ENVIRONMENT l

4/7/s2 -

n. - - - - - ~ -

TMI-1 Stocm Gan:rcter Tube Cracking SEAL WELD 0"

1"

/ TOP SURFACE TUBE SHEET

_ N ROLL TRANSITION 3"

4" '

100-5" g- .\\ so- -

so  ;-

1125 TOTAL DEFECTS BELOW 2" N E '

w i o so- 13

> 729 TOTAL DEFECTS BELOW 4" N

p So- .,

N $

D;i o  %$

(" TOTAL DEFECTS BELOW 6"

.;da

,' k "d(j NOTE: EXCLUDES ECT DATA

~

ADJACENT TO SEAL WELD 98 ,

?;

xx N xs e;: m BOTTOM SURFACE O 5 53 4 5 6 "

[2 18 24 TUBE SHEET DISTANCE DOWN TUBE /TUBESHEET - INCHES b

g a , y e-a--> . .,

7 i Potentially Defective Tubes (Projection of Eddy Current Data)

OTSG-A W

I

/ 10 TUBE

/ PERIPHERY 27%

27%

77% 20 % 20% 77%

(LANE) z- _x 10 TUBE -

- 40%

PERIPHERY 11 TO 20 TUBE N PERIPHERY I

y l

l l

I

, I l

4/7/82 l

l 8

Potentially Defective Tubes (Projection of Eddy Current Data)

OTSG-B  :

6%

i VERY FEW 50'/. INDICATIONS 40%

<2%

i (LANE) z- -x 15 %

10 TUBE PERIPHERY l

y l

l l

4/7/82 l l

l 1

)

9 l

FUTURE WORK / DECISIONS REMAINING e FINAL ECT AT ROLL TRANSITION AND UPPER END e RCS INSPECTION e FINAL RCS/SG CLEANUP METHODS / APPROACH e REPAIR, DEVELOPMENT /0UAL. TESTING OF S.G. REPAIR e FINAL TUBE SAMPLES / LABORATORY SIMULATION TESTS

• l 4/7/82

l 10 TMI-1 TECHNICAL BASIS FOR REPAIR LICENSING

~

e THE S.G. REPAIR APPROACH IS INITIAL - ROLL SEAL EXISTING TUBE TO TUBESHEET TO ISOLATE LEAKS / TUBE DEFECTS LONG RANGE - SLEEVE OVER DEFECTIVE TUBES (INCLUDING ROLL SEAL) TO ISOLATE LEAKS / TUBE DEFECTS /

INITIAL REPAIR, IF AND AS REQUIRED e S.G. DAMAGE IS UNIQUE IN INDUSTRY IN TWO IMPORTANT WAYS l

- LOCATED WITHIN THE UTS i

- DAMAGE MECHANISM OPERATES COLD / REACTOR SHUT DOWN e THE REPAIR APPROACH ISOLATES THE FAILURES AND RESTORES TH S.G. TUBE TO ITS ORIGINAL FUNCTIONAL / DESIGN BASIS CONDITIO 1

e BELIEVE NO INCREASED PROBABILITY FOR LARGE PRIMARY / SECON TUBE RUPTURE OR ACCELERATED DEGRADATION OF TUBES IN SERV e THERE ARE INSPECTION / SURVEILLANCE / TESTS TO MONITOR CO SATISFACTORY PERFORMANCE OF THE S.G.

l

{

i l 4/7/82

11 s

TMI-1 STEAM GENERATORS GENERAL INTERGRANULAR STRESS CORROSION CRACKING FAB HISTORY MET L,AB l

MATERIAL 1

1 STRESS v

-+- STRESS

~

+ OPERATING HISTORY ENVIRONMENT OPERATING HISTORY CHEMISTRY DESIGN IMPURITIES 4

FAILURE i

MUST EXPLAIN - TIMING OF CRACKING

- MATERIAL FAILURE MODE, I.E., INTERGRANULAR

- CONTAMINANT SOURCE FORM i

\ .

- AXIAL / RADIAL CRACK DISTRIBUTION l

i i

4/7/82 l

TMI-1 OTSG TUBE MAKING PROCESS o ALL TUBES MANUFACTURED BY PATC0 NO FORMAL PATC0 RECORDS AVAILABLE 1 GPUN/B&W REPS VISIT TO PATC0 (1982) s MPR TRIP REPORT TO PATC0 (1968) l o BASE MATERIAL SUPPLIED BY B&W TUBULAR PRODUCTS o GENERAL PROCESS BASE MATERIAL - ROUND HOLLOW BARS + 2" OD, 4 .088" WALL ONE COLD DRAW THRU ROCKER TYPE REDUCER DIE T0 a 1h" 0D,<v0.080" WALL FOUR COLD DRAWS OVER FLOATING MANDRELS THRU A DIE T0 40.625" 0D, v0.034" WALL TUBES CLEANED, ANNEALED IN HYDROGEN ENVIRONMENT AT 16500F 250F TUBES STRAIGHTENED AND CENTERLESS GROUND - MINIMUM WALL IS 0.034" o OTHER DATA EXTREME CARE TO PREVENT CONTAMINANT CONTACT WITH TUBE NDE TESTS INCLUDED UT, PT, EC, HYDRO, METAL COMPARATOR CHECK INTERMEDIATE CLEANING, ANNEALING AFTER EACH DRAWING OPERATION i

I Y*

4/7/82

OTSG POST WELD HEAT TREATMENT o

BASIC CYCLE TO PERFORM ASME CODE HT HEAT TO 1100-11500F HOLD FOR WELDS, 1 HR. PER 1 INCH OF THICKNESS FURNACE COOL TO BELOW 600 F MAX HEATING / COOLING RATES ~100 F/HR WELD THICKNESSES, 9h IN. AND 7 IN.

o ACTUAL RATES HEATING <20 F/HR FOR T7600 F

- 0 COOLING <.15 F/HR FOR Tz 600 F o FURNACE 85' x 18' x 18' ELECTRIC HEATING ELEMENTS - CAR FLOOR, ROOF, EACH WALL ARGON GAS CIRCULATED 4/7/82 U

FINAL FULL VESSEL PWHT IN OTSG ELECTRIC FURNACE IN BARBERTON OTSG - A 1200 -

ALL SHELL AVG 1100 _

ALL TUBE AVG 1000 .

/

800 -

g '

600 _

/- ~.

400 -

p' 200 -

I I I t i I I t i I I I I I I I I I I I I I I I I i O O O O N

O O

O

- n k~

TIE IN HOURS 4/7/82  %

OTSG POST WELD HEAT TREATENT HEATUP TIE TIME AT C00LDOWN TIE OTSG 200 F TO 1100 F G.T. 1100 F/850 F 1100 F TO 200 F A ~100 HRS +18 HRS / ~87 HRS + 128 HRS B ~ 68 HRS +13 HRS /*73 HRS a 129 HRS 14/7/82 3

! TIREE MILE ISLAND NUCLEAR 16 GENERATING STATION UNIT 1 STEAM GENERATOR A J -

S.

• ' C=L

( , . .

. . ... e.

e ,3 _ i

..:- e

. s  :

9

., I

(

. t  :  :

:7 3 l

)

t .

.= .

i t s . *.{: :: : * ,

• i 2 3 j! <0 * **.

. '" \

• a ,, ,

t

• d ,.: " . . ..

3 .4

'( ..

.  :  :: ( r. 7

..4  ;

e e

4 t ':::e

( .: .

:: ( :f c ,,, .

1:: .: .r w-

: :: ';. lp, ' . ' :.

4  :: J. 's 3:  ::: ' .: . '

".l A

. ) y ]. . *:

%: ,: .: .  :. <l l 2 A;. s i

to . l

.. I. .  :

• p y, .

gg.

t

.a j e a ag,... ; g

M *.

sC

.s

.a ,

3, *  ;,': y. e ..

0 :- .

• ' ~

..(~

5 )

t .';  ::::  : ;.

(:te:!'.  :: :$

c

.'cJ: .

c.

l, -

.. s:. \:;:et .

t  :

,"d s

.. ( .-

s t==,

3;;:-.

4*  :- ' / \ (. . . . ..

3, *f

a , ,c 6

.. g . . .s

a. ,

. .. a %.'. .

,c,=. =.

1. t3=r: *

<.- * ~ 3  :>,1

.g. t. -

n.1::

4y

lr:< * . .* ..> .ay:. .s

. . :: t p

• • 4-l Si:..

e; $ cs .

' s, . v;i

-. u: -};

j M LDCATIGIB F ALL RMS Fut BBE: W.AT IDERB _

M2329

.1545 n29:2

17 THREE MILE ISLAND NUCLEAR GENERATING STATION j UNIT 1 STEAM GENERATOR A l; b \

e

, t N

N

• 3

/s.

m h?

- s

.!.?.. W .

A a = h x o

- ,u k  %

E

~

MN NN l

X

% WM N

np:'t is . . , g. . .

r . . y'

, . O, Y EJ FOR TLBE HEAT Magots r2328 4/7/82

> THREE MILE ISLAND NUCLEAR 18 GENERATING STATION UNIT I STEAM GENERATOR 8 c 2. .::/.

=

,~ ::

=

{=

• o.

.o -

.t.

f t

..f

( , . .

a r,,

e * , **

, .I ".

• • T. ,. .

G G +-~.

. m; M. e

• 4 b , ,

, o g 4 ,.

! 1 tee . e' >

t'* *** .. O ** ,

• .. 9 q ,

9 EI, *e y O *

• 4

, b O g- 4 >

41, 18

• /

W

, , 4pi> g 3

• • • ..I

" . . 8 p= > .

3, . * .

=s,,, ,

4, -l .:-:: 1-+.

8 ' > ,,,

4 ll :

e

/ ,.f ..

c,e,

,,e#'h, ,y f.

8 3

e :

.c ,

< li

=

W=e:;, fi q'

s h . , f, D - e . .

,t , b . ,, 4 I>

a >,.

3. v. .. ,

. ,,als>

,,v

., > . ~:

.. ,. 8

,. . .ein ,

n>- . ,g t - :: . . . <

.e.so %g, og

's

. , , ,  :.,* *.-. ): , . , .. . ,

g

S , ,, .

a;.^*, .

s . in *

• e f**, *>

e c.:

.v .

?

e <

' *.

• e  % ,

.~.=.s*' "

assoas e.

• e 'C. -

q**

a:* .

esl, * '

- *E * *

• li j >

s s'

,*.' 3'

~

S

. .t .:,,, ~;.~ :.E:.'1Rm > '

i;::l* e

• e '  ;.:

.::::.1 s e  :- * .

e  :..f

e.  :

eJ* ? e . #  ::':.1

. :3. *. . e G.

e,

• **7%

... g .

4- g a s, im.'.. .% *::

.;2, t .4 ,

ff, , .,y

• =: * .

E T'ee* & g d 4

• * ' 4,. t ,y s"*. ,

7 .*.. . * '*

• E* -. > e') .

' * . 4

• . ' ,,,,,,, .S.'*b /

h. De. . 4 e

? *

". 7.

-ve

• 4*8 A

• . 4 b

> e.s '

s >

• l##.?

.::::f*8 e , # - *

• a ?. 4

.%' e' s .

- <-- ::ee.

..e*

• , E

• 1 * >

/.' >> ** 3.

. :'**$

• e ,

!. e o h*:;, ' ** i >t eo' n *

.::::::'* e o y

• f 3, ig ..e ,e,n' , 4'/.'

/.'

s . .

ea.~*

m  :::g, ** iC' *

• ' ". J. * . $ 5;, % ,{ 4 a f,4

./

.h {* *.J o'i s e ,4 ,,, . (. 4 ,,i *

,f

• 9ag.-

5 Q.--- ,

(. % .

enj e . .*4 ap:3, e

./. p ,= .". -

,' ,> es

• j e .mase p es

~* . . ' . 2,.s.

.l l.- %

e ,o* .

', , m% .

o 6

%Ye  ; e o *1, '. // . , b., **

' -4 i>> se .

. .}, .

• /. **n. ** .

• i, * . * ~5. "

./ .,** .

• e5 e

3 ,

s e ** "3 . / M* * ;3

$' c*

.* * % # ** ./* *

• o a>use .,3 4:

43,

~. e. . ..e* **. **

• • O

. '8B .

3 ,

i

• . \ s,n2.*:*

> ..I gy

u *o 7 an, . f -

-: ., e .

,e.

, 3, e

. ., em . e g &e M LOCATL*22 *F ALL 11M3 FOR M IGT M88ERS l'2798 f12BN h/7/00

\

l I

19 TISEE MILE ISLMS NUCLEAR GENERATING STATION UNIT I STEAM GENERATOR B

,4 ed *" .

7<t.

. l

\

1 j

,# h

/ *w h s,

/ ,y

/ v

/', w h

. V a

y e

o r

• F k

• MD y.

I

i ,.

,, ,e  ::

1 P

r, y >.

a se m x k b

v , . .s Y' "

h. ,., ,

\/

% ,DnD .

i

\ fy ,

\ "-

\=

\

v:

=

g* s, y

N .. + ,

N( N *

\

w -

itse unnes run amenw tuus FBR ltBC EAT Masots Inlet

'.; s *t M

20 HEAT VS DEFECT CORRELATION APRIL 2,1982

SUMMARY

o TUBE FAILURES ARE ASSOCIATED WITH SPECIFIC LOCATIONS IN THE GENERATOR NOT HEAT RELATIONSHIPS.

o THE DEFECT PATTERNS IN THE TWO GENERATORS ARE DIFFERENT AND THIS WILL NEED TO BE EXPLAINED BY A PARAMETER OTHER THAN HEAT NUMBER, o

HEATS OF MATERIAL EXIST WHICH HAVE HIGH DEFECT FREQUENCIES IN BAD AREAS AND THE SAME HEATS WILL HAVE LOW DEFECT FREQUENCIES IN GOOD AREAS.

l 4/7/82 l

l

21 STRESS' RELIEF DATA REVIEW, "B"'0TSG UTS o CENTER OF BUNDLE IN UTS IS 10 - 20 F HIGHER IN TEMPERA-TURE DURING EAT UP AND HOLD BUT IS 5 - 10 F LOWER IN TEMPERATURE DURING C00LDOWN o NO SIGNIFICANT TEMPERATURE VARIATIONS EXIST AROUND BUNDLE PERIPHERY o NO SIGNIFICANT DIFFERENCES IN TIMES AT TEMPERATURE EXIST AROUND THE PERIPHERY o MAXIMUM TUBE TEMPERATURE ACHIEVED DURING STRESS RELIEF WAS 1140 F o OVERALL THE TIMES AT TEMPERATURES INDICATE THE TUBES ,

WERE HELD IN TEMPERATURE REGIONS WHERE SENSITIZATION WOULD BE EXPECTED TO BE SEVERE I

4/7/82

22 l

TMI-1 Steam Generator Typical Cracks l

INCONEL TUBE e

1 fg

• rf7a- TYPICAL CRACKS ROLL TRANSITION ,

(._

9 -9 EEL TUBESHEET

$ 9 f 6 /'/

I CRACK CHARACTERISTICS: CIRCUMFERENTIAL NOT FULL ARC l

GENERALLY VERY TIGHT l

INSIDE INITIATED

\

I I4/7/82

23 TUBE ANALYSIS

SUMMARY

NO. OF NO, 0F ANALYSIS Tl]BES SAMPI_FS METALL0 GRAPHIC 8 38 BEND TEST 15 19 SCANNING ELECTRON MICROSCOPY (SEM) 15 15 ENERGY DISPURSIVE X-RAY ANALYSIS (EDAX) 15 15 AUGER ELECTRON SPECTROSCOPY (AES) 5 7 ELECTRON SPECTROSCOPY FOR CHEMICAL ANLYSIS (ESCA) 5 6 SCANNING TRANSMISSION ELECTRON MICRO-SCOPY (STEM) 5 7 ELECTROCHEMICAL POTENTIOKENITIC REACTIVATION (EPR) 4 5 HUEY TEST 1 3 SECONDARY ION MASS SPECTROSCOPY (SI.S) 2 3 ELECTRON DIFFRACTION 1 1 TRANSMISSION ELECTRON MICROSCOPY (TEM) 2 2 TENSILE TEST 3 3 RESIDUAL STRESS 1 1 S0DIUM AZIDE SPOT TEST 3 5 4/7/82 l

l 24 GPUN Failure Analysis investigation Team Babcock & Wilcox Tube failure analysis Lynchburg Research Center l

Babcock & Wilcox Corrosion testing l Alliance Research Center i

Battelle Columbus Laboratories Tube failure analysis Oak Ridge National Laboratories Corrosion testing l Metals & Ceramics Division i

Massachusetts Institute of Tube analysis for Technology sensitization j-l W'

Electric Power Research Institute Failure analysis review 1

ADDITION AL LABORATORIES .

Westinghouse Electric Research Independent tube

& Development Laboratories failure analysis

[j)cd6/[hd Nd >"'

g g; ,g A g n 6MN and $w 5d -

4/7/82

n, 25 TUBE UTILIZATION SUf1 MARY APRIL 2,1982 TOTAL TUBING AVAILABLE: 37.8 FT.

TOTAL TUBING EXAMINED: 13.1 FT.

TUBING ALLOCATED FOR TESTING:

o WESTINGHOUSE FAILURE ANALYSIS - .35 FT.

o TENSILE TEST OF DEFECT TUBE -

.67 FT.

o CORROSION TESTING -

7.10 FT.

o ROLLING / SLEEVING TESTS - 7.10 FT.

TOTAL 15.22 FT.

UNALLOCATED TUBING o PIECES W/0 DEFECTS 7.9 FT.

o PIECES WITH DEFECTS 1.6 FT.

TOTAL 9.5 FT.

e 4/7/82

26

SUMMARY

OF FAILURE ANALYSIS APRIL 7, 1982 o

ALL CRACKS ARE STRESS ASSISTED INTERGRANULAR CORROSION WITH INITIATION ON THE ID SURFACE o

EDDY CURRENT EXAMINATION HAS BEEN A RELIABLE INDICATOR OF CRACK LOCATION o

INCIPIENT CRACKS HAVE NOT BEEN DETECTED IN CLEAN SECTIONS (N0 E.C. INDICATIONS) 0F TUBING BY VISUAL AND DESTRUCTIVE EXAMINATION o

CARBON IN THE FORM 0F A HYDROCARBON APPEARS AS THE MAJOR CONTAMINANT ON FRACTURE SURFACES. SULFUR AND CHLORINE ARE PRESENT AS SECONDARY CONTAMINANTS o

RESIDUAL STRESS MEASUREMENTS IN ROLL AND ROLL TRANSITION REG SHOW NO STRESS PEAKS BUT RATHER A UNIFORM DISTRIBUTION o

CHROMIUM LEVELS IN THE GRAIN BOUNDARIES VARY FROM 8 WT. %

TO 20 WT. %

o THE INCONEL MICROSTRUCTURE APPEARS TYPICAL FOR STEAM GENERAT TUBING WITH DISCRETE CHROMIUM CARBIDE PARTICLES IN THE GRAIN BOUNDARIES o

SMALL AREAS OF INTERGRANULAR CORROSION SEVERAL GRAINS DEEP HAVE BEEN OBSERVED ON THE ID AND OD SURFACES AT RANDOM LOCATI o

NO RELATIONSHIP HAS BEEN ESTABLISHED BETWEEN MATERIAL HEATS AND DEFECTIVE TUBING o

MECHANICAL TESTING OF UNCRACKED TUBES SHOW THAT THE MATERIAL EXCEEDS MINIMUM SPECIFICATION REQUIREMENTS 4/7/82

I rI cU RE II -4 27 OTSG Longitudinal Section Elevations (Typ.)

PRIMARY SIDE (INSIDE TUBES)

TUBESHEET UTS) ELEV^ '

pEET AUXILIARY N emoninwami'w - 346'6" FEE 0 WATER -

[gjjph Illilgllla;i i N0ZZLES - ' * - -

-- 342'8" (AFW) ( Dl!l5

. 0.l.!llllM. - um'lI 339'9"

' N -lI.L . .Wh

= ~ i* l 336'5" STEAM OUTLET l ll '* Mr$

333'8" 330'6" MAIN FEEDWATER ll yIal li!)lr N0ZZLES (MFW)%,_ _,- '* - 327'4" l l qa nui

,, "" y - - 324'4" O .g ji SECONDARY S10E *-

y n , -- 321'1" (EXTERNAL TO TUBES) % g

9m jg . ,

1 Ir3 ,'" ' 317'9" s  ! i  !

W

. ",, - ll 314'6" 311'8"

,h*- 308'5" TUBE SUPPORT

, / '

=-

305'1" PLATES */

a-L,iu

- 301'10" LOWER .[. t ___

SECONDARY ~ , 0,,

HAN0 HOLE ~ LOWER

' l ' fM t v SECONDARY MANWAY

- 294'5" LOWER TUBESHEET (LTS) 1 REACTOR BLOG. FLOOR 281'0" 4/7/82

28 TMI-1 STEAM GENERATORS - STRESSES e STRESS MAXIMUM IN ROLL TPANSITION AREA,~34 KSIt e STRESS MAXIMUM ON OUTER EDGE OF GENERATORS e STRESS EXPECTED TO BE QUITE VARIABLE IN ROLL TRANSITION AREA -

e MAXIMUM STRESSES ARE AXIAL e STRESSES SAME IN UPPER AND LOWER TUBESHEET eSTRESSESMAXIMUMDURINGC00LDOWN,COLb e AT OPERATING TEMPERATURE - HOOP STRESS > AXIAL STRESS i

l 4/7/82

29 ll11llll lilli i ll l lllI11lll l-lllll111lll1111 llllll11 lilli ll I I !!Illlll ll!Illl!illlillli l ll!Illlll lllllll11 l l 11111 l l l l l l l l l l l l 111 l l l ! ! ! l 11 I .

l? lll1lllllllllll!I l lillllllllillllllllil'lliililll:"

( Ill!Illlli ,Il lli l l l i l l l l-l l l l l l 11 l l i 1111 l l l 11 l l , E

! @,? lllill Illi ll 'lllllllllllillllilillllli % t ylilli I  ? I lllr1llllllll111 lillli  ! "

E ll ll 1lll11ll ll gu ll e

lil lli l

i = ll l- l 11'11ll11 llll1lllll: Illll.j lill (d lllllll  !

lilli I! l l l l l l l l l l l l l 11 l l l l l l If l

11 l ll l l ll l' I Ill lli!

ill' ll llll1d l

lllllllllll {

14 lll l l ' i ,.

Il l l 1: 11. ,

IIt I Jl.l I .

I I I l l l T! I

..J__SJ_2 _ ',, Q i 1  !

llj ll1 (1 N . 11Al l I  :

l Ill  !

7 17 1 ll ll1 1 ,7e * *l l Ill i I 1 I ll l l'l!! ~

Illi ll ; .

Illi Ill '

IE illl 11llll11 11 ; e ll / ll illlilllllli!IlI[g"* l-lli l l llllllllll I Ill1 1

- ~

- J-i-'r~l

l l l 1; l

1 L - - 11.1 l 1 lillllill,lill:ll" b

..d' T l

I, lli 1 ill lillili' lill ,

f I L il i I ill ll l h i l **"

j _

lll I l '

lllll lll'111111lll l' i Ill ll I l ll cl 'l l i I I IL l Il 11ll llllll

+

l, 4 l l l 11 I I lil llll!!

k I } ~

E I '

I ll l ll lil l lil l'

{ 'll  ?

l ll Ill l Illi Ill  !

' W N

• 11  %  !

I l I lil'Ill l'Illli Ill l^

I 1 I I I I ll I III I ll Ill IIT l

1.1 ,

t ll!, c i 11 ttct 11 il i i t t- 1;istyi: ij " -r "3 s/6

[! i 4 3 1 A 1 $ 3, 3 i go unm wn 9 a, o voo IIIl l iI i lii t i.iiiiij ii'i t iiii.i;;;i  :!!i:i:i  !:i:i t ';!3

~ll ii isi i, !

_ i,1 1 ! i 4 i l l ! 13 i l i 2 *

• 3 1 li l

,1 ill 1 m __ m; -a ....

. ..,...........,,,,,,..,,,,,,iiii lilli Ill Illill lli Illilllllillllli!!!!lilliliiti P

G ue

OPERATING HISTORY OBSERVATIONS-1 o LOWER END GENERATOR ALWAYS SUBMERGED (WETTED), UPPER END ALTERNATE WET AND DRY WITH AIR (0XYGEN) INTERFACE o

WATER LEVEL IN THE PRIMARY SIDE OF OTSG WAS IN UTS FOR BETWEEN 31 AND 243 DAYS o SOME DIFFERENCES IN AMOUNT OF FLOW SINCE FEB '79 TOTAL PUMP HOURS OTSG - A = 681 HRS TOTAL PUMP HOURS OTSG - B = 393 HRS BACK FLOW IN OTSG - B FOR 10 HRS DURING SEPTEMBER '81 C00LDOWN 4/7/82 8

OTSG Level July 1981 November 1981 RCS PRESSURIZED RCS FILLED RCS

= 355 - Y & PRESSURIZEDy y LET DOWN o ,

P .

$ 350 -

g, l  !

d M /" '

C 345 s

/6 UPPER TUBE SHEET ~-

b Af '

i . i l

i l

340 I,;

! j g I/ ,*  ! 13TH SUPPORT PLATE m 335 -

\/~ ,

9 A '

2 33g _

RCS DEPRESSURIZED i I

t 325 - '

h LOOPS DRAINED TO

$ E 320 -

$ LOOP SEAL LEVEL LOOP SEAL LEVEL {

! A OSTG I

315 -

B LOOP PUMPED DOWN t FOR MAINTENANCE Y B OSTG 310 l I I I JULY AUG SEPT DCT NOV DdTTED LINE MEANS ESTIMATED LEVEL VENTING ARRANGEMENT CAUSES UP TO 27" ERROR 3

OPERATING HISTORY OBSERVATIONS-2 o POTENTIAL SULFUR SOURCES PRESENT SOME OIL INTRODUCED INTO RCS IN MAR '79 SULFURIC ACID ADDED TO RCS IN OCT '79 S0DIUM THIOSULFATE ADDED TO RCS AT VARIOUS TIMES OVER LIFE OF PL o

SODIUM THIOSULFATE THOUGHT TO BE PRIMARY CONTRIBUTOR ACCUMULATED IN BUILDING SPRAY PIPING - 1979 AS A RESULT OF VALVE LEAKAGE JUN, AUG, SEP '81 OPERATION OF SPRAY PUMPS ADDED SOLUTION TO BWST INJECTION INTO RCS OCCURRED DURING SEP '81 C00LDOWN 4 /7/82 U

l TMI-1 STEAM GENERATORS - SULFUR SAMPLES TOTAL SULFUR SYSTEM DATE SULFATE (PPB) (PPB AS S0is)

REACTOR COOLANT DECAY HEAT 7/31/79 -

1,500 8/02/79 -

<600 11/01/79 -

<660 12/04/81 -

730 1/18/82 -

400 2/04/82 -

100 B0 RATED WATER STORAGE TANK 1/20/82 -

. <100 REACTOR BUILDING SPRAY PUMP OUTLET .

1/20/82 -

15,000 3/17/82 2,876 -

3/20/82 764 -

INTERCONNECT BETWEEN BUILDING SPRAY 1/20/82 -

176,000 AND DECAY HEAT 3/17/82 2,465 -

3/20/82 752 -

SPENT FUEL P00L 1/18/82 -

400 3/17/82 246 -

3/20/82 149 -

4/7/82

l CORROSION TESTING

SUMMARY

o TEST WITH ACTUAL DECAY HEAT COOLANT ON SENSITIZED INCO AND STAINLESS STEEL BENT STRIPS. RESULTS - NO CRACKING IN TWO WEEKS o

TEST WITH ACTUAL DECAY HEAT COOLANT ON AN ACTUAL TUBE REMOVED WITH AN INCIPIENT DEFECT. RESULT - NO CRACK GROWTH o

34 ELECTROCHEMICAL CORROSION TESTS WITH VARIOUS CONTAM PRIMARY COOLANT ENVIRONMENTS AND VARIOUS SPECIMENS BORIC ACID (PPM) - 13,000, 5,000 THIOSULFATE (PPM) _ 100, 10, 1, 0 HYDRAZINE (PPM) - 200 MATERIALS - M5442, M2320 - ACTUAL M2320 - ARCHIVE TEMPERATURE - 550, 100 F ATMOSPHERE - AIR, HYDROGEN 4/7/82

35 PRELIMINARY CORROSION TEST RESULTS o

CORROSION TESTS IN ACTUAL PRIMARY COOLANT INDICATE IT IS CURRENTLY INNOCUOUS o

REDUCED SULFUR SPECIES CAN REPRODUCE THE TYPE OF CRACKING OBSERVED IN STEAM GENERATOR TUBES o THE DEGREE OF SENSITIZATION (I.E., PRIOR HEAT TREATMENT)

IS A KEY PARAMETER IN DEFINING THE MATERIALS SUSCEPTABILITY TO IGSCC o

THE PROPENSITY FOR A SULFUR CONTAMINATED PRIMARY COOLANT ENVIRONMENT TO INITIATE CRACKING VARIES INVERSELY WITH TH BORIC ACID AND LITHIUM HYDR 0XIDE CONCENTRATIONS o

CRACK INITIATION APPEARS TO BE THE RATE CONTROLLING PAR-AMATER o

CRACK GROWTH RATE IS VERY RAPID ON THE ORDER OF IMM/ DAY o

CRACKING APPEARS TO BE A LOW TEMPERATURE OCCURRENCE o

CRACKING TENDENCY IS REDUCED BY RAISING THE PH 1

l l l

4/7/82  !

I E

i KEY ELEMENTS IN EXP REi!IEH o IGSCC 0F I-600 OBSERVED AT 575 F IN SULPHATE CONTAINING WATER; UNLIKELY TO OCCllR UNDER PWR PRIMARY SYSTEN REDUCING ENIVRONMENT NOT ASSOCIATED HITH DEGREE OF SENSITIZATION

o

' IGSCC 0F I-600 OBSERVED AT 75- 225 F IN SllLPHUR OXYANION (E.G. THIOSil WATER; MORE LIKELY TO OCCUR IN P9R PRIMARY SYSTEM CRACKING IS RAPID SUSCEPTIBILITY DEFENDS ON SENSITIZATION, PH, TEMPERATURE, AND ELECTROCHEMICAL POTENTIAL o

PLANT AND MODEL BOILER EXPERIENCE IS ENTIRELY RELATED TO SECONDA o

NONE OF PRIMARY SIDE INDllSTRY EXPERIENCE IGSCC 0F I-600 ATTRIBUTED T BULPHUR SPECIES LI/7/82

HIGHLIGHTS OF STRESS ANALYSIS o

TUBING AXIAL TENSILE STRESSES LARGEST DURING C00LDOWN; MAY -

APPROACH YIELD STRESS o

SIGNIFICANT AXIAL TENSILE STRESSES ALSO EXIST DURING COLD SHU o

LOCALLY HI AXIAL TENSILE STRESSES POSSIBLE IN SEAL WELD Hf:Z AND NEAR ROLL TRANSITION o

AXIAL STRESSES GENERALLY LARGER AT PERIPHERY THAN IN CENTER OF TUBE BUNDLE t

4/7/82 0

SUSCEPTIBLE MATERIAL MICROSTRUCTURE 4

o FAB HISTORY SHOWS TUBING TO BE MILL ANNEALED PLUS STRESS RE 8o HIGHLY SENSITIZED o

MET EXAMS CONFIRM EXPECTED MICROSTRUCTURE o

CORROSION TESTS SHOW PULLED TUBES SUSCEPTIBLE TO CRACKING IN THIOSULFATE / BORIC ACID SOLUTIONS 1

I4/7/82 g

I AGGRESSIVE ENVIRONMENT

. S04 - AND S 0-23 CONTAMINATION PROBABLY PRESENT i

CHANGES IN S-SPECIES EXPECTED DURING HOT FUNCTIONAL -- DIFFICULT TO PREDICT SPECIES PRESENT AFTERWARDS I

l

! 4/7/82 l

3:

40 AQUE0US SULFUR SPECIES Sulfur Formula Structure Oxidation Number

• Name H2 S or S" -2 sulfide H3'3 22 2 S-S -1 H3'3 23 3 S-s-S -2/3 polysulfides

' (

, f H8, S-S-- -2/x 2x . .

N

/

S S 8rings O sulfur E

S0 23 0-S-S +2 thiosulfate O

S0 46 0- -S-S-S-0 +2.5 tetrathionate 0 6 S0 3

0- -0 +4 sulfite (sulfurous acid)

S0 +4 2 sulfur dioxide l -

S0 26 0- - -0 +5 dithionate l SO "

4 0-h-0 +6 sulfate l 0

• 0xidation number is the formal electrical charge assigned to the sulfur on the assumption that H is +1 and 0 is -2 in these compounds.

14/7/82

PROPOSED FAILURE SCENARIO

1. S04 -

AND 23S 0 -- (POSSIBLY OTHERS) ADDED DURING LAYUP 2.

REDUCED S-SPECIES FORMED DURING HOT FUNCTIONAL

3. WATER LEVEL DROPPED.

HIGH CONCENTRATION OF AGGRESSIVE S-SPECIES FORMED IN DRY-0UT REGION

4. CRACKING OCCURS IN DRY-0VT ZONE S.

CRACKING TERMINATES DUE TO REDUCTION OF CONCENTRATION 6.

CRACKING IS DISCOVERED WHEN OTSGs ARE PRESSURIZED 4/7/82 ,

FEATURES COVERED BY SCENARIO -

TIME OF CRACKlHG MODE OF CRACKING AXIAL DISTRIBUTION OF CRACKING RADIAL DISiRIBUTION OF CRACKING (OTSG-A)

CORROSION TEST RESULTS 4/7/82 2

I JMPLICATIONS OF SCENARIO SULPHUR REDUCTION NECESSARY TO PREVENT RECURRENCE

-- OXIDATION TO SOLUBLE FORM

-- REMOVAL VIA DEMINERALIZER

~

ATTACH OF OTHER PRIMARY SYSTEM COMPONENTS, IF ANY, MOST PROBABLE IN VICINITY OF WATER LINE LOCATION FOLLOWING HOT FUNCTIONAL

-- INCONEL X-750

-- SENSITIZED TYPE 3011 STAINLESS STEEL i

11 / 7 / 8 2 0

i

REACTOR COOLANT SYSTEM REVIEW I OBJECTIVES o

REVIEW REACTOR COOLANT SYSTEM COMPONENTS FOR CONTINUED SAFE OPERATION o CLASSIFY ITEMS FOR MATERIAL CONDITION, ENVIRONMENT EXPOSURE AND APPLIED STRESS o

SELECT CANDIDATES FOR INSPECTION AND TESTING THAT ARE REPRESENTATIVE OF WORSi CONDITIONS o MINIMIZE EXPOSURES o

EMPLOY STANDARD ACCEPTANCE TFSTING BUT SELECT SUSCEPTIBLE MATERIALS FOR DESTRUCTIVE METALLURGICAL EXAMINATION 14/7/82

05 l

l REACTOR COOLANT SYSTEM REVIEW PROGRAM PLAN O CLASSIFY ALL MATERIAL TYPES USING FABRICATION HISTORY AND LOCATION IN RCS O IDENTIFY ASSOCIATED STRESS LEVELS AND SAFETY CONSIDERATIONS FOR APPLICATION O EVALUATE RCS MATERIAL CORROSION SUSCEPTIBILITY O IDENTIFY POTENTIAL PROBLEM AREAS FOR RECERTIFICATION 1

INSPECTION AND TEST O DEVELOP INSPECTION PLAN O PERFORM INSPECTIONS AND EVALUATE RESULTS PROVIDING As NECESSARY ANY CONTINGENCY TESTING O DOCUMENT ACCEPTABILITY OF PRIMARY SYSTEM FOR SAFE RESTART l

l l

4/7/82 l

C6 REACTOR COOLANT SYSTEM REVIEW INSPECTION PLAN O DESTRUCTIVE METALLURGICAL - INCONEL 600 ANALYSIS - INCONEL X-750

- SS 304

- INCONEL 718 l

0 EDDY CURRENT

- I-600 NOZZLE TO SS FLANGE

- I-600 (NOT AXIALLY LOADED)

O ULTRASONIC TESTING - SS 304 - BOLTS INCONEL X-750

- SS 304 TUBING

- I-600 SAFE ENDS SS 304 WELDMENTS O RADIOGRAPH TESTING -

I-600 SAFE ENDS SS 304 WELDMENTS O PENETRANT INSPECTION -

SS 304 CLAD I-600 CLAD 0 FUNCTIONAL TESTS -

IN-CORE DETECTORS VENT VALVES

\

1 I

4/7/82 1

07 INSPECTION PLAN (CUNT *D)

O VISUAL EXAMINATION - CORE COMPONENTS

- PLENUM

- HOLD DOWN SPRINGS

- END FITTINGS

- FUEL RODS

- SPACER ASSEMBLIES

- CONTROL RODS

- SHELLS AND BOLTING RINGS

- BAFFLE PLATE REGION

- LOWER BOLTING RINGS

- LOWER VESSEL HEAD O OVERALL -

INSPECT OR TEST APPROXIMATELY 1000 ITEMS l

l 1

l 1

I 4/7/82

48 1

Repair Criteria  !

l (1? The maximum allowable primary-to-

! secondary leakage rate for normal operation shall be as low as reasonably achievable and allow plant operation within the radioactive effluent limits of the technical specifications.

l l

l l

4/7/82

49 1

Repair Criteria (2) Repaired tube shall sustain, with adequate margins, the design basis loads Loads Generic 177FA TMI-1 LOCA + 2641 lb + 2641 lb MSLB + 3140 lb + 3140 lb (being reanalyzed)

FWLB - 620 lb - 620 lb Normal cooldown: + 1107 lb + 1107 lb

+ = tension

- = compression

{

(

1 4/7/82

50 l

Repair Criteria 1

(3) The effects of both repaired and plugged tubes on the thermal and hydraulic performance of the plant and on the structurai and vibrational adequacy of the steam generator shall be evaluated and shall be within the acceptance criteria for both normal  !

operating and design basis accident l conditions as specified in the licensing l basis documents.

a i

4/7/82

51 l

Preliminary Repair Process O.ualification Criteria e Result in a process capable of providing a leak-tightjoint

• Produce a joint capable of carrying the design l

basis loads

• Maintain the tensile preload in the free standing portion of the tubes within allowable limits

• Result in minimal tensile stresses

• Produce an expansion capable of being non-destructively examined i

• Be adaptable to remotely operated tooling o Permit future sleeving l

4/7/82

1 52 Original Configuration of OTSG Tubes Seal weld -

1 A l

Vfff/,

, Original shop roll

~1 1/2" yjs 24" tubesheet 4

j e Crevice i

.002 to .008 radial em t

~ ,

Y 4/7/82

53 Ropair Configuration of OTSG Tube i

Seal weld -

, Original shop roll N ~ 1 1/2" p'a

~10" ~'

Crevice closed to (0.001 (radial) i:0 New leak limiting /

N '

4 load carrying hard roll expansion 95 s i

24" tubesheet '

Existing crevice i

0.002 to 0.008 (radial) v 4/7/82

54 Sleeved Configuration of OTSG Tube Seal weld -

A a a Original shop roll NY ,

~1 1/2"

~ 10', o Sleeve H .

~12" Crevice closed

-t to (0.001 (radial)

(

New leak limiting /

load carrying hard I

roll expansion

! v

I v

24" tubesheet '

, Ex.isting crevice j

0.002 to 0.008 (radial) y 4/7/82 m

55 Scope of Process O.ualification Design Variables ,

• Hone ID surface in area to be expanded

-avoids inclusion of contaminants

• Depth of roll approximately 10" max

-allows later sleeving

-leaves approximately 500 tubes or less to be plugged or otherwise repaired

• Crevice closure by low-torque roll, explosives, or hydraulics i

-proven techniques

-minimizes residual stresses 1

-inspectable

• Mechanical roll with 4-10% wall thinning

-carries axial load 1 -retains preload

-leak tight

-proven technique i

• Sound tube material below repair 4/7/82

56 Facts on Mechanical (Roller) Expansion o Residual stresses in roll transition zones can be reduced by increasing the end radius on the rollers o The optimum roller geometry has been determined to be:

-1 1/2" long rollers with largest standard available diameter

-1" effective length of roller with 21/2" l

end radii o Axial residual stresses are greater than those produced by hydraulic or explosive expansions o Residual hoop stresses are less than those produced by hydraulic or explosive expansions o Roll expansions produce thinning of the tube wall in the expanded area; industry standards (based on allowable metal strain) are 4 to 10%

o Roll expansions produce a net elongation of the tube due to the extrusion of the tube walls 1

4/7/82

57 4 I

._ . e. . . . .

n . . . .n. . . _n . . . . . - . ,. , . -

7 w' - .4 - :._- . .

Depth of

<..,.. - - "._ . O . ' ' w'e. q 4.f.

l l

+ = t. '- ~ - .. KL_ '. e c=d y+M . . . ~:G

. cold work

. b .J '

..ts ,. m...

l .

n.st',."Z.. .%./ cV.v,w.- .

u,. 4. .  ;,

.4.- . . .,,;.

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

e .. ~. ., u ., - ;. . .... . . ..

. 2 '. : -q.a.W x. . .c . _.. ..

U. c r . . ,: ..

. . . . -2

, = . * .g

, . ., s 10% Chromic 100X Figure 8 -

l Rolled portion of a tube showing the amount of cold work l present in the overlap area using standard rolls.

1

(

old work

n -;.... '. .

... F . f,, :'.o f. ......-c

-~ . .... . . . .

. . . . ,. .v . u. e

\ ,

m .

l . ,. ,. .. , x

. . . . .. . x. y . , ..., . .

am... . . > -

7,, . c . . , , .. s : 9 ..-: -

s ., ... ..... ., . . -,,m ,. e. .Z,.

.. o. . v ,.-01:; . n . .

-c ";c w .:,: *

,.  :. ,f:-Y ,w. - . . +. ..

+. . L. -c

..s\. nu a. . ..,.:.x.i. .:. . + cm:

. . .w.,.n.. . ..

10% Chromic 100X 1 - Figure 9 -

Depth of cold work produced by increasing the leading radius to 2-1/2 inches.

l i

4/7/82 I l

l

58 c

Experience with Mechanical Joints  !

i

./,

• Industry SG experience with mechanical joints

, i.

4 - Doel-2 Tube /tubesheet rolls About 100 rolled in 1980 0

(repair) & 1981

- Point Beach-1 Rolled sleeves (repair) About 12 sleeves in 1982

- San Onofre-1 Rolled sleeves (repair) About 7000 sleeves in 1981 l

- Obrigheim Tube /tubesheet rolls 12 years service (original)

- Palisades Hydraulically expanded installed commencing 1976 sleeves (repair) i l

• Other industry experience with repair hard roll l

- Big Rock Point RTR vessel /CRD housing 4" tube in 1979. No l

leakage l - Dyster Creek RTR vessel /in core flux 2" tube in 1975. No monitor tube leakage 1

- Gargliano - Ditto - 2" tube in 1966. No leakage o Standard heat exchanger manufacturing process 4/7/82

59 Preliminary Tube Expansion Process l Comparison Mechanical Hydraulic Explosive Roll Expansion Expansion Residual stresses ID Greater Base Equal OD Less Base Equal Effect on tugh Decrease

{ Preloa(7' increase Little change Load carryipg capability Greater than Base Greater than Leak tightness Greater than Base Greater than Supporting data based on:

eB&W Canada and B&W USA R&D and production work accomplished on both once-through and u-tube steam generators 1

fV g 7y,ht:]hg '

?

4/7/82 l l l

60 Points to Be Addressed by Qualification Program l

1 Adequacy of repair process j

• Leak tightness following thermal cycling

• Load carrying capability following l thermal cycling l

l

• Tubesheet hole ovality i

• Water or moisture in crevice

• Statistical leak tightness margin determination

• Roll torque / length vs leak tightness ,

• Roll torque / length vs load carrying l

capability

• nspeetability 4/7/82

61 Points to Be Addressed by O.ualification Program i

t

(

Effect of repair on total OTSG performance

• Primary water in crevice and tubesheet corrosion

• Change of tube preload

• Residual stresses in tube

• Effect of trapped contaminants

• OTSG performance with specific l

tubes plugged

• Confirm adequacy of existing operating and accident analyses 4/7/82

TMI-1 Tube Expansion O.ualification Program PREL STRESS

+ ANAtysis >

ANALYSES TUBE 4 OTSG LOAOS V

y INiilATE PREL LOAD REVIEW PROGRAM ~M

~ s LEAK TESTING

+ + OEFINE

~M -QUALIFICAil0N DES'

+

P ARE P FORM OATA PROCESS TESTING

- ~

REVlE TECHNICAL -+- OTSG REPORT JL REPAIR A A A ROLL h- PARAMETER > R PARE STUOY OCggp I

& CORROSION ON GOING TESTING " m CORROSION TESTING (IF REQUIRE 0) y V TOOLING OESIGN UPGRADE ON SITE

+ PROC URES,

& FABRICATION AS + DEMONSTRATION SETUP 8 REQUIRED TRAINING

& TESTING ,

b D

s N

O N

t

63 OTSG Tube Rolling Top View-Manipulator MANWAY I

I ROLL EXPANDER F VERTICAL TABLE O

o. ,R - PlVOT oo '

MANIPULATOR 1- f, /- ^""

_f}pq

==

fOo '

93 R CARRIAGE CLUTCH 1

l 4/7/82 I

l

64 OTSG Tube Rolling - Elevation W .

' ~

9 DRIVE MOTOR DRIVE MOTOR AND AIR CYLINDER RIGHT ANGLE DRIVE (FOR TABLE VERTICAL DRIVE) lh W

, , $W L R CARRIAGE (WITH ENCODER AND DRIVE MOTOR)

W h

~

\ .

N DRIVE WHEEL N

\

MANIPULATOR ARM ROLL EXPANDER PlVOT ,

l 4/7/82

65 Currently Planned Process Monitoring and Inspection i

l j Tube identification

• E/C manipulator record l

• Video record Depth of rolls

• Automated insertion tool e Tool location feedback

• Video Torque

• Transducer feedback

• Air pressure alarm

• Periodic calibration

• Frequent equipment inspections and cleaning t

l 4/7/82 1

I

66 OTSG Repair Program Overview I. Tube expansion / testing complete Oct. '82 II. Projected total repair exposure

< 500 man rem Ill.100% ECT examination in affected area IV. Plug or repair all tubes with inside diameter ECT indications in the roll, roll transition or tubesheet crevice V. Plug / stabilize all tubes with inside diameter ECT indications that are not within the tubesheet

VI. RCS cleanup to reduce the amount of sulfur on surfaces

, Vll. Sensitive leak tests following repair Vill. Sensitive and continuous leak rate i i monitoring during operation 4/7/82

l l

67 l

~OTSG Repair Program Plan / Schedule 1982 1983 MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR 6/82 OTSG/RCS INSPECTIONS N TUBE EXPANSION 6-7/82i QUALIFICATION N

i i iY Y 10/82 OTSG REPAIRS -

cPLUGGING OEXPANSION OLEAK TESTING i

=="

I -

6 i

l l

4/7/82

68

)

Preliminary Cumulative Man Rem Exposure

1. Actual OTSG exposure to date N70 man rem II. Estimated additional exposure ~230 man rem A. RCS inspection N 60R B. Eddy current testing N 10R C. Tube samples N 10R D. Tube plugging ~100R E. Tube expansion N 50R lli. Projected total OTSG repair exposure with 200 man rem contingency < 500 man rem l

I 4/7/82 l

69 OTSG Tubing Eddy Current inspection Program I. Objective oldentify scope and extent of tubing damage and soundness of tubing areas accepted for service

• Repetitive inspections to l detect new defects and/or defect growth II. Scope *100% of affected areas

' ~

• Statistical sample below affected areas

lil. Techniques

• Standard differential i

probe (multi-fre-l quency system - _,t26 M ~

increasedr s nisitivity M%.

tubing areas between lower and upper tubesheet roll transitions

• Absolute probe upper tubesheet roll transitions l and rolled areas (4 coils

- 2 orientations - 360 degrees coverage )

4/7/82

70 Standard Differential Eddy Current Technique O.ualification I. Metallurgical e 100% correlation on Analysis 29 E/C defects located at or below  !

the roll transition II.Other destruct e Laboratory testing of Testing ~ 13 feet of tubing verifies soundness of portions of removed tubes accepted by eddy current examination

~

Ill. Correlations eCircumferential among E/C coils vs pancake coils designs / e Standard differential vs techniques absolute techniques IV. Production eEvaluation of all E/C data intelligible signals interpretation irrespective of amplitude to account for crack 4

orientation and geometry 4/7/82

I 71 I

,A.,

( A

%.. . % n

$- n -e 1 g o. _ _ _ _ _ _ _ _ __ _ ___________g@g -

1

.~.

M

= g._________

.______ ___g = -

O W m. . n.t.

g .

Su> w

=c w "

A n

.O

- mm w

Sz e

~.

w e- -e a gg = g._ _ _ __ _ _ _ _ _g g

=mm '. -

O""

h

.~.

b g -

O O .~.

~

Q g) ..'.

w " ,

3 O -

pJ n m geg...........6. . . - _ _ _ . .g ; =t E

N e4

)- m. ...

O =.

W N - .

• in #

m =

0 m --I g. _ _ _ _ _ _..'c -_ _ _ _ _ _ _ .g2 =

• > e."

'g g = g e. - _ . . _ _ _ _ _ . _,, .

_ _ _ _ _ _ _ _ _ _g .eg=

0 m w

?e -

"t- -

g, te ~ == A~ we E km ,

$7~h ~

~

L[_ _$__q . 7

~

3~

_a g._k_ __2 ~

9 n. =

, %. g. _ gg..r_._~ 2. G .  :  :.e. .g %:,

m g _ S=_ . .=

O s e" :g

_ _ . .6._

.._ =_= g _ _ r-_.g -

e g e g = g. _ _ _ _ _ - _ __ _ _ _ _ g g g z e r r- >

r-

.O gw b mm n

._ g. _ _=.-g

-- ~ __ y,a e a _ _: =__ _..,

.g=ga--4:3 t

.g =. g. -

~g 4

mm

- , ,g. . .

ww ,~-=,e

~

e ,, e. _ _ _ .

.. .g g , A-e~

~ - >c,-

~

e.,

n. = A ss" - ~_

~

e-s y g:=,g=.e:i__=_______,-

- e-

. . _. : ;: == _ _==__.__ _ _! _ _g:=_-e- ,"_,s -

p=s O:p ,Q:

1 .

_ _ _ . _ _ _ _ . _ _ _ _ _ _ _ _ _e_ > _._ . o ; _ ,, n ,

" iT of af

- t a =,I A s

.s, ~

= i SN011V310N11N3HH03 A003 ilV31DH0'I'lV13W -lVHSIA AH01VH0 api 4/7/82

72 e

~

~

- 4 e- ,

o ,, =g_____ _ _ ,

______.g C ==g.______

g - ________g==- -

e 0

o h

N u ,a N O ae

= g. _ _ - - _ - -- _ - - _ _ _g~ :

O (J _m

.o .

m a ,,

3 .~. ~

u.n

1. - -e.

g g e___________. __ _ _ _ _ _ _ _ _ _g g O

E ,,

s I-f s g________

_ _ _ _ _ _ _.gg g

_~_

m N m . .

e M.

+

. e +-

- e.

O ~ - .

, o- .e: 4

~ e-e = =, .

g g __ _ _ g______

.= _ ____________

_ _ ,e=_ g ___ =e_____ _g g.__ g ~.

,_ 4 m ,__=___g a =.

p* e .~ . ____________e= 2. e

=

= e g._ _.~e __ _ _ _ _

__________-______g..

C. -

g-___.

e g._______________.

o E g e_ _ _ _ _ _ - _ _ _ _. _ _ _ 3. _ _ _ _ -_ _ g .e O e _ _ _ _ __ _g :

U =

. 2

~

, 6 5 es

=g ____

se  ?

e-

= __ __g g g m .

= ,.__________

u ,_w l __________g

.a: w

u. :c . .

cc m -

a c e. e, i ww  :- g .

a. m .e m.ee=~ . . . e ~. ., A ms ng -

g- - - , =g_ _ - - - _ _ _ _ - -- ... -~ e. e

! a_E__~_4__. _ G-S

_ _ __ _ _ _ _- _ _. _ _o :- O : g -O s -- ge i 3

-_ _ o .__--O

ina g%g a

N

i. I

~

SN011V310N11N3HH03 A003 lV31DHnllV13W-lVASIA AH01VH08VT 4/7/82 l

l

73 Correlations Among 1 Coil Designs I. Scope e Standard differential vs 4X absolute

~435 tubes full length l N4500 tubes partial length e Standard differential vs 3X pancake differential

~100 tubes full length II. Conclusions ein all cases there was good correlation einconsistencies can be explained by:

-low level signals ( < 1 volt) drop in and out by both

techniques

-resolution of multiple defects that are close together 4/7/82

74 TMI-1 OTSG Absolute Eddy Current Technique O.ualification

1. Metallographic -100% correlation analysis on 30 E/C defects

[ top 0.25 inch excluded due to alignment problems which we are correcting)

II. Tube /tubesheet -demonstrates mockup detection testing of simulated cracks located at the primary tube seal weld and below i

4/7/82

- -------- - --~~--- -- ~- -

uG -

g

- . - g ;

a.

s..

. '4 -

2,..

~

7

. a. *

.+

. .L _

. ~ ?-

m e

y-p u

. a2 k .

c _

i o _

'-,- M 6 ;c e 5,: . i 1

)

d n

m

_ - a _

. 2h

,hk b

e _

o r

_ j.$ ?,

P T

C

, .u11 / E

. 6 s e

gN N. t

l u

. - o

+ .

s

.b

[u -

.A k, ,.~

.'nt's

- .. i

c. ;w s-

' 4 J .

2

, :* .A

_2 ,

,y,

-u 4 ai J  ;

\

}s-

- h k% [;

I g

g' S. k tu N

76 TMI-1 Eddy Current Defect Mockup Absolute Technique Qualification 1

RDLL TRANSITIDN

/

...................... - ------ --- m--

~

~

D-1 D-2 D-3 D-4 D-5

......................I Y yV 1

! CLAD SE^L UNROLLED ELD i l TUBE

< TDP l 1/4" l

,,, SURFACE

l l1/s. TUBESHEET t

4--- 1 " (M IN) R D LL --+

Defects #1 thru 5 (D1 -D5) are ID defects

! located as shown Extent of defects varied from 20% thru 100%

of tubing wall thickness - defects are 3/16 inch long (EDM notches)

• 40% and greater defect depths were detected 4/7/82

77 OTSG Tube Plugging Plans 100 90 80 70 n'

SO s

5' p 50 5

40 8

30 ry _ 1, . 7 .:;,

20 .fyJ'jy, $566-376 (TOTJ t;& . ' DEFECTS-BEEDW 6-10" l 3. ' y't' 4 l *WHICH MAY BE ~ 104 (TOTAL DEFECTS BELOW 21 1/2" WHICH

', ~ clf PLUGGED)

• WILL BE PLUGGED)

U1 554 5 6 12 18 24 DISTANCE DOWN TUBE /TUBESHEET-INCHES

/

u/ Wr-&2 4/7/82

78 Analysis of Tube Plugging Affects on OTSG Performance

• Reactor coolant system flow rate

• Safety analysis for LOCA

• OTSG exit steam quality l

1 i

4/7/82

79 RC Flow Results o Calc RC flow w/o plugging: 109.86%

o Error in calc: 1.50 %

l 0 Resulting flow: 108.36%

o Tubes plugged: 500/OTSG l

o RC flow reduction: 0.25%

l 0 Resulting flow: 108.11 %

o Tech spec limit: -106.50 %

o Margin: 1.61 %

Conclusion The reduction in RCS flow is acceptable 1

4/7/82

80 I

LOCA Results Considerations e Boiler - condenser mode heat transfer eInitial RCS liquid inventory e EFW spray cooling e RC flow rate eCore cooling Conclusion No effect on licensed power level of 2568 MW T for up to 500 plugged tubes per SG l

4/7/82

81 OTSG Exit Steam Superheat (100% Power Results) e Normal superheat = 54 F e 300 tubes plugged in one SG iuniform Dist'n 1

-average exit superheat = 49 F e 300 tubes plugged ll25% of tubes plugged in a peripheral region l

-central region superheat = 54 F

-peripheral region superheat = 11 F

-average exit superheat = 49 F ,

l Conclusion The reduction in OTSG exit steam superheat is acceptable 4/7/82 o

^

l 82 i

Removable Plug Development

1. Objective einstall removeable plugs in tubes which may be ,

returned to service by sleeving II. T y p e

• Roll plugs similar to those i

used at San Onofre Ill. Qualification *100 thermal cycles (120 F to 650 F)

• Leak tests at AP = 2250 psig

• Rapid cooldown from 650 F

• Simulated circumferential crack in roll

• Ejection / pull-out tests IV. Results jg _p 4g *dropsAverage leak rate

/ minute at operating AP .U e 6200-12,000 psi plug fla6ezad[

puo y/xfa<

f ,

ejection pressure

+ 3510 lb average pull-out load V. Conclusion

• Roll plug qualified for intended use at TMI-1 1

4/7/82 I

83 Primary System Cleanup

1. Sulfur in RCS water has been reduced from 750 ppb to 100 ppb .

II. If analysis shows it is required, we plan to reduce the amount of sulfur on the surfaces of primary system components and OTSG tubes Ill. Cleanup method identification will consider:

eH O2 2 concentrations of 0,10,100, and 1000 ppm epH of 7.0, .8.0, and 9.0 with LiOH or NH 4 0H additive e Normal RCS chemistry i

, 4/7/82  !

84 Preliminary OTSG Pre-Service Testincj Plans

• ECT -Statistical baseline examination of the new expansion and transition

• Drip test -150 psi on OTSG secondary side (H 2 O)

• Bubble test -150 psi on OTSG secondary side (N 2 )

-Sensitivity ~.1 gpd/ tube

• Leak test -2155 psi on primary side A PE 125% of normal

( h 1500 psi)

-Sensitivity ~ 10 gpd (after 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />, current RCS activity level)

• Power -Natural circulation cooldown escalation Main feed pump trip (40%

testing power) l Turbine trip (100% power) l l

4/7/82

85 s

l OTSG In-Servsce Monitoring e Continuous leak rate monitoring using ativity, mass balance and/or chemical concentrations

-sensitivity 2 10 gpd

• after 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> of leakage 03% failed fuel

• condenser vacuum pump discharge activity e Basis for corrective action

-total leak rate

-rate of change of leak rate l

l

\

4/7/82

86 OTSG Repair Program Overview We expect that the overall OTSG repair program, including inspections, repair process qualification, primary system cleanup, leak testing and differential pressure testing, will provide assurance '

that the probability of abnormal primary to secondary leakage during operation is I very low.

1 4/7/82

l 87

SUMMARY

e THE REPAIR METHOD IS EXPECTED TO SHOW CONFORMANCE TO EXISTING LICENSE AND REGULATORY REQUIREMENTS e TECH SPECS FOR APPENDIX I MAINTAIN fl0RMAL OPERATIONAL CONDITIONS WELL WITHIN ACCIDENT ASSUMPTIONS e THE REPAIRED STEAM GENERATOR IS EXPECTED TO PRESENT NO SIGNIFICANT HAZARD TO STATION OR PUBLIC

, 4/7/82

._ .. . , , . . . ._ . _. . _ . - . - - - - . - - - - - - - - _ . 8 8.s. _

REMAINING WORK e DEVELOP, TEST, QUALIFY THE TUBE - TUBESHEET REPAIR METHOD AND PROCESS DETAILS e RESTORE ADEQUATE STATE OF CLEANLINESS e INSPECT OTHER PRIMARY SYSTEM INTERNALS e COMPLETE DETAILS OF FAILURE ANALYSIS AND

~ ~

TECHNICAL AND SAFETY ANALYSIS OF REPAIRED .

STEAM GENERATORS M' S $

W **

O e4

89 Current TMI-1 Operating and Effluent Limits

'o0 4 /

5X103 -

!,,'/ 4 1A

1. C 9 #+g,Y 7,Os A 4 p /g

1. f4 Ap #+p

• iii 3g3 '+, I/

4,,

7 4r 5X102 _ 6/g

. #p,<

i

=

s.,'"%

+

s

.'++

g 102 -

pp +j p

[ 1 GPM LEAKAGE LIMIT \

$ 50 -

x E

a 10 -

=

5 m

1.0 l l I ' I I I

.02.03 .05 0.1 0.3 0.5 1.0 5.0 FRACTION FUEL FAILURE (%)

4/7/82

90 SAFETY EVALUATION PARAMETERS - STEAM GENERATOR REPAIR e TUBE PLUGGING - AFFECTS PRIMARY SYSTEM FLOW

- AFFECTS LOCA ANALYSIS (HEAT TRANSFER) 5 1000 TUBES MAY BE PLUGGED WITHIN BOUNDS OF EXISTING SAFETY ANALYSIS / TECH SPEC'S e TECHNICAL SPECIFICATION LEAKAGE - UNIDENTIFIED 1 GPM

- STEAM GENERATORS 1 GPM l

e TECHNICAL SPECIFICATION RELEASE - 5 0.06 MR/M0 WITHOUT TREATMENT  ;

e DESIGN BASIS STEAM GENERATOR RUPTURE - 435 GPM e APPENDIX I - LIQUIDS 5 3 MR/YR -

- GAS 5 5 MR/YR e ALARA CONSIDERATIONS FOR SECONDARY SYSTEMS ACTIVITY, IF i ANY ACTIVITY PRESENT I

4/7/82

l 91 SAFETY REVIEW e WILL PERFORM A COMPLETE INTERNAL REVIEW UNDER THE PROVISIONS OF 50.59 1

e WILL HAVE THE INTERNAL SAFETY ASSESSMENT / REVIEW FURTHER EXAMINED BY

- THE GPUN GENERAL OFFICE REVIEW BOARD (GORB)

- AN EXTERNAL (T0 GPUN) FURTHER INDEPENDENT REVIEW GROUP e THE REVIEW WILL BE BASED ON MEETING ESTABLISHED NRC REGULATORY REQUIREMENTS AND EXISTING TMI-1 -

TECH SPEC'S a

4/7/82

92

_ CONCLUSIONS - STEAM GENERATORS e FAILURE MECHANISM THEORY IDENTIFIED e BASIC STEAM GENERATOR TUBE MATERIAL REMAINS GOOD e REPAIR WILL NOT DEGRADE THE ORIGINAL DESIGN MARGINS e CAUSATIVE CHEMICAL SPECIES DEPLETED BEFORE RESTART ,

e EVEN WITH LARGE NUMBERS OF TUBE FAILURES IN UTS - DESIGN BASIS TUBE RUPTURE ACCIDENT IS NOT APPROACHED OR EXCEEDED e OPERATIONAL EXPERIENCE SUGGESTS TIGHT /VERY SMALL LEAK PATHS CLOSE DURING OPERATION e METHODS EXIST TO CONFIRM CONTINUED SERVICEABILITY OF THE STEAM GENERATORS AFTER REPAIRS ACCOMPLISHED e ECT e ABSOLUTE LEAKAGE AND LEAKAGE TREND MONITORING e SECONDARY SAMPLING / RADIATION MONITORING e LIMITED PLANT THERMAL CYCLE TESTING 4

4/7/82

__ -__-_.- ___ --