ML20135A720

Provides Addl Info Re Request for Authorization to Utilize ASME Boiler & Pressure Vessel Code Case N-521

ML20135A720
Person / Time
Site:	Prairie Island
Issue date:	11/26/1996
From:	Wadley M NORTHERN STATES POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NUDOCS 9612040016
Download: ML20135A720 (29)

Text

_ _

Northern States Power Company Prairie Island Nuclear Generating Plant 1717 Wakonade Dr. East Welch. Minnesota 55089 i

November 26,1996 10 CFR Part 50 l Section 50.55(a)  !

l l

U S Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555

' PRAIRIE ISLAND NUCLEAR GENERATING PLANT i Docket Nos. 50-282 License Nos. DPR-42 l 50-306 DPR-60 Additional Information Related to the Request for Authorization  ;

i to Utilize ASME Boiler & Pressure Vessel Code Case N-521 The purpose of this letter is to provide additional information related to our November 1,

, 1995 request for NRC authorization to utilize ASME Boiler and Pressure Vessel Code i Case N-521, " Alternative Rules for Deferral of Inspections of Nozzle-to-Vessel Welds, inside Radius Sections, and Nozzle-to-Safe End Welds of a Pressurized Water Reactor (PWR) Vessel,Section XI, Division 1", during the Prairie Island Third Ten Year Inservice Inspection Interval for both Units 1 and 2. This letter also supersedes the November 1,1995 letter and the two related letters of July 25,1996 and August 9, 1996.

ASME Code Section XI requires that at least 25% but not more than 50% of the Nozzle-to-Vessel and inside Radius Sections be examined in the first period of the ten year interval. However, Code Case N-521 provides for deferral of the inspection of nozzle-to-vessel welds, inside radius sections, and nozzle-to-safe end welds of a pressurized water reactor vessel to the end of the inspection interval provided specified conditions /

are satisfied, rather than dividing the inspections between the first and third periods of the inspection interval. This request is submitted pursuant to 10 CFR Part 50, Section (),'

Q )f 50.55a(a)(3) and Section 50.55a, Footnote 6. This request is for deferral of inspections \

not for deletion. We are requesting this deferral pursuant to 50.55a(a)(3)(ii):

O Compliance with the specified requirements of this section would result in 1 dg 0,>g hardship or unusual difficulty without a compensating increase in the level of k quality and safety.

9612040016 961126 PDR ADOCK 05000282 G PDR

, .. i

• . USNRC

, NORTHERN STATES POWER COMPANY November 26,1996

, Page 2 l

Code Case N-521 was approved by the ASME Code on August 9,1993. However, the Code Case has not yet been endorsed in Regulatory Guide 1.147," Inservice inspection Code Case Acceptability ASME Section XI Division 1". Until the Code Case is generically endorsed by the Regulatory Guide, specific NRC authorization is required before it can be used.

Both Units of Prairie Island are currently in the 3rd 10-year ISI Interval. The interval began December 16,1993 for Unit 1 and December 21,1994 for Unit 2. Relief is sought for inspections of the primary system outlet nozzles during this interval. The code case addresses: nozzle-to-safe end welds, nozzle-to-vessel welds, and the inner radius sections of the nozzles. However, our need for relief is not the same for each of these groups.

1. Nozzle-to-safe end welds Although Code Case N-521 addresses these welds, we do not need schedular relief from the code requirements; these inspections have already been performed in the  !

1st period for Unit 1 and will be performed in the 1st period for Unit 2 in the refueling outage scheduled to start in January 1997. Therefore, whether or not authorization is granted to use the code case for the safe end welds will have no effect on the inspection schedule. Thus there will be no further discussion of the safe end welds. j

2. Nozzle-to-vesselwelds l

i The nozzle-to-vessel welds have been inspected recently (see Table 1). They were inspected to the 1980 Edition, Winter 1981 Addenda of ASME Section XI requirements rather than the 1989 Edition, No Addenda (the edition of the current 1 10-year Interval); schedular relief to the 3rd period of the interval will alleviate the need to repeat the inspections within a short period of time. Details are discussed i below. The resulting interval between inspections will be from the 3rd period of the 2nd interval to the 3rd period of the 3rd interval.

3. Inner radius sections l The inner radii have not been inspected since 1985 on either unit (see Table 1),

therefore direct application of deferment per the code case is sought. Details are discussed below.

N 521#4. DOC

e

... USNRC NORTHERN STATES POWER COMPANY l

November 26,1996 Page 3 TABLE 1

UNIT 1 Item Irspected Outlet Loop A Outlet Loop B

$- NV, Item # = IR, item # = NV, item # = IR, item # =

] 1S1-50, W-7 ISl-50, W-7 IR ISl-50, W-10 ISI-50, W-10 IR j Last inspected 1994 1985 1994 1985 Code Edition / Add 80/ws1 80/W81 80/ws1

) 80/ws1

Inspection Results NAD NAD NAD NAD T

d i

j- UNIT 2 4 Item inspected Outlet Loop A Outlet Loop B t NV, item # = IR, item # = NV, item # = IR, item #=

3- 2-ISl40, N-7 2-ISl40, N-7 IR 2-1S140, N-10 2-1S140, N-10 IR Last Inspected - 1993 1985 1993 1985 j Code Edition / Add 80/W81 80/ws1 80/w81 80/w81 inspection Results NAD NAD NAD NAD l OUTLET NOZZLES NOZZLE-TO-VESSEL WELDS I During the Prairie Island Unit 2,1997 refueling outage, which is the last refueling outage of the first inspection period, the two outlet nonle nonle-to-vessel welds would need to be scheduled for examination in order to meet the 25 percent minimum requirement of Section XI under the present ten-year ISI program. To do these j inspections equipment is required to be placed in the reactor vessel. For Unit 1, the

{ last refueling outage of the first inspection period has been completed. To do these i inspections per the code schedule, the unit would need to be shutdown between

] refueling outages in order to place the inspection equipment in the reactor vessel.

We believe that the significant additional cost for each unit to perform the examinations of the reactor vessel outlet nonle nonle-to-vessel welds, in separate outages from the other examinations, is not a practical expenditure of resources in light of the following points:

1. All of the nonie-to-vessel welds were last examined in 1994 (Unit 2,1993), per ASME Section XI,1980 Edition, Winter 1981 Addenda, with no observed indications. We consider the results of these past examinations to be representative of the present reactor vessel nonle conditions at Prairie Island;

2. Although the scheduled welds are to be examined with automated equipment, some personnel radiation exposure will occur. As part of our overall ALARA program, any amount of personnel radiation exposure that can be saved is important. In addition, radioactive waste will be reduced; N-521#4. DOC

USNRC NORTHERN STATES POWER COMPANY j November 26,1996 Page 4 4

} 3. It makes good sense to combine all reactor vessel examinations into one automated effort from both ALARA and cost considerations. Using the same i vendor, equipment, personnel, and procedures to examine all the welds at the same time will enhance the quality of the examination results and produce a much smaller possibility of error.

i

4. It would provide increased safety to move the equipment into and out of the vessel only once in the interval.

Code Case N-521 allows deferral of the inspections of nozzle-to-vessel welds of a PWR

[ vessel to the end of the inspection interval if the following conditions are met:

5 a) No inservice repairs or replacements by welding have ever been performed j on any of the Nozzle-to-Vessel Welds, inside Radius Sections, and Nozzle-l to-Safe End Welds.

2- b) None of the Nozzle-to-Vessel Welds, inside Radius Sections, and Nozzle-to-Safe End Welds contains identified flaws or relevant conditions that currently j' require successive inspections in accordance with IWB-2420-(b).

j c) The unit is not in the first inspection interval.

l; Both Prairie Island units rneet the provisions outlined in the Code Case. In addition, 2

these welds were inspected twice in the second interval, once in the first period and once in the third period. Thus, deferring the third interval inspection from the first period

{ to the third keeps with the Code philosophy for detecting age-related degradation

! during a specified interval of time. We therefore request authorization to use the Code

! Case to defer the first period inspections of the nozzle-to-vessel welds to the 3rd period -

!, of the inspection interval.

l Justification for the conclusion that full compliance with the specified requirements

[ would not result in a compensating increase in the level of quality and safety:

The reactor vessel inspections done in 1993 and 1994 were performed per the 1980 Edition, Winter 1981 Addenda as augmented by Regulatory Guide 1.150 Revision 1. To have met the requirements of the third ten year interval the inspections were required to be performed to the 1989 Edition, No Addenda of -

the Code as augmented by Regulatory Guide 1.150, Revision 1. Regulatory.

Guide 1.150 contains the recording and reporting requirements. There is no difference in extent of examination between the 1980 Edition, Winter 1981 Addenda and the 1989 Edition, No Addenda of the Code. There are differences between the two codes regarding allowable indications, but since the 1993 and 1994 inspections did not detect any recordable indications on the outlet nozzle to vessel welds, there is no need for reevaluation of any indications to the newer Code.

i N 62184. DOC

j ,, J USNRC NORTHERN STATES POWER COMPANY

, November 26,1996 Page 5 4

Another difference between the two codes is in the requirements for 3xaminers

certification. Examiner certification requirements in Appendix Vil to Section V is i new to the 1989 Edition. However, this edition allows the examiner to perform j inspections under the examiner's current certifications until the certifications

! expire; at the time of expiration the examiner then requalifies under the

requirements contained in Appendix Vil to Section V of the 1989 Edition, No j Addenda. Therefore, even if the examinations had been performed under the

! requirements of the 1989 Edition, No Addenda, the examiners' certifications j could have been unchanged from what they were for an examination under the l 1980 Edition, Winter 1981 Addenda.

1

. Therefore, there would have been no significant differences in the 1993 and 1994 nozzle-to-vessel weld examination results had the examinations been j performed to the 1989 Edition, No Addenda instead of the 1980 Edition, Winter l 1981 Addenda.

i The level of quality and safety will not be decreased by the use of Code Case i N-521 to defer the outlet nozzle-to-vessel welds to the 3rd period of the 3rd

interval since these welds were inspected in the 3rd period of the 2nd interval. ,

l j' OUTLET NOZZLES INNER RADIUS SECTIONS '

e j! During the Prairie Island Unit 2,1997 refueling outage, which is the last refueling

!' outage of the first inspection period, the two outlet nozzles inner radius sections would l need to be scheduled for examination in order to meet the 25 percent minimum i requirement of Section XI under the present ten-year ISI program. To do these I inspections equipment is required to be placed in the reactor vessel. For Unit 1, the I

i. last refueling outage of the first inspection period has been completed. To do these  ;

l. inspections per the code schedule, the unit would need to be shutdown between '

l refueling outages in order to place the inspection equipment in the reactor vessel.

4 1

i We believe that the significant additional cost for each unit to perform the examinations i of the reactor vessel outlet nozzle inner radius sections, in separate outages from the j other examinations, is not a practical expenditure of resources in light of the following j points: 4 j 1. Four of the six inner radius sections (2 inlet,2 outlet, and 2 safety injection nozzles) were last examined in 1994 (Unit 2,1993), per ASME Section XI,1980 Edition, Winter 1981 Addenda, with no observed indications. We consider the L results of these past examinations to be representative of the present reactor vessel nozzle conditions at Prairie Island;

[

! 2. Although the scheduled outlet nozzles inner radius sections are to be examined f

with automated equipment, some personnel radiation exposure will occur. As j M-52154. DOC i

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

w ..

.' USNRC NORTHERN STATES POWER COMPANY ,

l November 26,1996 j

j Page 6 4

part of our overall ALARA program, any amount of personnel radiation exposure i

that can be saved is important. In addition, radioactive waste will be reduced; 3

j 3. It makes good sense to combine all reactor vessel examinations into one

! automated effort from both ALARA and cost considerations. Using the same vendor, equipment, personnel, and procedures to examine all the inner radius j sections at the same time will enhance the quality of the examination results and j . produce a much smaller possibility of error.

4. It would provide increased safety to move the equipment into and out of the vessel only once in the interval.

1 Code Case N-521 allows deferral of the inspections of inner radius sections of a PWR vessel to the end of the inspection interval if the following conditions are met:

i j a) No inservice repairs or replacements by welding have ever been performed i on any of the Nozzle-to-Vessel Welds, inside Radius Sections, and Nozzle-l to-Safe End Welds.

b) None of the Nozzle-to-Vessel Welds, inside Radius Sections, and Nozzle-to-

!- Safe End Welds contains identified flaws or relevant conditions that currently 1 1 require successive inspections in accordance with IWB-2420-(b). l

! c) The unit is not in the first inspection interval.

i l Both Prairie Island units meet the provisions outlined in the Code Case. We therefore

request authorization to use the Code Case to defer the first period inspections of the

inner radius sections to the 3rd inspection period of the interval.

4 l Justification for the conclusion that full compliance with the specified requirements would not result in a compensating increase in the level of quality and safety:

Technical Evaluation

An evaluation has been performed by Westinghouse to appraise the susceptibility l~ of the inner radii of the outlet nozzles to flaw development and growth. A summary

. of this evaluation and a discussion of the methodology used is attached to this letter.

The evaluation concludes that deferring the inspections of these inner radius sections does not decrease the quality or safety.

i n This evaluation assumes that existing flaws of significance would have been

[ detected in the 1985 examinations, which were performed to the requirements of the i 1980 Edition, Winter 1981 Addenda would have been detected. The ultrasonic

examination techniques used in 1985 are essentially the same as are currently j being used in our ISI program. Flaws large enough to be reportable are easily

, detectable with these techniques and it is highly improbable that a flaw of such size could have existed and not been detected with the techniques used.

{  ;

w.sm. .we i l

gaa a zam *Eua se'A-.a eiw

. USNRC NORTHERN STATES POWER COMPANY November 26,1996 Page 7 Industry History The inspection of reactor vessel nozzle-to-shell welds, their nozzle inner radius sections, and associated nozzle-to-safe end welds was originally distributed by Code requirements between the three inspection periods of an interval on the basis of monitoring these areas uniformly during the interval. The cost in terms of radiation exposure due to removing and replacing the core barrel caused the Code to be changed such that only 25 percent, to a maximum of 50 percent of the nozzles were to be inspected in the 1st period and inspections were dropped for the 2nd -

period. This resulted in the core barrel being pulled only once because the normal arrangement of the two outlet passages in the core barrel allowed direct access to the nozzles. The remaining nozzles were inspected in the 3rd period with the removal of the core barrel required. The industry believes that with the present large population of operating reactors, the examination of nozzles of a given vessel being distributed between an outage in the 1st and an outage in the 3rd period can' be eliminated and concentrated in an outage in the 3rd period of the interval. .The examinations of all nozzles in all vessels within a given ten-year interval are rather uniformly distributed across the industry. Each reactor is representative of the operating conditions throughout the population of the reactors for a particular nuclear steam supply system design. The number of nozzles inspected within a ten-year inspection interval will be the same, either with all nozzles of a given reactor being inspected during one outage in the 3rd period, or the inspections being distributed between outages in the 1st and 3rd periods. Therefore, we believe that inspecting all of the nozzles of a given reactor at one time will provide the necessary assurance'of detecting the presence of an industry wide degrading condition.

In this letter we have made no new Nuclear Regulatory Commission commitments.

Please contact Jack Leveille (612-388-1121, Ext. 4662) if you have any questions related to this letter.

ff'h

Michael D Wadley Plant Manager Prairie Island Nuclear Generating Plant c: Regional Administrator - Region Ill, NRC Senior Resident inspector, NRC NRR Project Manager, NRC J E Silberg

Attachment:

Technical Basis for Outlet Nozzle Inner Radius inspection Exemption N-52154. DOC

.- . ~~ . _ ~ . .- _ - - . - ~ - - - - . - - - - - - -..

7

) .

7

,/. ,

i Technical Basis for Outlet Nozzle Inner Radius Inspection Exemotion:

j Prairie Island Units I andl

1.0 INTRODUCTION

The purpose of this note is to provide an assessment of the structural integrity of the Prairie j

[ Island Units 1 and 2 reactor vessel outlet nozzle inner radius sections, and the need for their j j inservice inspection. Two complementary approaches will be used, inspection history, and  !

fracture assessment. Each of these approaches has led to the same conclusion, which is that 5 . inservice inspection of th'e nozzle inner radius does not significantly improve confidence in the

, structural integrity of the reactor vessel. Extending the inspection interval for the nozzle inner radius beyond 10 years will not affect structural integrity. Even if the interval were 3 extended to 20 years or beyond, the conclusion would not change. Therefore it will be shown that extending the interval between inspect:ons for the Prairie Island outlet nozzles will not

pose any technical problems. The geometiy of the outlet nozzle for the Prairie Island units is

! shown in Figure 1.

1

2.0 INSPECTION HISTOBX l The outlet nozzle forgings for both Prairie Island Units 1 and 2 were examined by both UT j and MT (magnetic particle testing) prior to the cladding being applied, per the requirements of the material specification. After cladding, the nozzles were required to be liquid penetrant l tested to ensure the integrity of the cladding. The inspection after the hydrotest includes UT as well as radiography, for acceptance to the ASME Code,Section XI, and Section III.

For inservice inspection of the outlet nozzle inner radius region, a contact examination is l carried out from the inside of the reactor vessel. The details of the previous inspections have ,

l been discussed earlier.

i Underclad Cracks. The only indications of any type which have been found are small underclad cracks, which exist in all vessels manufactured by Creusot i Loire and Rotterdam Dockyard. These small cracks were created by l the cladding process, and have a maximum depth of about 0.12 inches.

. Crack growth evaluations and multiple inspections of these indications j in a number of reactor vessels have shown that these flaws do not grow during service and have no impact on the integrity of the vessel. i 1

1- A survey was conducted of all Westinghouse plant owners to determine if flaws have been found in the inner radius of any of the nozzles in the primary system. The results of over 300 UT examinations which have been completed on these nozzles showed that no indications

have been found. The survey included nozzles in pressurizers, and steam generator primary i

! nozzles as well, and confinned the expected findings. i l

2

= - ,

a j t- .. .

3

• =

+ .

, While this finding is not itself sufficient to prove there is no need for further inspection in this -

area,it is consistent with the other findings reported here,in that no concerns are evident with flaws in this region at the beginning of service. There are no mechanisms of damage other than fatigue for the outlet' nozzle. Stress corrosion cracking has never been ovserved in either stainless or low alloy pressure vessel steel with no oxygen present. Therefore, the only scenarios of concern are for a flaw which was not found in the preservice examination to grow during service or for a flaw to initiate during service and grow. Initiation is assessed quantitatively by the fatigue analysis carried out to certify the design acceptance of the vessel.-

' Crack growth has been assessed as part of the fracture assessment below, in Section 3.3.'

3.0 - FRACTURE ASSESSMENT A fracture evaluation was carried out for the Prairie Island outlet nozzle inner radius,~ and i I

documented in Reference 1, in the form of flaw evaluation charts, which show the largest allowable flaw which would meet the ASME Section XI flaw acceptance requirements of ,

paragraph IWB 3600. Although this report was prepared in 1984, it is based on a detailed series of finite element stress analyses and fracture analysis methods which are consistent with j the present state of the art. Furthermore, the requirements of Section XI IWB 3600 have not changed since that time. Therefore the conclusions of references I and 2 remain valid today, that is that there is a reasonable tolerance for the presence of flaws in the outlet nozzle inner radius region.

3.1 Allowable Flaw Size This is seen in the flaw chart of Figure 2,in ter ns of allowable flaw depth as a percentage of the nozzle corner ligament, which is 17.3 inches for the Prairie Island units. The details of the  !

construction of this flaw chart are described in Appendix A, In Figure 2 we see that the allowable flaw depth is presented as a function of flaw shape (a/f), including the effects of l fatigue crack growth. For a continuously long flaw the allowable is equal to the value from i the IWB 3500 standards table,2.5 percent of the wall thickness. For flaws with more realistic l shapes, the allowable depth is much greater 0.67 inches for a 6:1 flaw (a/t = 0.167) and 1.6 i inches for a semi circular flaw (a/t = 0.5).

3.2 Critical Flaw Size Another view of the tolerance of the outlet nozzle for the presence of flaws can be obtained by calculating the critical flaw size for this region as a function of flaw shape, and these results are presented in Figure 3. The critical depth for a continuous flaw in this region was calculated to be 3.7 inches, or 21.6 percent of the ligament thickness for the dominating normal and upset transient, Loss of Power. For the governing emergency and faulted condition, the Large Steamline Break, the critical depth was 2.2 inches. The critical flaw l l

l

\

depth for other flaw shapes is much larger, as seen in Figure 3. Therefore it may be concluded that the structural integrity of the outlet nozzle will not be affected by the presence of any flaws smaller than those discussed above at the inner radius.

' 3.3 Fatinue Crack Growth The fatigue crack growth predicted for a ran,ge of flaw sizes an.<t shapes is shown in Table 1.

It may be readily seen that the predicted crack growth is very small, even with the assumption that the surface flaw is exposed to the PWR water environment. Since service experience has -

shown that few (if any) flaws have even been found to break through the cladding, this is a very conservative assessment.' It is clear from Table 1 that a range of flaw sizes will not grow

- beyond the code allowables in the 40 year design life.

l M CONCLUSIONS The assessments discussed here have shown that deferring the outlet nozzle inner radius exams will not decrease the level of quality and safety for Prairie Island Units 1 and 2. l Inspections which have been done have not led to the discovery of any indications in the entire service history of Westinghouse plants, or Pressurized Water Reactors regardless of manufacturer or designer. The fracture assessment showed the outlet nozzle inner radius has a large tolerance for flaws. There are no mechanisms for the development of flaws or the -

significant growth of existing flaws during service, so the risk of failure is not decreased by inservice inspection. The case for extending the inspection interval for these nozzles is so st:ong that efforts are underway to consider elimination of this inspection requirement from Section XI of the ASME Code. )

l i

l l

l 1

l

.. .. 4 Technical Basis for Outlet Nozzle Inner Radius Insnection Exemntion-Prairie Island Units 1 and 2 j I

1.0 INTRODUCTION

The purpose of this note is to provide an assessment of the structural integrity of the Prairie Island Units 1 and 2 reactor vessel outlet nozzle inner radius sections, and the need for their inservice inspection. Two complementary approaches will be used, inspection history, and fracture assessment. These approaches lead to the conclusion that inservice inspection of the nozzle inner radius does not significantly improve confidence in the structural integrity of the J

reactor vessel. Extending the inspection interval for the nozzle inner radius beyond 10 years will not affect structural integrity. Even if the interval were extended to 20 years or beyond, the conclusion would not change. Therefore it will be shown that extending the interval between inspections for the Prairie Island outlet nozzle inner radius sections will not pose any technical problems. The geometry of the outlet nozzle for the Prairie Island units is shown in Figure 1.

2.0 INSPECTION HISTORY The outlet nozzle forgings for both Prairie Island Units 1 and 2 were examined by both UT and MT (magnetic particle testing) prior to the cladding being applied, per the requirements of the material specification. After cladding, the' nozzles were required to be liquid penetrant tested to .j ensure the integrity of the cladding. The inspection after the hydrotest includes UT as well as radiography, for acceptance to the ASME Code,Section XI, and Section Ill.

For inservice inspection of the outlet nozzle inner radius region, a contact examination is carried out from the inside of the reactor vessel. The details of the previous inspections are discussed in the body of the letter.  ;

)

Underclad Cracks.

The only indications of any type which have been found are small underclad cracks.

These small cracks were created by the cladding process and were detected by special ultrasonic testing techniques developed for this specific application. (An industry effort was developed following the 1979 discovery of these cracks in another Westinghouse reactor.) Field tests were performed during scheduled ISI outages for both Prairie  ;

Island units in 1981. Reflectors detected during these examinations were evaluated per ASME Section XI requirements and found to be acceptable. Crack growth evaluations performed to address these flaws have shown that these flaws do not grow significantly during service and have no impact on the integrity of the vessel. Corroborating these evaluations are the special inspections performed for two reactors to appraise the  ;

underclad cracking following 15 years of operation since the initial cracking was  ;

detected. No growth was observed. The author knows of no Westinghouse PWRs worldwide that have experienced problems with underclad cracks.

A survey was conducted of all Westinghouse plant owners to determine if flaws have been found in the inner radius of the nozzles in the primary system. The results ofover 300 UT examinations which have been completed on these nozzles showed that no recordable indications have been found. The survey included nozzles in pressurizers and steam generator primary nozzles, and confirmed the expected findings. In addition, this author is unaware of any recordable indications in the nozzle inner radius region of any plant of Westinghouse design worldwide.

While this finding is not itself sufficient to prove there is no need for further inspection in this area, it is consistent with the other findings reported here, in that no concerns are evident with flaws in this region at the beginning of service. 'Ihere are no mechanisms of damage other than fatigue for the outlet nozzle. Stress corrosion cracking has never been observed in either stainless or low alloy pressure vessel steel with no oxygen present. Therefore, the only scenarios of concern are for a flaw which was not found in the preservice examination to grow during service or for a flaw to initiate during service and grow. Initiation is assessed quantitatively by the fatigue analysis carried out to certify the design acceptance of the vessel. Crack growth has been assessed as part of the fracture assessment below, in Section 3.3.

3.0 FRACTURE ASSESSMENT A fracture evaluation was carried out for the Prairie Island outlet nozzle inner radius, and documented in Reference 1, in the form of flaw evaluation charts, which show the largest allowable flaw which would meet the ASME Section XI flaw acceptance requirements of paragraph IWB 3600. Although this report was prepared in 1984, it is based on a detailed series of finite element stress analyses and fracture analysis methods which are consistent with the present state of the art. Furthermore, the requirements of Section XI IWB 3600 have not changed since that time. Therefore the conclusions of references I and 2 remain valid today, that is that there is a reasonable tolerance for the presence of flaws in the outlet nozzle inner radius region.

3.1 Allowable Flaw Size This is seen in the flaw chart of Figure 2, in terms of allowable flaw depth as a percentage of the nozzle corner ligament, which is 17.3 inches for the Prairie Island units. The details of the construction of this flaw chart are described in Appendix A. In Figure 2 we see that the allowable flaw depth is presented as a function of flaw shape (a/l), including the effects of fatigue crack growth. For a continuously long flaw the allowable is equal to the value from the IWB 3500 standards table,2.5 percent of the wall thickness. For flaws with more realistic

l A

shapes, the allowable depth is much greater. The calculated depths are 0.67 inches for a 6:1 flaw '

]. (a/l = 0.167) and 1.6 inches for a semi circular flaw (a/l = 0.5).  !

3.2 Critienl Flaw Size 4

Another view of the tolerance of the outlet nozzle for the presence of flaws can be obtained by

, calculating the critical flaw size for this region as a function of flaw shape, and these results are

. presented in Figure 3. The critical depth for a continuous flaw in this region was calculated to be 3.7 inches, or 21.6 percent of the ligament thickness for the dominating normal and upset

. transient, Loss of Power. For the governing emergency and faulted condition, the Large Steamline Break, the critical depth was 2.2 inches. The critical flaw 3

depth for other flaw shapes is much larger, as seen in Figure 3. Therefore it may be concluded j that the structural integrity of the outlet nozzle will not be affected by the presence of any flaws smaller than 2.2 inches at the inner radius.

1

, 3.3 Fatigue Crack Growth i-l- The fatigue crack growth predicted for a range of flaw sizes and shapes is shown in Table 1. ' It may be readily seen that the predicted crack growth is very small, even with the assumption that l 3

the surface flaw is exposed to the PWR water environment. Since service experience has shown  !

p that few (if any) flaws have even been found to break through the cladding, this is a very )

j conservative assessment. It is clear from Table i that a range of flaw sizes will not significantly

+

grow m 20 years or more.

i 7 LQ CONCLUSIONS 4

The assessments discussed here have shown that deferring the outlet nozzle inner radius exams i j will not decrease the level of quality and safety for Prairie Island Units 1 and 2. Inspections ]

which have been done have not led to the discovery of any indications in the entire service i history of Westinghouse plants, or Pressurized Water Reactors regardless of manufacturer or i designer. The fracture assessment showed the outlet nozzle inner radius has a large tolerance for j . flaws. There are no mechanisms for the development of flaws or the significant growth of

{ existing flaws during service, so the risk of failure is not decreased by inservice inspection. The  !

]' case for extending the inspection interval for these nozzles is so strong that efforts are underway l

_ to consider elimination of this inspection requirement from Section XI of the ASME Code. I i

I 3

m_ _._- ._ _ _ . . _ _ _ _ _ _ _ _ _ _ _ . _ _ . _ _ . _ . . _ _ _ . _ _ _ _ . ___.-. _-- .

. s.

i REFERENCES i l

i 1. W. H. Bamford, et al, " Background and Technical Basis for the Handbook on Flaw Evaluation j for Prairie Island Units 1 and 2 Reactor Vessels," Westinghouse WCAP 10562, December 1984.

L l 2.' WCAP-10363, " Handbook on Flaw Evaluation for Prairie Island Units 1 and 2 Reactor Vessels,"

by W. H. Bamford, et. al., December 1984.

l

!- 3. Newman, J. C. Jr. and Raju, I. S., " Stress Intensity Factors for Internal Surface Cracks in -;

Cylindrical Pressure Vessels," ASME Tranc , Journal of Pressure Vessel Technology, Vol.102, r 1980, pp. 342-346.

l

4. C. B. Buchalet, and W. H. Bamford, " Stress Intensity Factor Solutions for Continuous Surface Flaws in Reactor Pressure Vessels," in Mechanics of Crack Growth, ASTM, STP 590,1976, pp.

385-402.

l .

l S. S. S. Palusamy, " Fracture Mechanics Evaluations of Prairie Island Units I and 2 Including the

. Effects of Potential Underclad Cracks in Reactor Vessel Outlet Nozzles", Westinghouse Electric l l Corp. WCAP 9722, July 1981. - t i

6. W. H. Bamford, "The Effects of Vessel Cladding on Structural Integrity", to be t l presented / published at the 1997 ASME Pressure Vessels and Piping Conference, Orlando, FL, {

July 1997.  !

l j 7. H. A. Schimmueller and J. L. Ruge, " Estimation of Residual Stresses in Reactor Vessel Specimens by Stainless Steel Strip Electrodes" Internatinnal Conferance on Residual Stresses in Weld Construction and Their Effects. The Welding Institute, London,1977 pp. 251-258.'

?

l f 8. D. E. McCabe, " Fracture Resistance ofIrradiated Stainless Steel Clad Vessels", in Effects of Radiation on Materials: 14th International Symposium, ASTM STP 1046,1990.

l t -

i l'

4

, n .

- -, ,e , - r .-- - --

4 e_AB.4 .A eA ,a eth 424__A,w_du._J# uma .A-._4Au- Jew-4-_.. W sh_ ,h A. & .a

• Eau , .A.e..a* .JaAAE 3..A 4 _a4..,h,- C4'd 2. h 6A4 n ap IL.-- _4 4 h.A.mAJ A._.

. e a e d

.-  % rA ~ '

^

l7 l .

4 N

I +E*

4 1 .

E 2 2 D 4 g .. g a *5 "5 -

4 i

4 5

~

i A c-1 d J352.8 \

s  %

229 _

871E

, tiets A

1 4

4, Figure 1 Geometry of the Reactor Vessel Outlet Nozzle for Prairie Island Units I and 2

~ ~

=~ ~ ~ ~'

)::i:[ ;-- ill il E~~ ~-l-N jfl~~ -111 18 g-g55 gyggj;; :::::::::: .__

---=== =

16 ~

I?~ - -- ~~~ - ---

l?? :::::::: ZT ::::: -

--~

1 1--------

[14  ::: _...___ .____. _ ._._: 5  ???f = - -- ---

__ _  :::-- --- 77_

k 12

~

~

~

T ~~

'~-

E

~ -----

I ' ~~

10

-]

a ---

===pa "

d =EE

, ?

---

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

8 _ ______  :: i: WITHIN THIS ZONE THE SURFACE N

" " ==

__ p FLAW IS ACCEPTABLE BY CODE b6 u.

1_.--_-._'

~  ? ~~--~~~'-:: ANALYTICAL CRITERIA (IWB 3600)

.e p _

=.__

me N

.____-l_ II. ~iz_.____._

^~~----

2III .23 CODE ALLOWABLE UMIT 2

~~~' ^~-~ ~--'- -

O 0 0.1 0.2 0.3 0.4 0.5 FLAW SHAPE (a/A)

Figure 2. Surface Flaw Evaluation Chart for Code Allowable Flaws per Section XI IWB 3600 at the Outlet Nozzle Inner Radius for Prairie Island Units 1 and 2

Loss of Power Large Steamline Break .

so , _ _____ _________._

= ,_ __

f _. _ _ ____

f (5 -

45  ??I

__.~~

5??55&-----

/-- ----}--??- d 1______n

____._.___r:: ___

40 2----- 7_

:_::::7> ==== _..

".."'j m ---.l ----

~~ .. .__

f .__

y -

^

? i Q ao  :::::::x:

)

r--

I e ,

r 9- EEE:Y ---

?

25 JL = . . . .__ ___ _. _ ___ -. _ ..__._._  ::

o 20 3 m -

b1s E

7 WW-- __.

- ~'- ~ ~~ ~~~ '~ ---- ------

10 .._

' ~~~' ~-~~ ~

--- - -~~

5 11-~---

~

~~

O 0 0.1 0.2 0.3 0.4 0.5 FLAW SHAPE (a/A)

Figure 3. Surface Flaw Evaluation Chart Showing the Critical Flaw Depth at the Outlet Nozzle Inner Radius for Prairie Island Units 1 and 2

__..___..__.__..-___.._______________._.__.______.__.mm_- -- _ _ _ ._ _ - , _ _ _ . - -

• _ -m-- - _ __m - ___

Table i Fatigue Crack Growth Results for the Outlet Nozz!c Inner Radiue:

Prairic Island Units 1 and 2(1)

Continuous Surface Flaw  !

l Initial Depth Depth After Year 10 20 30 40 0.130 0.302  !

0.10 0.193 0.461 0.125 0.185 0.290 0.443 '0.663 0.25 0.390 0.588 0.861. 1.22 0.375 0.572 0.840 1.20 1.64 0.50 0.748 1.07 1.49 1.99 L 0.75 1.09 1.50 2.01 2.61 1.0 . 1.41 1.90 2.48 3.16 1.25 1.72 2.27 2.91 3.69 Semi-elliptic Surface Flaw, a/l = 0.5 ,

Initial Depth Depth After Year i i

10 20 30 40 0.125 0.134 0.142 0.151 0.160 '

O.250 0.266 0.282 0.288 0.317 0.375 0.402 0.430 0.462 0.497 0.500 0.543 0.592 0.649 0.718 0.750 0.840 0.936 1.039 1.151 -

1.0 1.112 1.23 1.35 1.48 1.25 1.38 1.52 1.66 1.81  !

l l

1

c ,

3

! APPENDIX A - FRACTURE ASSESSMENT DETAILS To complete the fractum assessment, all the design transients for the Prairie Island Units 1 and 2 were i considered, as listed in Table 2-1 of Reference 1 and attached here for completeness as Table A-1. A j detailed finite element analysis was completed for this region of the reactor vessel, and each transient i was analyzed. The stress distributions through each cross section ofinterest were obtained from these >

i results and the maximum and minimum hoop stresses for each transients were used in the fatigue crack l growth analysis.

4 i - The material properties used in this evaluation, the reference fracture toughness curves and the fatigue i

crack growth law, have not changed since the analyses were completed for reference [1].

4

) nere are no other sources of stresses which would affect the outlet nozzle corner region. There are no 3

structural welds, and no repairs have been made. Therefore there are no weld residual stresses. The

, piping loads are dissipated before the corner is reached, because of the reinforcement of the nozzle bore i as shown in Figure 1. The other source of stress is from the cladding residual stress, but these stresses

are so low as to be negligible for the temperature range ofinterest, above 200*F. To investigate this

concern, a series of experimental measurements of clad residual stresses were made at a range of 1

temperatures, as documented in reference (6). Tests were made on sample blocks removed from a commercial reactor vessel nozzle dropout. He hole drilling method was used to measure the residual r j stresses through the thickness, starting with the cladding and progressing into the base metal. The results j are shown in Figures A-1 through A-3. Each figure shows the results obtained for two samples tested at j the temperature noted. Twenty-five locations were measured, and the results were very consistent

between the two samples for each temperature. At 70*F, the stresses in both the clad and base metal l reach a maximum of 20 ksi, but near the clad-base metal interface the stresses are compressive. At l 200*F, the stresses in both the clad and base metal peak at about 10 ksi, while at 400*F, the stresses are i even lower, peaking at about 5-8 ksi. As the temperature increases, the clad residual stresses are  ;

continually decreasing, as we approach the stress-free temperature corresponding to the conditions when .

I the clad was originally deposited. These residual stress measurements were made only in the direction ,

j parallel to the clad-base metal interface, but the residual stresses perpendicular to the interface are i generally of the same levels, as shown in Figure A-4, from reference (6). These results show that the clad residual stresses trend toward low values as the temperature increases.

1

! Separately, a series of flawed bend bars was tested with and without cladding, to see what the effect of j the clad and residual stresses would be on the failures. The test set-up and specimen design is shown in Figure A-5, and the results are shown in Figure A-6. The results are presented in terms of the mouth opening of the specimen at failure, which is proportional to the crack driving force. The clad specimens j failed at consistently lowe- .k mouth opening displacements at the lower test temperatures. This is J

because of the residual stres <hich increase as the temperature decreases. At temperatures above i 150*F, there is no effect of the cladding on the results, showing that the residual stresses have no measurable effect.

i e

1 i

j Since the lowest temperature of the transients of concern in evaluating the nonle inner radius is 210*F, j the effect of clad residual stresses can be neglected. This behavior has been studied for irradiated

. specimena by McCabe, who reached similar conclusions (8).

I 1

The governing normal and upset transient was found to be the Loss of Power transient, and the governing l . emergency and faulted condition was the Large Steamline Break. The stress distributions for each of j these transients are tabulated in Table A-2. The temperature for the Loss of Power transient never drops below 500'F, and the corresponding lowest water temperature for the large steamline break is 210 F as

, shown in Figure A-7. The initial RTwor fo the nonle forgings is SF for the Unit i outlet nozzles and -5F for the Unit 2 nonles(l).

i The effect ofirradiation from the core on the outlet nonle regions is negligible, since these nonles are

, far above the core region. Therefore the fracture toughness will always be on the upper shelf for these j and all other transients, so the fracture toughness was set at 200 ksi fin.

(

The expression for stress intensity factor was taken from the work of Raju and Newman [3] for a surface

, flaw in a cylinder. This expression has been shown to be conservative for a nonle comer flaw through later work which actually modelled a nonle corner flaw in a Westinghouse PWR inlet nonie. The

results for the PWR nonle corner flaw were used in the flaw evaluation charts of reference 1, but these

) expressions are not published, so the results shoivn here were obtained using the Raju-Newman approach, so they can be easily verified. The nonle crotch dimension is used as the thickness of the -

cylinder.

j The fracture assessment reported here is a direct application of the rules of Section XI Appendix A. No I exceptions were taken to any of the guidelines contained therein. The acceptance criteria of Section XI )

! have not changed since its inception in 1974, and the material properties have not changed since the  !

l fatigue crack growth law for water environments was last revised in 1980. )

{

! I i

i- l

$ i

\

i ,

3' 4

5 r

4 . .

j Table A-1 Summary of Reactor Vessel Transients from Table 2-1 of(1)

NUMBER OF OCCURRENCES NUMBER TRANSIENT IDENTIFICATION SPECIFIED USED IN THE

. . ANALYSIS Normal Conditions

1 Heatup and Cooldown at 100*F/hr (pressurizer cooldown 200'F/hr) 200 200

^

, 2 Load Follow Cycles (Unit loading and unloading at 5% of full power / min) 18300 18300*

3 Step load increase and decrease of 10% of full power 2000 2000

. 4 Large step load decrease, with steam dump 200 200 5 Steady state fluctuations 6

, Infinite 10  ;

Upset Conditions 1

6 Loss of load, without immediate turbine 80 80

, or reactor trip 7 Loss of power (blackout with natural circulation in the Reactor Coolant System) 40 40 8 Loss of flow (partial loss of flow, one l pump only) 80 80 j 9 Reactor trip from full power 400 400 Faulted Conditions 10 Large Loss of Coolant Accident (LOCA) 1 1 11 Large Steam Line Break (LSB) (other transients described in Section 4) 1 ,

1 Test Conditions-12 Turbine roll test 10 10 13 Primary Side Hydrostatic. test conditions 50 50 14 Cold Hydrostatic test 5- 10  !

• 6000 of these transients were used in the analysis of the outlet nozzle region, based i on plant records, as documented in reference [3].

Table A 2 Stress Distributions for the Governing Transients: Outlet Nozzle Inner Radius for Prairie Island Units 1 and 2 i

Loss of Power Large Steamline Break Location (x/t) Stress Location (x/t) Stress -

0 45.35 0 67.2 0.59 44.9 0.59 63.1 i

.166 37.0 .166 58.8

.332 37.2 .332 55.0 l

.399 37.2 .399 50.8

.474 36.9 .474 43.9

.542 36.3 .452 35.6 l .613 35.8 .613 27.2 l .682 35.4 .682 19.7

.751 35.2 .751 13.8

.824 35.2 .824 9.8

.896 35.7 .896 8.2

.965 36.8 .965 9.4 l.0 36.8 1.0 9.8 l

s

-p i

40.0 I i T l

l I .

i

. l g

l Fusion Line I

g Base Line

'20.0 l l l

= l "B -

l i I l I l

2 g I l g 0. 0 - -

L l 4 1 6

I E i

.; . I  !

i l -

! l 2 1 -

-20.0. l l I 1 I

I i

I l

I I I I

-40.0  !

I '

0.00 0.10 0.20 0.30 0.40 DEPTH Z (in.) 1 l

I Figure A-1 Residual Stresses Parallel to. the Clad-Base Metal Interface as Measured at 70*F on Specimens H and I l

l I

i l

1 4

40.0 I i I l'

' I .

, l l '

I  ;

~

l

I l/ Base Line Fusion Line 4

i I L.,0 l , g l -

l O l l I 5 g, I l E

I l

~ I l

0.0 _ t*_ M i _ <

T. w -w a g '

$ ,1 I b I F

$ l w - 1 l

5 I i

! l -

-20.0 , I .

g I I i I

. l l

l l

l 1

-40.0 I l 0.00 d.10 0.2b d.3d 0.40 DEPTH Z (in.)

Figure A 2 Residual Stresses Parallel to the Clad Base MetalInterface as Measured at 200*F on Specimens E and F

40.0 _

i l

n' l

! - l

! I  !

I l Fusion Line l

Base Line l

20.0 . g i

l n I B

1 i d l l

I l g

c i I c I l 1 0.0 ~

g h '

f~I 1

l .

T ~

-  :: l l l l 2 I

-20.0

• I .

l l I I I

' l I

1 1

I I

-40.0 ' . . , ' '

0.00 0.10 0.20 O.3O o'.'40 DEPTH Z (in.)

Figure A-3 Residual Stresses Parallel to the Clad-Base Metal Interface as Measured at 400*F on Specimens B and C

e 4

'a ,

e

• ll t.e ne ptIges.e I" 8 i i I I I I I I I E8 i i I il 1 i i i

.i. . - -

.i. . _

- t ...t.

(lestk Analysis

,r % _ ,, ,----[-3-i,[.,,,tel _

ARet tilPF ter specimen Aner 110PF lasik Analysis

[.

j 41 0 Anneel.

eRet Anneel

- 41 0 Anneel l ter eRet5palmen Anneel - ,

_ x.

3

.N ,

/ -

x.

-\ /i l

i l

/

! 21 l

\ ,

s

/

ae - tselwsAnne. / -

f .0 ~

,/ I"" #

1 1 , pecore let .ei I Anneel \

/

m. _ i l

l

/ -

4 i

I F. 0 -

( -

F. 0 - -

8 00  !

-2. 0

- ~

~Q ~ ~

l Fusbn l Fusion d _ '

8

-!1. 0 I

-11.0 .

Second First 5enend Fird I toyer of -.l,. Layer of -. -tayer of -.g.-Layer of-al

-m0 - Cl**8'al l cleosing -: -me - cleesint l cm l ~~~I

-290 I ll t i I I i i 1 -no I I I i 1 1 I I I o fe 0 00 a l. 4 24 EN te te EM EM E T2 15 15 L e L t. 4 24 RN te te EM OM E Tt R S Distence from Surtue, r iIn. I Dhtence trem Surtece. a lla. I

a. Stresses Perellet le Directlen of note t. Stresses Transversele Dirstlen of woes Figure A-4 Residual Stresses in a Stainless Steel Clad Block, Before and After ll50*F Anneal (Measurements are the Average of Two Specimens)(7)

1

1. I i  ;

4 i I

N s . /b

, 'N '

. x' ss s

, s s.

N. -

,' i

~~

f (' s ,

,/ '1-

,'-e' ' - SURFACE FLAW  !

FOtJn POINT LOADEO DEND BAR DN. 9380A37 308 L Stainless 5 teel Cladding Notch,0.05 in. Wide, 60* Inc. Angle - 0, g) -

0.20 wwxxT axxxw

.. m .

v r 9

2.0 OA

!= 9.0 d

!= 2.5 =!

A533 Gr B Cl 1 '

l Base Plate ,

(Half Size) ( Full Size)

All Dimensions in Inches.

1.0 in. = 2.54 cm i

• l l

Figure A-5 Test Set-up and Specimen Geometry for Bend Bar Tests to Measure Clad Effects vs. Temperature l

l

e .

l 1

1 4

i a

2 l Curve 751375-A 0.7 i i , , ,

i No Break --

0. 6 1 0

/

/

Estimated

30.5 - _

$ /

_4 ,

Estimated /

s 0.4 / -

E I

j f-

__ 0. 3 - -

E tr 1 5 '

0 O. 2 - I -

I i

/ o A533 Gr B Cl1 0.1 -

o 308 L Cladding -

3 .

O I ' ' ' '

-50 0 50 100 150 200 250 i Temperature, 'F i

1 l

Figure A-6 Final Mouth Opening Displacement Versus Temperature for Clad and Unciad Bend Bars

's l' 1 .

I . -

t, .

i i

! l l

, 700.0 i

),

i l 1

4 600.0 _ ,

100 Secs 00 Secs l

500.0 - '

000 Secs '

t l 4

!s 400'.0

%u 300.0 _

. 3000 Secs 200.0 -

l 100.0 -

0.0 , i i 1 0.0 0.2 0.4 0.6 0.8 1.0 Figure A-7 Temperature History due to Large Steamline Break Clad Model-Nonle Corner, from Ref. 5, Figure 4-20