ML20205R706
| ML20205R706 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs  |
| Issue date: | 03/26/1987 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20205R051 | List: |
| References | |
| NUDOCS 8704060419 | |
| Download: ML20205R706 (9) | |
Text
__ . _ ____ - - _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .
[o uruy'o,, UNITED STATES 8 o NUCLEAR REGULATORY COMMISSION
$ ,I wAsHWGTON. D. C. 20666 Y
%, * * . 4 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION BALTIMORE GAS AND ELECTRIC COMPANY CALVERT CLIFFS NUCLEAR POWER PLANT, UNIT 1 DOCKET NO. 50-317 FLAW INDICATION DISCOVERED IN THE MAIN STEAM PIPING
1.0 BACKGROUND
As part of a licensee initiated program to monitor piping for erosion / corrosion effects, ultrasonic thickness measurements were taken on the No.12 Steam Generator Main Steam Line (EB-01-1005-05) at the second elbow downstream from the flow restrictor. Initial readings were taken on a grid spacing of 3 inches by 3 inches. This particular elbow is inside the containment at elevation 61 feet -0 inches. The inner pipe diameter is 34 inches, the wall thickness is a nominal 1 inch, the material is ASTM A-155, Grade KC70, Class I, and the design minimum wall thickness is 0.95 inches per ANSI B31.1-1967, the piping code in effect at the time of plant construction.
Downstream of the field girth butt weld joining the elbow to the hori-zontal pipe, reduced wall thickness readings were recorded on the under-side of the horizontal pipe. Wall thickness readings below 0.95 inches to as low as 0.86 inches were found immediately adjacent to the butt weld.
The thin wall pipe section can be characterized as a band 1/2-inch wide and 24 inches long adjacent to the butt weld. The length of this band is 22.5% of the inner pipe surface circumference. The size of the band was defined and confirmed during a second ultrasonic examination. The wall thickness for the remainder of the pipe examined ranged from 1.00 inch to 1.12 inches.
Five other locations on the main steam line were also ultrasonically tested to determine if similar conditions existed. Pipe wall thicknesses were found to meet the minimum wall thickness criteria. The licensee reviewed the radiograph of the affected pipe taken during plant construc-tion. The film showed a dark band in the same area as that indicated by the ultrasonic test examination. A new radiograph was taken. Examination of the films showed no significant differences between the old and new radiographs. The licensee believes this condition has existed since construction because the indication is essentially the same shape and size after more than 11 years of operation.
The probable cause of this thin wall section (less than that allowed by the code) was grinding of the edge of the pipe to achieve proper fit-up for welding during initial construction. Seamed piping normally is diffi-cult to fitup and occasionally grinding is required to provide a smooth, continuous surface between pipe sections. There is no evidence of erosion or corrosion.
8704060419 870326 PDR ADOCK 05000317 P PDR
- 2. 0 TECHNICAL REVIEW CONSIDERATIONS The approach taken is to treat the thin wall section of the pipe as an inservice flaw. The licensee is currently required by 10 CFR 50.55a(g)(4)(1) to meet the requirements of Section XI of the 1974 Edition of the ASME Code with Addenda through the Summer of 1975.Section XI, Article IWA-3000 " Standards for Examination Evaluations," requires this type of material and weld flaw (type C-F) be evaluated by comparison to the original construction requirements as provided in ANSI B31.1-1967.
This flaw does not meet the original tonstruction requirements of a minimum wall thickness of 0.95 inches.
Due to this flaw not meeting the IWA-3000 requirements of Section XI of the 1974 ASME Code, the licensee submitted letters dated December 16 and December 19, 1986, that requested the ASME Code requirements be updated for this flaw only to the requirements of the 1983 ASME Code with Addenda through Summer 1983.
The size of this flaw was found to exceed the normal acceptance standards of IWB-3514 of Section XI of the 1983 ASME Code. For the flaw to be deemed acceptable for continued service, the flaw analytical evaluation criteria of IWB-3600 is required to be met. This was not the case as the licensee identif.ied in these submittals.a Section XI 1983 ASME Code requirement, specifically IWB-3610(b), that could not be met and requested relief from this requirement for one operating cycle, that is, until the next refueling outage.
, 3.0 ASME CODE UPDATE AND RELIEF REQUESTED l The licensee requested relief from IWB-3610(b) of Section XI of the 1983 ASME Code. Supporting technical information was provided in Reference 1.
The staff reviewed this information as related to the design, geometry, and materials of construction of the pipe. The licensee's Inservice Inspection (ISI) Program is based on the requirements of ASME,Section XI, 1974 Edition including Addenda through Summer 1975, and remains in effect until the next 10-year ISI interval begins on April 1,1987.
In accordance with 10 CFR 50.55a(g)(4)(iv) the licensee, in a letter dated December 19, 1986 (Reference 2), requested use of the ASME Code Section XI, 1983 Edition through Summer 1983 Addenda for the evaluation of the pipe containing the flaw. The affected pipe is in the Number 12 Steam Generator Main Steam Line (EB-01-1005) at the second elbow downstream from the flow restrictor, and is adjacent to and downstream of ISI weld number 34-MS-1205-8.
l 3.1 Code Update Requested l l
The licensee requests to update the Section XI ASME Code requirements for this flaw only.from the 1974 Edition with Addenda through Summer 1975 to the 1983 Edition with Addenda through Summer 1983. 10 CFR 50.55a(g)(iv) permits portions of editions or addenda of the ASME Code to be used provided that all related requirements of the respective editions or addenda are met.
i l
l _ _ _ . .- . .- .. -- .
l The basis for this requested update in the Code requirements is that i the 1983 Edition provides superior analytical flaw evaluation -
techniques to those specified by the 1974 code. In addition, the licensee is required by 10 CFR 50.55a(g)(4)(ii) to update to the 1983 Code Edition for the second 10 year inservice inspection interval.
This interval commenced on April 1, 1987. Hence, during the majority of this operating cycle, the licensee would be required to comply with the 1983 Edition vice the 1974 Edition of Section XI of the ASME Code. Finally, the NRC staff has determined that in accordance with 10 CFR 50.55a(g)(iv) all related' requirements of the 1983 Edition of the ASME Code are met with the exception of the circumferential (hoop) stress limits of Article IWB-3610(b) of Section XI of the 1983 ASME Code. The licensee has requested relief from this requirement j for one operating cycle for this flaw only as compliance with this j hoop stress limit was deemed to be impractical. Therefore, the Commission agrees with the licensee's request of December 19, 1986 to update to the requirements of the 1983 Edition with Addenda through Summer 1983 of Section XI of the ASME Code for this flaw only.
3.2 1983 ASME Code Requirements Subparagraph IWB-3610 states that a flaw that exceeds the size of allowable flaws defined in IWB-3500 may be evaluated by analytical procedures such as described in Appendix A to calculate its growth until the next inspection or the end of service lifetime of the component. The component containing the flaw is acceptable for continued service during the evaluated time period if the following are satisfied:
(a) the criteria of IWB-3611 or IWB-3612; (b) the primary stress limits of NB-3000, assuming a local area reduction of the pressure retaining membrane that is equal to the area of the detected flaw (s) as determined by the flaw characterization rules of IWA-3000.
For IWB-3610(b) the primary stress limits are those of ANSI B31.1-1967, the piping code in effect at the time of plant construction.
- For IWB-3610(a) the licensee chose to meet IWB-3612 which allows the l following: A flaw exceeding the limits of IWB-3500 is acceptable if I the applied stress intensity factor and the flaw size a satisfy the f
following criteria:
(a) For normal conditions:
K y < Ky ,/ 410 where Ky = the maximum applieu stress intensity factor for normal (including upset and test) conditions for the flaw size a
f (defined in IWB-3611)
. l K,=
y the available fracture toughness based on crack arrest for the corresponding crack tip temperature (b) For en.ergency and faulted conditions:
Ky<K1c/ J2 where Ky = the maximum applied stress intensity factor for the flaw size a under emergency and faulted conditions f
bc=theavailablefracturetoughnessbasedonfracture initiation for the corresponding crack tip temperature 3.3 Code Relief Requested The licensee requests relief from the provisions of IWB-3610(b) for one operating cycle. Because the discovery of this indication occurred at the end of the outage, the licensee considers the immediate repair or replacement of this pipe impractical.
3.4 Licensee's Basis For Relief .
The licensee bases the request for relief on nondestructive exami-nation (NDE) results, a fracture mechanics evaluation, a limit load analysis, stress analysis and schedule. Relief is requested for only 3' one operating cycle. Before entering Mode 2 from the next refueling outage for Unit 1, the licensee will ensure that the affected piping will meet all the applicable criteria of ASME Section XI, 1983 Edition through Summer 1983 Addenda.
3.4.1 Nondestructive Examination Results The ultrasonic tests and radiographs support the contention that the indication is confined to the Number 12 Steam Generator Main Steam Line (EB-01-1005-05) and existed preservice. The probable cause of the indication was grinding for weld preparation.
- 3.4.2 Fracture Mechanics Evaluations The fracture mechanics evaluation was performed by Southwest Research Institute and submitted by the licensee for review by the staff as Attachment 8 to Reference 1.
Flaw growth was determined for a 2 year period.
Information was provided for the calcalation of applied stress intensity factors.
3.4.3 Stresses in the Pipe 1
Attachment 8 to Reference 1 presents a limit load analysis from which the minimum wall thickness required to prevent l
l
. . , . . _~ _
l 3
deformation is determined. An analysis of the st'resses in the pipe is contained in Attachments C and D to Reference 1.
This analysis was performed to determine whether the primary stress values in the thin wall pipe section met the original construction code, ANSI 831.1-1967.
4.0 STAFF EVALUATION The analytical evaluation was performed in accordance with the fracture mechanics criteria in ASME Code Sectibn XI IWB-3600. The fracture mechanics evaluation depends upon the size of the flaw, the location of the flaw, the amount of flaw growth, the fracture toughness of the material and the stress distribution at the flaw location during the limiting transient.
The transient conditions evaluated include the design transients and the corresponding number of design cycles as listed in Tables 1 and 2 i of Attachment B, Reference 1. Final flaw size is an accumulation of incremental amounts of growth due to all transients. Flaw growth rate is predicted by the following equations where da/dN is in inches / cycle and AKy is in ksi Jin.
da/dN = 3.795 x 10 10(AKy )
and Ky is given by Kg = (o, M ,4 ab Hb) J"*/9 E-where M M and Q depend upon crack geometry. Values of M,, Mb and Q are in AppeEdixbA of ASME Code Section XI.
The longitudinal stress is the appropriate stress for o in the fracture mechanics evaluation. Both the longitudinal stress and" hoop stress are normal to specific dimensions characterizing the size of the flaw. The longitudinal stress is normal to the 24 inch dimension and the hoop' stress is normal to the 1/2 inch dimension. The longitudinal stress controls due to the geometric shape of the flaw and the bending stress in the longitudinal direction. Thus, the flaw is treated as circumferential as opposed to axial. Also, there is uncertainty with regard to the.1/2 inch dimension as to the specific wall contour within the 1/2 inch band, that is, the flaw could be characterized as being substantially less than 1/2 inch.
The licensee's evaluation does not include a specific bending stress.
However, the value of M is approximately 20% greatIr(1.65) used in the calculation of KthanthevaluegivenbyF Appendix A of the ASME Code Section XI. The licensee uses a value of og (16.93 ksi) in the evaluation. This value of stress is nased on a pressure of 1693 psig, a value roughly 90% greater than the hot standby pressure of 900 psig. Using Attachment C, an estimate of the bending stress, as can be made. At most, the bending stress is 40% of the longitud 1 Mal stress. Thus, although the bending stress has not been l
L--.
l specifically identified in the evaluation, it is more than compensated for by the values of M,and o, used.
Based on the above information, a crack with an initial depth of 0.230 ;
inches (nominal pipe wall thickness of 1.08 inches minus the remaining metal thickness of 0.85 inches) could grow to a depth of 0.233 inches over the next 2 years. The staff agrees that crack growth over the next 2 years would be less than 0.003 inches.
Based on the above values, the calculhted value for the stress intensity factor K, is 24.5 ksi Jin. This value bounds the values of K computed g for normal and upset conditions which range between 15 and 22 ksi 41n and for emergency and faulted conditions which range between 15 and 23 ksi lin. If the value of 100 ksi din for fracture toughness of the pipe is taken as both Ky , and KIc, then Acceptance Conditions Criteria Calculated Normal and upset, K /K >3.16 4.08 Emergencyandfaultd8,k >1. .08 Ic I Thus, the fracture mechanics evaluation of ,the flaw in the pipe i demonstrates that it is acceptable because it meets the criteria specified in IWB-3612 of Section XI of the ASME Code.
A limit load analysis would yield the pipe wall thickness required to support either the associated longitudinal stresses or circumferential (hoop) stresses required to prevent pipe deformation. The limit load analysis is based on an actual material yield stress of 38,300 psi, a pipe wall thickness of 0.85 inches, and a pipe outer radius of 18.08 inches.
Reference 3 provides the following equation with regard to longitudinal ,
stress: l
-Load = (2/43) o n (R,2 - ( R, - tain) )
For a pressure of 850 psig (normal operating pressure) the actual wall :
thickness of 0.85 inches is more than five times greater than the hinimum wall thickness required. At a pressure of 1200 psig (emergency and faulted) the actual wall thickness of 0.85 inches is more than three times greater than the minimum wall thickness required.
With regard to circumferential stress, a limit load analysis is based on the following equation:
Dy , = P (R, - t,9 ,)/tmin For a pressure of 850 psig (normal operating pressure) the actual wall l thickness of 0.85 inches is more than twice the minimum wall thickness required. At a pressure of 1200 psig (emergency and faulted), the actual wall thickness of 0.85 inches is 55% more than the minimum wall thickness ,
required.
l l
r- - - - - - - -.,n,------ , - - - - --, -- .,n,. ,-v ------ --.-_a ,v---n, , . . - - , ~ ~ - , m- -
g ,. , , .
, The licensee has also performed an approximate analysis of the p~imary r l l stresses in the region of the defect, as required by ASME Code Section XI, 1983 Edition. The results were evaluated in accordance with the original construction code, ANSI B31.1-1967; the allowable for faulted condition I
! was based on ASME Section III Code Case 1606. '
! I i The analysis was based on the minimum measured well thickness assumed to i be uniform, and was performed for both the longitudinal and the circum-forential (hoop) directions. Since a uniform wall thickness was assumed, I
- the stress intensification factor duo'to the wall thinning was discounted.
L This stress intensification occurs as a result of the difference in )
i expansion (due to internal pressure and temperature) between the thinned i and normal wall regions of the pipe. This defect (actually a groove in the pipe wall) has the effect of inducing localized longitudinal bending stresses. In addition, a highly localized stress region also exists which i depends on the shape of the corner at the joint between the thinned and l normal pipe walls. l l
F The licensee has not performed an analysis to quantify these effects, other than to determine the pemissible operating pressure with the reduced wall thickness. The stress analysis does demonstrate that primary circumferential (hoop) stress limits are not met. IWB-3610(b) of Section ;
) XI of the ASME Code pequires the primary stress limits, which determine i the minimum wall thickness, be met.
- Also included was an excess-reinforcement argument to demonstrate that the l pipe has adequate margins to operate, at least until the next refueling outage, up to pressures of 900 psig. The excess reinforcement argument 4 was proposed on an ad-hoc basis to demonstrate that the structural
! integrity of the affected region would be preserved under the worst i expected loading conditions. It is based on the reinforcement methodology j as shown in ANSI B31.1-1967 for branch connections. The staff has reviewed q these calculations and has concluded that, within the present context, this approach is not acceptable, since the application of this methodology to a pipe wall with a shallow groove is not appropriate.
)
5.0 CONCLUSION
4
) Based on the material reviewed, the licensee is permitted for this flaw only to update the Section XI requirements of the ASME Code from the 1974 il Edition with Addenda through Summer 1975 to the 1983 Edition with Addenda j through Summer 1983 and is granted relief from IWB-3610(b) of Section XI l of the 1983 ASME Code until the next refueling outage pursuant to l 50.55a(g)(6)(1). With respect to the relief granted, the staff has 1 determined that the requirements of the code are impractical and that pursuant to 50.55a(g)(6)(1), the relief is authorized by law and will not endanger life or property or the common defense and security and is l otherwise in the public interest, giving due consideration to the burden i upon the licensee that could result if the requirements were imposed on
! the facility. The basis for the granting of relief is:
)
(1) The'affected pipe retains adequate fracture toughness and indication growth is negligible over one operating cycle.
l
(2) Under the worst condition a limit load analysis demonstrates the actual pipe thickness (0.85 inches) is at a minimum 50% greater than the minimum wall thickness required for collapse due to net section yielding.
- a. The longitudinal stresses are relatively low under all operating conditions, indicating there is adequate margin with respect to the yield stress. l 1
- b. There appears to be adequath margin with respect to the tensile instability (burst) pressure.
(3) The affected pipe will be replaced or repaired at the next refueling 1 outage.
(4) It is unlikely that the indication is service induced and it is an isolated case.
Date: March 26, 1987 Principal Contributor: Bob Wright t
e t
v y
REFERENCES
- 1. Letter and attachments from A. R. Thornton, Baltimore Gas and Electric Company, to Ashok C. Thadani, U.S. Nuclear Regulatory Commission, Accession Number 8612290169 861219, December 16, 1986.
- 2. Letter and attachment from A. R. Thornton, Baltimore Gas and Electric Company, to Ashok C. Thadani, U.S. Nuclear Regulatory Commission, Accession Number 8612190140 861216, December 19, 1986.
- 3. Kumar, V. , German, M. D. , and Shik, C. F. , "An Engineering Approach for Elastic-Plastic Fracture Analysis," EPRI Report NP-1931, July 1981, p.
4-6.
)
1 l
l i
l s
1
)