ML17059B542
| ML17059B542 | |
| Person / Time | |
|---|---|
| Site: | Nine Mile Point  |
| Issue date: | 05/08/1997 |
| From: | NRC (Affiliation Not Assigned) |
| To: | |
| Shared Package | |
| ML17059B543 | List: |
| References | |
| NUDOCS 9705160109 | |
| Download: ML17059B542 (72) | |
Text
gg'0 REJIG
+4 po I
~'i Cl0 C
0 4'/
0
~O
++*++
UNlTED STATES CLEAR REGULATORYCOMMIMON WASHINGTON, D.C. 005554001 AF U
ON Y T F
C 0
R GUA ROU CK 0
0-0 1.0 ~UO I.l Purpose, This safety evaluation (SE) assesses the reinspection results and analyses submitted by the Niagara Hohawk Power Corporation (NHPC and the licensee) for its core shroud at Nine Nile Point Nuclear Station, Unit I (NHP1) to determine whether the, core shroud meets the structural integrity requirements of the American Society of Hechanical Engineers Boiler and Pressure Vessel Code (ASHE Code),Section XI.
The NRC staff considers the core shroud repair by means of shroud stabilizer assemblies (also called tie rod assembl-ies) to be an alternative to an ASHE Code repair under 10 CFR 50.55a(a)(3)(i).
Since the assumption for the repair is that the circumferential welds are fully cracked, the vertical welds need a
minimum ligament to maintain the structural integrity of the shroud.
The NRC staff reviewed the analysis of the vertical welds to determine whether the results supported the alternative repair.
The alternative repair and Code requirements are discussed in Section 2.0 of this SE.
1.2 Background
By letters dated January 6 and January 23,
- 1995, NHPC submitted an application for repairs to the NHPI core shroud.
The shroud repairs and the use of stabilizer assemblies (tie rods) were submitted as an alternative,= as discussed above.
The NRC staff approved this repair in its SE dated Harch 31, 1995, according to which NHPC was also to address certain issues by the upcoming spring 1997 inspection (refueling outage 14).
NHPC committed to submit its plan for reinspection of the core shroud repair assemblies and the core shroud following refueling outage 13.
Also, as noted in the SE, the NRC staff recommended that NHPC perform certain activities to qualify the ultrasonic testing (UT) techniques used ta inspect weld H8 and to develop an effective method to locate the segment welds of the top guide support ring.
The NRC staff also noted in the SE that NHPC would reinspect all reported indications on the top side of the H8 weld during refueling outage 14.
The reinspection would verify the postulated crack growth of these indications.
During the post-installation inspection of the shroud repair, NHPC identified conditions that differed from the intended repair design and reported them in 9705i60i09 970508 PDR ADQCK 05000220 PDR
its letters of March 25 and April 30, 1995.
NHPC committed to take appropriate corrective actions during refueling outage 14 to restore the shroud stabilizers to their original design.
NMPC proposed the final design modifications in its letter of August 14, 1996, as part of the long-term corrective action plan.
In its SE of Harch 3, 1997, the NRC staff reviewed the anomalies in the installed shroud repair hardware and approved the proposed corrective actions.
By 1etter dated October 4, 1995, NHPC submitted its inspection plan for the core shroud and its repair assemblies.
The plan described the inspections projected for the core shroud stabilizer assemblies, the repair anchorage, the H8 weld, the top guide ring segment welds, and the vertical welds.
In its letter of February 7, 1997, NMPC confirmed its intent to conduct shroud inspections in the spring 1997 outage according to the inspection plan it previously submitted.
NMPC also presented additional information about the inspection plan, addressed the issues in the SE of March 31,
- 1995, and provided the fracture mechanics analysis that determined the next required ultrasonic inspection of the HS weld.
In its letter of February 28,
- 1997, NHPC updated the plans for inspecting the vertical welds described in the October 4, 1995, letter.
The inspection scope discussed in the letter was based on a draft version of the Boiling'Water Reactor Vessel and Internals Project-07 "Guidelines for Reinspection of BWR Core Shrouds,"
EPRI Report TR-105747 (BWRVIP-07).
NMPC stated that its current plans are in conformance with the BWRVIP-07 guidelines, with one exception related to the expansion criteria for inspection of vertical welds.
.NHPC also corrected statements made in its letter of February 7, 1997, about the scope of inspection of the H8 weld.
On March 3, 1997, the NRC staff issued its SE of the proposed inspection program.
It found that NMPC addressed the issues noted in the SE of March 31, 1995.
It also found NHPC's plans acceptable, but stated that the inspections for this outage should be conducted in accordance with the BWRVIP-07
~
guidelines without the exception.
Although the NRC staff has not completed its review of these guidelines, it accepted their use for this NHPl refueling outage.
NHPC inspected core shroud vertical welds in accordance with the BWRVIP-07 guidelines.
On Harch 20 and April 2, 1997, it notified the NRC staff (by telephone) of the. inspection findings.
NHPC found vertical weld cracking that exceeded the screening criteria of BWRVIP-07.
Additionally, its inspections of the four tie rod assemblies showed that'the tie rod nuts had lost some preload (clamping force) and that the lower wedge retainer clips on three tie rods were damaged.
In its submittal of April 8, l997, supplemented by letters of April 25 (two letters),
26, and 27 (two letters),
and 30, 1997, NHPC reported the findings of its inspections.
It presented a root cause analysis, corrective actions, design documentation to establish the acceptability of.the vertical weld
- cracking, a weld reinspection
- schedule, and details of the actions taken to restore the tie rods to the as-designed condition. and to modify the lower wedge retainer clip design.
\\
,r
2.0 URHN In accordance with ASHE Code Section XI, 1983 edition, Summer 1983 Addenda, NHPC chose to perform corrective measures for the cracked circumferential shroud welds on a pre-emptive basis.
The repair for cracked welds to satisfy the repair rule of Section XI, IMA-4000 would be to remove the defect and conduct a repair by welding.
This is impractical since the core shroud is irradiated to an extent that might. cause the shroud to crack further ifit were welded.
Therefore, as an alternative to IMA-4000, NHPC chose to repair the core shroud with mechanical tie-rod assemblies that serve as a replacement for the circumferential welds.
The NRC staff has approved the tie-rod assembly as an acceptable alternative repair in an SER dated March 31, 1995.
The current request, which is the subject of this SE, is to modify a component of the tie-rod assembly (i.e.,
the lower wedge retainer clip) pursuant to 10 CFR 50.55a(a)(3)(i) and institute a revised installation procedure to ensure that pre-load of the tie-rod assemblies will not be lost by unanticipated motion of the lower toggle bolt supports.
The design assumptions for the alternative repair are that circumferential, welds are cracked through-wall for 360'nd that the tie-rod assemblies will accommodate the vertical and lateral loads for the "stacked cylinders."
The cylindrical sections are fabricated from rolled plates.
Two vertical welds are used to join the sections, completing the cylinder.
Therefore, vertical weld structural integrity is needed to ensure that the cylinders will remain as cylinders.
The crack growth rate used as a part of the fracture mechanics analysis for assessing the needed weld integrity is considered bounding if the plant water chemistry is maintained in accordance with the EPRI BMR Water Chemistry guidelines.
However, the NHP-1 current TS is not consistent with the BQR Mater Chemistry guidelines.
NHPC has committed to submit an application for a license amendment within 60 days to address this matter.
The NRC staff has reviewed the modification and the fracture mechanics analysis that ensure the structural integrity of'the vertical welds for the requested operating period of 10,600 hot operating hours and finds that the alternative repair is acceptable pursuant to 10 CFR 50.55a(a)(3)(i).
This approval is predicated on the condition that NHP-1 is operated in accordance with the BWR Mater Chemistry guidelines, Electric Power Research Institute technical report TR-103515,
'BMR Mater Chemistry guidelines-1996 Revision."
'I Oetails of the review are discussed in the following sections.
3.0
~.Figures 1-1 and 1-2 show the location of the shroud welds.'able 2-2 shows the findings for vertical weld inspections.
NHPC did not present a table of findings for the horizontal welds, but provided examination indication maps
'ilfigures and tabies attached to this NC staff SE are extracted, srithout change, frcl reports aubnitted by INPC.
Therefore, figure and table raahers do not necessarily correspond to the section taabering within this SE.
~
~
I
4 and discussed the results in a qualitative way in Appendix C of the report by General Electric Nuclear Energy (GENE), titled "Assessment of the Vertical Meld Cracking on the NHP1 Shroud,'ENE-523-B13-01869-043
.(portions attached)..
The scope,
- methods, and findings of the inspections are discussed hereinafter in Sections 3.1, 3.2, and 3.3, respectively.
3.1 Inspection Scope As discussed in Section 1.0, NHPC stated that its current plans were in accordance with the BWRVIP-07 guidelines, but with one exception.
The exception concerned the inspection of vertical welds, For the vertical weld inspections scheduled for the spring 1997 outage, NMPC proposed to modify the aethod for sample expansion presented in BWRVIP-07 as Option B.
According to Option B, if the cumulative cracking in either the original sample or the expansion sample exceeds 10 percent of the equivalent length of weld inspected, the inspection scope is to be expanded to verify the minimum required uncracked length for each vertical weld that is not structurally replaced by existing hardware or the repair or both.
NHPC proposed to expand the inspection scope to verify the minimum required uncracked length for each vertical weld not structurally replaced by existing hardware or the repair or both=only after finding the cumulative. cracking in either the 50-percent expansion sample or the 100-percent expansion sample greater than 10 percent of the equivalent length of weld inspected.
'I The NRC staff did not grant the exception, finding that NHPC did not present adequate technical bases to deviate from the BWRVIP-07 guidelines.
The NRC staff's judgment was that the provisions of Option B should be followed and
- hence, a finding of more than 10 percent of the equivalent length of weld inspected being cracked would merit additional inspections to further qualify the degree of cracking.
NHPC inspected the NHPl vertical welds according to the sampling option of BWRYIP-07.
This option specified a visual inspection of 25 percent of the equivalent total vertical weld length from either the outside diameter (OD) or inside diameter (IO).
The ring segment welds (labeled in Figure 1-2 as V5 and V6) were excluded from the vertical welds requiring inspection based on a GE analysis of the ring segment welds submitted to the NRC staff for review by letter dated February 7, 1997.
The initial inspection of the vertical welds with enhanced visual techniques found cracking over the entire OD length of the V10 weld.
NHPC then expanded the inspection plans to establish the minimum required uncracked ligaments on the vertical welds that are required to meet the shroud stabilizer repair design-basis assumptions.
It performed the inspection using an enhanced visual inspection (EVT) method supplemented by UT.
NHPC found extensive cracking on the OD of vertical welds V9 and V10.
As a
- result, NHPC performed a complete baseline inspection of the accessible parts t
1 I
l
- ~ of most horizontal and vertical welds to assess the overall material condition of the NHPl core shroud.
NHPC examined the accessible areas of the ring but was unable to locate the welds V5 and V6.
NHPC inspected the horizontal welds H2, H4, H5, H6a,
- H6b, arid H7, even though inspection of horizontal welds was not required because a preemptive repair had been installed; Such repairs are designed assuming the circumferential welds are fully cracked (through-wall for 360 degrees),
obviating the need to inspect them.
The inspection scope is sunmarized in Table 2-2 and Appendix C.
The scope was sufficiently comprehensive to identify degradation that could invalidate design assumptions for the tie rod repairs.
In expanding the scope of inspections, and inspecting from the ID and the OD of the weld, NHPC determined whether each vertical weld contained the minimum ligament needed for maintaining the structural integrity of the shroud.
The relation of the results of vertical weld inspections to the design assumptions for the tie rod repairs is discussed previously in Section 2.0'f this SE.
3.2 Inspection Methods Table 2-2 and Appendix C indicate the inspection method used for the various welds.
EVTs are performed with a video camera qualified in tests to detect a
0.5-mil wire. 'isual inspections cannot determine the depth of a flaw.
They are used to detect surface flaws on either the ID or OD.
Ultrasonic inspections are volumetric examinations. used to detect flaws and determine their size.
Accessibility of the weld determines which method is used.
Table 2-2 shows that each of, the vertical welds was examined from the OD and ID either by EVT dr UT or both.
Supplemental eddy current inspections were also used to determine if defects were present on the ring segments containing welds VS and V6.
These welds are difficult to locate by UT or visually because the support ring was machined after fabrication and welding.
The UT for examining the core shroud at NHPl was performed with a GE delivery device, the suction cup scanner.
Three transducers were used including a 45 shear, 60'ongitudinal, and an OD creeping wave.
The inspection device was demonstrated and qualified to detect and determine the size of cracks on a core shroud mock-up at the EPRI Nondestructive Examination (NDE) Center.
This device has been used to perform core shroud inspections at a number of other BQRs.
The crack size (length and depth) measurement uncertainties for this device were determined in accordance with guidelines in the "Reactor Pressure Vessel and Internals Examinations Guidelines (BWRVIP-03)" (EPRI TR-105696), October 1995.
In measuring length, uncertainties arise from the delivery system and the NDE technique.
The qualification of the system at the EPRI NDE Center included quantification 'of measurement uncertainties.
The uncertainty factors from the delivery system and the NDE technique were determined to be 1.106 inches and 0.364 inches, respectively.
The combined uncertainty factor of 2.94 inches t(1.106 inch + 0.364 inch) x (2 ends)]
was applied to each ultrasonically measured flaw length including the calculated critical crack length.
The 60'ongitudinal wave transducer was used in depth measurement.
The uncertainty factor associated with the depth measurement was determined to
I P
I l
be 0.108 inch (no contribution from delivery system),
and was applied to each UT depth measurement.
The EVT performed for the core shroud inspections was qualified by demonstrating the capability of resolving a 0.5-mil-diameter wire.
The uncertainty factor for the length measurement by the EVT was determined to be 1.2 inch.
To account for the measurement uncertainty, 2.4 inches (for two ends) were added to each visually measured flaw length.
3.3 Inspect)on Findings NHPC presented detailed descriptions and examination indication maps of the vertical weld cracking.
The results are summarized in Table 2-2 and are included in Appendix C (attached).
The inspections showed significant cracking on welds V4, V9, and V10.
Hinor cracking was found on welds V3, V12, V15, and V16.
No cracking was found on the accessible areas of welds V7, V8, and Vll.
Melds V5 and V6 were not located.
The weld with the worst cracking.was V9. It had numerous indications on more than 90 percent of the weld and up to 80 percent of the depth.
Meld V10 had numerous indic'ations on more than 80 percent of the weld.
NHPC tried to examine the ring segment welds V5 and V6, welds known to be difficult to examine given current technology.
Using eddy current and visual inspections, NHPC could not locate the specific welds.
- However, EVT of the accessible outer.surface of the ring segment did not find any cracking.
The NRC staff approved NHPC's inspection plans for these welds in its SE of Harch 3, 1997.
Those plans contained an analysis that showed that if NHPC could not locate these welds, the shroud repair would remain effective even if the welds were significantly cracked.
NHPC's evaluation showed that the structu'ral integrity of the ring would be maintained even with cracking up to 95 percent of the segment welds.
However, the tie rod repair may be less effective if ring segment welds are totally cracked and the two neighboring horizontal welds, H2 and H3, are 'also completely cracked.
Under this condition, the'stiffness of the core shroud may change and, consequently, affect the amount of the thermal preload applied by the tie rod assemblies during normal operation.
The inspections verified that sufficient intact weld length exists in the H3 weld to support operation, even if the ring segment welds were significantly cracked.
NHPC submitted indication maps of the horizontal welds and a qualitative description rather than a table of quantitative results.
The installed stabilizer assemblies were designed to ensure shroud integrity even if all the horizontal shroud welds were cracked through the wall around the full circumference of the shroud.
Thus, the condition of the horizontal shroud
.welds did not need a quantitative assessment to support continued operation.
The findings of the inspections
- are, however, included in Appendix C
(attached).
The structural significance of the indications is assessed in Section 4.2 and the potential for bypass leakage in Section 4.3.
k I
>>I
4.0 UA ON F
SP ON 4.1 Root Cause GS NHPC determined that the cracking in the vertical welds was caused by intergranular stress corrosion cracking (IGSCC).
NHPC's findings agree with research and operating'xperience with this phenomenon.
It is well known that the core shrouds of all boiling water reactors (BWRs) are susceptible to IGSCC.
The following relevant factors affect the cracking:
operating time, coolant conductivity, material carbon content, plate orientation, fabrication-related surface cold work,. neutron fluence, residual stresses resulting from welding and fabrication, and operating stresses.
The NHPl shroud is susceptible to IGSCC because it is made of high-carbon Type 304 stainless steel that is sensitized by welding and subjected to residual stresses from welding and fabrication.
NHP1 is classified by industry guidelines as a Category C plant, a category that. contains the most susceptible core shrouds.
Category C plants have, core shrouds made of Type 304 stainless steel and more than 6 hot operating years, regardless of reactor coolant conductivity.
The location of the cracking in the NHP1 shroud is consistent with IGSCC as shown by inspection data.
The cracking was located at heat-affected zones (HAZs) sensitized by welding and in areas with residual stresses in which
'fabrication-related welding or grinding +as apparent.
In a few limited instances, cracking was found to extend up to 1.5 inches away from the weld perpendicular to the longitudinal axis of the weld.
A few crack indications not associated with HAZs were found in the base metal.
.However, these cracks initiated from the cold-worked areas resulting from grinding to remove the attachment welds (lugs) during construction.
~
'he cracks are expected to stay confined to the HAZs or in the severely cold-worked area.
Radiation, when exceeding a threshold value, can sensitize aaterials, making them susceptible to IGSCC.
However, offsetting this effect, is the reduction of tensile residual stresses outside the affected areas.
Unless high tensile stresses, such as those that result from grinding or cold work are present, there is not enough driving force to propagate a crack beyond the affected areas.
The depth of the 'cold-worked layer in the base metal resulting from grinding is generally very shallow and, therefore, the cracking in the cold-worked areas will also not grow very deep.
NHPC contended that although some cracks extended beyond the HAZ, they did not differ from those seen in other BWR plants in the past.
To check the contention, the NRC staff reviewed inspection data for circumferential welds, considering data for the H5 weld at Brunswick Unit 1 as an example.
The inspection report for that plant of June 21, )995 (Brunswick Steam Electric Plant No. 1, Docket No. 50-325/License No. OPR-71 NRC Generic Letter 94-03,
'Intergranular Stress Corrosion Cracking of Core Shrouds in Boiling Mater Reactors," letter dated June 21, 1995, from R. Anderson of Carolina Power and Light Co. to the NRC) for the 10th refueling outage states that H5 was visually examined 100 percent of the ID and 45 percent of the OD.
Five of the cracks on the OD were circumferential, 0.5 to 3 inches long.
The remaining
l
>~
I I
cracks were axial, extending less than 9 inches from the toe of the weld.
Long circumferential cracks and short axial cracks, less than 4.5 inches long, were on the ID.
Saseline depth measurements were taken by UT in 'two locations to benchmhrk future inspections.
The depth ranged, from less than Oe3 inches to Oe6 inches.
That inspection report further states that during the 10th refueling outage,
'punch marked cracks" on the ID were visually reinspected for length.
In addition, two areas, inspected by UT during the 9th refueling outage were ultrasonically reinspected to determine crack growth and were compared to previous sizing data.
The report concluded that there was no change in depth or length.
This conclusion supports the contention that cracks are not likely to propagate outside the cold-worked area and that the crack growth rate is slower than the NRC's bounding rate of 5 x'10 s inch per hour.
The NRC staff consider'ed the question of the adequacy of base metal inspections.
HAZs are typically narrow and adjacent to the weld.
NMPC reported that the HAZs associated with the vertical welds near the beltline extended about 0.8 to 1.0 inches from the weld fusion liqe.
NNPC visually examined about 2.5 inches of base metal on each side of each vertical weld.
Considering the extent of the examination together with the fact that the driving force for cracks is greatly reduced outside the HAZ, the NRC staff concluded that NHPC examined enough area outside the HAZ to determine the condition of the base, metal.
The NRC staff considered'the effect of radiation fluence on the cracking observed.
Irradiation increases the susceptibility to IGSCC, and the fluence was in a range for which irradiation-assisted stress corrosion cracking (IASCC) conditions can occur.
The qstimated fluence at welds V9 and V10 was in the range of 2 to 4.5 x 10 n/cm
(>I HEV).
NHPC will check whether IASCC was a contributing mechanism to the cracking by analyzing boat samples of cracked material.
This work will be completed and the results will be submitted to the NRC for review.
This issue does not need to be resolved immediately because the crack growth rate used takes into account the effect of fluence.
Crack growth rates are evaluated in Section 4.2.
v The NRC staff finds that NNPC's analyses reasonably explain the observations; they showed that the welding and fabrication processes caused the cracking pattern seen on the vertical welds.
They also showed that the pattern of crack depth is consistent with the calculated fluence axial and radial profiles.
That is, both welds V9 and V10, the welds with the wors't cracks, are between welds H4 and H5 and are exposed to neutron irradiation along their entire length.
The axial cracks in weld V9 were found to be the deepest near the H4 weld where the fast neutron flux is highest.
4.2 Structural Integrity Assessment The methodology for assessing the structural integrity of a flawed core shroud axial weld, is broken down into a number of steps.
First; the size of the
. flaw at the time of the analysis (the initial flaw size, a ) is determined by u1trasonic or visual inspection of the weld,and this flaw (s conservative1y assumed to be present completely through the thickness of the core shroud
I I
I
wall.
This value, arepresents the best estimate of the flaw size.
- However, some uncertainty exists in a, because of the method used to determine the flaw size.
When the uncertainty due to the inspection methodology (discussed in detail in Section 2.0) is added to the best estimate of the initial flaw size, a new value, a, is determined.
This value, ao, is taken as the initial flaw size in the structural analysis since it represents a
bounding value for the size of the existing flaw.
Then a crack growth rate (da/dt) is determined.
For this analysis, a bounding value of the IGSCC growth rate was assumed (5 x 10 s inch per hour) and multiplied by the amount of operating time to be justified (t) to determine a bounding value for the flaw size at the end of the proposed operating time (the final flaw size, a<).
It is this. final flaw size, a that must.be shown to be less than some critical flaw size determined by prescribed loading conditions (with appropriate safety factors) according to structural integrity assessment methodologies.
The flaw sizes based on reported measured values and the uncertainties associated with these measurements are discussed in Section 3.3.
Section 4.1 discusses assumed crack growth rates.
Section 4.2 discusses structural integrity assessment methodologies.
Section 4.3 summarizes NMPC's structural integrity assessment, and Section 4.4 discusses the NRC staff's independent analyses.
4.2.1 Evaluation of Crack Growth Rates GE, in its analysis for NMPC, assumed
- a. bounding crack growth rate of 5 x 10 inch per hour.
This crack growth rate bounds laboratory test data for a variety of water chemistries, bounds field experience with IGSCC of other components of the reactor coolant system,'and has been used by the NRC for flaw growth evaluations.
Even after considering potential IASCC contributions, the actual crack growth is expected to be lower than this bounding value.
GE predicted the crack growth rate for the vertical welds using several methods.
It considered the effects of radiation on crack growth rates, comparing predictions of rates for unirradiated material to those for irradiated materials at the same values of reactor water conductivity, electrochemical potential (ECP),
and initial sensitization.
The comparison showed that at the fluence levels at BWR shrouds, the predicted crack growth rate is similar for both the unirradiated and the irradiated materials.
GE attributed the similarity in rates to offsetting effects.
Although irradiation increases the susceptibility of the material, it also relaxes the weld residual
- stresses, the driving force for crack growth.
Ultimately, GE concluded that the crack growth rate over the range of shroud finances is bounded by the NRC bounding crack growth rate of 5 x IO inch per hour.
The
'NRC staff agrees with GE's determination.
The NRC staff considered the effect of the distribution of residual stresses on the crack growth rate.
Similar results to those reported by MPM Technologies in its 'Analysis of Nine Mile Point Unit I Shroud Welds V9 and Weld V10 Cracking," Report No. HPM-497439, April l997, were obtained by researchers performing NRC-sponsored work at Argonne National Laboratory
I I
I
I
(
I (ANL).
Researchers at ANL analyzed the distribution of residual stresses from welding in shroud welds.
The work at ANL, unlike that of MPH, did not include residual stresses induced by fabrication, but did find residual stresses from welding to be highly tensile at the surfaces and compressive in the center of the wall of the cylinder.
Additional observations include that actual conditions like "out of roundness" would bias cracking to a particular surface.
Further, if stresses were uniform across the thickness, cracks would grow at about a 2:I aspect ratio.
Even assuming multiple crack initiation sites, the aspect ratios found in the cracking at NMP1 imply that the stresses decrease sufficiently through the wall to reduce the stress intensity.
The decease in stress intensity is clearly more than sufficient to show that 5 x 10 inch per hour is a conservative rate of crack growth.
The NRC staff examined the question as to whether crack growth rates in vertical welds differ from the rate for horizontal welds.
The rates are expected to be similar for the following reasons:
The cracking on both the horizontal and vertical welds is caused by the same mechanism, IGSCC, in analogous locations, that is in weld HAZs, which are areas of high residual stress.
The residual stress patterns are similar for both kinds of welds.
They, are tensile on the outer surfaces and decrease, even become compressive, within the wall and with distance away from the weld.
At NMPl, no field data on crack growth are available for vertical welds.
Of the horizontal welds, field data are available only for the HB weld.
During the current outage, NMPC located a flaw found by UT during the previous (13th) refueling outage as well as an additional flaw in the same area.
Through UT, NHPC determined that the flaw was of legs through-wall depth than in the 13th refueling outage.
NHPC concluded, after reviewing the previous data, that the earlier sizing was very conservative.
NHPC drew no conclusions from these data about the crack growth rate at NMPl,.and the bounding rate of 5 x 10 inch per 'hour was used in NHPC's analyses.
The NRC staff examined the question of whether the water chemistry at NHPI conforms to the requirements under which the bounding rate is applicab'le.
This rate was derived from laboratory test data.
The reference for the data is a letter from,NHPC to NRC, "Responses to NRC Staff guestions Provided During Telephone Conversations of April 22,
- 1997, on Core Shroud Cracks and Repair," dated April 27, 1997.
The crack growth tests for sensitized stainless steel were performed at a range of conductiyities from 0.3 to 1.5 microSiemen/cm.
The data from these tests are. bounded by the crack growth rate of 5 x IO s inch per hour.
In its letter of April 30,
- 1997, NHPC committed to continue operating. in accordance with the EPRI Water Chemistry Guidelines.
These guidelines are also incorporated into BWRVIP-07; that document assumes that plants meet the requirements specified by the EPRI Water Chemistry Guidelines Action Level l.
According to Action Level I, if the conductivity exceeds the limit of 0.3 microSiemen/cm, the conductivity must be reduced to that value or lower within 96 hours0.00111 days <br />0.0267 hours <br />1.587302e-4 weeks <br />3.6528e-5 months <br />.
Although the conductivity limit in NMPl Technical Spe'cification 3.2.3 is 5 microSiemen/cm (when steaming rate are equal to or greater than 100,000 pounds per hour),
NHP1 has been operating at a much lower
~ conductivity, less than 0.10 microSiemen/cm, during the past three cycles and less than 0.3 microSiemen/cm during the last seven cycles.
Therefore, the
I
~
p bounding crack growth rate applies to NHPI because this rate is conservytive with respect to Action. Level I in the EPRI Guidelines.
By letter dated Hay 7,
- 1997, NHPC has committed to submit a license amendment within 60 days that will address the difference between the current reactor. coolant system chemistry requirements of NHPl TS 3.2.3 and the coolant chemistry criteria referenced for core shroud crack growth rates as described in NHPC's letter of April 8, 1997.
The NRC staff finds this commitment acceptable.
4.2.2 Structural Integrity Assessment by Linear Elastic Fracture Hechanics and Limit Load Analysis Hethodologies Either of two analysis methodologies, limit load analysis (LLA) or linear elastic fracture mechanics (LEFH), may be applied to demonstrate the structural integrity of flaws in ductile materials such as Type 304 stainless steel.
Either of these methods seeks to determine the length of flaw that can be tolerated (or, conversely, the amount of uncracked ligament that, must remain) for the location to support the applied stresses under prescribed loading conditions.
These loading conditions are divided for the purpose of this analysis (and by the ASHE Code) into two general categories:
(I) normal operation or upset conditions and (2) emergency or faulted conditions.
The significance of this is that the ASHE Code analysis procedures require that a
safety factor of 3.0 be applied to the loads determined for normal operation or upset conditions, while a safety factor of 1.5 is to be applied for emergency or faulted conditions.
The first methodology, LLA, determines the amount of uncracked ligament that must remain to keep the location from being pulled apart by full-section plastic yielding.
An LLA is usually applicable to materials that retain a
high level of fracture toughness and demo'nstrate'o crack extension under loadings up to those that cause plastic yielding.
The results of this method are then compared to the results of an LEFH analysis to determine which method controls the evaluation (i.e., gives the smallest acceptable flaw size).
LEFH is a conservative method for assessing the potential for crack growth in a ductile material.
LEFH assumes that the material will behave in a nonductile fashion (which reduces its flaw tolerance) and compares the stress intensity due to the,flaw and the applied loading to the material's fracture toughness (a quantitative measure of the material's flaw tolerance).
LEFH would normally be expected to be appropriate for materials that.have lost ductility, for example, from irradiation damage.
4.2.3 'HPC's Evaluation of Structural Integrity GE performed the analysis that determined the structural margin of the vertical welds.
The assumptions for the shroud repair are that horizontal welds are cracked through-wall.
Under this assumption, vertical weld cracks are acceptable, as long as the. crack lengths are less than the allowable flaw sizes or as long as the structural integrity of the vertical weld can be shown.
Typically, the allowable crack sizes are large and approach or exceed the length of the weld itself.
J I
I The analysis determined that the pr'imary stress that could cause vertical weld failure would result from the internal pressure.
Consistent with ASIDE Code practice (Appendix C,Section XI), the analysis considered internal pressure as the only load for axial cracks.
It considered the internal pressures under all conditions normal, upset, and accident events with the appropriate ASIDE Code safety factors.
The analysis did not take credit for the horizontal weld integrity in determining the allowable vertical weld flaw sizes.
The horizontal welds are assumed to be cracked completely through the wall, and the cylinder between any two horizontal welds is assumed to be "stand-alone."
The calculations also assumed simultaneous cracks at.the diametrically opposite welds in a given cylinder.
The analysis applied both LEFH and limit load to determine the allowable flaw sizes.
Calculations for the required minimum ligament allowed for crack growth and inspection uncertainty.
In the analysis, GE first determined whether the vertical welds met the screening criteria'pecified in BWRVIP-07.
GE then performed a more detailed fracture mechanics analysis to demonstrate the integrity of vertical welds with indications that did not conform to the. screening criteria.
Allowable shroud vertical crack lengths were calculated on the basis of both LEFH and LLA.
The high fluence region in which there is potential irradiation embrittlement is limited to the shroud section between horizontal welds H3 and H6a.
Therefore, both LEFH and LLA were.used for vertical welds in this region.
LEFH was found to be governing -for welds V9, V10, Vll, and V12, and LLA was governing for welds V7 and VS.
The vertical welds in other regions were governed by LLA.
Table 5-2, Columns 3 and 4, (see enclosure to this SE) shows the allowable crack length as well as the required uncracked ligament lengths for each vertical weld.
In accordance with BWRVIP-07, allowances for crack growth and uncertainty in the inspection were added to the required uncracked ligaments (Column 5 of Table 5-2).
Cracking of radial ring welds was separately evaluated and found to have negligible impact.
The only effect of radial ring weld cracking is on the
, thermal preload, but the analysis found that even with 90 percent of the weld assumed to be cracked, the effect on preload was insignificant.
Column 6 in Table 5-2 lists the values of uncracked ligament lengths determined from the UT and EVT of the various shroud vertical welds.
A comparison of these lengths with those required by the conservative screening criteria lengths (Column'5) shows that each of the vertical welds except for V4, V9, and V10 meets the required uncracked ligament length criteria.
'The analysis showed that each weld except for V4, V9, and V10 is acceptable for continued operation for at least a fuel cycle of 16,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />.
These three welds were analyzed further.
In this more refined analysis, credit was taken for uncracked ligaments'for part-through cracks after accounting for crack growth.
~
~
Of the three welds, V9 was determined to be the most limiting case and was, therefore, evaluated first.
The analysis applied the required safety margins consistent with the ASHE Code Section XI criteria for piping.
The analysis justified continued operation for at least 10,600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> (approximately 14-1/2 months).
Meld V4 was evaluated by LLA because it is located above the top guide support ring.where the fluence is low enough that the material is not embrittled.
Table 5-2 shows the required uncracked ligament.
The analysis showed that the limit load margins are satisfied for the V4 weld for a period of at least 10,600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br />.
The structural analysis concluded that continued operation at NMP1 is justified for an additional 10,600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> (about 14-1/2 months).
The allowable hours of operation ensure that the required ASHE Code margins are maintained, considering both LEFM and LLA.
The NRC staff finds this conclusion acceptable.
The analysis was performed according to methodology approved by the NRC staff.
The methodology is detailed in the General Electric Nuclear Energy document GENE-523-113-0894, Revision 1,
'BWR Core Shroud Inspection and Flaw Evaluation Guidelines,"
March 1995, and has been used for circumferential welds.
It is incorporated by reference in,BWRVIP-07.
4.2.4 Independent NRC Staff Structural Integrity Analyses The NRC staff has performed an independent, quantitative assessment of the largest axial flaw in the NHPl core shroud.
The NRC staff evaluated this flaw using both LLA and LEFH techniques.
The NRC staff found that the LEFM
- analysis, presented below, controls the evaluation.
The NRC staff accepted NHPC's best estimate for the largest initial flaw length of approximately 74 inches.
The NRC staff accounted for the uncertainty in the sizing of the initial flaw by adding the uncertainty factor in GE's ultrasonic testing method (1.5 inch) to each end of the best-estimate flaw.
Therefore, the NRC staff used a value of 77 inches for ao, the bounding initial flaw size.
The NRC staff then applied the bounding crack growth rate (5 x 10 s. inch per hour) for 10,600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> (approximately 14-1/2 months) to determine a bounding final crack length.
The NRC staff then applied the stresses that were reported to result from faulted loading conditions with the safety factor of'.5 (as these were reported to be the most challenging loading conditions) and used an analytical equation that contained a small correction factor for long cracks that was not included in GE's LEFM analysis.
This resulted in a calculated stress intensity of about 175 ksi4in at the tip of the bounding final crack length.
The NRC staff compared the result with the fracture toughness of 150 ksi/in reported by GE from test results for material removed from a foreign reactor.
Further, the NRC staff has determined that additional test data for weld material from another foreign nuclear power plant indicate similar fracture
. toughness to that discussed.
In addition, an % astic/plastic fracture analysis was performed by Structural Integrity Associates with the fracture toughness of the base metal reduced by about a factor of 3 to bound weld metal properties.
The analysis by Structural Integrity Associates produced results
~
~
C that were similar to the LEFH analysis performed by GE.
On the basis of the information reported by NMPC and the NRC staff's analysis, the bounding final crack length for the limiting core shroud flaw is slightly larger than would be acceptable under LEFH analysis.
This independent evaluation by the NRC staff used a handbook solution that is not as precise as NMPC's finite element analysis.
The results from the analysis is that the assumed material toughness is slightly exceeded.
Considering the test data from the two foreign nuclear'ower plants yielding similar values for fracture toughness, the results of the elastic/plastic analysis performed by Structural Integrity
'ssociates and the uncracked ligaments for which credit was not taken in NMPC's analysis, the NRC staff concludes that the more refined analysis performed by NMPC appears reasonable.
Finally, the NRC staff confirmed that the LEFH analysis was controlling by determining the acceptable amount of remaining ligament according to LLA to be 7.0 inches, mu'ch less than that required by %he LEFH analysis.
Furthermore, evidence exists that the materials removed from shrouds that have been tested are failing in a ductile manner, indicating that LLA or elastic/plastic analysis may be more appropriate analysis methods.
4.3 Assessment of Potential for Bypass Leakage NMPC evaluated the potential crack opening and associated leakage for the V-9 and V-10 vertical welds, as documented in General Electric Nuclear Energy Report GENE-B13-01869-043, Revision 0, "Assessment of the Vertical Meld Cracking on the NMP1 Shroud," April 1997.
The evaluation also considered whether the leakage would be detectable:
The NRC staff's review of NMPC's leakage evaluation is presented below.
4.3.1 Leakage Estimate By conservatively assuming that both the V-9 and V-10 welds were cracked through-wall, NMPC estimated that the postulated vertical weld cracks would provide 3-square inches of leakage flow area (approximately a 0.003-inch opening a1ong the entire length of the welds).
On the basis of this leakage flow area, the total calculated leakage from the V-9 and V-IO welds was estimated to be 200 gpm at 100-percent rated power and 100-percent rated flow.
This statistic is equivalent to approximately 0.11 percent of the core mass flow.
In a letter dated Harch 31, 1995, the NRC staff presented its evaluation of the estimated leakage with the tie-rod assemblies installed and with the postulation of through-wall cracking of all the horizontal welds..
The NRC staff found that with the tie-rod assemblies installed, the estimated leakage was 0.33 percent of core flow at 100-percent rated power and 85-to 100-percent rated core flow.
The NRC staff notes that the postulated 360'hrough-wall cracks in the horizontal welds with the tie-rod assemblies installed yield a higher estimated leakage than the total estimated leakage from the V-9 and V-10 welds cracked through-wall.. NMPC estimated that the total leakage resulting from the tie-rod assemblies installed and the postulated through-wall cracking of the horizontal welds and the V9 and V10 welds was 0.65 percent of the core mass flow at 100-percent rated power and l00-percent rated core flow.
The NRC staff does not consider this a
~
)
.significant leakage rate since it is less than I percent of the total core flow and is, therefore, acceptable.
The NHPl emergency core cooling system (ECCS) consists of the single-train feedwater coolant injection (FWCI) system, the automatic depressurization system (ADS), and the two-train core spray (CS) system.
The FWCI system requires limited offsite power to be functional.
During a postulated loss-of-coolant accident (LOCA), the CS system transfers water from the suppression pool to the reactor pressure vessel (RPV) where the water is sprayed upon and cools the core and returns to the suppression chamber by way of the break.
CS system water is introduced at the top of the core using circular spargers located inside the core shroud and containing distribution nozzles pointed radially inward and downward to produce fine spray droplets directed onto all fuel assemblies.
On the basis of this description of the CS system, the NRC staff notes that the estimated total leakage from the installation of the tie-rod assemblies and the postulated through-wall cracks in the horizontal welds and the V9 and V10 welds does not affect the performance of the CS system.
Because NHP1 is a BWR-2, it does not have jet pumps and does not depend upon reflood for a postulated recirculation line or steam line break.
- Thus, leakage of the shroud is not an issue for a recirculation line or steam line break since the core would be cooled by spray cooling.
Therefore, ECCS performance is not affected by the increase in postulated leakage as a result of the V9 and V10 welds.
4.3.2 Detection NHPC evaluated whether the postulated through-wall cracking of the V9 and V10 welds would be detected as a result of changes in reactor operating parameters.
The nuclear industry has established that shroud cracks may be detected if the estimated leakage produces a power anomaly of 2 percent in the rated power.
The NRC staff notes that approximately 3-percent leakage is required to produce a 2-percent power reduction.
For the estimated leakages discussed
- above, the combined leakage from (I) the installation holes from the tie-rod assemblies, (2).the postulated cracking in the horizontal welds, and (3) the postulated cracking of the V9 and V10 welds would not be detectable.
Any change to other leakage indicators, such as recirculation loop temperature and core support plate pressure differential, would also be too small for the reactor operators to detect.
NMPC postulated that if vertical welds V-9 and V-l0 did develop a larger leakage flow area, i.e.,
a crack opening of 1.5 inch for the entire length of the weld, the power decrease would be more than 2 percent and, therefore, would be detectable.
NHPC maintains a Special Operating Procedure NI-SOP-2, Revision 5, "Unexplained Reactor Power Change" that was revised in 1994 to account for potential shroud cracking or displacement.
In this procedure, depending upon where the indication is detected (i.e., below the core plate, between the core plate and the top guide, or above the top guide), the operator is iristructed to either scram the reactor or commence a normal reactor shutdown.
At the NRC's public meeting on April l4, l997, NMPC stated that operators were retrained on this procedure every 2 years.
~
~
On the basis of the estimated leakage rates and the above information, the NRC staff concludes that shroud cracking in the horizontal or vertical welds would not produce effects that could be detected.
However, if the crack openings were of significant size along the entire length of the welds (0.25 inches for all horizontal welds or 1.5 inches for both the V-9 and V-10 welds), the'RC staff concludes that these cracks would be detectable, ahd NHPC has suitable operating procedures in place to shut the reactor down.
5.0 R
U
'.l Summary Description of the 1995 Shroud Repair In its SE dated March 31, 1995, the NRC staff approved NNPC's submittal of the shroud repair.
A summary description of the repair is given in the following paragraphs.
The shroud repair consists of four stabilizer assemblies (also called tie rod assemblies) placed at approximately uniform intervals around the shroud (azimuths 90', 166',.270',
and 350').
The stabilizer assemblies were designed to structurally replace the circumferential welds in the core shroud.
Each assembly functions to vertically hold the shroud to the shroud support cone and to horizontally support the shroud at the top guide and core plate elevations.
There are also other horizontal supports that would prevent unacceptable horizontal movement of any shroud cylindrical segment in the event the horizontal shroud welds should fail.
i ~
,The stabilizer assemblies provide'ertical restraint and represent an alternate load path between the top of the'hroud and the shroud support cone.
This load path consists of the upper support, tie rod, C-spring, lower
- support, and toggle bolt.
Differential thermal expansion from the different materials used for components of this load path, produces a thermal preload at plant operating conditions.
Under postulated failures of the horizontal shroud welds,'he thermal preload is sufficient to hold the cylindrical segments of the shroud in place for all normal operating conditions, such as the vertical upward force applied to the shroud by the coolant flow and pressure.
The vertical load path is also designed to have a vertical spring rate that both prevents unacceptable vertical load during plant upset thermal conditions and provides acceptable dynamic response during a plant seismic event.
The shroud is restrained horizontally by linear springs at the top guide and the core plate elevations.
At the top guide elevation, the linear spring consists of the upper spring, upper wedge, upper 'contact, and the upper support.
At the core plate elevation, the spring consists of the lowet wedge, lower contact, and lower spring.
The horizontal.spring rates of these springs were designed to produce acceptable horizontal dynamic response during a seismic event.
The horizontal displacement of the shroud during all events must be limited by these springs to ensure control rod insertion and pr'event unacceptable leakage.
In addition, in the event of failure of the horizontal shroud welds, unacceptable displacement of other cylindrical sections of the shroud would be prevented by displacement limiters, such as the mid-support and top support.
~
~
'h During the initial installation of the repair hardware, the attachment features were machined into the shroud head and the shroud support cone.
The lower support toggle bolts were inserted through the shroud support cone and
. tightened'to,40 foot-pounds.
In each
- assembly, the tie rod nut at the top was also tightened to a specified amount.
The lower wedge was machined to provide a 0.01-inch compression of the lower spring.
The mid-support was machined to provide a 0.25-inch horizontal displacement of the tie rod.
The lower wedge is a component that was machined on the basis of actual site measurement to fit between the RPV and the lower spring with a small (0.010 inch) compression of the lower spring at room temperature.
The lower wedge is held in place by a latch device.
The latch (also referred to as a retainer clip) is a wishbone-shaped piece that is intended to prevent relative motion between the lower wedge and the lower spring.
Similar latches are also used to prevent relative motion at the mid-support and at the upper spring.
The lower support is an assembly that connects the shroud repair hardware to the shroud support cone.
The upper spring jacking bolt was adjusted to slightly compress the upper spring.
All moveable features were locked in place with mechanical devices such as crimps or spring retainers.
The lower spring wedge latches are one type of spring retainer.
Thus, it was intended that, after
'nstallation, the upper and lower springs would be compressed between the
- shroud, and.the RPV and the tie rod would be tight with a,small preload.
The lower support and toggle bolt assemblies were tightened to the shroud support
- cone, but (as would be discovered during the spring 1997 outage) their position within the support cone holes had not been properly specified in the
. installation procedures.
They could be at any location within the hole depending on'he hole's size and shape.
As stated earlier, the NRC staff reviewed NHPC's submittal relating to the shroud repair hardware and found it acceptable.
5.2 NRC Staff Evaluation of the Stabilizer Assemblies 5.2.1 Root Cause Evaluation of the Shroud Repair Deficiencies ind Corrective Actions During the spring 1997 outage, NMPC performed post-operational inspections on the core shroud stabilizer (tie rod) assemblies, consistent with its commitments and NRC staff requirements.
Deficiencies were found in.the tie rod repairs, which included, to varying degrees, loose tie rod nuts on all four tie rods, damage to the retainer clips on the lower spring wedges, and mispositioned mid-supports.
NNPC performed root cause evaluations, additional
- inspections, and testing of the tie rods.
NHPC described these matters to the NRC staff in a letter dated April 8, 1997, forwarding GE's analysis of the
.repair.
~
I g
1 The tie rod nut at the 270 location was first discovered to be loose.
The nut-locking device was normal and, in accordance with the original design and installation requirements, the nut could not be moved without removing the locking feature.
However, there was essentially no preload between the nut and the tie rod.
After the locking feature was removed, the nut was turned with less than 25 foot-pounds of torque.
The rotation of the nut before tightening it at 25 foot-pounds was equivalent to an axial clearance of approximately 0.1 inch between the nut and the tie rod.
NHPC's inspection and analysis confirmed that the'ut had loosened because the toggle bolts had moved slightly within the oversized holes in the shroud support cone during plant operation.
Although the holes are round when viewed perpendicularly from the shroud support cone, their projection on a horizontal plane is oval-shaped due to the angle of the support cone.
Thus, the toggle bolts are allowed to move up the inclined surface of the shroud support.
NNPC calculated the maximum tie rod looseness that could have been caused by movement of the toggle bolts within the holes in the shroud support cones.
The maximum upward movement (as measured by turning the tie rod nut on the 90'ie rod) was 0.151 inch; whereas, the differential thermal expansion in the tie rod is 0.155 inch, which is greater than the looseness at the shroud support cone.
Therefore, although the initial installation mechanical preload was lost, some thermal preload wou1d have remained.
Significantly more thermal preload remained at the other three tie rod locations where the measured upward movement ranged from 0.054 inches to 0.093 inches.
Visual inspection revealed that the lower support wedge latch at the 90'ocation was broken.
A piece of the latch that was missing was later found on the lower support cone at approximately. azimuth 330 After examining photographs and video tapes, NHPC stated that fatigue did not appear to be the failure mechanism.
There was no visible evidence of plastic deformation,
~ which would be necessary for a single-overload type of failure.
The failure surface appeared to be consistent with a stress corrosion failure under high stress.
Results from a metallurgical evaluation at a GE testing laboratory are expected to be available by Nay 20, 1997.
Pending completion of those examinations, NNPC believes that the most likely failure mechanism is stress corrosion, although failure from overload has not 'been ruled out.
As the NRC staff understands this information, this conclusion 'appears reasonable.
Videotape inspection of the other three lower wedge latches indicated that they are all in one piece, but the latch at the 350'ocation appeared to be bent.
In addition, the lower spring wedges showed minor evidence of local hard contact with the latches.
Since the latch material, alloy X-750, is
'arder than the lower spring wedge material, Type 316 low-carbon stainless steel, the lower spring wedges are likely to show surface wear before the latches..
A similar latch is used in each mid-support assembly, and two similar latches are used in each upper support assembly.
NNPC observed the latches on the middle and upper supports, and all 12 appear to be in good condition.
Because of design differences, these latches cannot be loaded as severely as the lower wedge latches.
The contact force between the RPV and the shroud repair is much smaller at these locations than the contact force at the lower wedge.
In
I
~
e I r addition, these latches are not loaded during plant heat-up.
The lower wedge at the 90'ocation had dropped to the bottom of the post on the lower spring.
The lower.wedge at the 350'ocation appeared to be approximately 1/8 inch below its normal position.
The other two wedges were at their normal positions.
Contact sliding marks have been observed on the RPV wall at the 166'ocation above the lower and upper contact points of the upper spring assembly.
Sliding marks are not evident at any other contact points for the upper springs located at.the 90', 270',
and 350'ocations.
Visual inspection of all mid-support contacts confirmed that there was contact with the RPV surface in the cold condition.
During normal operation, the mid-support compression on the RPV increases because of th'ermal conditions.
Thus, the functions of providing a load path from the tie rod to the RPV and increasing the natural frequency of the tie rod assemblies were maintained.
During inspections and tests related to the root cause investigation of the lower support wedge latch
- failures, two mid-supports at azimuths 90'nd 166 had to be repositioned.
New mid-supports have been fabricated and installed with the original design preload at these locations.
The mid-supports at the 270'nd 350'ocations have also been verified for proper preload.
Thus, the required support configuration will be maintained during future operation.
The root cause for the tie rod degradation affecting both the tie rod nut and the lower latch, is attributed to the movement of the toggle bolts within oversized lower support bolt holes.
The insta11ation procedures did not contain specific criteria for locating the toggle bolts during installation of the lower support.
The lower support toggle bolts are nominally 4.000 inches in diameter.
.The measured electric discharge machining (EDM) holes in the shroud cone, ranged from 4.090 inches to 4;110 inches.
Since the position of the lower support within the machined holes was not procedurally controlled during installation, the relative position of the bolts within the holes was randomly located.
During heat-up, the expansion of the shroud and tie rods generated a force sufficient to overcome the installation friction forces and slightly move the lower support toggle bolt'ssembly up the shroud cone.
This translated into a vertical movement of the tie rod, loading the latch on the lower spring wedge and causing its failure.
These latches were not designed to accommodate differential movement between the RPV wall and the lower spring wedge during n'ormal and transient conditions in the event the lower wedge got stuck at the RPV wall.. This also caused the loss in preload on the tie rod nut, described earlier.
The NRC staff has reviewed NMPC's analysis of the tie rod degradation and agrees with NMPC's conclusions.
in the original design of the lower wedge, it was assumed that the wedges would slide on the RPV wall and accommodate differential thermal expansion between the tie rod assembly and the RPV.
However, the actual frictional force between the wedge and the RPV was higher than anticipated and sufficient to prevent movement of the wedges at the RPV interface during thermal growth of the tie rod assembly.
This caused the retainer clip (or latch) to stretch
~
g I'L
.0 between its attachment points on the tie rod and the wedge.
A portion of the retainer clip was overstressed and failed because of stress corrosion cracking..
On the basis of its review (as discussed above), the NRC staff believes NHPC's preliminary root cause determination is reasonable.
- However, even if the failure resulted from overstress, it appears that the modification would address the failure.
5.2.2 Hodification of the Lower Support Latch Deqign The lower support latch (or retainer clip) has been redesigned to accommodate movement during normal and transient conditions.
The redesigned latches were installed before core reload.
The new design is substantially more flexible and takes into account the various potential sliding cases.
These cases include combinations of sliding at the RPV wall/lower wedge and lower wedge/lower spring interfaces.
The largest latch displacement is postulated during the initial heat-up and hydrotest where the wedge is projected to slide at the pring interface but remain locked at the RPV interface.
During a potential cooldown to a ambient conditions the wedge is projected to slide only at the RPV interface.
Finally, during a heat-up to full power, the wedge is assumed to slide at the spring interface with no movement at the RPV surface.
The calculated latch displacement for this worst-case scenario has been determined to be 0.182 inches.
The calculated stress in the new latch for this displacement has been determined to be 60 percent of the allowable stress.
In addition, the latches have been evaluated for potential damage due to SCC.
Utilizing a Stress Rule Index (SRI) methodology, NHPC has demonstrated that for the worst-case sliding, stress corrosion will not occur in the next operating cycle.
However, if'a most probable wedge/RPV/spring interface sliding scenario is assumed, then no corrosion is projected for the remaining life of the plant.
5.2.3 'RC Staff Evaluation of the Structural Analysis of the Redesigned Lower Support Latch On the basis of its review of the new retainer clip design, the NRC staff raised several concerns.
One concern was that the slightly sloping surface on the lower wedge would preferentially cause the wedge to slide in one direction and get stuck in the other direction.
NNPC, in response, clarified that the purpose of the angle on the lower wedge is to facilitate moving the lower wedge into position and creating the contact force with the RPV wall.
This slight slope on the wedge does not influence whether sliding occurs at the
'edge/spring interface or the RPV wall/wedge interface.
Therefore, during plant operation, sliding is expected to occur at whichever surface has the lowest friction factor.
As discussed in GENE Report Bl3-0I739-22, sliding-is likely to occur at the wedge/spring interface because the machined surfaces and dissimilar materials will result in a low friction factor.
In response to the NRC staff's inquiry about the feasibility of lowering the lateral load at the wedges to facilitate sliding, NMPC discussed the considerations in determining the lateral design loads on the wedges.
The intent of the lower spring design was.to ensure that positive contact, exists
I f
N
I
~
~
l during shutdown conditions, and the 0.010-inch radial contact at the lower spring represents a minimal condition that meets this requirement.
Recognizing that the measurements for matching the lower wedge have to be taken remotely from the reactor refueling floor, the 0.010-inch interference represents a value that confidentlj ensures contact at the lower spring support.
Mithout radial support at the lower wedge, the only potential-problem is vibration.
- However, as demonstrated at the 90'ie rod location during the last operating cycle (which lost contact),
and by analysis, NHPC concluded that vibration of the tie rod assembly is not an issue.
NHPC's responses as discussed
- above, reasonably address the NRC staff's concerns.
The maximum possible differential vertical displacement of the lower wedges and the probable wedge movement has been determined.
Since the potential exists that the wedges may not always slide at the spring interfaces due to unanticipated forces on the wedge during various operational transients, the design of the latches has been based upon the maximum estimated vertical displacement.
The maximum displacement assumed for the latch design exceeds the maximum vertical displacement of the tie rod hardware at the lower spring elevation as shown in Table 4.1 of GENE Report 812-01739-22.
During certain plant operating and test conditions (e.g., during a hydrotest),
the radial contact force on the we'dges is likely to be minimal.
Under such conditions, the wedges could potentially slide circumferentially along the RPV surface and remain stuck in that position.
In response to NRC staff concerns that this could impose an additional torsional moment on the latch during subsequent plant operation, NHPC indicated that these considerations have been factored into the new latch design.
The ratio of the width of the wedge to the potential gap between the wedge and the RPV wall is very large.
Therefore, potential angularity of the wedge in a loss-of-contact event is extremely small and any. applied force on the wedge will cause it to readjust the orientation of the wedge to distribute the contact forces across the entire face of the wedge surface, and not allow an edge to become stuck.
Even if it is assumed that the wedge would become stuck on one edge, the latch mechanism has lateral clearance with the wedge and the spring interface slots, so that the latch will. reposition itself without any torsional loads being applied.
The new latch design stresses have been calculated to be substantially lower than the original latch design stresses during various postulated operating,
- transient, and accident conditions.
The membrane plus bending stresses in the new latch design are lower by a factor of 8.6 in comparison with the original latch stresses.
The corresponding reduction factor for the membrane,
- bending, and peak stresses is 12.8.
This indicates a substantial improvement in the design margin for the new latches.
5.2.4 NRC Staff Analysis of Shroud and Tie Rod Assembly Section 4 of GE Report GENE 813-01722-40, "Shroud Repair AnomaliesNine Nile Point Unit I, RF014," dated April 1997, contains an evaluation of the safe operation of the shroud and the tie rod assembly with the existing ligaments in the horizontal and vertical welds.
NHPC's assessment indicates that the safe operation of NHP1 was not impaired with the as-found condition of the tie
~
~
i
)
~
~
l rod assembly and an assumption that all horizontal welds were 360'hrough-wall cracked.
It is known, through actual inspection of the shroud, that in the regions where inspection has been performed, no through-wall cracks or 360'rack lengths had been found in any of the horizontal welds.
Therefore, NHPC's evaluation is considered conservative.
NHPC has analyzed the as-found condition of the shroud vertical welds and has established that the plant can be operated safely.
A conservative.interval for reinspection of the welds has been established.
NHPC has committed to reinspect the tie rods, which will include checking the latch devices and checking the tie rod nuts for tightness, after approximately 10,600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> of operation.
On the basis of its review of the structural integrity of the re-designed latch and the core shroud vertical welds discussed earlier, the NRC staff finds the inspection interval acceptable.
5.2.5 Revised Installation Procedures for Tie Rod Assemblies As described in NHPC's Document NHP-SH0-003, Revision I, 'Lower.Hedge Latch Replacement and Tie Rod Torque Checks," including Special Process Control Sheet
- SPCS801, Revision I, NHPC has developed improved installation procedures to ensure that the tie rod assembTies are installed with no gaps at the upper end of the support cone holes.
The improved procedure requires that each tie rod be jacked at three locations during tie rod nut torquing to remove any gaps associated with installation tolerances.
Jacks are placed under the lower support, on the RPV side of the-lower support, to push it up the shroud cone.
This removes the clearances between the toggle bolts and the shroud side of the cone holes.
In its letter to the NRC dated April 27, I997,,"Responses to.NRC Staff guestions Provided During Telephone Conversations of April 10, Il, and 24,
- 1997, on Core Shroud Cracks and Repair,"
NHPC states that there are no plant operating, transient, or test conditions that will cause the toggle bolts to slide down the shroud support cone.
'The temperature of the shroud/RPV annulus would have to be higher than the RPV and shroud temperatures, in addition to a complete loss of preload in the tie rods to cause such a movement.
This combination of events does not occur during operation.
After installation, and in accordance with the revised procedures, inspections were completed on each tie rod assembly to verify proper contact by verifying the absence of gaps.
These inspe0tions revealed that the middle support was no longer in contact with the RPV on the 90'nd l66'ie rod.
This resulted from the movement of the lower support assembly up the cone toward the shroud.
The middle support dimensions were obtained and new middle supports were installed before reload.
Other locations on the tie rod assemblies with the potential for gaps and non-conforming conditions were inspected.
No additional deficiencies were noted.
La ~sP C
IIE In its letter of April 30,
- 1997, NHPC also stated that it will propose an inspection plan for the next scheduled outage and submit the plan to the NRC at least 3 months.before the outage is scheduled to begin.
This plan should
I I
K provide details regarding the inspection of the shroud repair components; the shroud repair anchorages; and the shroud's horizontal, vertical, and ring
- segment welds.
The plan will specify the inspection methods to be used, including the provisions for sample expansion.
v.e auxuuuar 7.1 Inspection The NRC staff has determined that the scope and methods used for the inspections of the NHPI core shroud during the spring 1997 refueling outage met or exceeded the requirements of BWRVIP-07, and are, therefore, acceptable.
7.2 Structural Integrity The NRC staff has concluded that results of the analyses show that the core shroud maintained the required
'ASHE Code margins'uring the last operating cycle.
The review considered the loss of pre-load for the tie-rod repairs.
Additional analyses submitted by the licensee pursuant to 10 CFR 50.55a implemented by ASHE Code Section XI for the vertical welds have been shown to have sufficient ligaments to support the tie-rod repair for at least 10,600 additional hours of operation.
The staff performed an independent analysis and determined that the licensee's analyses appear reasonable.
7.3 Tie Rod Repair On the basis of its review of the root cause analysis of the shroud repair anomalies, the NRC staff concludes that the tie rod was loosened by the movement of the toggle bolts within oversized bolt holes.
NHPC's new installation procedures include measures to prevent tie rod looseness and maintain tie rod vertical forces as intended in the original design.
The root cause of the latch failure was larger-than-anticipated vertical displacements of the latch, which overstressed the latch and most likely subjected it to SCC.
The new latch has been redesigned to accommodate larger vertical displacements while maintaining its original function of locking the wedge to the lower spring structure.
On the basis of a most probable sliding scenario at the wedge/RPV interface, no failure is projected for the remaining life of NHP1.
The calculated stresses are within the ASHE'Code allowable values and the latch has been analyzed to be resistant to stress corrosion for a minimum of 2 years assuming worst-case displacements in the latch.
NHPC's root cause evaluation and corrective actions offer reasonable assurance that the tie rods will perform their intended function. The root cause evaluation will be confirmed in the near. term by ongoing evaluations.
Future inspections will monitor the conditions of the tie rod assemblies.
The NRC staff has determined that the modified tie-rod repair is an acceptable alternative to a repair in accordance with ASHE Code Section XI, pursuant to 10 CFR 50.55a(a)(3)(i).
The alternative provides an acceptable level of quality and safety because the ASHE Code margins of'safety will be maintained for the operating period.
This NRC approval under 10 CFR 50.55a(a)(3)(i) is contingent upon NHPC (1) maintaining reactor coolant chemistry within the
. ~
I
~
~
1 guidelines set forth in Elect'ric Power Research Institute technical report TR-
.103515,
'BWR Water Chemistry Guidelines-1996 Revision," in accordance with its commitment by letter dated April 30,
- 1997, and (2) submitting, within 60 days, an application for a license amendment that addresses this matter in accordance with its commitment by letter dated Hay 7, 1997.
Failure to satisfy either of these conditions will render this approval null and void.
8.0 Letter from L.B. Marsh, U.S.
NRC. to Niagara Mohawk Power Corporation dated Harch 31,
- 1995, and attached Safety Evaluation by the Office of Nuclear Reactor Regulation relating to the Core Shroud Repair at Nine Mile Point Unit 1.
2.
Letter from Niagara Mohawk Power Corporation to U.S.
NRC (NHPIL 0927),
dated March 25, 1995.
3.
4.
Letter from Niagara Mohawk Power Corporation to U.S.
NRC (NNPIL 1067),
'ated April 30, 1995.
Letter from Niagara Mohawk Power Corporation to U.S.
NRC (NHPIL 1111),
dated August 14, 1996, with enclosure providing final design documentation for modifications to the core shroud repair assemblies installed in March 1995.
Letter from NRC to Niagara Mohawk Power Corporation, dated March 3 1997, with enclosed Safety Evaluation by the Office of Nuclear Reactor Regulation relating to the Hodifications for Correcting Deviations identified in Core. Shroud Repairs at Nine Mile Point Unit 1
(TAC No; M90102).
6.
Letter from 'Niagara Hohawk Corporation to NRC, dated January 6, 1995, with enclosed design report of Core Shroud repair hardware.
Supplemented by letters dated January 23, 26, February 14, 24, 28, Harch 7, 9, 13 (2 letters),
14 (2 letters),
23, 27, 28 and 30 1995.
7.
Letter from Niagara Hohawk Corporation to NRC, NMPIL 1200 dated April 8,
- 1997, (Enclosure 2)
GENE B13-01739-40 "Shroud Repair Anomalies Nine Mile Point 1, RF014," April 1997.
8.
GENE-B]3-01739-22, Revision 0, Nine Mile Point 1, "Design Report for Improved Shroud Repair Lower Support Latches," April 1997.
9.
10.
Niagara Hohawk Document NMP-SHD-003, Revision 1, "Lower Wedge Latch Replacement and Tie Rod Torque Checks," including Special process Control Sheet SPCS¹01, Revision 1.
Safety Evaluation of Niagara Mohawk Power Corporation (NMPC)
Reinspection Plans for Nine Hile Point Unit 1 Core Shroud, NRC, Harch 3, 1997;
t I
~
f
I
~
~
~
I I
11.
12.
13.
- 14. "BWR Vessel and Internals Project,. Reactor Pressure Vessel and Internals Examination Guidelines (BWRVIP-03)," EPRI Report TR-105696, October 1995.
J
'BWR Vessel and Internals Project, Guidelines for Reinspection of BWR Core Shrouds (BWRVIP-07)," EPRI Report TR-105747, February 1996.
"Reactor Pressure Vessel and Internals Examinations Guidelines (BWRVIP-03)," EPRI TR-105696, October 1995.
'Assessment of the. Vertical Meld Cracking on the NHPl Shroud," Report No. GE-NE-523-B13-01869-043, Rev. 0, April 1997.
Prepared for Niagara Hohawk Power Company by, GE Nuclear Energy.
15.
16.
. 17..
18.
19.
20.
21.
22.
23.
24.
25.
'Shroud Repair Anomalies, Nine Mile Point Unit 1, RF014," Report No.
B13-01739-40, Rev. 0, Class III, April 1997, Prepared for Niagara Hohawk Power Company by GE Nuclear Energy.
"Nine Nile Point Unit 1 Core Shroud Cracking Evaluation," Prepared for George Inch, Niagara Hohawk Power Company, by Richard E. Smith, Altran Corp., April 3, 1997.
"Analysis of Nine Hile Point Unit 1 Shroud Melds V9 and Meld V10 Cracking," April 1997, Report No. MPH-497439, HPH Technologies.
Safety Evaluation 96-018, pursuant to 10 CFR 50.59 Rev.
1.
"Concerns Regarding Operation of Nine Nile Point 1," letter dated April 9, 1997, from D. Lochbaum, Union of Concerned Scientists to S. Bajwa, NRC.
"Potential Unanalyzed Operation of Nine Nile Point Unit 1 Core Shroud Vertical Cracks," letter dated April 17, 1997, from D. Lochbaum, Union of Concerned Scientists'to S. Bajwa, NRC.
'BWR Core Shroud Inspection and Flaw Evaluation Guidelines,'ENE-523-113-0894, Rev.
1, General Electric Company, Harch 1995.
Prepared for the BWRVIP Assessment Committee.
'BWR Mater Chemistry Guidelines, 1993 Revision, Normal and Hydrogen Mater Chemistry,"
EPRI TR-103515, February 1994.
'Response to April 17, 1997, letter from Union of Concerned Scientists to the NRC,'etter dated April 25, 1997, from NHPC to NRC.
'Response to Union of Concerned Scientists letter dated April 9, 1997, Regarding Nine Nile Point Unit 1," letter dated April 26, 1997, from NHPC to NRC.
"Responses to NRC Staff guestions Provided during Telephone Conversations of April 10, ll, and 24,
- 1997, on Core Shroud Cracks and Repair," letter dated April 27, 1997, from NHPC to NRC.
I
~
k I
I
u I
l
~
I l 26.
"Responses to NRC Staff guestions Provided during Telephone Conversations of April 22 1997, on Core Shroud Cracks and Repair,"
letter dated April 27, 1997, from NMPC to NRC.
27.
Brunswick Steam Electric Plant No. 1, Docket No. 50-325/License No. DPR-
~ 71 NRC Generic Letter 94-03, "Intergranular Stress Corrosion Cracking of Core Shrouds in Boiling Mater Reactors," letter dated June 21,
- 1995, from R. Anderson of Carolina Power 'and Light Co. to the NRC.
28.
Letter from Niagara Hohawk Power Corporation to U.S.
NRC (NHPIL 1217),
dated Hay 7, 1995.
Dated:
May 8, 1997
~
~
I
~
I I
I J
~
)
~
ENCLOSURE Figures, tables; and Appendix C from 6eneral Electric Nuclear Energy Document 6E-NE-B13-0l869-043, Revi si on 0.
I
~
l I
I
)
~ ~
1 H-2 H.S n
.S4 Not!0 ecole 0
> r>>
V-'F l450 Vf2 V~55 e
figureM HMFlQeead%chf Lceagogg
(
l I
g g
~
V
~
Table22 Summary ofReceat Shroud Verfhal%eld hspedions (RFO I4)
V3 Y.frin V4tin Vi)
%Aid Length On) hLspcctkon
.Cte<<rage~
15 Left l5" 22" Left 11'i Not located 9" Left 11 r t
M"Left 98" OD OD Nh Nh OD Nh RJlw Length 18" ID Rtght HAZ 0.$'D, Ri t HAZ 22 "1D LeftHAZ, lb~ID t HAZ Nh No Indications V 10 90.12 90.12 IDand KVT~1 OD OD KVT~1 OD lndicatices mos 90K right HAZ Minorcraclcing ce OD 1<<ft aide and on ID both sides Numerous indications ce OD, LeftHAZ
~ minor Saws ce ID,Right HAZ Cracking ce OD, Right Caching ce ID,Left and Right HAZ Flaws detected on > gOS on OD, Right HAZ Flaws detected ce a IONce OD, Left HAZ V01]
Veg 100/i OD 50'li ID 100'li OD
$0'/i ID IDand OD KVT-1 KVT 1 No lndicatiotm 6" OD, Right HAZ V 1S 22.13 1 1" Left 1 1" Ri t
~
UT 6"
ID, HAZ X2"ID HAZ V0]6 22.13 100'li 103" Left K" Rght OD OD EVT 1
.7S" OD, LeftHAZ S" IDLeftHAZ
'i IDRight HAZ 3i IDLett HAZfncn~ ah@
~
The cd re jons indicated on <<ach side ofthc weld <<rc ret n<<ce/sari JT coinc~ bene<< tbc mspcct g
~~~~~<<~b,~g aym<Sea b
ca aea~acla,mi ~e t ~V 4pment J<<ngth.
7
1
hppeadiz C
~~~Id Mpectioa Summary
0 I
v
~
)
'I
lhe Mowing is a weld by weld Nanmary detailing the <<cope of Qspections and sesuits of the shroud examirntions performed to date.
%@M V4 Pedormed ttltrason/c examination of<<pproxhnately I$ hcles ofeach Nde ofthe weld &em the
<<heed OD surface, hpproximately IS" ofSaw was detected on the ID~bee <<nd 0.8" ofSaw
~N the OD sauce.
%AdVA performed ultrasonic examination ofayproximateiy l1"ofthe MtHAZand 22" ofthe zight HAZ.
. )D Saws were detected over the entire examined hngth of the MHAZand IS" ofSaw was detected on the IDofthe rightHAZ.
MV;7 performed ultrasonic araminstion ofapproximateiy 9" ofthe MtHAZand Il"efthe right HAZ.
No Saws were detected during the examination.,
%VcMU4 Petaled ttltrasonic eraminstion of<<pproximately M"of the kft HAZ and 9E'f the sight HAZ.No Saws were detected during the examination.
%CM U-9 e
h 5
8 4
ROD dl f ~lylh~laet ofboth the le and right HAZs as well as EVT&nn both the ID and the OD. Visual cracking was detected over yester th'an%5'f the right HAZon the OD and minimal crwhing was detected on the IDin both the leA an right HAZs. Minorcr3cking w'as aIso detected on the OD in the leh HAZ.
The cracks detected visually on the shroud ID surface were found to be predominantly trNNverse the weld whereas the cracking detected visually on the shroud OD surface was tnostIy parallel to 4e weld with components that SNL)QRd tr<<nsverse to the 'weld. Ultrasonic examinations of essentially the entire length ofthe weld was perfortned Sem the shroud OD surface <<nd detected ggmerous gnaws over the khgth ofthe MHAZ emanating 50m the <<hroud OD <<ttrface. Two Nnail Saws on the IDsurface were detected in the right HAZ.
%aMV-l0 Performed Ultrasonic examination &0!5the shroud OD surface for approximately the entire $gggtth ofboth the le8 and right HAZs as weil as EVT &om both the ID and the OD. Flaws were detected on greater than SP/~ ofthe right HAZon the OD surface and greater than jp/q of the left HAZ gevealed flaws on the OD surface.
The EVTeraminteon revealed cracking in the ~ <<nd right
0 el
~
r;r) 4
Gsh'~l~lN~g, ~.o
~
0AZs on the OD surface formost othe hngth ofthe weld and on the in both the leffand righ HAZs. We cracks detected visually on the ahroud ID Ntrface were found to be predominantly eansvene to the weld whereas the ++Acing detected visuaHy on the ahroud OD Ntrface was mostly yaraM to the weld with coegenents that branched transverse to the weld.
'%'eM V 11 EVTeaminations were performed on 8e accessible weld length &nn both the ID and W OD of bothy Mand right HhZs. No cradaag was detected during.the catamination.
%eM V-9 KVTexamhm6ons were perfcemed on 8e accesNRe veld length &an both the IDand 8e OD of both the leff and right HAZs, One 6" crack was detected on Se iength OD atrface kn Ce nght HAZ. No other cracking was detects
%cld V'15 Ultrasonic automation was performed Gem the ahroud OD auface on approximately I 1 inches of both the leff and right HAZs, One 6" Saw was detected in the left HAZ on the H) Nuface and
'oral ID fhws totaling X2" in total length was detected an the ID in the sight HAZ, No Saw 4etected in ather HAZwas greater than lan through wa11..
%CM V~]6 Ul
'. 'l WfM AIR 6 8 ROD 8'~lytD.5" fl lt HAZ. Two Saws were detected on the ID surface. One fhwwas S" in length, 10/i through wali.
'Qe other ID Saw'in the IeffHAZwas detected Som the acan on the right HAZand was 3" long and 30/o through watl. hpproximately 22 inches ofthe right HAZwas examine Som the shroud PD surface.
One fhwwas detected on the IDwhich measured 4" in length and 2lNthough wall.
hn EVT atamination of both HAZs Sam the ahroud OD surhce revealed one crack in the M HAZ.
Recent Inspection Results forShroud Horhonta1%eMs h addition to&ihroud verticaI we1d inspections, Ce horizonta welds H4,M,H-$, H4a, H-65, and H-7%ere a1so inspec~ for NuLlytical purposes, to cva1uate the overall integrity ofthe grad sating 4$$lnnptions of ~
case
. cracking of 4e ical
%olds.
y%
'L f,)
'h r
4 a 3
~ t"- %V@HH4 ei rrvalM>eoMS, italo Ultrasonic etamination was performed from the shroud OD turface ofapproximately 24 inches ofthe upper HAZ adjacent to add VA. Approximately 7 inches of intermittent Saws were detected on the OD surf'ace, with the deepest aea haviag a thro~ wall depth of~ incl~
%dd H4 ggrasonic examination from the shroud OD amrface %as performed oa ly 60N ofthe Sower HAZ, 5) andlor OD Saws were detected intermittently throughout the examituttion areL Some IDlaws were detected in the upper HAZ. 'y 32 inches ofthe upper HAZwas
. attrasoaicaHy examined. 3 iaches ofshallow OD Saws were detected in the upper HAZand one 6
@eh hag ID Saw was detected with the znaximum through wall depth ofMhades..
AnEVT etaminatioa ofthe OD was performed ofover 70N ofthe upper and lower HAZs. Cacus ave detectedinboththe upper and lower HAZs.
%tMH-5 ZHtrasonic eamination from the shroud OD Ittface was performed on approximately 30 Je ofthe upper aad lower HAZs. OD and ID Saws mm detected in the upper HAZce1y. No Saws vere detected in the lower HAZ, EVIofapproximate)y'60% ofthe ahroud OD atrface revealed craclrs kntetmittently ia both the upper and lover HAZs. Most ofthe Saws detected visually oa the OD Iurface were oriented papendicular to the weld. No Saws~ detected in tbe tapper HAZat the hfersectioas ofwelds V9 or Vl0.
'%V@MHAA Ul U U
3 3 Ih hyyh M3 Uhh fUUUUyhhly 33U
'f the circumference from the shroud OD atra.
FIaws~ detected ce the OD Ntrface ofthe
)over HAZonly. No Saws were detected in the upper HAZor ce the IDofeither HAZ.
3
. %Veld BOB Ul
'U U
hhhlh hyy ~ U U33 f~ly33'/
ofthe circLtaderence from the shroud OD surface. Htws were detected ce the OD surface ofthe
'pper HAZonly. No Saws were detectedin the lower HAZor ce the IDofeither HAZ,
~ I
%Vdd H-7 Ultrasonic examination ms pafonned for the Ihoud OD Ntrface ce ~~ HAZ on lyhUh fly
.ll 3 hh Uh 'Ulh
%cMH4 Ultrasonic examination ms pafonned for the shroud OD Ntiface ce the ~ HAZ oa Npproxittmtely MYiof the circumference. h Saw ~ch was identified by UI'ut~ a prior CA
0 I
s E
r',e sO ), eaasde was heesed as well ae addtdossd gg~ g sbe eases areL s fhw was sdoasoaieetly
'aired to be oflesser thzough wI11 depth thorn h A@013. Azeview'of the previous data hdicates that the previous %zing pcrfoI?ned was very coaservttive.
h5 EVT wLs pclfo7?Dcd on Ipproximately 30 Ji ofthe circumference &am the shroud OD Nrrface. Of the Sve Nnall cracl~
dually detected during RF013 only I was visWe during this inspection.
The inspection in the area ofthe other four wms hampered by the placement ofa Tie Rod NIpport which prevented a good EVTinspection. Gals were visually detecIed ia three new locltions in the upper HAZ. The hrgesI ofthese aacks (9"-l2")is locaIed predominantly in the zing mgmeat Upper HAZand zuas jato the weld toe and back into the zing Iegaaeat.
R hnEVT.examination was perforzned in one aea 26 inches hag. No hdicltions were sets durhg the cxNminwon
~
( a, "3y.
Vabh 54 Aa~hru Sh f rtheN eMScMntUahl
.Shroud VertIha1%eQs 0) 0)
0)
'%eM ASwaMe Tinwgimal
%1M ED
- Isaiah, la ZLPM 4JaMLet8 44)
Nhhaaa ewphef
+aaeat, IL 09 M3LQpaeat Jedsdlag arch Immy (tao geae) aaC'sspeake Oaercalaty, h.
ote 1 C<)
AnBable Sqibalent
- Qaerachod
'gsaiat 4eaith, h.
V.),VW uS V 1,V4
)LS0 V4, N.32 VSAO V lo V 33, 4340 4410 V.$2 0
V )S, QL3)
V036 l.14 aw-
. lan S3AO 040
%ee 3
.))
Sotc 2 (VA N (V~1) 5,6 (V4 os 2 (V+
ÃW2(V 30 3.7S (V 3 l) 25.1S (V-12 Not! 4 (V+35) 58 (V-3 otes
- g. Based on crack grcerth otlAh and UThspceSon aucefah
- NXiMVS, hch ithach crack Ny forIength iRag.
4'ee¹ ~admen¹ based on farther entuation relied In Subsection K3.
~e mhimum Sgament forEVTknspedon Is huger Co account forgreater Nncertahg 4 the equal hgeeSon. The'aa~hg Isctor apped S equal aRx&h
- f5e ~u}valent length ttkt$0bttscthg crck gro%8) phd inspection eacercah5 h~ ~~ch ~ greater Chan the ~abed Qgiment et~ In and thus aeeptabh.
~;ca',4