ML20042F578
| ML20042F578 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 05/03/1990 |
| From: | Shelton D TOLEDO EDISON CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 1802, NUDOCS 9005090095 | |
| Download: ML20042F578 (28) | |
Text
,
2 U
.. j., -
o TOLEDO' EDISON A Centenor Energy Cornpany
' DONALD C. SHELTON veemean-waer Docket Number 50-346-l License Number:NPF-3 Serial Number 1802 May 3, 1990
]
United States Nuclear Regulatory Commission Document Control Desk Vashington, D. C. 20555 Subjec t :
High Pressure. Injection /Hakeup Nozzle and Thermal Sleeve Program Davis-besse Huclear Power Station Unit 1 Gentlemen The purpose of this letter is to provide the status of the High Pressure Injection (HPI)/%geup nozzle program at Davis-Besse Nuclear Power Station' Unit 1 and to request NRC approval of operation for Cycle 7!and beyond based on program results. The'HPI/ Makeup nozzle program represents Toledo' Edison's follovup actions addressing the discovery of the failed HPI/Hakeup nozzle-thermal sleeve during the fifth refueling outage (SRF0) in 1988. The thermal sleeve, which had failed due to-thermal fatigte, allowed makeup wated to impinge on the mouth of the nozzle.
Liquid dye penetrant _(PT) the nozzle revealed indications in the, cladding in the area whe; inspection.of re the thermal sleeve had failed. Manual ultrasonic (UT) examination performed ats the time found no evidence of flaws extending into the base metal. : Conservative-analysis indicated that significant flav growth was not expected with an effective thermal sleeve in place. The failed thermal sleeve vas replaced,-
minimum makeup-flow vas increased, and control over makeup flow was improved to reduce thermal cycling of the thermal sleeve. These actions.are discus' sed in Toledo Edison's September 14,- 1988-letter to the NRC (Serial Number 1580).
The'NRC approved restart and operation for one cycle (Cycle 6) based on these actions (NRC letter to Toledo Edison dated October 4, 1988, Log Number 2725) and required Toledo Edison to identify follovup actions to be carried out during the sixth refueling outage (6RF0).
Toledo Edison's June 19, 1989 letter to the NRC-(Serial Number 1664) identified planned actions for the 6RF0. The 6RF0 actions included re-routing makeup flow to an alternate HPI nozzle,_ visual inspection of the thermal sleeve, and development of enhanced UT techniques and inspection of the' e
9005090095 900503 i
PDR ADOCK 05000346 P
pdc 0h i
THE TOLEDO EDISON COMPANY EDISON PLAZA 300 MADISON AVENUE TOLEDO, OHIO 43652 -
,jDodkit Numb r 50-346 License Number-NPF-3 Serial-Number 1802 Page 2 IIPI/ Makeup nozzle and the alternate nozzle.
Although these inspections were not expected to discover indications in the thermal sleeve or flav growth in the nozzle, contingency plans were developed for thermal sleeve replacement and veld overlay repair of the nozzle.
.The enhanced UT development program and system capabilities _are discussed in-Toledo Edison's letters to the NRC dated February 20, 1990 (Serial Number 1768) and April 16, 1990 (Serial Number 1778),'and in the enclosed report. By letter dated November 8,1989 (Serial Number 172tJ) Toledo Edison requested NRC approval of the veld overlay repair contingency. Veld overlay. repair also was discussed in a January 24, 1990 meeting with the NRC which is documented in -
letter Serial Number 1768; l Toledo Edison has completed the 6RF0 actions discussed in letter Serial Number 1664. The enhanced UT er. amination of the HPI/ Makeup nozzle confirmed that the
{
cladding flaws ideatified during the SRF0 had'not propagated into the base metal.
Consequently, veld. overlay repair of the-nozzle is unnecessary.
Accordingly,: Toledo Edison withdraws its request for NRC approval of veld overlay repair (Serial Number 1726). The visual inspection of the thermal sleeve-identified no thermal fatigue indications and thermal sleeve J
replacement also was unnecessary.
The inspection results are< documented in the enclosed report (Attachment 1),
i i
The enclosed report includes-a-structural re-evaluation of the former j
HPI/Pakeup nozzle. The re-evaluation considers those design transients that are consistent with the HPI duty of the nozzle, since makeup flow has been re-routed to an alternate nozzle. The re-evaluation assumes that an. intact-q thermal sleeve is in place and conservatively assumes a pre-existing flav which penetrated 0.125 inch into the base metal at the most structurally I
limiting location. The assumed flav size is consistent with the enhanced UT system detection capabilities. The.re-evaluation concludes that the assumed j
flav at the most structurally limiting location vould be expected to grow i
less than 20 mils over an additional 40-years of service.
l 4
Toledo Edison plans to perform an enhanced UT examination of the former
,b
!!PI/ Makeup nozzle during the 7RFO.
Provided the 7RF0 inspection reveals no indications, a further inspection vill be conducted in conjunction with the i
next 10-year interval inservice inspection.
In summary, Toledo Edison has addressed the implications of the failed HPI/ Makeup nozzle thermal sleeve discovered during the SRFO. The actions j
taken have eliminated cold makeup water as a potential driving force for-1 thermal fatigue of the former HPI/ Makeup thermal sleeve by re-routing makeup i
flow to a previously unaffected nozzle. The improved control of makeup' flow i
and the increase in the minimum makeup flow provided since the SRF0 are 3
expected to increase the lifetime'of the thermal sleeve in the alternate
~
nozzle.
i a
'Dodkit Number 50-346 3
License Number NPF-3 Serial Number 1802 Page 3 Toledo Edison recognizes the need to assure long term thermal sleeve integrity and vill continue to address this issue.
The improvements made in makeup flow control provide reasonable assurance that thermal sleeve life is greater.than four operating cycles in makeup service.
Toledo Edison therefore concludes that restart'and operation of Davis-Besse Unit 1 for Cycle 7.and beyond is justified.
Toledo Edison' requests NRC approval of operation for Cycle 7 and beyond by May.
30, 1990. Approval by this date is requested to support the anticipated entry-into Mode-4-(Hot-Shutdown).- Should you have any questions concerning this-t' matter, please contact Mr. R. V. Schrauder, Manager --Nuclear Licensing, at (419) 249-2366.
I Very truly yours, M
PVS/ssg Attachment ec:
P. H. Byron, DB-1 NRC Senior Resident Inspector A. B.-Davis, Regional Administrator, NRC Region III T. V. Vambach, DE 1 NRC Sen.ior Project Manager-e
k E*
- Dock 3t Numb 3r 50-346
'Licensa Numb 3r NPF-3' i
. ~, '
Serial. Numbe r 1802 HPI/ MAKEUP N0ZZLE'AND. THERMAL SLEEVE
-REPORT OF ACTIONS TAKEN DURING-THE~ SIXTH REFUELING OUTAGE TOLEDO EDISON COMPANY 8-DAVIS-BESSE NUCLEAR POWER STATION, UNIT 1 APRIL, 1990
'I INTRODUCTION e
During the fifth-refueling outage (5RFO) in.1988, a combined ASME Section i
XI'and pre-fueling visual inspection discovered. loose parts'in ther.
Davis-Besse Nuclear Power Station (DBNPS). Unit 1 reactor vessel which were.'
l later determined to be a portion of the High Pressure Injection-i (HPI)/ Makeup nozzle thermal sleeve. Approximately three-inches of the thermal sleeve had broken off from the end of the sleeve which' protrudes t
1 into the reactor coolant system (RCS) cold leg (Figure I-1).
The absence.
4 of the broken end of the thermal sleeve permitted cold makeup water 'to-impinge direstly on the cladding of the nozzle and the adjacent portion of
.the RCS cold leg piping. Dye Penetrant inspection of the HPI/ Makeup nozzle revealed linear indications in the. area of the nozzle exposed to the cold-makeup water following the thermal sisave failure.
1 A conservative structural analysis' concluded that,. with the thermal sleeve 4
intact, the potential crack growth is acceptable for the 40-year lifetime i
of the unit.1 The failed thermal sleeve was replaced, minimum makeup _ flow-was increased, and control over minimum makeup : flow was improved to reduce the potential for thermal' f atigue of the' thermal -sleeve. the.cause of the failure. Toledo Edison informed the Nuclear Regulatory Commission (NRC) of 1-these actions.(Reference 1).
The NRC approved restart of operation'for one l'
fuel cycle. based on these actions and required Toleto Edison to identify l
follow-up actions planned for the sixth refueling outage (6RFO) in 1990 (Reference 2).
Reference 3. informed the NRC of the planned actions for 6RFO. The planned actions included visual' inspection of the thermal sleeve,. enhanced' ultrasonic examination of the HPI/ Makeup Nozzle, and piping, modifications to utilize an alternate HPI nozzle for. normal makeup flow.
j This report documents the completion of these activities and provides the
[
basis for restart and future operation of Davis-Besse Nuclear Power i
Station, Unit 1.
J WP DZ C/95 1
Docket Number 50-346
'LicGns2 Numb 2r NPF-3 t
a.
Serial Number-1802 L.
HPI/ MAKEUP N0ZZLE AND THERMAL SLEEVE REPORT OF ACTIONS TAKEN DURING THE SIXTN REFUELING OUTAGE f
s TOLEDO EDISON' COMPANY DAVIS-BESSE NUCLEAR POWER STATION.. UNIT.No. 1 I
APRIL 1990.
e L
i I
i i
l t
J s
r l
WP DZ C/95
't i
I
. ~.
il l,,i II h
B i
unw Makeup /HPI Nozzle Original Thermal Sleeve Design
- G
.080" lo.015 Diametrical Clearance Between Thermal Sleswo and Nozzle I.D.
-. Sale Eral
- 2.000" DIA REF.
~
~
'~
%~
Froan lil'l J _.
M 4 th e 1 liard Roll - *--+
4
-.Approxituately 12 //8 IlLF RCS Papeng FIGURE l-1
=.
L I-
,2e Dockot Numb 3r: '50-346-
' License Numb 3r NPF-3:
Serial _ Number 1002 II BACKGROUND Following the 5RF0 discovery of the f ailed HPI/ Makeup nozzle Al thermal sleeve, the f ailed thermal-sleeve and the identical thermal sleeve in HPI nozzle A2, located in the adjacent RCS cold leg, were removed and subjected -
to detailed laboratory examination.
The laboratory examination concluded that:
1 The primary failure mechanism was the result of inside diameter (ID)
]
initiated axial thermal f atigue cracks. ; ' The observed cracks were most i
severe near the outlet end of the1 thermal sleeve where flow turbulence and mixing produced high cyclic thermal stresses. The failure is believed to have resulted from the interaction of the hot, turbulent reactor coolanc water, and the cold makeup water, in the end of the thermal sleeve.
1 The broken thermal-sleeve fragments had resided near the core a -
maximum of 414 days when discovered, placing the earliest time-of f ailure_ at January 1987, during Cycle 5.
The broken thermal sleeve was in service for more than four fuel cycles, prior to failure.
.l There were no flaws in the intact-thermal sleeve removed from HPI-I nozzle A2.
This thermal sleeve had not been subjected lto makeup s e rvice.
The thermal sleeve failure exposed the blend radius area, near the mouth of nozzle Al, to the cold makeup water / hot reactor coolant' interaction _ The resulting thermal cycling is the probable cause of the linear indications j
found in the cladding by the SRF0 dye penetrant _ examination _of this region.
H A manual-ultrasonic examination (UT) found no evidence that the indicated
. flaws had propagated into the base metal. Conservative analyses concluded that significant flaw growth would not be expected with the thermal sleeve
_3 in place.
During SRF0 both thermal sleeves were replaced with the B&W Owners Group
~
design dw eloped in 1984 (F.igure.II-1), which permits insertion from i
-outside ne RCS pipe and provides more structural stability and resistance j
to cracking at the safe end (Reference 1).
LThe new thermal sleeves are the same as the original design at the discharge end and would respond similarly_to'a fluctuatinB temperature field at that location. With an j
intact thermal sleeve in place, the cold makeup flow / hot reactor coolant l
interaction is separated from the nozzle and reactor coolant piping inside surfaces, effectively removing the potential for thermal' cycling of these i
regions.
Toledo Edison committed to increese minimum makeup flow, provide for accurately setting this flow, ar.a to clearly establish administrative control of the flow. The NRC approved' restart from the 5RF0 and operation for one cycle without the cladding indications being removed. As a condition of restart, Toledo Edison committed to continue evaluation and; inform the NRC of plans to re-examine, re-evaluate and repair the nozzle, as required during the 6RFO.
i WP DZ C/95 3
~... _.....
b l
. ~.
I 1
nur
- ";a i=s
Makeup /HPI Nozzle Replacement
![g[
[
Thermal Sleeve Design 370
"?
i i;
~~
I~som IIPI l
j c'
i 1-i
'D 17 5!8' EF j
1 Thermal Sleeve i
l
[
FIGURE 11-1 HCS Piping
Dockot Numb 3r 50-366 license Number NPF-3 Serial Number 1802 III
SUMMARY
OF ACTIONS TAKEN Reference 3 identified the following planned actions for 6RFO:
1)
Modification of piping to re-route normal makeup flow through an alternate HPI nozzle.
2)
Visual inspection of the HPI/ Makeup nozzle thermal sleeve, 3)
Enhanced ultrasonic (UT) examination of the HPI/ Makeup nozzle (A1)-
from the outside diameter, and 4)
Baseline enhanced UT examination of the alternate HPI ncezle (A2).
Additionally, a preliminary design and methodology for external weld overlay reinforcement of the nozzle, and means for removal and replacement of the thermal sleeve were developed, as a contingency, in the event that the planned inspections had revealed additional defects.
t Plant Modification 89-0066 re-routed the normal makeup flow path from HPI/ Makeup nozzle Al to HPI nozzle A2 on the adjacent RCS cold leg (Figure t.II-1 ).
The re-routing ensures that neither the thermal sleeve in nozzle Al nor the nozzle itself will be subjected to thermal conditions related to normal makeup flow. Nozzle,Al will retain HPI duty and HPI nozzle A2 will become the combined HPI/ Makeup nortle.
The' nozzle A2 thermal sleeve was replaced during the 5RFO. The makeup flow path modification has no effect on HPI system operation or performance.
The inside diameter of the HPI/ Makeup nozzle Al thermal sleeve was visually inspected using fiberoptics. The inspection was similar to that performed during the SRF0 following discovery of the thermal sleeve failure (Reference 1).
Indications found included scratches, which are believed to have resulted from attempts to insert tubing through the thermal sleeve to r
drain the RCS piping, and marks from the installation roll process. These indications have been evaluated, and determined not to be detrbmental to the lifetime of the thermal sleeve. No indications of thermal fatigue cracks were found.
Subsequent dye penetrant examination was considered unnecessary.
Toledo Edison in conjunction with B&W Nuclear Service Company developed the enhanced ultrasonic examination (enhanced UT) system which was used during the 6RF0 to assure the structural integrity of the nozzle. Toledo Edison submittals (References 4 and 6) documented the development program and the capabilities of the enhanced UT system.- The enhanced UT system utilizes automated scanning and an enhanced digital data acquisition and analysis system to examine the nozzle and adjacent reactor coolant piping from the outside diameter.
The system was shown to be capable of detecting, locating and sizing thermal fatigue type flaws penetrating through the inner clad surface into the base metal within the regions of the nozzle and adjacent RCS piping, which had been potentially affected by cold water following the thermal sleeve f ailure.
The demonstrated ability to detect all flaws penetrating into the base metal by 0.125 inch or more ensures detection of flaws well below the depth where structural integrity of the nozzle would be compromised.
WP DZ C/95 5
NU/-PI FLOW C0:\\FIGJRATION
~
it : C ?
M h
DfG "bzb B
i i
3 343 2
Ym n?L 5
MU6420
"?-
N 5
m6422 m32 FROM NAKEUP i
REVISED pygg L
FLOW PATH g
p l
ORIGIl4AL
_ e MAKEUP POINT
~b C
N0ZZLE l
LJ ac4crog
/ /-
T ML 5
w a
1 o
o r,
"i d
HP59 HP57 l
HP2A Y
L.0.
o v
O R
W c) s i
m REVISED V/I 7'
HPI i-2 PUNP MAKEUP N0ZZLE HP58 HP56 HP2B22 POINT L U-A2 s,
m, FIGURE lil-1
Dackst Numb 3r 50 346
'Licznso Numb 2r NPF-3
!erial Number 1802
'l The 6RFO inspection plan includeo repeating the 3RFO manual ultrasonic axamination prior to the enhanced UT.
If the enhanced UT. detected flaws in e nozzle the repeated manual UT would assist in dispositioning the flaw 1 pre-existing and not detectable by manual UT or es new. developing l
since the 5RFO. As in the $RFO. the manual UT detected no flaws, f
the enhanced ultrasonic extmAnations detected no flaws in nozzle Al or i
nozzle A2 either in the fladding or penetrating in*,0 the base metal.
longitudinal wave examinations of the clad layer, performed specifically f or the purpose, did not datect any discernible flaw Indications.
This suggests that any cracks represented by the dye penetrant indications found in the $RF0 are confined to the clad layer and are most likely very shallow.
Based upon the enhanced UT results, the structural integrity of the nozzle was again. reviewed-for acceptability for the remainder of the plant design.
life. Toledo Edison contracted Structural Integrity Associates to perform' a fracture mechanics evaluation based upon a 3 D finite-element analysis and an assumed flaw depth based upon.the confirmed limits of detection of-i the enhanced UT system. The methodology la similar to that used to cupport Toledo Edison's request for NRC approval of. weld overlay repair as a contingency (References 4 and 5).
The etructural evaluation of HPI nozzle Al considers those; design transients that are consistent with HP1 duty of the nozzle since makeup flow has been re-routed to an alternate HPI nozzle.
It assumes an intact thermal sleeve and a pre-existing flaw penetration through the cladding and 0.125 inch into the bass metal at the location of maximum stress.
The evaluation concludes that the flaw will propagate less than 20 mils over an additional 40 yeare of service life. The ASHE Code Section XI maximum flaw size is not challenged and ASME structural reinforcement requirements are met with margin.
In summary. Toledo Edison has addressed the impl'ications of the the failed HPI/Hakeup nozzle thermal sleeve discovered during the $RFO. The actions have eliminated cold makeup water as a potential driving force for thermal f atigue of the thermal' sleeve in nozzle Al and.the nozzle itself by re-routing makeup flow to a previously unaffected nozzle. The increased j
minimum makeup flow is also expected to enhance the lifetime of the thermal sleeve in nozzle A2.
Toledo Edison is confident the improvements made in l
makeup flow control provide assurance that thermal' sleeve life is greater l
l than four operating cycles in nakeup service. Toledo Edison therefore concludes the restart and operation of Davis-Beasa Unit'l with the existing
~
l cladding indications in nozzle Al is acceptable for Cycle 7 and beyond, l
+
t i
)
l l
l VP DZ C/95 7
i 5
h
l Dock]t Numb]r 50 346 Liccnso Numb 3r NPF-3 Serial Number 1802
.i IV MODIFICATION OF MAKEUP FLOW PATH During the 6RF0 Toledo Edison re-routed the makeup flow path from the i
combined HPI/ Makeup nozzle Al to HPI Nozzle A2 located in the adjacent RCS I
cold leg. The flow path modification (Reference 7) is shown in Figure III-1.
t The makeup flow path modification was performed to eliminate any possibility of cold makeup flow effects upon the thermal sleeve in nozzle Al or the nozzle itself. Nozzle Al had served in combined makeup and HPI duty since plant operations began in 1977. Ths thermal sleeve in nozzle Al I
was replaced during the 5RFO. and was exposed ;o only a single fuel cycle of operation in combined makeup /HPI duty. As can be seen from Figure III-1, the modified flow path has no effect upon the.high pressure injection system operation.
The physical re-routing of the 2 1/2 inch schedule 160 piping involved the addition of-approximately 7 feet of pipint and fittings in Number 2 Mechanical Penetration Room (Room No. 236). The makeup tie-in downstream of the valve HP2A was capped off and a new connection in the alternate HPI line downstream of valve HF2B was established. All work on this modification is planned to be completed prior to return to power from the 6RFO.
V THERMAL SLEEVE INSPECTION A visual inspection of the inside surface of the thermal sleeve installed in the HPI/ Makeup nozzle Al was performed during 6RF0 (Reference 9).
The inspection was performed using fiberoptic viewing equipment with the results recorded on video tape.
Review of the video tape records by B&W Nuclear Service Company (Reference-10) concluded that 1)
The examination constituted a complete visual inspection of the inside diameter of the thermal sleeve and 2)
There are no abnormalities that could be interpreted as being deleterious to the continued serviceability of the thermal sleeve.
Indications reported from this inspection included primarily longitudinal scratches plus marks apparently from the rolling process used in installation.
These n. arks included occasional smali_ gouges and small rounded depressions. The shallow scratches are believed to have been caused by attempts to force tubing through the thermal sleeve in order to drain a local section of the RCS piping. The indications were subjected to engineering review in accordance with Toledo Edison procedures.
It was concluded that none of the indications were representative of. or precursors to, thermal f atigue cracking.
Based upon these reviews, it was concluded that the thermal sleeve was suitable for return to service, and that a further dye penetrant examination was unnecessary.
WP DZ C/95 8
l
h Sockst Humb2r 50-366 Licsnst Numb?r NPF-3 Serial Humber 1802 Attachment
+
VI H0ZZLE INSPECTIONS i
l' Volumetric. inspections of the HPI/ Makeup nozzle Al and HPI nozzle A2 were planned and executed during the 6RF0 using the enhanced ultrasonic
. examination system discussed earlier.
The enhanced UT system employs automated scanning using a. Puma Model 262 robot to manipulate the transducer along predetermined scanning, trajectories. The ultrasonic data were collected and displayed ~for.
analysis using the B&W Nuclear Service Company ACCUSONEX data acqb.'.sition and imaging system.
The capabilities of the system were enhanced during the development program i
(Reference 6) which selected the scanning techniques and' confirmed the required resolution. The resolution was demonstrated by using a series of test blocks and a replica mockup of the nozzle in which both cracks and' narrow notches simulating cracks were installed. Blind tests were-also incorporated in the development program. The development program proved the system's capability to detect, locate, and size flaws in the zone of concern in the HPI/ makeup nozzle and in adjacent portions of the reactor coolant pipe. This zone of concern encompassed the regions potentially-s exposed to cold makeup water following the thermal sleeve failure discovered during the 5RFO.
The enhanced UT system was demonstrated to have high resolution for detection of small flaws penetrating into the base metal. The system can reliably detect and locate flaws witbin the zone of concern which penetrate into base metal by 0.125 inch or more.. fhis detection threshold is well below the depth where structural integrity of the nozzle would~be compromised.
Reference 8 is a detailed report of the ultrasonic examinations of HPI/ makeup nozzle Al and EPI nozzle A2.
Manual Scauning of Nozzle A1 Prior to conducting the automated scanning. a manual examination was.
performed in an attempt to duplicate the manual examination performed during the 5RFC. No recordable indications were detected during this repeat examination, matching the results of the last examination. All of the manual scans were performed in the same manner as the previous scans with the exception of the 35 degree longitudinal wave scan from the outside diameter blend radius. This particular scan was omitted because it had been determined during the enhanced UT development program to be ineffective for detecting cracking in the nozzle inside blend radius region.
b WP DZ C/95 9
k n
d Dockat Numb 3r 50-346
' liccnse Numb 3r NPF-3 Serial Number 1802 Enhanced UT of Nozzle A1 The examinations performed on nozzle Al using the enhanced UT system were organized in several different scanning patterns.
These patterns were designed to detect both radial and circunferential cracking in the zone of concern in the nozzle bore. including the inner radius and extending around the reactor coolant pipe to a distance of 8 inches from the nozzle centerline.
The scan types included axial scans from the nozzle taper, and radial, circular, and tangent scans from the reactor coolant piping surface.
Figure VI-1 illustrates the HPI nozzle examination region.
Figures VI-2 through VI-7 illustrate the various scanning patterns.
During the automated scanning, the transducer is indexed approximately every 0.050 inch between traverses.
The axial taper scans, (Figures VI-2 & VI-3), are capable of detecting circumferential flaws in the nozzle bore. These scans were performed in segnents that together extended 360 degrees around the nozzle for complete coverage. A 45 degree shear wave,1.5 MHz, 0.375 inch transducer was used for this examination. An examination with a 45 degree longitudinal wave, 1.5 MHz, 0.375 inch transducer was also used to supplement the shear wave examination to enhance detection of flaws contained in the cladding.
The radial scans performed from the reactor coolant pipe surf ace are capable of detecting circumferential1y oriented flaws on the pipe inside surface. (Figure VI-5).
These scans were performed in segments that together. extended 360 degrees around the nozzle for complete coverage. A 45 degree shear wave, 1.5 MHz, 0.375 inch diameter transducer was used for this examination. These scans covered the region extending from the nozzle inside blend radius back to eight inches from the nozzle centerline.
The circular scans performed from the reactor coolant pipe surface are capable of detecting radially oriented flaws on the pipe inside surface.
(Figure VI-4). These scans were performed in segments that together extended 360 degrees around the nozzle for complete coverage. A 45 degree shear wave, 1.5 MHz, 0.375 inch diameter transducer was used for this examination. These scans cover an area extending from slightly under the 0.D. blend radius to 8 inches from the nozzle centerline.
The tangent scans performed from the reactor coolant pipe surface are capable of detecting circumferentially and radially oriented flaws on the pipe inside surface extending from the nozzle'inside blend radius to a point beneath the tangency point of the outside blend radius back to eight inches from the nozzle centerline. (See Figures VI-6 and VI-7.)
In addition, this scan is capable of detecting axial flaws in the nozzle bore.
I The tangent scans were performed by directing the transducer along parallel paths which are tangent to the nozzle bore at increments of 30 degrees.
The scanning pattern is illustrated in Figures VI-6 and VI-7.. Each of the j
scanning increments are eslied rotation angles, i.e.,
30, 60, degrees, etc.
Three different transducer angles are required at each rotation angle for WP DZ C/95 10 I
i
.,Dockot Numb 2r 50-346 a
Licensa Numb 3r NPF-3 i
Serial Number 1802 maximum coverage.
These angles' include 57 66 and 75 degrees. The transducer angles were generally used with the shear wave mode at each rotation angle around the nozzle. The 240 degree angle proved the exception since access was restricted due to interferences with the reactor coolant pipe whip restraint and the HPI/ makeup line check valve. For this case, the scan plan was modified by performing a radial scan using a 66 degree shear wave over the range of 220 to 255 degrees to detect circumferentially oriented flaws on the pipe side in this region. The scan plan was also modified to ensur'e detection of axial flaws in the nozzle, which would otherwise have been detected using the tangent scan. To J
provide the required coverage, the 57. 66 and 75 degree transducers were scanned at the 60 degree rotation angle with the scan offset to cover the 330 degree side of the nozzle. These supplemental. scans provide the same coverage and detection capabilities that the tangent scan at the 240 degree angle was intended to provide.
In addition to the shear wave scans, the downstream half of the nozzle (relative to reactor coolant flow) was examined using' tangent scans with I
57, 56 and 75 degree longitudinal waves to enhance detection of flaws contained in the cladding layer.. This region was believed to have the d
highest probability of having any significant cracking due to the interaction of hot reactor coolant and cold makeup-flow.
The longitudinal wave examination did not detect any discernible flaw indications.
Therefore, the flaws previously identified by dye penetrant testing during the SRF0 are believed to be confined to the clad layer and are most likely very shallow.
Results of Enhanced UT of Nozzle Al e
There were no indications detected with any of the scans which are considered to be service induced. This applies to both the examinations with the shear wave and longitudinal wave modes.
The only significant indications detected were in the nozzle-to-reactor coolant pipe weld.
These were small and volumetric in nature. Of these, there were three veld indications which exceeded 202 Distance Amplitude Correction (DAC) level based upon the calibrations established on the ASME calibration block for the reactor coolant piping.
The ASME Section XI l
recording threshold is 50% of DAC.
These indications had peak amplitude readings of 237 272. and 822 of DAC and were determined to be acceptable.
One of the segments in the axial taper scans which was scanned with the 45 degree longitudinal wave did show indications at the clad layer. However, these indications were determined to be non-planar and were actually.
detected with the 22 degree shear wave which accompanies the 65_ degree longitudinal wave.
These indications are typical of those from the clad l
layer detected with shallow angle shear waves at high gain levels. While the exact source of the reflections is unknown. possibilities include small inclusions in the clad layer..the metallurgical interface between the base l
metal and the clad layer, et scoustic variations in the clad layer itself.
(
It is noteworthy that these indications are located several inches from and are unrelated to the cladding flaws detected by dye penetrant testing in the $RFO.
The reason that the indications only appeared in one scan segment is that a higher gain setting was used to acquire data for this i
segment than the other segments.
WP DZ C/95 11 e
t w
mem
- ~ ~...
r
' Docket Nustbar 50 366 License Number NPT.3 Serial Number '1802 e
h HPI' NOZZLE EXAMINATION. REGION h
at Prat f
i i
1 L
t tasan evu
.a,.
s
._ N i
a, J
i I
l
/
t
/
7 j
..../-
( tenna, m.s.vu cou.am
+
i t
/
/
/
5 s.,
/
/
/
/
/
I RC P!85 I
J 4
t I
F i
s J
4 4
4 FIGURE VI-1 k
12 3
i
~
' Docket Numb 3r 50 346 License Numb 2r NPF-3
~ '( ~7 "*
- p ~" 'O O,* '
" * * 'q Serial Number 1802 i
1
( For o.t.ction of Circumf.r.ntisk F1.w. In sr.. Bor.)
l s:AN CQVERASE e~~,~
3
(
i l
I l
I 1
I scaw PATTenn l
l l.
i I
1 p
rsnien onesense l
I steet cassence J
J L' s,e,,.
ini-.n..
I An.ular!
l Increment r
l i
1 4
4 so. in.1..
in.i.<. >,
4.
u.v. n...
sn..e o.s..uon Tr.n..u..e c
os
.e
- u. inon rr..u.n.y 1.s -
FIGURE Vl-2
~
1
- e 13 1
' Docket Numbar 50-346 License Numb 3r NPF-3 Serial Number 1802 a, X.L, L
.N u O,.,c.8 A G t ocu
/
FROM NOZZLE
/
s.,N s
SIZING SCANS et et
_I
\\'.,'
D l
6' \\
)
~
95 43 DETECTION SCAN 1
---(
l
's l
4 FIGURE VI,3.
14
e
' Dock 3t Numb 2r 50 346 7
t,1csnss Number NPF-3 Serial Number 1002 CIRCULAR SCAN 1
From RO P1De Surface
( Pipe Sice Aacial Flaws )
Restel SCAN PATTERN increment /
F1nten Angle A'
TANoENCY POINT
~~~,
/
SNoutoER CYLIWER i
- SCAN PATTERN
\\
/
s.
^
w,==
Redtel Steet s
st4N COYEMAst i
e Start Angie Asesel Finten-s e
Nootnal Bose Angle ( s )
45 Wave Mode Sheer.
1 Detection Trenecucer Dieseter i
3.'8 inch
)
Frequency 1.5 MHz Siting Treneeucer 01eeeter 3/8 inch Frequency 5 MHz Siting Technicus.
Tip Diffraction FIGURE VI-4 15 A.
' Docket Numb 3r 50-346 i
Licenso Number NPF-3 Serial Number 1802 qg q
3 w
c 3
v
- - s' From AC Pipe Surface
?
- Pice Side Circumferential FloweJ s
/
4 N _,tet
.e
/
SCAN f2ERN f
/
TANSFNCY F0!NT 3
3,p Zi m t p
SNOULDEA CYLINORA
..e.
8
'**SCLN PATTRAN s
i
~.
\\
.E i
l
\\
s s
Restel Start SCAN COVERAGE Start Angle i
moetel,tatem i
Noetnita Bees Anglet e )
45 Weve Hoce Shece Detection Treneoucee Oletteier a
3/8 inen i
Frequency 1.5 MHz Sizing Trenaeuche l
01eestem 3/8 inch l
1 Frequency 5 MHz Sizing Technteve Tip Oltreaction
+
FIGURE.VI t r
1 16
. i
\\-
-n..
Dock?t Numb 3r 50 366 Licsnso Numb 3r NPF.3 Serial Number 1802 TANGENT SCAN l
I t
[
(Axlel Flows in the N0Z21e Scre)
( Recial one.Carcumferentiel Flows in the Knuckle)'
'h i
i a
TANoENCY POINT
~~%
l
./
SNOUL.0ER N
/
g SCAN COVERAGE
/
CYLINDER g
DORE
\\
1 g
\\
/
\\
/
g 81nten meetWe
\\
/
+
/
\\e
\\
/
N
/
.t.,.eesue
/
s
/
i s
,, i.4 # /
/
i
\\
s,.
s
/
N
- Finton 01stense Storat Dietence -
/
i I
cistenee
(.
scan PATTERN Ineeseent L
Noetnel Se88 A"91 *I "
veve MOOe Sheer Detection Trenoeucer 8
Dimeeter i
1/2 inch L
Frequency 1.5 MHz i
S1 zing Trenaeucer.
1 Dioetter t
1/2 inch i
Frequency 1 5 MMt l
l FIGURE VI-6 I
t 17 i
\\
. _[.,. -,
i k --
ki.
-'~'"
" ~
IL_
i NECS
- *1 " ?
nea.
TANGENT SCAN COVERAGE
!;~;
asrs Y Pr i nen 0;43 37L VERTICAL SECTION HORIZOf JT AL SEC11014 G
(
\\
T A r4 G E N C Y POIt4 T E
TANSENCY POINT su X-m y-r T-2.-
I q
g
/
/
au
.7*
ew BW = BEAM WIDTH zw-m.
.m_
ww.=
w m
,.rmmw.#J-.i
_.--m-
a.a.--rew'*--
Docket Numbor 50-346 License Numbor NPF-3 Serial Number 1802 It was concluded that any indications in the cladding discovered during the
$RF0 following failure of the HPI/ makeup thermal sleeve had not reached the clad-base metal interface in the regions examined.
The use of the enhanced ultrasonic examination techniques add confidence that there has been no significant growth of any flaws which might have been initir ed following the thermal sleeve failure. The ability of the enhanced UT s stem to resolve any flaws penetrating more than 1/8th inch into the base metal offers a conservative bound for an assumed flaw depth for re-evaluation of flaw growth potential in future operation.
Baseline Examination of Nozzle A2 HPI nozzle A2 will become the normal makeup flow path following 6RF0 as a result of the modification to re-route the normal makeup flow path. A baseline examination of this nozzle was performed using the enhanced UT system in a manner similar to that performed for nozzle A1.
As for nozzle A1, no service related indications were detected in nozzle A2.
VII N0ZZLE STRUCTURAL REVIEWS Although the enhanced ultrasonic examination showed no evidence of flaws penetrating into the nozzle or reactor coolant pipe base metal, a conservative review of the potential crack growth rate was made using fracture mecnanics techniques.
The analytical methods used were improved over those used in the evaluations submitted folloving discovery of the thermal sleeve failure in 1988 (Reference 1).
The new considerat'ons include the following:
The HPI/ Makeup nozzle Al will no longer serve the combined functions for makeup and high pressure injection. Due to the re-routing of the normal makeup flow path, the nozzle will serve in HPI duty only. The evsloation of future flaw growth takes into account the functional transients associated with this duty.
The peometry of the nozzle was determined to be slightly different than shown on the original B&W drawing.- As part of the development of the enhanced ultrasonic examination system the cladding was found to be thicker in some locatiens and to have a sharper inside blend radius. These cladding dimensional changes have been taken into account. Figure VII-1 illustrates these cladding dimensions.
The development of the enhanced UT system resulted in an improved resolution of small flaws penetrating the cladding into the base metal. Although no flaws were detected in the base metal in nozzle A1, the initial flaw depth assumed'was that depth at which there is full confidence of detection (Flaw depth extending 0.125 inch into the base metal).
q l
WP DZ Clo5 19
F Dockot Number 50-346 Liccnse Number NPF-3 Serial Number 1802 Stress Analysis A fracture mechanics analysis based upon a finite element stress analysis of the nozzle was performed by Structural Integrity Associates ($1A), and is described in Reference 11.
Through wall stresses were determined using the finite element model and the appropriate combination of internal pressure and HFI initiation thermal transients.
The fracture mechanics evaluation was performed using the linear elastic fracture mechanics. option of the pc-Crack computer code. The two aspects of the evaluation include allowable flaw size determination and crack growth evaluation.
Minimum Flaw Si e t
The allowable flaw site evaluation was based upon a fracture mechanics model of the nottle assuming a semicircular flaw penetrating 0.125 inch into the base metal at the location of maximum stress. The material fracture toughness for.the ASTH A105 Carbon Steel nottle base material was taken to be 200 ksi-Vinch divided by a safety factor of square root of 10 or 63.2 ksi. Vinch. Based upon this material fracture toughness, the allowable flaw depth in all cases was found to be through wall.
Patiaue crack krowth Fatigue crack growth analysis was also provided using pc-Crack and the bi'.inear law in Appendix A. ASME Code Section X1, 1977, 1978. Summer Addenda.
An initial crack si.te conservatively represeating the maximum cladding thickness plus 0.125 inch base metal crack penetration, was assumed for the analysis as illustrated in Figure VII-1.
The number of startup and shutdown cycles for the 60-year design life was assumed to be 240, and the number of HPI injection transients was assumed to be 80.
The resultant crack growth at the most structurally limiting location was determined to be less than 20 mils for an. additional 40. years of life.
Structur:1 Review Conclusions The f racture mechanics analysis rertGts show that flaws substantially larger than that conservatively ~ assumed can be tolerated without challenging the ASNE Code.Section XI allowable flaw site.
Additionally, a very large margin remains for flaw growth before the minimum nozzle reinforcement required by ASME Code,Section III, would be reached.
- WP DE C/95 20 1
Dbcket Numb 3r 50 366 i
Liesnso Numb]r NPF-3 Serial Number 1802 8
The ASME Code,Section III, reinforcement requirements are based solely upon the carbon steel base metal.
(Cladding is not considered part of the Code pressuro boundary.) The initial flaw depth appropriate to this concern is therefore just the 0.125 inch assumed penetration into the base metal..On this basis, reviews of the HPI nozzle have shown that a fracture zone defined within the carbon steel base metal is acceptable without i
challenging minimum reinforcement requirements (Reference 5).
With less the 20 mils growth projected for an additional 40 years of life, & very substantial margin remains before the ASKI code reinforcement limit coulo be reached.
Thus, the calculated flaw growth rate shows large structural margins remain to accommodate operation /for the remainder of the plant operating lifetime.
VP CZ C/95-21 1
y;
~
l
' Docket Numb 3r 50 346
?
License Number NPF-3 Serial Number 1802 POTENTI AL FLAW DEPTH CONSIDERING LIMITS OF DETECTION-ENHANCED UT SYSTEM REACTOR COOLANT P!PE BORE SEMI-CIRCULAR CRAC 7
65 INCH 0
PROFILE-1.0 INCH MAXIM' M PENETRATION J
ym l.0 INCH t
SEMI-CIRCULAR CRAC N LOCATION OF ELO MAXIMUM STRESS N0ZZLE k
BASE
. METAL 0.125 INCHN UT DETECTION LIMIT FOR FLAWS PENETRATING INTO CLAD DEPTH MEASUREMENTS.
N0ZZLE BASE METAL
~~
\\
- 0.20 INCH THICKNESS =0.65 INCH 1
r g
\\
AT N0ZZLE S!!0ULDER REGION:
m THICKNESS =0.20 INCH u
h 5
FIGURE Vll-1 i
I I
22 e-
._____._,......-,_<.._.,..,,..,.,-.....-,-.._-...r
.e Docket Numb;r 50 346 j
9
, e, Licznse Numb 2r NPF-3 j
7, Serial Number 1802
]
Attachment I l
i VIII CONCLUSIONS AND PLANS FOR THE 7TH FUEL CYCLE j
The program of inspections of the HPI/ Makeup nozzle Al and its thermal sleeve have shown the thermal sleeve to be intact and in place, and the nozzle to be capable of service for Cycle 7 and beyond. No indications of service related flaws have been found in either component.
The enhanced ultrasonic examination of the nozzle was qualified specifically for the purpose in an enhanced UT development program, resulting in confidence et detecting. locating and sizing if necessary, any flaws penetrating from the inside clad surface 0.125 inch or more into the nozzle base metal. The lack of any indications of such flaws provides confidence that processes are not at work with the thermal sleeve in place j
which would cause rapid growth of flaws which might have been initiated due to exposure of the nozzle blend' radius region to. cold makeup water following the thermal sleeve failure discovered in July, 1988.
Conservative fracture mechanics analysis, using improved finite element-models and taking into account improved information on the configuration of the nozzle as installed, has demonstrated a much larger than previously estimated ASME Section XI allowable crack size, which is now understood'to be essentially through wall.
Conservative crack growth rates predicted for nozzle A1. in its planned role for high pressure injection service only, show flaw growth for an additional 40 years of service to be less than 20 mils. This forecast crack growth offers no challenge to either the allowable flaw size or to the minimum reinforcement requirements of Section III of the ASME Code.
The normal malvup flow path nas been re-routed from the original HPI/ Makeup nottle Al to c"3tle A2 in the adjacent reactor coolant loop cold leg piping.
Th>; re-routing removes any potential future thermal effects of cold makeup vater on the makeup thermal sleeve in nozzle A1. extending the lifetime of this thermal sleeve indefsnitely.
Use of nozzle A2 for future normal makeup duty, provides essentially a new thermal sleeve, and offers a maximum life expectancy in normal makeup duty. The future life expectancy of this thermal sleeve, considering the increased attention placed upon maintenance of a higher minimum makeup water flow rate and stabilization of makeup flow from the 5RFO. is expected to easily exceed the four full fuel cycles operation experienced before failure of the original thermal sleeve.
The Toledo Edison Company is continuing to investigate' mechanisms which affect the lifetime of the makeup thermal sleeve and alternatives which might be pursued to ensure long-term reliable-operation. Toledo Edison's j
goal is to arrive at a practical long-term solution which can be implemented by the ninth refueling outage.
l WP DZ C/95 23
. w 4.
JC Docket Number 50-346 6
License Number NPF-3 Serial Number 1802 The existing HPIiMakeup nozzle A1. was examined ultrasonically during the-SRro, and has been examined again with the enhanced UT system developed specifically for the purpose.
In neither case were there service related flaws detected in the nozzle base material.
Toledo Edison plans to re-examine nozzle Al at the next refueling outage using the enhanced UT system. With the thermal sleeve in place. there is no known mechanism for significant thermal cycling leading to thermal fatigue of the nozzle, and crack growth rates due to anticipated design transients have been shown by analysis.to be minimal. Should Toledo Edisen find no indication of service-induced flaws in the nozzle base materici in the next examination.
It will be concluded that the failure of the thermal sleeve discovered in the $RF0 had a minimal impact on the nozzle structure. No further inspection would be planned for nozzle 1.1 until the next 10-year interval in-service inspection. If at that stime, no service-related indications are found, no further routine ultiasonic examinations would be performed.
The Enhanced UT System was also utilized to obtain baseline data for HPI nozzle A2.
As expected, considering the operating history of this HPI nozzle, no indications of flaws penetrating the clad layer were found.
No further ultrasonic examination of nozzle A2 is planned.
VP DZ C/95 24
- o dockst Numb 2r 50-346
- ~
+.4 Liccus? Numb 2r NFF-3 e
Serial Number 1802 IX REFERENCES 1)
Letter from Toledo Edison Company to the United States Nuclear Regulatory Commission. Docket No. 50-346. License No. NPF-3. Serial Number 1580, dated September 14, 1988. "High Pressure Injection / Makeup Nozzle Thermal Sleeve."
2)
Letter from NRC to Toledo Edison dated October 4, 1988, Log Number 2725. 'High Pressure Injection / Makeup Nozzle and Thermal Sleeve in Davis-Besse Nuclear Power Station. Unit 1. Toledo Edison Company.'
3)
Letter from Toledo Edison Ccmpany to tbn United States Nuclear Regulatory Commission. Docket No. 50-346. License No..NPF-3.
- t-isi Number 1664, dated June 19. 1989 "High Pressure Injectionit:akeup Nozzle and Thermal Sleeve."
4)
Letter from the Toledo Edison Company to the United States Nuclear Regulatory Commission. Docket Number 50-346. License Number NPF-3 Serial Number 1768, dated February 20. 1990. "High Pressure Injection / Makeup Nozzle and Thermal Sleeve Program."
5)
Letter from Toledo Edison Company to the United States Nuclear Regulatory Commission. Docket No. 50-346. License No. NPF-3. Serial Number 1726 dated November 8, 1989. ' Proposed Alternative to ASME Section XI: Makeup /High Pressure Injection Nottle Potential Veld overlay."
l 6)
Letter from Toledo Edison Compe.ny to the United States Nuclear Regulatory Commission. Docket No. 50-346. License No. NPF-3 Serial Number 1778. dated April 16, 1990. " Development of Enhanced Ultrasonic Examination System for High Pressure Injection Nozzles."
l 7)
Modification Package 89-0066. " Change Normal Makeup Flowpath from RCS
[
l Loop 2-1 to RCS Loop 2-2".
8)
Inspection Report HPI Nozzle HP58 and 59 Enhanced UT Examination Report.
9)
Inspection Report: Visual / Surface Data HP59 Thermal Sleeve. Document Identifier 51-1178912-00 dated March 29, 1990.
- 10) Letter from R. C. Mason, Jr., B&W Nuclear Service Company, to John Holden. Toledo Edison Company. dated March 24, 1990 ' Review of HP59 Thermal Sleeve Video Tapes. "
- 11) Report No. SIR-90-032. April. 1990 " Fracture Mechanics Evaluation of Davis-Besse HPI Nozzle Considering Increased Clad Thickness and a Sharp Blend Radius Corner." Structural Integrity Associates. San Jose, CA.
WP DZ C/95 25