ML20078Q554
| ML20078Q554 | |
| Person / Time | |
|---|---|
| Site: | Quad Cities  |
| Issue date: | 12/14/1994 |
| From: | Schrage J COMMONWEALTH EDISON CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), Office of Nuclear Reactor Regulation |
| References | |
| RTR-NUREG-0619, RTR-NUREG-619 NUDOCS 9412220244 | |
| Download: ML20078Q554 (7) | |
Text
-
-)
.E'N Commonw:alth Edison
[
)- 1400 Opus P! cs Downers Grove, Illinois 60515
\ O 7
.O December 14, 1994 )
Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Document Control Desk ;
Subject:
Quad Cities Station Units 1 and 2 NUREG-0619 Feedwater Nozzle Inspections NRC Docket Nos. 50-254 and 50-265 I
References:
(1) R. Janacek to D. Eisenhut letter dated January 22,1981 (2) R. Janacek to D. Eisenhut letter dated February 23,1981 (3) T. Rausch to D. Eisenhut letter dated October 6,1981 (4) T. Rausch to D. Eisenhut letter dated November 6,1981 (5) T. Rausch to J. Keppler letter dated November 20,1981 Gentlemen and Ladies:
In References (1) through (5), Commonwealth Edison (Comed) committed to various actions at Quad Cities Station Units I and 2 in order to resolve the concerns identified in NUREG-0619 <
"BWR Feedwater Nozzle and Control Rod Drive Return Line Nozzle Cracking." These committed actions included both plant modifications and examinations.
The examinations committed to by Comed at Quad Cities Station were in compliance with !
the examination schedule prescribed in Section 4.3.2, Table 2 of NUREG-0619. This table was used to specify examination frequencies for the visual examination of the feedwater system sparger, and the liquid dye penetrant testing (PT) and ultrasonic testing (UT) of the feedwater nozzle blend radius and ;
bore at Quad Cities Units 1 and 2, and Dresden Units 2 and 3. Based upon this table, Comed is required to perform a PT of the feedwater nozzle blend radius and bore during upcoming refuel l I
outages at Quad Cities Station (QlRIS and Q2R13, currently scheduled for March 1998 and March 1995, respectively).
The purpose of this letter is to request NRC approval to use the enhanced, automated UT technique to examine the feedwater nozzles during the upcoming refuel outages (QlRIS and Q2R13. l anmquarwnnumem riec.wpr
)
I 941222O244 941214 l PDR ADOCK 05000254 q'k'I )
P PDR
t USNRC December 14, 1994 in lieu of the PT required by Section 4.3.2, Table 2 of NUREG-0619. The use of the enhanced, automated UT from the nozzle outer surface, instead of the PT from the inside of the reactor vessel, will avoid significant radiation exposure associated with the removal of the feedwater spargers and performance of the required PT. The technical basis and justification for the use of the enhanced, automated UT technique to examine the feedwater nozzles, in lieu of the PT required by NUREG-0619, is described in the Attachment to this letter. Based upon the schedule for the start of Q2R13, Comed respectfully requests approval to use the automated UT technique in lieu of the PT required by Section 4.3.2, Table 2 of NUREG-0619 at Quad Cities Station by March 5,1995.
To the best of my knowledge and belief, the statements contained herein are true and correct, in some respects, these statements are not based on my personal knowledge but obtained information furnished by other Commonwealth Edison employees and consultants. Such information has been reviewed in accordance with Company practices and I believe it to be reliable.
We trust that the information is satisfactory; however, should you have any questions or desire any additional information on this issue, please do not hesitate to contact this office.
Sincerely, A / a~-
ohn L. Schrage Nuclear Licensing Administrator Attachment cc: J. Martin, Regional Administrator - Rlli C. Miller, Senior Resident Inspector - Quad Cities R. Pulsifer, Project Manager - NRR Office of Nuclear Facility Safety - IDNS i
l k :\nla\ quad \fw mu21e\cov rict2, wpf
I ATTACHMENT QUAD CITIES STATION UNITS 1 AND 2 NUREG-0619 EXAMINATION REQUIREMENTS FEEDWATER NOZZLES PURPOSE In response to concerns identified in NUREG-0619. "BWR Feedwater Nozzle and Control Rod Drive Return Line Nozzle Cracking," Commonwealth Edison Company (Comed) committed to implement the examination schedule prescribed in Section 4.3.2, Table 2 of NUREG-0619. This table was used to specify examination frequencies for the visual examination of the feedwater system sparger, and the liquid dye penetrant testing (PT) and ultrasonic testing (UT) of the feedwater nozzle blend radius and bore at Quad Cities Units I and 2, and Dresden Units 2 and 3. Based upon this table, Comed is required to perform a PT of the feedwater nozzle blend radius and bore during upcoming refuel outages at Quad Cities Station (QlRIS and Q2R13, currently scheduled for March 1998 and March 1995, respectively).
The purpose of this document is to request NRC approval to use the enhanced, automated UT technique to examine the feedwater nozzles, in lieu of the PT required by Section 4.3.2, Table 2 of NUREG-0619. The use of the enhanced, automated UT from the nozzle outer surface, instead of the PT from the inside of the reactor vessel, will avoid incurring significant radiation exposure associated with the removal of the feedwater spargers and performance of the required PT. It is also Comed's understanding that the Boiling Water Reactor Owners' Group (BWROG) is preparing a generic submittal to the NRC Staff proposing the substitution of PT with enhanced UT along with reduced UT frequencies.
TECilNICAL JUSTIFICATION IIistorical examination results of feedwater nozzles to date have indicated that thermal cracking is no longer an issue at BWRs. Since the issuance of NUREG-0619, no incidents of feedwater nozzle inner blend radius or nozzle bore cracking have been reported at BWR plants where vessel cladding in the nozzle areas was removed. The fact that no relevant UT indications have been detected at Quad i Cities or Dresden Stations in the nozzle blend radius and bore areas (since the installation of the GE- i designed, triple-sleeve, double-piston thermal sleeve) reinforces the above industry findings. The I results of periodic manual UT conducted at Quad Cities Units 1 and 2 since the replacement of the feedwater thermal sleeve are listed in Table 1.
Status of NUREG-0619 Commitments in addition to the examinations required by NUREG-0619, Comed has completed the following actions at Quad Cities Station, Units 1 and 2, as described in References (a) and (b).
- 1. Comed removed the nozzle blend radius cladding during the Unit 1, Fall 1982 refuel outage 1 1
(QlR6) and during the Unit 2 Winter 1979/1980 refuel outage (Q2R4) (Reference (a)).
- 2. Comed iastalled the General Electric (GE)-designed triple-sleeve double-piston sparger during QlR6 and Q2R4 (Reference (a)).
l
.l.
L Anla\quadifw mmic\ temp.wpf
- 3. Comed evaluated the operation of the low-flow controller. Based upon this evaluation, Comed concluded that the existing equipment and operating procedures are adequate to minimize the crack growth concerns of NUREG-0619 (Reference (c)).
- 4. Comed evaluated the rerouting of the Reactor Water Cleanup (RWCU) discharge piping.
Based upon this evaluation, Comed concluded that the marginal gain achieved by this modification did not warrant the high costs required to implemented it (Reference (a)).
Section 4.7.2 of NEDE-21821-A shows that rerouting the RWCU discharge piping results in only a negligible usage factor improvement with a non-leaking sparger. The installation of '
online leakage monitors at Quad Cities Station, Units 1 and 2, assures early detection of significant seal leakage, thereby minimizing any usage factor improvement associated with the RWCU discharge piping rerouting. The NRC reviewed and approved Comed's evaluation of the potential RWCU discharge piping reroute in Reference (d).
- 5. Manual UT of each nozzle blend radius and bore areas has been performed every second refuel outage. These UT's have not detected any relevant indications.
Discussion of Examination Techniques For plants with triple-sleeve double piston spargers, NUREG-0619 specifies routine PT examination every ninth refueling cycle (or 135 startup/ shutdown cycles), in addition to UT examination every second refueling cycle, to monitor small thermal fatigue crack initiation and growth. The accuracy and repeatability of the UT technology at the time the NUREG was issued was undefined. In addition, the NUREG states that "should future developments and the results of inservice UT examinations demonstrate that UT techniques can detect small thermal fatigue cracks with acceptable reliability and consistency, these techniques could form the basis for the modification of the inspection criteria."
Since the issuance of NUREG-0619, improvements in UT, both manual and automated, have occurred. Currently available automated UT techniques are now capable of detecting and sizing small fatigue cracks (0.25" deep or less) with reliability and consistency. This detection limit is l adequate, based upon the expectation that high-cycle-mixing shallow cracks will progress very i quickly to the depth of 0.25" deep, at which poir.t crack propagation is expected to attenuate out. l Attempts to detect flaws that are less than 0.25" deep do not provide significant safety margins. I The PT technique is not the most optimal examination technique, given the current state of examination technology. Implementation of the PT examination technique, in accordance with ,
NUREG-0619, will result in substantial radiation exposure to plant personnel. Removal of a '
feedwater sparger from the nozzle will be necessary, and PT must be performed on the inner surface of the nozzle. If cracks are detected, the remaining three spargers must be removed and examined. The task of removing sparger(s) to gain access for the PT will likely involve other '
activities such as draindown of the reactor vessel below the feedwater nozzles, installation of a temporary work platform and shielding, surface decontamination and preparation, sparger(s) replacement, and removal of temporary work platform and shielding. Removal and replacement ,
of spargers and thermal sleeves for PT will also increase outage duration and cost. During the l l
Spring 1978 Unit 2 refuel outage at Quad Cities Station (Q2R3), approximately 126 person-rem j was expended to perform a PT examination on the accessible areas of three feedwater nozzles, and the entire bore plus inner radius areas of the fourth nozzle with the " interference fit" sparger <
removed. l 2-knnla\quaefwnonlehemp.wpf l
i i
Based upon the acceptability of the enhanced / improved UT technique, the high radiation exposure associated with PT examination of the feedwater nozzles, and good historical examination results, Comed has concluded that enhanced UT examination is acceptable and the preferred alternative to the PT examination. Therefore, Comed requests NRC approval to implement the enhanced UT in lieu of the required PT during QlR15 and Q2R13.
Description of Alternative Actions in lieu of performing periodic manual UT and PT in accordance with NUREG-0619. Section 4.3.2, Table 2, Comed proposes the following examination plan:
- 1. Perform an enhanced automated UT on the nozzle blend radius and bore areas during refuel outages QlRIS and Q2R13. The enhanced automated UT will be capable of accurately and reliably detecting and sizing flaws less than or equal to 0.25" deep. Comed intends to utilize the GE GERIS-2000 UT system for this enhanced UT during refuel outage Q2R13.
The GERIS-2000 system has been qualified on full scale mockups of BWR feedwater nozzles that contained notches of various sizes and fatigue crack implants. Computer modeling will be performed to assure that the specific configuration of the Quad Cities' feedwater nozzles is within the bound of the aforementioned qualification. The enclosed GE report, "GERIS-2(XX1 ULTRASONIC 1hSPECTION OF FEEMif TER NOZZLES" (Enclosure), describes the GERIS-2000 system in more detail. The GERIS-2000 has been i approved for use on NUREG-0619 feedwater nozzle examinations at Plant llatch, Units 1 i and 2. )
l
- 2. Complete a plant specific fracture mechanics assessment to verify that an assumed crack of i 0.25" deep, which is the current conservative detection limit of an automated UT system, I would not grow to exceed the allowable crack depth for the remainder life of the units.
This assessment will be completed prior to refueling outage Q2R13.
- 3. In the event relevant indications are discovered, an engineering evaluation, which includes cracks characteristics and growth rate will be used to determine unit operability. For subsequent examination frequencies, Quad Cities Station intends to follow the recommendations of the pending submittal from BWROG to the NRC Staff, " Alternate BWR Feedwater Nozzle Inspection Requirements." This submittal will provide generic basis for the substitution of the PT specified by NUREG-0619 with enhanced UT and the reduction in examination frequencies.
l REFERENCES l l
(a) R. Janecek (Comed) letter to D. Eisenhut (NRC), dated February 23,1981.
(b) T. Rausch (Comed) letter to D. Eisenhut (NRC), dated November 6,1981.
l (c) NUTECll Report 'Feedwater Nozzle Leakage Assessment For Dresden Units 2 & 3 and Quad Cities Units 1 & 2", dated December 1984 (COM-12-304 and COM-22-301)
(d) D. Vassallo (NRC) letter to D. Farrar, dated April 16, 1984.
3-knrda\ quad.fu rmle\ temp.wpf
TABLE 1 FEEDWATER NOZZLE IRS & BORE MANUAL UT RESULTS UNIT COMPONENT YEAR EXAM. RESULTS 1 N4A,N4B,N4C,N4D 1982 No Relevant Indications (Baseline) i N4A,N4B,N4C,N4D 1986 No Relevant Indications N4A,N48,N4C,N4D !
1 1989 No Relevant Indications 1 N4A,N4B,N4C,N4D 1992 No Relevant Indications 2 N4A, N4B, N4C, N4D 1980 No Relevant Indications (Baseline) 2 N4A,N4B,N4C,N4D 1983 No Relevant Indications 2 N4A,N4B,N4C,N4D 1986 No Relevant Indications 2 N4A,N48,N4C,N4D 1990 No Relevant Indications 2 N4A,N48,N4C,N4D 1993 No Relevant Indications i
j 4
I I
l l
k:\nlaiquadifwmulle\ temp.m pf
l l
l l
ENCLOSURE l GENERAL ELECTRIC NUCIEtR EVERGY i
I GE-NFrC3100016-01 l l
GERIS-2000 ULTRMONICIMPECITONOFFEEDIVATERN071115 l
l l
k:\nla\quadfunoutc\covrict2.wpf
1 l
GE-NE-C3100016-01 ,
, Class l GERIS 2000 ULTRASONIC INSPECTION OF FEEDWATER NOZZLES August 23,1994 S.C. Mortenson GENERAL ELECTRIC NUCLEAR ENERGY 11515 Venetory Road, Suite 150 Huntersville, N.C 20078 Approved: W.
J. . SejV, ffanager ,
I pectibr(Services fkU"f45@$l'l'I7
b GE-NE-C3100016-01 Class I IMPORTANT NOTICE REGARDING CONTENTS OF THIS DOCUMENT Please Road Carefully The only undertakings of the General Electric Company (GE) respecting information in this document are contained in the Comed Purchase Order 354017 between Commonwealth Edison Company (Comed), and the GE, titled "lSI Services for Quad Cities Units 1 and 2 (R13)', etfective February 17,1994, as amended to the date of transmittal of this document, and nothing contained in this document shall be construed as changing the contract. The use of this information by anyone other than the Commonwealth Edison Company, or for any purpose other than that for which it is intended, is not authorized: and with respect to any unauthorized use, the GE makes no representation or warranty, and assumes no liability as to the completeness, accuracy, or usefulness of the information contained in this document, or that is use may not infringe privately owned rights.
,i fwgodwp.wpd i
GE-NE-C3100016-01 Class I
{
ABSTRACT l l
l
\
As a result of feedwater nozzle cracking observed in Boiling Water l Reactor (BWR) plants, several design modifk:stions were !
Implemented to eliminate the thermal cycling that caused crack I
initiation. BWR Plants with these design changes have !
successfully operated for over ten years without any recurrence of !
cracking. To provide further assurance of this, the U.S. Nuclear l Regulatory Commission (NRC) issued NUREG-0619, which l established periodic ultrasonic testing (UT) and liquid penetrant -
testing (PT) requirements. While these inspections are useful in confirming structural integnty, they are time consuming and can lead to significant radiation exposure to plant personnel. In particular, the PT requirement poses problems because it is difficult to perform the inspections with the feedwater sparger in place and leads to additional personnel exposure. Clearly an inspection program that eliminates the PT examination and still verifies the absence of surface cracking would be extremely valuable in limiting costs as well as radiation exposure. This report describes the application of advanced UT techniques to assure integnty of BWR feedwater nozzles.
The inspection methods include: 1) scanning with optimized UT techniques from the outside-vessel well for inspection of the nozzle inner-radius regions; and 2) scanning from the nozzle-forgirg outside-diameter for inspection of the nozzle bore regions.
Advanced methods of imaging UT data using recorded radio frequency (RF) data have been developed that show crack location, and depth of penetration into the nozzle inner surface.
These techniques have been developed to the point that they are now considered a reliable alternative to the PT requirements of i NUREG 0619. !
w n. < m li i
GE-NE-C3100016-01 Class I TABLE OF CONTENTS
- l. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . 1 il. ULTRASONIC INSPECTION TECHNIQUE . . . . . . . . . 3 A. OVERVIEW . . . . . . . . . . . . ......... 3 B. GENERAL ELECTRIC REMOTE INSPECTION SYSTEM (GERIS 2000) . . . . . . . . . . . . . . . . . . . 6 C. EXAMINATION TECHNIQUES . . . . . . . . . . . . 9 '
D. ULTRASONIC SUBSYSTEM . . . . . . . . . . . . 11 E. DATA ANALYSIS . . . . . . . . . . . . . . . . . . 11 '
F. FLAW SIZING . . . . . . . . . . . . . . . . . . . . 13 Ill. ULTRASONIC TECHNIQUE QUALIFICATION PLAN . . . 14 A. OVERVIEW . . . . . . . . . . . . . . . . . . . . . 14 B. QUALIFICATION NOZZLE MOCKUPS . . . . . . . 16 C. QUALIFICATION TEST RESULTS . . . . . . . . . 17 IV. CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . 18 V. REFERENCES....................... 19 1
I W 23. * ...
m .
GE-NE-C3100016-01 Class I LIST OF FIGURES Fiours Descriotion Pace 1 Feedwater Nozzle Examination Zones . . . . . . . . . . . . . 3 2 GERIS Data Analysis Displays . . . . . . . . . . . . . . . . . . . 5 3 GERIS Nozzle inspection System ................ 6 4 Nozzle-Mounted Scanner . . . . . . .. . . . . . . . . . . . . . . . . 7 5 Safe-End-Mounted Scanner . . . . . . . . . . . . . . . . . . . . . . 8 6 Nozzle Inner-Radius Data . . . . . . . . . . . . . . . . 12 7 A-scan Display . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1
August 23.1994 iv
L .
GE-NE-C3100016-01 ,
ClassI ;
c l.
- 1. INTRODUCTION :
Several boiling water reactor (BWR) plants experienced extensive feedwater nozzle cracking in the mid-to-late 1970's. The Nuclear Regulatory Commission (NRC) responded with NUREG 061g .
(Reference 1). This NUREG discusses design modifications and ;
establishes guidelines for periodic ultrasonic (UT) and liquid j penetrant (PT) testing. ' included in the design modifications were -
removing the cladding from the inner-radius and bore and.
installing a new thennal-sleeve design. No new cracks have been observed since these modifications were incorporated.
Automated UT examinations provide a technical basis for supplanting the PT exams required by NUREG 0619. An ;
additional advantage of automated UT is the possibility to reduce the frequency of UT examinations in the future.
The original designs of feedwater nozzles were susceptible to cracking initiated by thermal fatigue and propagated by subsequent 1 operational temperature and pressure cycling. The crack initiation was the result of rapid temperature cycling associated with the turbulent mixing of relatively cold feedwater with hot reactor water (caused by laakaga around the thermal-sleeve), coupled with the presence of stainless steel cladding on the inner surface of the low alloy steel (LAS) nozzles. As a result of extensive evaluation (Reference 2), General Electric (GE) developed new thermal-sleeve designs with reduced la=kaga and recommended that the
' cladding be removed from the nozzle inner-radius and boro, thereby reducing the high-cycle fatigue susceptibility and improving ultrasonic inspectability. In addition, changes in sparger design, specific system modifications and changes in operational procedures were implemented to further mitigate crack initiation .
and growth. These system modifications have been implemented at most BWRs. The NRC notes in NUREG 0619 that these changes are responsive to the issue and are an acceptable ;
approach to nozzle crack mitigation. To account for unexpected ,
crack initiation, or the presence of previous indications, NUREG 0619 also requires a crack growth evaluation and periodic i volummetric and surface examinations. ,
) I l
August 23,1994 1 l i
I
- GE-NE-C3100016-01 CI:ss l A hypothetical flaw 0.250" depth into the base metal is used for all NUREG 0619 evaluations (Reference 3).
Removal of the cladding has the additional benefit of enhancing i the ultrasonic inspectability of the nozzle inner-radius and bore.
Automated-UT scanners and data-acquisition techniques now provide the means of acquiring ultrasonic data in large quantities -
over a relatively short period of time. These methods have been developed and tested on full-size mockups of nozzles welded to >
sections of the reactor pressure vessel (RPV) with the goal of determining, which combination of the parameters is best suited for crack detection and sizing. There are numerous parameters to be considered, separately and in combination, so the development and qualification program is based on extensive testing using full-scale mockups, supported by computer modeling of the nozzle ;
geometry. The improved methodology gives reliable and :
quantitative measurements of crack locations and depths.
t This inspection technique implements special methods of '
ultrasonic examinations of the nozzle inner-radius and bore surfaces to provide data on crack location and depth The i methods for these examinations include scanning with optimized !
techniques from the vessel wall, the nozzle-to-vessel-blend radius ,
and the nozzle-forging outside-diameter (OD). Advanced methods of imaging UT data using recorded RF data show crack locations, and depths of penetration into the nozzle inner surface. The i
program objective is to be as quantitative as practically achievable with existing field constraints and considering personnel radiation-exposure-reduction objectives (ALARA). This inspection methodology is considered by GE to be a reliable alternative to the ,
PT requirements of NUREG 0619.
l August 23,1994 2
1 1
- GE-NE-C3100016-01 Cl;ss I (l. ULTRASONIC INSPECTION TECHNIQUE -
A. OVERVIEW Early in-service inspections (ISI) of nuclear plants used manual UT methods and manual data-recording. This work was performed in high-radiation zones where scan time was necessarily kept to a minimum. Nozzle inner surfaces were clad and the inspection entailed complex geometries with long metal paths. These conditions, combined with the problems associated with precise positioning, reduced confidence in the examination results. These and other factors, prompted NUREG 0619, which requires periodic intemal PT inspections of feedwater nozzles.
GE has developed a program using the GE Reactor inspection System 2000 (GERIS 2000), and has significantly improved flaw detection and characterization. The UT techniques have been developed and tested on full-size Reactor Pressure Vessel (RPV) and nozzle mockups of various sizes and designs. Notches and crack implants in these mockups, located in various zones along the nozzle inner-radius and bore, range in depth from 0.105" to 0.750". Figure i shows the various inspection zones of the nozzle inside surfaces.
"N p_
M N0Z2LE SArt [NO 'A'
+ Y - /
s , a.-n .u. J,
.B' Figure 1 Feedwater Nozzle Examination Zones August 23,1994 3
l
- GE-NE-C310001601 Cicss I Technique development was based on extensive mock-up testing to determine which combination of parameters is best suited for flaw detection (i.e., the ability to resolve small flaw signals from noise signals). These parameters include, but are not limited to:
location of the UT beam entry surface; beam and rotation angles; ultrasonic beam modo; and the angle at which the beam intersects the flaw. ;
The data analysis software includes a 3-D graphics pdc that superimposes peak UT data points on a nozzle image. Refer to Figure 2. Data may be viewed from various orientations and ,
magnifications. Access to digitized RF A-scan data is available for viewing any selected data point, which greatly aids in the classification and evaluation of UT data. In the A-scan display shown in the upper right hand comer of Figure 2, the amplitude '
(height) of the returned signal is shown en tins vertical axis, and the relative time-of-flight is shown on the horextal axis. The signal amplitude is affected by the relative orientation of the UT beam direction and the flaw aspect. The locations of the various flaws are indicated on the plan view of the nozzles, and the relative amplitude of the retums are color-coded for ease of interpretation.
Extensive mockup testing has shown that this nozzle inspection method is more effective than previous techniques, because:
The system design provides improved signal-to-noise ratio responses and improvements in flaw detection capability; Enhanced data acquisition, storage and retrieval capabilities; as well as A , B , C , volummetric side-view-and volummetric end-view- scan displays; and adjustable- -
color scales provide for quantity data for a more reliable baseline for comparison with the results from future inspections.
With these improved techniques, GE has demonstrated increased examination accuracy and reliability on full-size nozzle mock-ups.
The integrity of the feedwater nozzles can now be more thoroughly assessed by automatic ultrasonic inspections.
August 23,1994 4
GE-NE-C310001641 Class I ;
tierra flotrl i
.80001 W .:
i
< : i 9 - :
- i
( y l
...u t
~
'*'": a'.:'. % a t:', 7,r, = 1,':. . ....,
l !
,. . , . . . [
T. ;
i i
- . ::::: ' " ~ ' -
M
- I i
-y
! k g IM@eu N[1 ~_
,..n...
j ,.. m .
]
Figure 2. GERIS Data Analysis Displays August 23,1994 5
4
~' GE-NE-C3100016-01 CIIss l
' ]
I B. GENERAL ELECTRIC REMOTE INSPECTION SYSTEM (GERIS 2000) l Referring to Figure 3, an automated scanner moves UT transducers rad: ally and circumferentially around the outside Surface of a nozzle and the adjacent surface of the RPV. The UT data is collected and stored in a digital format. The complete RF .
waveform is digitized and recorded on optical disks.
1 M
~/ J
_. Wj M ltiMPI J I cowfaamest aata acouismom l [
nar* = =. i 1 '
.p.
- f !1l=PI
[2]
Figure 3 GERIS 2000 Nozzle Inspection System
' i Nozzle Mounted Scanner For scanning Zone 1 and 2A as shown in Figure 1, the nozzle-mounted scanner (Figure 4), mounted on a channel track clamped around the nozzle OD cylindrical surface, provides the means of performing a remote ultrasonic examination. The nozzle device includes the nozzle tractor, scanner arm and transducer package.
August 23,1994 6
i l
- GE-NE-C3100016 01 Class I The nozzle tractor has a main body with two motor-driven magnetic wheels and two hinged-end sections, each with one motor-driven magnetic wheel assembly. A pendulum and resolver are mounted l
on the main body to give the angular position of the nozzle tractor.
The reciprocating scanner arm is attached to the nozzle tractor and j extends perpendicular to the nozzle track for scanning the nozzle-to-vessel welds and nozzle inner-radius, i
1 The scanner arm consists of a frame, stepping motors, a worm-gear-driven resolver, and a ball-screw-driven plate that holds the l ultrasonic transducer package. The scanner arm is held to the vessel wall with two spring-loaded guide rods on the inboard end and two magnetic wheels mounted at the outboard end of the scanner arm.
f SCAANER raiwsoucta j PACKAGE 4 ,
7 a l. \ '
i g
( f - % j) inicion 9
'N /
/
\s /
wo m t N
taic FRONT stot Figure 4 Nozzle-Mounted Scanner The transducer package consists of a combination of various transducer wedges indvidually mounted in a frame. The wedges produce beam anglos as required by the examination technique.
i
)
i August 23,1994
)
7 '
l
' GE-NE-C3100016-01 Class i Safe-End Mounted Scanner The safe +nd mounted scanner (Figure 5) attaches to a channel !
track clamped around the nonle safe end, and provides the means of performing a remote ultrasonic examination of the nonle inner bore surface. The safe-end mounted scanner is designed for scanning Zones 28,3,4 and 5.
SCANhtR A&M vggg(4
)
f u. . e ,,ciu n _
=-
NN Crf(::::::: i
. A g, t sArt tuo TRACE FPONT SIDE Figure 5 Safe-End Mounted Scanner The safe-end mounted tractor consists of a main body with two motor 4 riven magnetic wheels and two hinged-end sections, each with motor-driven magnetic wheel assemblies. A pendulum and resolver are mounted on the main body to give the angular position of the tractor. The reciprocating scanner arm is attached to the tractor and extends perpendicular to the safe-end track for l
scanning the nonie-bore region.
August 23,1994 8
- ~ . - - _ _ . - - - - - - . . . ,
GE-NE-C3100016 01 l Cliss1-C. EXAMINATION TECHNIQUES l
1 Zone 1 Examination (Nozzle Scanner).
The nozzle inner-radius is examined from either the vessel plate or the nozzle OD blend radius surfaces.
Methods for examining from the vessel plate use refracted sheer waves that pass through the nozzle-to-vessel weld. The sound beam and rotation angles are calculated assuming that the incident point of sound beam is located on the vessel surface and the axial- ;
reflector location is typically at the conter of the nozzle inner-radius. The sound beam is designed to intersect the inner-radius at a favorable angle. Sound beam and rotation angles are dependant on the individual nozzle design.
Methods for examining the nozzle inner-radius with refracted shear waves from the nozzle OD blend radius were developed and qualified. The sound beam and rotation angles are calculated by determining the optimal incident point of the sound beam on the nozzle OD blend radius with the axial reflector located at the center of the nozzlis inner-radius. The sound beam and rotation angles l are dependant on the particular nozzle confguration. '
Zone 2A Examination (Nozzle Scanner).
The Zone 2A area of the nozzle bore is examined with refracted !
shear waves from the surface where the nozzle OD blend radius ,
merges with the cylindrical surface of the RPV. This scan is performed with the nozzle scanner. .
The sound beam and rotation angles are calculated assuming that ,
the incident point of the sound beam is located on the nozzle OD ;
blend radius adjacent to the vessel surface. The axial reflector ,
location is at the intersection point of Zones 2A and 28. Scanning from this area of the nozzle OD blend radius affords favorable conditions for flaw detection in areas where detection was once :
difficult. Depending on nozzle configuration, coverage generally 1 extends from the Zone 1 region well into the Zone 28 area.
R l
l Augum 23,1994 9
l 1
l l
\
GE-NE-C310001601 '
Chul Zone 28 Examination (Safe 4nd Scanner).
The Zone 2B area of the nozzle bore is examined with refracted shear-waves from the nozzle OD cylindncal surface using the safe .
end scanner (see Figure 5) Up to three separate (quarter, mid and three quarter point) refracted shear waves can be utilized to examine their respective af eas on the nozzle bore. The terms quarter, mid and three quarter point are designations for the three examination areas of the nozzle bore in Zone 2 (i.e., quarter-point
. being 1/4 of the Zone 2 length from Zone 1).
l Zone 3 Examination (Safe-End Scanner). l The Zone 3 area of the nozzle bore is examined refracted shear-waves from the cylindrical surface of the nozzle OD cylinderical surface using the safe end scanner (see Figure 5). The beam is directed perpendicular to the nozzle axis. The sound beam geometry is similar to that used in a circumferential piping examination.
l Zone 4 Examination (Safe End Scanner). ,
The Zone 4 area of the nozzle bore is examined with refracted shear-waves from the nozzle taper and taper-toelend radius ;
surfaces using the safe end scanner (see Figure 5). Nozzles with ;
small blend radius dimensions are normally examined with ma,nual techniques.
Zone 5 Examination (Safe-End Scanner). !
The Zone 5 area of the nozzle / safe-end bore is examined from the cylindrical surface of the nozzle / safe-end OD. This scan is performed with the safe end scanner (see Figure 5).
f l
l l
.l l
l August 23,1994 l 10 1
P GE-NE-C3100016-01 'l Clzss !
]
~ Q. ULTRASONIC SUBSYSTEM i
The UT subsystem has a multiplexed logarithmic UT flaw detection '
instrument. For each channel, the complete RF A-scan is digitized ,
and stored on optical disks. The system stores the RF signal in logarithmic format that has an instantaneous dynamic range that is
- greater than 85 dB. This allows the recording of the peaks of low :
and high amplitude UT signals at the same time with out clipping ;
UT signals, such as would occur with linear systems that have an !
instantaneous dynamic range generally less than 45 dB. l
' The system can record a complete set of RF waveforms from as l many as 16 channels concurrently while scanning at 2.0 inches (51 3 mm) per second and taking data overy 0.15 inches (3.8 mm). The j pulse sequence for each of the channels is controlled by the HP i 730 workstation. Each channel is equipped with adjustable gate t length to accommodate various types of angle 4eam examination ,
conditions. The system stores all RF data in raw form and if necessary, the data is distance-amplitude-corrected (DAC) through -
software and the original RF data file is never altered. The A- and C-scans are displayed during calibration and data acquisition.
- l i
E. DATA ANALYSIS l The GERIS 2000 analysis system utilizes advanced interactive ,
color graphics to evaluate and assist in characterizing indications from service, fabrication and geometric related UT reflectors.
Coordinated A , B , C , volummetric side-view , volummetric end-view- and 3D-scans are provided on a high resoir
- inn (1280 by 1024 pixels) color display. Several channels wits, mi of the above mention scans may be displayed at one time. These graphic displays have an adjustable color scale that provides the best resolution of flaw detection and characterization down to the I
material noise. The presentation of the data can be readily changed from or to gray scale.
Real-time interaction between all displayed views (of a given scan) is automatic and provides analysis personnel with quick ;
, coordination and identification of data. ;
i August 23,1es4
, 11 i
l
GE-NE-C310001641 l Class I j l
In addition, the workstation displays peak-amplitude-UT data superimposed on a 3-D wire-frame model of the nozzle or coinponent being examined (see Figure 6). Three-dimensional ,
graphic tools include component viewing at any perspective or ,
magnification. These capabilities assist in accurate UT data j evaluation by analysis personnel. An on-line data base of transducer / scanner position, UT and A-scan data can be accessed I during data analysis.
i
,/ ll ) :D
/
- ~ :
~ =
liiil55EEsiI 1 3
. e-- ~mm i
Figure 6 Nozzle Inner-Radius Data The real-time A-scan is readily accessible and can be viewed for data analysis. Figure 7 is an A-scan display of a 0.170" deep nozzle inner-radius EDM notch in GE's clad <emoved feedwater :
nozzle mockup. A-scan data can i:9 viewed either statically or l dynamically. Dynamic viewing enhains data evaluation by ,
providing the echo-dynamic character *stics of recorded data. l The A-scan presentations available are RF and video. The RF ;
presentation provides the phase of the UT signal, which can be a valuable tool for signal characterization. Additionally, the ,
presentation can be toggled between logarithmic and linear. The !
logarithmic presentation provides an instantaneous dynamic range ;
that is greater than 85 dB, and allows the viewing of the peaks of low and high amplitude UT signals at the same time. l August 23, t994 12 l
l 1
GE-NE-C310001641 Class I Tipdiffraction sizing techniques are incorporated .dilizing the 1 digitized A-scan. Tip-diffraction sizing affords better accuracy in flaw depth measurements that is needed to supplant the NUREG 0619 PT exams.
l ;
Figure 7 A-Scan Display.
F. FLAW SIZING The sizing method used is totally dependant on the so-called tip-diffraction phenomenon - sound energy encountering the tip of a defect will be radially scattered. This radially scattered sound retums to the transducer. Sound energy also reflects from the flaw comer and retums to the transducer. From the relationship between the time of arrival of the reflected signal from the flaw comer and the tip diffracted signal, and the angle of incidence of the sound energy on the flaw, the depth of the flaw can be derived.
The crack-tip signal is generally small in amplitude in comparison to the peak amplitude from the comer reflection. Depending on the interaction of the sound beam with the flaw and type of flaw, the crack-tip signal can be reduced by as much as 20 decibels (dB) or more in amplitude from that of the comer reflection.
l l
August 23. t 994 13 l
l
GE-NE-C3100016-01 Cl:ssI Ill. ULTRASONIC TECHNIQUE QUAUFICATION PLAN ,
A. OVERVIEW The ultrasonic qualification plan was based on testing full-scale mockups, (one has the clad removed). The plan incorporates a data sample set developed using ASME Code Section XI, Appendix Vill as a guideline. Appendix Vill does not presently contain specific rules for qualification of inner radius examination methods for unclad nozzles, but it was used as e guide to evaluate data, define sizing methodology and devise field inspection procedures.
The primary purpose of this demonstration was to qualify the UT equipment and techniques. Development of the full protocol for an Appendix Vill qualification is an industry effort that is on-going at the present time. Existing mock-ups, with flaws placed in the various inspection zones, were considered sufficient in the absence of protocol. Automatic data recording and retrieval capability allowed for subsequent reviews of inspection data, as required, to confirm the validity of the detection and sizing results.
The qualification plan focused on specific portions of Appendix Vill that are applicable to the feedwater nozzle inspection. Flaw depths in the range of 0.105" to 0.375" were in the sample set, encompassing the 0.250" basis in NUREG 0619. The flaw sizing statistical measurement criteria provided by Appendix Vill was applied to the flaw samples to generate a measurement that can be compared to the actual notch depth. Since the data was automatically recorded, it is available for subsequent scrutiny and review.
Appendix Vill, Supplement 5, " Qualification Requirements for inside Radius Examinations," provides rules for extending a qualification for examination of the clad-base metal interface on the vessel (Supplement 4) to a nozzle inside radius by using a mockup i
containing some additional notches. Supplement 5 states that the specimens shall comply with Supplement 4, except that the flaws may be either cracks or notches. For the case of the nozzle inner-radius examination, notches are considered equally representative.
I However, to verify the capability of the UT techniques to detect and size actual fatigue flaws, two fatigue cracks were implanted in the
! inner-radius of another nozzle mockup. The similarity between the August 23, t 994 14 l
l l
l GE-NE-C3100016-01 Class I ,
. fatigue cracks and EDM notches was demonstrated. Because this '
qualification plan addresses only the unclad nozzle inner-radius, the requirements for the size and number of flaws were adopted :
directly from Supplement 4 of Appendix Vill. l The minimum sample set requires at least seven flaws for detection qualification, and an additional three flaws to qualify the sizing technique (i.e., a total of ten are required). While Supplement 4 .
allows flaws up to 0.750" in depth to be used in the sample set, the GE qualification plan had a maximum flaw depth of 0.375", a more ,
conservative condition than required. However, manual sizing ;
techniques, which may be used to supplement the automated .
sizing data, were verified on notches with a maximum depth of f O.750".
Some notches are slightly wider than nominally specified by Appendix Vill, but this is not considered detrimental to the ;
qualification. The influence of notch width showed no significant 4
', difference in detection or sizing results for the GE technique. The notches were not filled; however, this was not expected to impact the ultrasonic examination. The flaw configuration in the feedwater nozzle mock-up had flaws in all inspection zones for the -
qualification data set. Flaws were radially oriented as specified in i Appendix Vill, which is also in accordance with fracture mechan'es ,
predictions and field experience with nozzle cracks. '
Field service personnel that perform the on-site examinations were trained and demonstrated their proficiency in the UT techniques using the samples in the clad removed feedwater nozzle mockep. ;
Those performing data analysis received a practical examination using recorded data, similar to that used under SNT-TC-1 A qualification programs. .
August 23,1994 ,
15 P
)
l
_ _ . _ _ _ . . . . _ . _ _ _ . - . _ _ _ _ _ - __ ~_.,_ _ - --_._. _ . _ _ . __ - _ _____ _ . _ _ . . . , ..
GE-NE-C310001641 Cl ssI B. QUALIFICATION NOZZLE MOCKUPS l
i CLAD REMOVED FEEDWATER MOCKUP (EDM NOTCHES)
The clad removed feedwater nozzle mock-up used in the l
qualification testing was fabricated by GE of components from a ,
cancelled BWR. !
The forging was machined to represent a barrel-type feedwater i nozzle that had the clad removed. The scanning for Zone 2A was performed on the nozzle-to-vessel weld surface. This surface was hand welded and ground, and was typical of the nozzles that experienced nozzle-inner-radius cracking in the field. This nozzle-to-vessel weld configuration was selected because scanning from '
hand-ground surfaces is the most difficult.
The EDM notches were distributed throughout the examination ,
volume. Notches were located in each inspection zone to demonstrate the individual testing technique for both detection and '
sizing. The notches are radially oriented as specified in Appendix ,
Vill, which is also in accordance with fracture mechanics predictions and field experience with actual nozzle fatigue cracks.
UNCLAD FEEDWATER NOZZLE MOCKUP (FATIGUE CRACK '
IMPLANTS)
To confirm the detection and sizing of fatigue cracking, two fatigue crack implants were welded into an unciad feedwater nozzle forging inner radius. The unciad feedwater nozzle mockup used for -
implanting these fatigue cracks is from another cancelled BWR. 'ihe fatigue cracks were generated in material specimens the same as the nozzle forging. '
(
- August 23,1994
, 16 ;
a
i
'- GE-NE C3100016 01 Cl:ss I ,
C. QUAUFICATON TEST RESULTS DETECTON As discussed earlier and shown in Figure 1, the inner-radius and bore are divided into different inspection zones. To effectively examine these zones, specially developed techniques are used where examinations are performed from the vessel plate, nozzle OD blend radius, nozzle OD, nozzle taper and safe end. All the techniques that are presently used by GE were included in this qualification testing. Individual nozzle geometnes will determine which of the above techniques will be applied.
A comparison of the EDM notch data and fatigue crack detection - i data showed that the amplitude responses of fatigue cracks and EDM notches of comparable depths were similar. :
In summary, the fatigue cracks were equally detectable as the :
EDM notches. This coupled with the extensive qualification with the EDM notches provides confirmation of the effective of the UT system and techniques to detect fatigue cracking in the field.
SIZING Sizing capabilities were qualified with the GERIS 2000 using the ,
methodology of Appendix Vill. Manual techniques may be used to !
supplement the automated data, if necessary, during a field l examination. For this qualification program, the EDM notches and the fatigue crack implants were sized with automated data from both directions. The sizing of flaws from both directions increases the data base for number of samples by a factor of two and better assesses sizing capabilities. The sizing soceptance c"iteria !
outlined in Appendix Vill was demonstrated.
In summary, the EDM notches and the fatigue crack implants were ;
successfully sized. The sizing results were acceptable to Appendix !
Vill acceptance alteria, therefore the UT techniques are fully validated for crack sizira in the field.
August 23,1994 17
-,,--.,,c-e -
1 GE-NE-C3100016-01 i Class l l i
IV. CONCLUSIONS i
The qualification program implemented by GE successfully F demonstrated the GERIS 2000's detection and sizing capabilities.
Qualification was performed on EDid notches and fatigue crack implants in full-scale mockups. In addition, the sizing results were acceptable to Appendix Vill criteria. .
I GE techniques developed over the past few years provide the -
means of performing routine quanitative inspections. These techniques have been developed to the point where they are now considered a reliable alternative to the PT requirements of ,
NUREG-0619. ,
a l
I P
f 6
August 23,1994 18 .
1 1
1
GE-NE-C3100016-01 Class I V. REFERENCES
- 1) NUREG 0619, "BWR Feedwater Nozzle and Control Rod Drive Return Line Nozzle Cracking", November 1980.
- 2) NEDE-21821 Class lil, " Boiling Water Reactor Feedwater Nozzle /Sparger Final Report", GE Nuclear Energy, March 1978.
- 3) Generic NRC letter 81-11 to all Power Reactor Licensees from Darrell Eisenhut, February 28,1981.
- 4) GE-NE-508-038-0394 Rev 1, "GERIS 2000 Ultrasonic inspection of Feedwater Nozzles," GE Proprietary, GE Nuclear Energy, April 94.
August 23,1994 19
--