ML20114D643
| ML20114D643 | |
| Person / Time | |
|---|---|
| Site: | Catawba  |
| Issue date: | 09/02/1992 |
| From: | Tuckman M DUKE POWER CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9209090222 | |
| Download: ML20114D643 (20) | |
Text
.
Il 1
l Duke twu romiumy At3 Tl w irk
' Ca:anto %dn:r Gewo!wn Dwartment fhe h viden!
WW Conrcid lioud iAaljkil42R5 n ice lo?k SC 2?'4]
(bu])Mid41s t< t i
3-
?.
i DUKEPOWER i
September 2,1992 i
i j
U. S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, D. C. 20555
Subject:
Catawba Nuclear Station, Unit 1 Docket No. 50-413 Technical Specification Amendment Supplemen' Steam Generator Repair Criteria i
On August 24, 1992, Catawba Nuclear Station submitted a proposed Tect r; cal l
Specification amendment to the NRC. The proposed revisions change the Steam Generator repair criteria for Catawba Unit 1 Cycle 7 operation. This amendment is intended to apply.
i to Cycle 7 operation only. In the "05000413/LER-1992-009-02, :on 920722,turbine Bldg Sump Sys Pipe Rupture Occurred.Caused by Inoperable Turbine Bldg Sump Radiation Monitor.[[Topic" contains a listed "[" character as part of the property label and has therefore been classified as invalid. Obtained from Sump & Analyzed & Flow from Temporary Pumps Terminated|August 24,1992 letter]], Catawba committed to submit i
further technical justification for the proposed ~ amendment at a later date. Please find enclosed the additionM technicaljustification that supports the proposed smendment request.
i r,
j Very truly yours, l
w I
M. S. Tuckman MHH/TECHSPEC.SUP i
l Attachments i
i-i 920a090222 920902 PDR' ADOCK 00000413 A
q P
PDn
\\g I
- +,,,n,-
4 U. S. Nuclear Regulatory Commission September 2,1992 Page 2 1
xc: (w/ Attachments)
Mr. S. D. Ebneter Regional Administrator, Region II Mr. Heywood Shealy, Chief -
Bureau of Radiological Health South Carolina Department of Health 4
Mr. Robert E. Martin, Project Manager Mr. J. Stang, Project Manage Mr. W. T. Orders NRC Senior Resident Inspector American Nuclear insurers -
clo Dottie Sherman, ANI Library The Exchange, Suite 245 270 Farmington Avenue Farmington, CT 06032 M&M Nuclear Consultants 1221 Avenue of the Americas New York, New York 10020
-IN?O Records Center Suite 1500 1100 Circle 7.'. Parkway
- Atlant' Gorgia 30339 l
m w.
mm
-w m
- --ve-4
rw wm w -r er"
'te*
8.M
'f "1T T-T
.. = _
U. S. Nuclear Regulatory Commission September 2,1992 Page 3 M. S, Tuckman, being duly sworn, states that he is Vice President of Duke Power Company, Catawba Nuclear Site; that he is authorized on the part of said Company to sign and file with the Nuclear Regulatory Commission this revision to the Catawba Nuclear Station License No. NPF-35, and that all the statements and matters set forth therein are i
true and correct to the best of his knowledge.
j 0;b.
%Um M. S. Tuckman, Vice President Catawba Nuclear Site r
Subscribed and sworn to before me the f e day of
% nf
,1992.
I "3f f]; t v h ^
st a&n Notarp Public
/
My Commission expires:
Alav ; { c90co A
1 5
'T h
4 i
CHANGES O TECHNICAL SPECIFICATIONS l
1.
The proposed amendment modifies TS 3/4.4.5 to allow a tube to remain in service if the flaw indication signal amplitude is less than or equal to 1.0 volt, regardless of the depth of wall penetration.
For flaw indications in excess of indicated i
1.0 volt, but less than 2.5 volts, the tube can remain in service provided an RPC inspection of the indication does not detect ODSCC or any other degradation mode exceeding technical specification repair limits.
Crack indications above 2.5 i
volts will be repaired by sleeving or plugging, and do not require RPC confirmation.
Based upon an end of cycle voltage distribution for cracklike indications, additional tubes may be removed from service based on a conservative estimation of potential primary to secondary leakage in the event of a steam line break event, such that the leak rate will be limited to 1.0 gpm (maximum) in the faulted loop.
2.
Application of a More Restrictive Leak Rate Limit TS 3/4.4.6 is modified to allow a primary to secondary leak rate of 150 gpd (0.1 gpm) per steam generator, 0.4 gpm (576 gpd) to al for all steam generator.
Implementation of this leak rate limit will reduce the potential for excessive leakage during both normal operating conditions and a
postulated steam line break event in the Catawba steam generators.
BACKGROUND INDUSTRY EXPERIENCE In general, the degradation morphology occurring at the - tube support. plate intersections at plants it the U.S. ran be described as axial ODSCC.
Axially oriented macrocracks can occur at one of more azimuthal locations around the circumference of the tube. The macrocracks are comprised of short, nearly collinear microcracks l
separated by ligaments of material.
Typical microcrack length is l
less than 0.2 inches.
corresponding macrocrack;can be as long
'~ ~
as the suoport plate tn-ness Minor to moderate intergranular attack (IGA) can occur in addition to axial ODSCC.
l CNS EXPERIENCE Duke Power Company first found indications of Tube Support Plate (TSP) Outer Diameter Stress Corrosion Cracking (ODSCC) during the l
Page 1
i 4
i i
End of Cycle -(EOC)
- 4. refueling outage at Catawba Unit 1.
During j
that refueling cutage, 17 tubes were plugged due to ODSCC at the j-support plates.
In April of 1991, during the EOC5 refueling l
outage, 158 tubes were plugged due to tube support plate ODSCC.
Based on' both Catawba-specific data and industry data, acibull
(
dis! ibutions were developed to preaict the occurrence of future j
indications. The results of that study indicated-that, during Unit t
l's next scheduled outage (the ?OC6 outage), approximately 300 to 40 tubes would require repair Locause of tube support plate ODSCC.
In the current outage (IEOC6) bobbin coi_1 inspections.of the steam l
l generator tubes were completed by August 8, 1992.
The inspection
[
found approximately 7000 indi. cations which affected approximately
~
4500 tubes.
When an indication is found using the bobbin coil technique, the Motorized Rotating Pancake Coil (MRPC) is used to i
confirm the existence of the indication.
Use of the-MRPC on a sample population of Catawba Unit 1 tubes confirmed the presence of j~
indications in approximately 23% of those tubes sampled.
This effort was completed, and the data was available, on August 110, i
1992.
Using this conformation data and the current criteria i
required by the - Catawba Technical Specifications, Catawba - has projected that approximately 1020 tubes.would require repair.
?-
With this data available, Catawba management decided on August 11, 1992, to pursue the possibility of amending Unit l's Tech Specs to permit the use of an interim plugging criterion.
On August 11, 1992, Duke requested Westinghouse to begin its analyses to-support j
such a change.
That same day, Duke also contacted the NRC Staff to inform them of the results of the steam generator inspection and i
analyses.
A TS change submittal was submitted on August 24, 1992 which included marked up TS pages and a No Significant Hazards Analysis.
i LICENSING BASTS l
In establishing the interim tube plugging criterion, the general l
approach is to verify the existence of acceptable margins to tube l
burst and to excessive steam - genera tor tube leakage during all j
plant conditions.
The NRC Regulatory Guide 1.121, " Bases for Plugging Degrade'd-PWR Steam Generator Tubes," issued for comment, addresses-tubes with j
through-wall-cracking.
The Regulatory Guide utilizes safety
[_
factors on loads-for tube burst and collapse that.are consistent l
with Section III of the ASME Code. _ Per paragraph C.3.d(1). sf RG 1.121, the applicable analytical and loading criteria in thinned or unthinned tubes.with through-wall cracks are:
i 1.
Through-wall cracks in tubes should not propagate and result l-in tube rupture under accident condition loadings.
i Page I i
L
.n m
_.m
~. -., _ -
t 2.-
The maximum permissible crack length of the largest single-crack should be.such that the associated burst pressure is at least 3 times the normal operating pressure differential.
3.
The leakage rate limit under normal operation set forth in the-plant technical specifications should be less than the leakage limit determined for the largest pe missible crack.
ANALYSIS
SUMMARY
In developing the interim tube plugging criterion (IPC),
the general approach, as outlined in Regulatory Guide 1.121, " Bases for Plugging Degraded PWR Steam Generator Tubes," is to verify - tube integrity under various postulated accident load combinations, the existence of acceptable margins to tube burst, no excessive steam generator tube leakage during all plant-conditions and establishing an operating leakage limit such that the leakage associated with the longest permissible single crack is not exceede..
Tube integrity under postulated LOCA
+- SSE (Safe Shutdown Earthquake) and Main Steam Line Break (MSLB)
+ SSE has been evaluated and is assured with the IPC. Specifically, some tube intersections near Tube Support Mate Wedges have been excluded from the application of the IPC (due to the potential for SG-tube deformation during LOCA + SSE load calculation).
Based on data gathered from Model Boiler. Tests, and pulled tube dara, a correlation of bobbin coil voltage to burst pressure has bea developed. The end of cycle limit is based on the capability of providing a factor of safety of 3 times normal operating differential pressure to tube burst. This burst pressure, _ which corresponds to 3750 psid, is equivalent to 4.1-volts. This voltage is then restead to allow for defect growth during the_ cycle 1and uncertainties in the growth rate as well as eddy-current-uncertainties. Thus a 2.5 volt limit is established for required-repair regardless-of-RPC confirmation. This'is equivalent-to the full APC limit.
Because of the complex data distributions, a Monte Carlo Analysis was performed to determine the post MSLB primary to secondary-leakage, which. took into account the beginning-of-cycle; indication distribution based on bobbin coil voltages, the projected growth-rates during the cycle based on the 541 largest defect. indications identified initially during the. 1EOC6 refueling outage, eddy
~
current analyst variability, probe wear uncertainty and postulated e leakage at steam line break conditions as a function ~of bobbin coil voltage. The'results of this analysis-conservatively indicate that t
the maximum expected primary to secondary leakage af ter ' a L Main Steam Line Break would be approximately 0.54 gpm; This leakage,.in addition to the primary to secondary leakage allowed for continued
- operation in the proposed-Technical Specification yields a total of 0.94 gpm which is bounded by the 1 gpm total primary to secondary Page 3 l
H,
j J
{
-a j
leakage assumed in t.he Catawba FSAR.
t A reduced allowance for primary to secondary leakage to_. O.1 gpm l
(150 gpd) will be established with this proposed IPC to - allow
]
detection of the largest single permissible crack.
1
)
CDNCLUSIONS i
the 1
The analysis performed for Catawba - Unit 1 demonstrates that j
requirements of Regulatory Guide 1.121 are_ fully satisfied by the proposed IPC at both beginning and end of cycle for Catawba Unit 1 Cycle 7.
Additionally, no leakage is expected to occur during normal operating conditions.upon application of this IPC. The results of a dose analysis, performed for a postulated main steam-line break _ out 'ide of-containment remain within the limiting dose i
consequence (Exclusion Area Boundary - Thyroid Dose) presented in j
the Catawha FSAR.
i l
MARGIN COMPARISONS The barst _ pressure capability for a 1.0 volt. IPC can be favorably l
compared with that for a 40% depth plugging criterion. Because' the uncertainty on the 40% depth burst pressure is not well defined, j
the comparison will be based' on nominal data only. The nominal r
correlation burst pressures at 1.0 volt is 6920 psi. The nominal j
burst pressure at 40% depth for a 1.5 inch long uniformly thinned a
tube (general basis for 40% limit) is 6110 psi. Thus at nominal values, the 1.0 volt burst capability exceeds that associated with the 40% thinned repair basis.
i Because Cctawha Nuclear Station is the first plant -with 3/4 inch SG l
tubes _ to request consideration of a voltage _ based IPC it-is expected that comparisons will be made to IPC's granted for plants
. with 7/8 inch tubes.
Specifically, the Beginning of-Cycle _ Margin for-a 3/4 inch tube with a one volt threshold for RPC confirmation provides a margin of 1.4 to tube burst from the 3 times normal 4p.
This is the same BOC margin to tube burst provided by the11 volt threshold for the 7/8 inch tube IPC's which operate at a slightly higher normal operating differential pressure' across the SG tubes.
A similar analysis to that described _above, has been-performed to determine end.of. cycle' margins.
The end : of - cycle _ margin-is calculated _ taking projected defect growth rates (plant; specific) and eddy-current uncertainties into account. The results of this analysis are tabulated in Attachment 2. Because the 3 times normal l
op provides the-largest challenge to tube integrity the_EOC margin
{
comparison will~be based on this value. As can be seen by the: data i
presented in Attachment 2,.both Catawba;and the 7f8 inch tube data-
-provide equivalent margins to tube burst at end of cycle.
Page 4
.. - ~..
9 TECHNICAL JUSTIFICATION To preclude unnecessary repairs of tubes in the Catawba steam t
generators an interim tube plugging criterion has been developed.
The plugging criterion involves a correlation between eddy current-bobbin probe signal amplitude (voltage) and tube burst capability.
Leakage during normal operating conditions is not expected to occur upon implementation of the interim plugging criterion.
Any potential for primary to secondary leakage during a postulated SLB event is assessed relative to conseivative leakage thresholds at various bobbin coil voltages.
" Bases for Plugging Degraded Steam Generator Tubes," describes a method acceptable to the NRC Staf f to meet General Design Criteria 14, 15, 31, and 32 by reducing the probability and consequences of a steam generator tube rupture through determining the limiting safe conditions of degradation of steam generator tubes, beyond which they must be repaired emoved from service.
The inter..o c @e plugging criterion is based upon the conservative assumptions t. hat the tube to tube support plate crevices ure open (negligible crevice deposits or TSP corrosion exists), and that the tube support plates are displaced under SLB condition loadings.
The OD SCC is thus considered to be free span degradation under normal and SLB accident conditions and the principal requirement for tube plugging considerations is to provide margin against. tube l
burst per RG 1.121.
The open crevice assumption leads to maximum leak rates compared to packed crevices and also maximizes the potential for TSP displacements (which can uncover cracks) under accident condition loadings.
TUBE INTEGRITY ASSESSMENT For the faulted plant condition evaluation, the pcatulated events considered are Loss of Coolant Accident (LOCA), Steam Line Break (SLB), and Feedline Break (FLB) in combination with Safe Shutdown Earthquake.(SSE).
It is shown that potential through-wall cracks which may exist as a result or implementation of the interim tuba plugging criterion are not expected to propagate and result in tube rupture under accident condition loadings.
The potential effects of a postulated LOCA + SSE event on the continued maintenance of tube integrity and the ability of the Catawba steam generators to perform their intended safety function are addressed below and are found to be acceptable.
The potential effects of a postulated SLB
+ SSE or FLB + SSE event on the continued maintenance of steam generator tube integrity are enveloped by the R.G. 1.121 criterion requiring the maintenance of a factor of safety of 3 times normal operating pressure differential to tube burst with the presence of a through-wall crack.
This is the case as the combined maxitmim tube bending stress at any elevation during a SSE event is less than the tube material yield strengt-h (using lower tolerance limit Page 5
properties),
Outer diameter stress on the order of the yield j
strength of the tube material is required before any significant ef f ect on tube burst strength is realized (WCAP 7832A).
Tube burst l
capability following a combined SLB + SSE or FLB + SSE event upon
^
implementation of the interim plugging criterion is evaluated below l
and is found to be acceptable.
1 Combined SLB + SSE/FLB + SSE Loadinas i
l During a-postulated accident, lateral support provided to the tube L
by the TSP can induce bending stresses at the TSP intersection, j
which vary from tension to compression around the tube i
circumference.
Compressive stresses have the potential to reduce the tube burst capab311ty due to crack opening.
Test results show that bending stresses on the order of the tube material yield strength at operating temperature are required - to significantly affect tube burst capability.
Per RG 1.121, burst capability during accident conditions is required to be at least i
2650 psi.
Also per R.G. 1.121, the bending stress due to SSE must be coupled with the pressure induced primary membrane stress when
Based on the results of the seismic analyses, the maximum bending stresses at the top TSP have been shown to be less than the tube i
yield strength at temperature.
As noted-above, bending stresses l
approaching the material yield are required to affect burst.
i Therefore, the postulated effects of a-combined SLB + SSE do not l
adversely affect the burst capability of the Catawba tubing.
f Combined LOCA + SSE Loadings In addressing the combined. ef fects of LOCAL + SSE on the - steam j
generator ccmponent (as required by GDC 2, and RG 1.221), it_has j
-been postulated that local tube collapse may occur in the steam generators et some pla.its.
The tube support plates - may become deformed as a result of radial loads at the wedge supports at the periphery of the plate due to combined.LOCA and SSE loadings.
There are two issues associated with local steam generator tube i
collapse.
First, the potential collapse of steam generatoritubing adjacent to wedge groups reduces the RCS flow area through the l
Second, there is a potential-that partial through-wall cracks in tubes could progress to through-wall cracks, or that existing L
through-wall cracks might open up, during tube-deformation-resulting in secondary to primary in-leakage, which similarly may_
cause an increase'in peak clad temperature (PCT),
In order to estimate the level of tube deformation resulting from the _ combined LOCA + SSE lcadings, results from a' crush test program for Model D steam generators.are utilizec.
The force / deflection-Page 6
=6 t--
+-
.--c-
,.-,-w---e we-e--
r yi.
p,
.,e-i
.w y
-,,y-,-y----rwwyAw-y-+.--
7,..w--,-
w
,w
++w--
results represent inelastic behavior of the plate and tubes.
In order to make use of this data, an approximation was made between the elastic analyses that determine the plate loads and the i
inelastic crush test.
This approximation -is based on the area under the force deflection curve for the crush test versus the area corresponding to elastic plate response.
Since the leak-before-break criterion has been approved for.the Catawba Unit i reactor coolant loop piping, large break LOCA forces can be excluded, and only small break forces require evaluation.
Conservatively, large break forces were used for the LOCA + SSE-evaluaticn.
The results of the analyses indicate that for the flow distribution baffle elevation, no tubes would be excluded from-IPC implementation.
However, a limited number of tubes adjacent to the wedge groups for the remainder of the plate locations, may deform or collapse and secondary to primary in-leakage may occur.
The IPC cannot be applied to these locat_ons, and they will be plugged or repaired by sleeving if TSP degradation is detected at e.levations Other than the FDB in these tubes.
For all other steam generator tubes (tubes not located in wedge areas), the possibility of secondary to primary leakage in the event of a LOCA + SSE event is not significant.
Any secondary to primary leakage is expe::ted to be less chan the current primary to secondary operating leakage limit.
Steam generator tube integrity and operability are enhanced with the reduction of leakage allowed from 500 gpd to 150 gpd per steam generator.
Furthermore, secondary to primary leakage would be less = than primary to secondary leakage fcr the c ame dif ferential pressure since - the cracks tend to tighten under secondary to primary dif ferential pressure.
Additionally, the presence of the tube support plate is z-expected to reduce the amount of in-leakage as the annulus between the TSP hole and tube functions as a leak limiting orifice.
TUBE BURST CAPABILITY DISCUSSION The criterion of RG 1.121 to maintain a factor of=3 times normal
-operating pressure differential on tube burst-is _ inherently satisfied during normal operating conditions.
Based on Catawba eddy current data, the tube support plate - elevation ODSCC is situated within the thickness of the tube support plates.
Steam generator tube denting- (due to TSP corrosion) and cracking potentially initiate and progress at high temperature within-the TSP.
Since the tubes and support plates are in an equilibrium situation during normal operation, and since the causes of the cracking occur-at the + ube/ support plate intersections, it is clear that the crack 4 ag wealu be situated within the plates during normal operating conditions and - that tube buret - cannot occur - due to degradation within the TSP.
A w
Page 7
-_d
. ~ - - =.
J Catawba Pullad Tube Burst Test Results In general the burst testing of-the Catawba tubes is judged to l
have been concluded prior to true bursting of the tube.
Little or no crack tearing (representing ductile failure of t.he' tube material)'was observed for many of the burst tests, and therefore, adjustments must be applied to the burst test results in ordor to yield burst pressures indicative of the point of ductile failure in the tube.
Where burst test data is discussed, adjusted burst pressures will be given in parenthesis.
Table 2 outlines the Catawba burst test results.
Table 2 Catawba Unit 1 Tube Burst Test Data-Tube T9P Field Post-
-Burst Maximum Bobbin Pull Pressure Crack Volts Bobbin Depth Volts R5C112 2 hot leg 0.48 0.25 9700 (10800)
N/A R5C112 3 hot Jeg 1.82**'
5.06 4150 (N/A)*
93%
R10C6 2 hot leg 1.46 2,07 6000 (7100) 72%
R10C6 3 hot leg 1.31 5.34 4850 (5740)
N/A R7C47 3 hot leg 1.57 4.13 N/A*
84%
R20C46-2 hot leg 0.42
-0.82-8600 (N/A)*
N/A R20C46 3 hot leg 0.79 1.04 7200-(N/A)*
N/A l
R10C69 2 hot leg NDD NDD 9400 (10340)
N/A R10C69 3 hot leg 1.48 3.31 J 5000-(N/A)*
N/A Burst adjustme_nt expected to be greater than 1.25, therefore these tubes not included in burst-data base.
1 Based'on the large-discrepancy between-the pre and post pull data it appear-that - this sustained-additional--damage during the tube pulling process. Ther2 fore basing a leakage threshold on the data provided by this tube is conservative.
3/4" Model Spqiler Burst Data A set of
'5~ laboratory induced stres0 corrosion crackino
.ube specimens are used in the 3/4" tubin" data base in the-deve.
..aent of the interim plugging criterion.
Page 8 e+
-ye-a
,p.y
,>y--
g y
y, p--
ypmy--
.y
-=
CATAWBA TUBE WALL DEGRAtATION CHARACTERIZATION To date, 5 tubes have been removed from the CataNba Unit 1 steam
(
generators and destructively examined. The corrosion morphology of the Catawba pulled-tube is consistent with the evidenced TSP degradation morphology.
The maximum length of intergranular corrosion evidenced from Catawba pulled tubes is 0.5 inch, centered about the midpoint of the tube support plate.
1 Testing of model boiler specimens for free standing tubes at room temperature conditions show burst pressures in excess of 5475 psi for indications of outer diameter stress corrosion cracking with voltage measurements as high as 11 volts.
First testing of Catawba pulled tubes iridicate burst pressure of up to 7100 psi foi 1.46 volt indice.tions.
Based on the exi sting 3/4" data base for free span tubing, and-correcting for the influence of Catawba operating temperature on material properties, and for the-minimum strength levels, - the safety requirements for tube burst margins during both normal and accident condition loadings for Catawba can be satisfied with bobbin coil indications with voltage _ levels less,than 4.1 volts, independent
' the depth measurement. Alternate crack morphologies could correspond to 4.1 volts so that a unique crack length is not defined by the burst pressure to voltage correlation.
To address such uncertainties in burst capability, the structural limit is based on the lower 95% confidence level on the burst correlation.
Again, conservatism is-added by atssuming the crevice area will behave as free span degradation.
CATAWBA GROWTH RATES Voltage growth rates for the Catawba steara generators have been
= defined in terms of percent growth per cycle and are applied at the voltage amplitude requiritg tube plugging.
To Address the potential for variations in future cycle length, an average voltage growth allowance of '45% is applied to establish.the plugging or repair limit.
The average growth allowance is determined. from the eddy current results (the 541 largest indications) of Unit 1 Cycle 6.
These results for Catawna Unit 1 show a 25% average growth -(0.18 volt absolute growth) over the entire range.of v.21tages and about 1L% average growth (0.14 - volt absolute growth) for indications above 0.75 volts at.the beginning of the last cycle.
The Cycle 5 growth data were developed for 126 indications at TSPs plugged in 1991 at EOC5.
For BOC indications s 0.75 and ' < 1. 0 volts, Cycle 5 and 6 show essentially the same average growth rate (about 40% and 29% respectively)..This difference is most likely due to the larger number of indicationa used for the growth rates in Cycle 6 than for Cycle 5.
Based on the similarity between Page 9
~.
- - - ~ - -
i Cyc'les 5 and 6 growth, the Cycle 6 growth rate data can be eyected to be representative of Cycle 7 anticipated growth as used for the l
Catawba Unit 1, IPC tube integrity assessments.
I The voltage based structural limit must be reduced by the allowances for crack growth during the next cycle plus eddy current-uncertainties to obtain a voltage based tube plugging limit which supports the margins to tube burst cutlined in RG 1.121.
For a EOC voltage structural limit of 4.1 volts, with values f or. average i
growth per cycle, and assumed eddy current uncertainty of 45t and l
20% respectively, the plugging limit is determined to be 2.5 volts.-
For the interim, the voltage based plugging limit is set at 1.0 volts, thus, a 1.5 volt margin is realized upon application of the.
interim plugging limit of 1.0 volts.
Applying the g owth value of i
0.62 Volts and an NDE uncertainty of 1F% to the interim plugging limit results in a nominal expected EOC 7 voltage of 1.8 volts.
A 2.3 volt nominal margin to burst capability of.3-times the normal operating pressure differential is provided for EOF 7 conditions.
1
(
EDDY CURREN'~ INSPECTION CAPABILITY l
j Detectinn of Indications Discussed l
The probability of detection of indications at tube support plate elevations has been previously evaluated for both bobbin coil and-l RPC probes at another plant.
This evaluation combined field l
inspection results with pulled tube examination results for TSP indications.
4 Ava!'able data for the IGA / SCC mode of tube degradation has been i
eva~.ated for detectability. The overall pulled tube data base for IGA / SCC indications shows voltage levels as high_or higher than obtained for ODSCC at comparable depths of-indications.
Degradation of this type is considered to be readily detectable, j
Pulled tubes from European plants with more significant-levels of-IGA exhibited bobbin voltages as high as 11 v_olts.
The - burst l --
characteristics of this tube are calculated to exceed the-Reg Guide.
1.121 minimum limit of 3
.imes the normal operating pressure i
differential.
l Based on the previous evaluation, the probability of-detection of stress corrosion cracking 40%
average (actual) depth ia approximately 100%.
Very short (approximately 0.2") cracks could have greater than 40% depth for detectability.
It should be noted that the Technical Specification 40% p3ngging limit represents an NDE determined _ depth, dependent on tub _e wall degradation type, which may represent-actual tube degradation.in the ranne of 55%,
based _ on an assumed uncertainty of 15% for phase angle based. depth calls.
i-1005 Eddy Cu nent Inspection Addressing RG 1.83. considerations, upon implementation of the criterion, all future refueling outages implementing IPC or APC Page-10 4
~
- _ -. - ~
wil'1. involve a 100% inspection of all hot-leg tubes down to the.
lowest cold leg intersection with identified ODSCC in order to 3
monitor the progression of ODSCC.
/
A-RPC inspection will be conducted for all flaw like bobbin probe o
indications exceeding a signal amplitude of 1.0 volt but less than-j_
2.5 volts.
The RPC results are to be evaluated to establish that
]:
the principal indications can be characterized as ODSCC.
If j
indications other than ODSCC are identified, these indications will be evaluated against a 40% depth requirement for tube plugging.
a The RPC inspeccion recommendation is consistent with a threshold value below which SLB leakage is expected to be negligible 1and l
other types of degradation _ (wear, cold leg thinning, etc. ) are not i
expected to have a significant effect on steam generator tube j
integrity.
3 l
Catawba Eody_purrent Methodoloctv 1
t The Catawba Unit 1
EOC 6 bobbin coil data was analyzed in l
accordance with the Eddy Current = Analysis Guidelines, Catawba Nuclear Station Unit 1,
Rev. 2 dated July 9, 19924 All signals 4
indicative of degradation were reported regardless of depth and-l with no minimum voltage threshold.
All data were analyzed with 2 independent reviews ( primary and secondary analysts).
Tha results of the primary and secondary analysts were compared and resolved by.
{
a team of 2 resolution analysts, Every HL call was then remeasured to - obtain a voltage value consistent with hastinghouse recommendation for measuring bobbin voltages for ODSCC degradation at HL TSPs.
This process was simply i
a measurement exercise, to obtain c voltage value related to a i
specific normalization, channel (550/130 Khz MIX 5), and signal isolation.
This was not a reanalysis as the presence of..the
{
degradation at each reported TSP had already'been determined and-j-
was not changed.
Tne remeasurement was performed in accordance i
with the jigt. leo Tube Suonort Re-Sizino Analysis Guidelineq, Rev.
O,-
i dated August 12, 1993.' An-analyst remeasured each HL TSP call and j
generated bobbin coil graphics depicting the call.
Each call was reviewed by a team of 2 resolution analysts who concurred with the accuracy of the; measurement, and assured all hL TSP calls were-resized.
4 h
j Every HL_ TSP call was then remeasured two more times:
i Once again on the current EOC 6 bobbin data with a 400/130_Khz mix and also from-the EOC5 1991 data with a__ 400/100 Khz mix.
These remeasurements were. performed to obtain ODSCC growth trending j
information.
The remeasurements -were performed in accordance with j
the Hotleo Tubf Sucoort Re_-31zino Analysis Guidelines for Growth Trending, Rev. c-dated August.12, 1992. _These measurements-were i
i also performed by an analyst with two resolution analysts reviewing-them for accuracy.
l
-n j
Page 11 4 -
I
{
1 To'obtain growth trending information over 2 cycles, a set of HL TSP calls pluggM after the EOC 5 analysis in 1991 were remeasured using the EOC 5 1991 data using a 400/100 Khz mix.
Resizing n..d resolution were performed in identical fashion to the growth trending described in the paragraph above.
Assumed Eddy Current UnrArtainties The NDE uncertainties on bobbin probe-voltage include such factors-as probe wear, calibration standard, and measurement repeatability.
The eddy current uncertainty allowance in messuring voltage is conservatively set at 20% for the full APC.
This large NDE uncertainty allowance f actors in both the contribution from analyst variability and probe wear.
A-16% NDE uncertainty is utilized for the development of the IPC.
TUBE LEAKAGE CONSIDERATIONS Although, following the implementation of the bobbin probe s3 nal 9
amplitude plugging criterion, tubes are not expected to burst under normal or accident conditions.
and are expected to retain sufficient margin, it cannot be assured that the cracks will not leak during all p;.
t ccadiH anc Operating Plant Leakau _Rxnerience Pulled tube examination results from.other plants indicate that through-wall cracks can potentially occur below 10 volts but that the associated crack lengths are short with no measurable leakage at operating conditions.
Leakage at operating conditions has:noe been identified for bobbin coil voltages less than 6.2 volts in a-3/4" tube in the field.
A total of three suspected tube leaks are attributable to OD.6CC at the TSPs in operating European steam generators.
Two leakers-(crack indication signal amplitudes of about 6.2 and 20 volts occurred at one plant in the same operating cycle.
In this case, a total cf 5 tubes were suspected to be contributing to the leakage of less tnan 140.gpd.
The other crack indication which leaked-had a signal amplitude of 7.7 volts and an indicated depth of 924 thro'igh-wall; the leakage also included leakage at roll transitions I
fron. other tubes and was 63 gpd.
No - field leakage has been
(
reported below 6.2 velts and no measurable field leakage was observed for the pulled tube f rom Catawba at 1.82 volts with through wall degradation.
A reasonable judgement for the leakage threshold at SLB conditions is approximately 1.82 volte with a significant likelihood of exceeding 6.0 volts for operating plant leakage detectability.
Accident Condition Loadings Lea.kagg Page 12
--.-n-.,
y
-.w,[
s.w, Y
4 For Catawba steam generator interim tube plugging criterion for TSP -
elevation ODSCC, the total SLB leakage limit of 1.0 gpm is met by determining the end of cycle SLB leak rate for tubes left in-l service with crack indications.
As 'noted above, a reasonable judgement for the zero leakage threshold _ at steam lina break l
conditions is approximately 1.82 volts.
This tube, removed from L
the Catawba plant, leaked at approximately 0.0025 gpm at SLB conditions.
Based on_the 3/4" pulled tube data base, a 3.5. volt indication conservatively would be bounded by leakage a.0.0044 gpm (1
liter /hr) at SLB pressures.
Expected SLB leakage for indications exceeding 3.5 volts EOC, is conservat'vely assumed to:
be 0.041 gpm (10 liters /hr).
Any interim criterion utilizing.a lower bobbin voltage based plugging limit which would predict EOC-voltag-s under 1.8 volts would result in no primary to secondary leakage at SLB conditions.
Petermination of Expected Leakage Durino SLB Conditions As noted above, a number of. tube intersections with -vol" age-signals that sa*isfy the interim tube plugging criterion wjl! remain in
- survice antil impected. and reevaluated during subsequent refueling outages.
It a-conservatively assumed that_ soma. of this number will have the potential for leakage during a postulated steam line break -
although leak rate testing of pu32ed tubes has not demonstrated potential f or - leakage from crack - -indications with l
voltages under-1.8 volts.
Therefore, leakage-rate during steam line break conditions is to be evaluated for the-tubes remaining in service at each outage.
Uncertainties in voltage signal, growth allowance, and leakage rate versus voltage are accounted for using Monte Carlo techniques.
An end of cycle voltage population is assessed (accounting 'or uncertainties) -for the -cotential for leakage during a postulated steam line break event.
Tha current distribution of number of indications'versus voltage has been obtained for the Catawba steam generators.- Also, the most l
recent change in voltage was obtained on an effective full power-year (EFPY) basis.
The-current voltage distribution-is then combined with the volt &ge growth rate. measurement, using Monte Carlo techniques, to establish an end of -
cycle _ voltage L
distribution.
Uncertainty in the voltage signals-for the' current-inspection is accounted for in a statistical manner via Monte Carle simulations utilizing distributions for the NDE uncertainties.
An interim criterion which uses a bobbin - probe. voltage based plugging limit which would predict EOC voltages less-than 1.8 -volts for flaw like indications would not result in. leakage at. SLB :
conditions.
Based upon.the probabilistic _ determination of EOC voltages using Monte Carlo, and _ applying the. leak rate _ values listed above for the three voltage thresholds, an EOC SLB leak rate of 0.54.gpm has been calculated which does not adversely af fect
- radiological consequences.
- The - deterministic leakage has been-determined to be.15. gpm.
Page 13 l.
l' i
-~ ~
[
i l
LIMITING CRACK LEAKAGE LEVELOPMENT The RG 1.121 acceptance criterion for establishing-an operating leakage limit requires that plant shutdown be initiated if the leakage associated with the longest permissible single crack is i
exceeded and, thus,.provides for leak-before-break.
The longest j
permissible crack is the length that provides a-factor ut safety of 3 against bursting at. normal operating pressure differential.
For 3
3/4" tubing, a voltage amplitude of 4.1 volts for-typical ODSCC corresponds to meeting this tube burst rcquirement at the lower 95%
confidence level on the burst correlation.
Alternate creek morphologies can correspond to 4.1 volts so that a uni que crack-4 length is not defined by the - burst pressure versus voltage correlation.
Consequently, typical burst pressure versus through-i wall crack length correlations are' used below to deline the
~
" longest permissible crack" for evalt9 ting operating leakage limits.
i The single through-wall crack lengths that result in tube burst at~
i 3 times normal operating pressure differential and SLB conditions are about 0.48 inch and 0.76 inch, respectively.
A maximum operating leak rate of 150 gpd will be implemented at the Catawba 4
Unit 1 plant.
Primary to secondary leakage up to 150 gpd provides for the detection of a 0.4 inch long through-wa'_1 crack, based on mean leak rates and a 0.6 inch long through-wall-crack at the 95%.
confidence level-leak rates.
- Thus, for cracking that may be occurring at location within the TSP elevation, the Catawba 150 gpd limit provides added conservatism for plant shutdown prior to j
reaching critical crack lengths for SLB conditions at leak rates less than a 95% confidence level and for three times - normal
+
operating pressure differential-at less than nominal leak rates.
l Therefore, leak-before-break' is enhanced for tubes with ODSCC l
occurring within - the TSP intersections with the 150 gpd leakage
-limit.
d l
DOSE ANALYSIS i
Of the accidents analyzed in Chapter 15 of tha FSAR, only.the Main Steam Line Break Analysis is effected by IPC. The Catawba Technical Specifications limit operating leakage to 0.4' gpm (576 gpd) from all steam generators, - 150 gpd in any one steam generator. The calculated leakage during a postulated SLB event is approximately 1 gpm total leakage from all steam generators.
Assuming 1.0 gpm total leakr.ge (0.64 in the faulted loop, 0.3 in the non-faulted-loops), the results of a dose evaluation for a main steam line rupture outside the containm_ent structure indicate that
'he limiting dose, Exclusion Area Boundary Thyroid Dose, is within the value presented in the Catawba FSAR.
CONCLUSION Page 14 1
)
i J
,_1
The' steam generator tube plugging criterion for the tube support plate elevation degradation observed in c.he Catawba steam-generators is summarized in Table-1.0.
The recommended tube l
plugging criterion is based upon bobbin coil inspection voltage signal amplitude, which is correlated with tuba burst capability.
The criterion is developed to preclude free span tube burst if it is postulated that TSP displacement would occur under accident condition loadings.
The interim tube plugging limit provides RG 1.121 tube burst margin.
The interim criterion-plugging limi' '
c (much more conservacive than the structural plugging limit) is expected to result in the majority.of the end of cycle (EOC) voltages being below the SLB leakage-threshold.
A sma.t1 number'of-tubes can be expected to grow at faster than expected rates.
However, the leakage potential from the crack indications in these tubes is expected to be less than the 1.0 gpm total leakage assumed in the FSAR accident analyses during a pcstulated SLB, thereby maintaining offsite doses to less than a small fraction of the 10CFR100 guiuelines.
Table 1.0 INTERIM STEAM GENERATOR TUBE PLUGGING CRITERIr"T FOR TUBE SUPPORT PLATE ELEVATION ODSCC BOBBIN SIGNAL ACTION l
VOLT 6G_E s 1.0 LIMITED *
> 1.0 but less than 2.5 RPC **
2 2.5 PLUG-OR REPAIR i
If it is found that the pot'ential for steam line break leakage at end-of cycle conGitions for tubes planned to be left in service exceeds 0.7 gpm in any steam generator, then additional tubes will be-plugged to reduce steam line break leakage potential in-that steam generator to below 0.'7 gpm.
If additional tubes are to be plugged or repaired in-crder to show compliance with the 1.0 gpm leakage limit, the largest bobbin coil voltage flaw indications would be plugged or repaired.
Plug or-repair if' indication of ODSCC are detected.
The 4.1 volt signal amplitude, which comprises the structural requirement for a free span tube to have a burst capability equivalent to 3 times the normal operning pressure differential (3750 psi), _must be reduced f or NDE mehsurement uncertainties _ and Page 15
4, j
i l
projected crack growth between inspections-to esta*lish ~the
-allowable structural limit for tube plugging.
The development of
[
the 4.1 volt structural. limit is based upon a lower 95% prediction l-interval reduction of burst data; with burst pressures adjusted to lower tolerance limit strength properties at operating temperatures l
(650 F).
The voltage limit with this same-basis is 10.95 volts at-the SLB pressure differential ot' 2650 psi.
i In addition to meeting the structural criter'-
ase of the interim plugging criterion also limits normal operr :. ;o - and SLB leakage to l
acceptable levels during all plant
.co.u. ' l an-For normal 1
operating conditions, no leakage has been t.;crted below 6.2 volts.
i I
The threshold voltage for detectable plant leakage -is estimated at I
about 6.0 volts.
For SuB conditions, pulled tube data indicate-that a threshold voltage of about 1.8 volts could result in through-wall cracks long enough to les;'t at SLB conditions.
1 i
p l
l l
l Page 16 w
v 1
Projected End of Cycle 7 Tube Burst Margins oatio to SLB ECT Volt Lower'95% B.P.
Ratio to 3 x delta P 3/4 7/8 3/4 (3750) 7/8 (4380)
Ratio 3/4 7/8 Ratio 90%-Confidence Level Data 1.8 4660 5640 1.24 1.29
.96 1.76 2.13
.83 99% Confidence Level Data 3.37l(7/8) 5013 1.14 1.89
_2.81'(3/4) 4186 1.12
.98 1.58
.84 99% Confidence Level Data IPC = 0.84 volt f1.14 1.0 1.62-1.89
.86 2.58 l(3/4) 4284 5013 1.14 99% Confidence Level Data IPC = 0.70 volt 2.39,(3/4) 4366 5013 1.16 1.14-1.02 1.65-1.89
.87
'99% Confidente Level Data IPC = 0.50 volt 2.11 (3/4) 4482-5013 1.2 1.14 1.05 1.69 1.89
.89 Page 17 L_