ML20217H911
| ML20217H911 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 10/10/1997 |
| From: | Langenbach J GENERAL PUBLIC UTILITIES CORP. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 6710-97-2456, NUDOCS 9710170118 | |
| Download: ML20217H911 (15) | |
Text
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
o (GU L".Ts'T WCW
[ll,f,[,ytfousso teimwmi October 10, 1997 6710 97 2456 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555 Gentlemen:
Subject:
Three hlile Island Nuclear Station, Unit 1 (Thil 1)
Docket No. 50 289 Operating License No. DPR 50 Additional information to Suppon Te.hnical Speci0 cation Change Request (TSCR) No. 268 - OTSG Inspection Criteria for 12R Examinations This letter provides additional information regarding Thil 1 Technical Specifkation Change -
Request (TSCR) No. 268 in response to a request from the NRC during conferenc, : alls on October 8, and October 9,1997. This information is clarifying in nature and therefore does not affect the safety evaitation justifying the proposed license amendment and does not change any of the conclusions supponing the no significant hazards consideration analysis pre vided in TSCR No. 268.
Also enclosed as attachment 2 is the Cen10cate of Service for this request certifying service to the chief executives of the township and county in which the facility is located, as well as the designated official of the Commonwealth of Pennsylvania,13ureau of Radiation Protection.
Sincerely, 24100 E Y2 Kf GA James W. Langenback Vice President and Director, Thil h1RK Attachments: (1) Additional Information Regarding Technical Specification Change Request No. 268 (2) Cenificate of Service for Additionallnformation Supponing Technical 170'301 Speci0 cati n Change Request No. 268 i
AcDI i
cc:
Administrator, NRC Region 1
-f A Thil Senior NRC Resident inspector Thil 1 Senior NRC Project hianager j
((
9710170118 971010 PDR ADOCK 05000289 P
hiETROPOLITAN EDISON C0h1PANY JERSEY CENTRAL POWER & LIGilT C0h1PANY AND PENNSYLVANIA ELECTRIC COhiPANY TilREE hilLE ISLAND NUCLEAR STATION, UNIT 1 Operating 1.icense No. DPR&
Docket No. 50 289 Additional Information Regading Technical Specification Change Reque:t No. 268 -
COhih10NWEALTil OF PENNSYLVANIA
)
) SS:
COUNTY OF DAUPillN
)
This additional information is submitted in support of Licensee's request to change Appendix A to Operating License No. DPR-50 for Three hiile Island Nuclean Station, Unit 1. All statements contained in this submittal have been reviewed, and all such statements made and matters set forth therein are true and correct to the best of my knowledge.
GPU NUCLEAR INC, BY:
34RW N-M6 ice President and Direhr, Thil Sworn and Subscribe before me this$_h' dayo
_ L> c,1997.
/ I u~, w,
~
umeak
~ wan na,m avis liQgif, f(UT,%y G04 /,bb QN Qi b
NITACilM ENT I Additional information Regarding Technical Specification Change llequest No. 268
Additional Information Regarding TMI-l Technical Specification Change Request No. 268 NRC Question:
II hat assurance is there that the volumetric ID IGA that is allowed to remain in service will not leak ifsubjected to postulatedMSID accident conditions?
GPU Nuclear Response:
GPU Nuclear letter of October 9,1997 described the vaumetric Inside Diameter (ID) Inter-Granular (IGA) indications that were characterized by Eddy Cunent Testing (ECT) examination during the Thil-1 Cycle Refueling (12R) Outage, and provided information to support a reasonability argument that the indications remaining in service would maintain leakage integrity during postulated hiain Steam Line Break (hiSLB) accident conditions.
Figures 1,2, and 3 show the distribution of Volumetric ID Indication lengths for the circumferential and axial indications along with the distribution of axial lengths versus circumferential length of indications found during the Thil-1 12R Outage ECT.
The purpose of this response is to provide additional information and analysis to demonstrate that
- ufricient leakage mttedty would be maintained under the most bounding postulated conditions.
As described in our Octobt V iW response, in-situ pressure tests were performed on three tubes with circumferential indicataa m 0. c 0.29 and 0.24 inches in circumferential extent and two volum:tric indications of 0.27 x 0.25 :nd 0.14 x 0.30 inches (axial by circumferential extent respectively). The test pressures were 4,400 psi which induced tube axial loads between 1,035 and 1,044 lbf. No leakage was measured during any of these tests which have a minimum detectable leakage measureme t sensitivity of.0005 gpm.
Ahhough certain indications wem selected as most limiting for in-situ testing, additional less limiting indications were preser in the tube sections being tested and were pressurized along with the limiting flaws. The total number ofID volumetric IGA indications tested was six (6). These ranged in axial extent from 0.12 to 0.27 inches and in circumferential extent from 0.16 to 0.30 inches. The rotating pancake coil (RPC) voltages ranged from 2.36 to 9.2 volts. The test results all showed no leakage at a maximum test pressure of 4,400 psi. Details of the tested flaws are given in the Table 1.
In addition, fracture mechanics calculations were used to determine the extent that the test of the axial flaws at 4,400 psi could be used to bound crank growth that would cause leakage in circumferential flaws. These calculations show that a 0.25 inch circumferential indication will not leak under hiSLB load conditions. Since there are only 25 indications len in service that exceed this size, the leakage calculations conservatively bcand this condition.
In addition to the in-situ pressure tests and the fracture mechanics calculations, a leak test has been conducted with tubing representative of Thil-1 OTSG tubing and with manufactured ID IGA indications. These indications were characterized by ECT and behave similarly to indications in
- crvice in the Th11-1 OTSGs. The indications have been sized to be significantly larger than all indications len in service. This test was run at primary side conditions of 595 F,2750 psi and
'6710 97-2456 Page 2 of 5 secondary side pressure of 90 psi. The material was axially loaded to produce the strain that would be induced under htSLB loads and the flaws did not leak under those conditions.
Furthermore, the PICEP code was used to calculate the flaw size of a circumferential flaw which would produce the minimum detectable leakrate under the test conditions. If this flaw is conservatively assumed to exist in all volumetric IGA indications, a total leakrate can be calculated for this flaw size at htSLB loads of 3,140 lbf. The reference load of 3,140 lbfis taken from B&W Topical report B AW 10146, " Detection of htinimum Required Tube Wall Thickness for 177 FA Once Through Steam Generators," dated October 1980. The results of these calculations demonstrate that approximately 460 indications could remain in service without exceeding the limits of a small fraction of 10CFR 100 limits.
A. FRACTURE h1ECHANICS ANALYSIS OF LlhilTING FLAW SIZE Fracture meci anics analysis methods were applied to daermine the extent to which in-situ mnsure testing of axial flaws bounds the leakage integrity of circumferential flaws under axial
!cs Jue to the postulated hiSLB. Stress intensity factors were computed using models for an Nrkal axial inside surface crack and a part-through, finite-length circumferential inside whe crack in a cylinder, it was conservatively assumed that the ID IGA acted like a sharp crack perpendicular to each principal stress direction. The stress intensity factor is a perameter which represents the interaction of the nominal stress field with the crack tip so that it can be used as the driving force at the crack tip.
The results indicate that the stress intensity factor for a 0.25" axial flaw in the in-situ pressure test exceeds the stress intensity factor for a 0.25" circumferential flaw under the accident load at a common depth greater than 60% through wall. This means that the in-situ pressure test assured leakage ir. ;grity for 0.25" long circumferential flaws for depths breater than 60%
through-wall using similar size axial flaws as a gauge. Failure of the remaining ligament ahead of the flaw is not expected to occur. The material fracture toughness is expected to be much greater than the applied stress intensity which is on the order of 16.0 ksi(in)" at 60% through-wall. Ligament failure is not credible for either flaw orientation with shallower flaw depths.
B. LEAK TESTING OF TUBE SAhiPLES GPU Nuclear has performed " hot" leak testing on an OTSG tube sample containing one large and one smaller ID IGA flaw. The flaws were produced in the laboratory by exposure to sodium tetrathionate. (Sodium tetrathionate produces reduced sulfur species similar to those produced by the sodium thiosulfate which initiated the Thil-1 steam generator tube damage in 1981.) The leak testing was performed in accordance with EPRl's " Guidelines for Leak and Burst Testing of Steam Generator Tubes."
l The sample tubing was material typical of Thil OTSG tubing with a yield strength of 46. I ksi.
The tubing used in the Thil-1 generators has yield strengths ranging from approximately 42 ksi to 57 ksi. The larger flaw had a 0.540" bobbin coil eddy current voltage of 4.51 Volts, and axial and circumferential extents of 1.01" and 0.51", respectively, by hiRPC examination. The smaller flaw size was 0.30" axial by 0.20" circumferential extent. It had a.540" bobbin voltage
- 6710-97-2456 Attachment i Page 3 of 5 of 2.'97 Volts. (The eddy current probes and techniques used to establish these values were equivalent to those used in the Thil-1 generator tubes during the 12R Outage.) These extents are large relative to the 0.25" axial and 0.35" circumferential extents of the largest volumetric ID IGA remaining in service for Cycle #12 operation.
The leak testing was conducted at the following conditions:
Primary Side Temperature:
595120 F Primary Side Pressure:
2,750 100 psig Secondary Side Temperature:300 F Secondary Side Pressure:
90 20 psig Axial Tube Loads:
948 lbs,1402 lbs,1755 lbs,2400 lbs,2700 lbs, and 2843 lbs.
The axial tube loads included a load imparted by a jack, as well as an end load imparted by the internal pressure of the primary side water against an end cap installed on the sample. The tube sample began to strain signincantly when it was axially loaded above 2,700 lbs.; it was pulled until the strain equaled the maximum strain value that would be induced in any tube in the steam generators during a postulated MSLB.
This testing is conservative from the standpoint that the prescriptive condition under which the peak axial tube load occurs during the MSLB scenario actually occurs at a tube temperature of 235 'F (BAW 10146," Detection of Minimum Required Tube Wall Thickness for 177 FA Once Through Steam Generators, dated October 1980). The tubing strength (including that in the area of the tested flaws) would be greater at the lower temperature of 235 F as compared to the strength of the tubing at the 595 F test temperature.
No leakage occurred during the tests.
C. BOUNDING LEAKAGE CALCULATIONS Deternination Of Minimum Detectable Crack Size The minimum detectable crack size was determined from the minimum detectable leakage flow (0.0005 gpm at 80 deg F) by running the PICEP code under the following conditions and varying the circumferential crack length until the calculated leakage was equal to the minimum detectable leakage:
Differential Pressure = 2,500 psi Axial Load = 1,000 lbf Temperature = 80 F The result was a crack length of 0.017 inches.
Leakage During Main Steamline Break Conditions A hypothetical main steam line break (MSLB) accident that results in the maximum possible axial tube load of 3,140 lbf and a maximum possible differential pressure of 2,500 psi was applied to a tube with the 0.017 inch crack length and a Reactor Coolant System (RCS)
r "6710-97-2456 Page 4 of 5 temperature of 450 F. The result was a leakage of 0.036 gpm (corrected to 579 F,2,100 psia conditions).
If this leakage rate were applied unabated for 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />, the integrated leakage would be a total of 4.3 gallons per indication. For 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> it would be 21.6 gallons per indication. Since the 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> Exclusion Area Boundary (EAB) leakage limit, to not exceed a small fraction of 10 CFR 100 limits, is 3,228 gallons and the respective Low Population Zone (LPZ) leakage limit for the duration of the postulated MSLB accident is 9,960 gallons as approved by the NRC's Safety Analysis Repon (SER) for TMI-l Technical Specifications Amendment No. 204, then approximately 460 indications could have the minimum detectable crack length at the start of the event and not exceed allowable leakage limits. This is limited by the 9,960 gallon limit for the course of the MSLB accident.
Conservatisms There are numerous conservatisms applied in this evaluation that, if removed, would greatly increase these margins. The conservatisms are described below:
1.
The differential pressure used in determining the minimum detectable crack length was 2,500 psi versus the actual test conditions of 4,400 psi. This was done as a result of PICEP code pressure input limitations. If the higher pressure were used, it would reduce the minimum detectable crack size in approximately a proportional manner to a crack length on the order of about 0.010 inches. This would increase the number of tubes that could leak under the postulated MSI B accident without exceeding the dose limits.
2.
The application of a 3,140 lbfload for the entire event duration is extremely conservative since this load will only exist during the period of time that the overcooling is occurring, which was assumed in the analysis (B&W Topical report B AW 10146, " Detection of Minimum Required Tube Wall Thickness for 177 FA Once Through Steam Generators,"
dated October 1980) to be 20 minutes. Lower loads would be expected after 20 minutes.
The assumption that the maximum load persists for hours results in much higher leakage than would actually be expected.
3.
The analysis of the 3,140 lbf MSLB axial load assumed the break of a 36 inch main steam line and a 1650 gpm Emergency Feedwater (EFW) flow rate for 20 minutes. TMI has 24" main steam lines and has cavitating venturis which would physically limit the EFW flow rate to much less flow. Plant specific MSLB an@ses indicate that the actual maximum axial tube load for TMI would be 1,310 lbf. If the leakage were calculated on the basis of the 1,310 lbf axial load, there would be substantial margin to the results discussed above.
A reduction in these large conservatisms would translate into a much larger number of tubes allowable with a minimum crack length (or a smaller number of tubes with a larger undetected crack size) that would satisfy the TMl-1 licensing basis.
PICEP Sensitivity Analyses Leakage sensitivity analyses were conducted on the reference case which was defined to be a 0.017 inch circumferential crack length with an axial load of 3,140 lbf, a differential pressure of 2,500 psi and a temperature of 500 F. Input sensitivity variations included Young's Modulus,
'6710-97-2456 Page 5 of 5 -
Yield Stress, surface roughness, temperature, axial load, and differential pressurt
. ; results
- are shown in Table 2. As can be seen, none of the variations resulted in more thau ;. actor of 2 increase in the leakage. This is a small vsriation in contrast to the large variations associated with the conservatisms in the assumptions as discussed above. The assumption of a test pressure of 2,500 psi vs the actual test pressure of 4,400 psi alone ofTsets the sensitivity-variation. The case which provided the most sensitivity was a reduction in the yield stress from 42,000 psi to 35,000 psi. Data on TMl-1 specific tubes indicates that the yield stress varies from approximately 42,000 psi to a high of 57,000 psi. A reduction of Young's modulus from 2.89 E+07 to 2.0 E+07 resulted in a maximum increase in leakage of about a factor of 1,8. The measured Modulus for TMI tubes varied from 2.94 E+07(at 75'F) to 3.1 E+07 (at 600 F) which bounds the base input assumption.
D, CONCLUSION Since axial tube loads that approach MSLB conditions cannot be achieved for in situ tests, GPU Nuclear has used in situ pressure test results combined with leakage calculations, structural calculations and leak testing of manufactured flaws to conclude that TMl OTSG tubing will maintain the required leakage integrity under MSLB conditions.
Figure 1: Volu. etric ID Indications m
Circumferential Length Distribution 350 r
E300--
,g 16250
.9 2 200 0150 g
p100-S 50 3-0
=
0 0.15 0.3 0.45 0.6 0.75 0.9 Circumferential Length (Inches)
~
~
pigure 2 VOIS*
1 ic ID Indications oisteb" "
350 g300 i
O
~
% 250 O
~
$200 "o 1 5 0 B
o 100 E
[
~
0 0
0.15 0.3 0.45 0.6 76 Axial Length n< oches)
1~3 I
)
Figure 3: Circ vs Axial Length ID Indications in Service, Both OTSGs 0.6 30.5
_8 20.4 8'
d%"
3 0~3 -
h-E0.2-s_
y
.8 0.1 o
, e,,
.w 0
0 0.1 0.2 0.3 0.4 0.5 0.6 Axial Length (Inches) o0 g -
. ~...... -. - _ -... - - - -. -. -..
-.. - =.. - ~....,.. -.. - -..
... ~. - - - - - _.
t f
TABLE 1
}-}
TMI-1 IN-SITU PRESSURE TEST _ LIST l
t io VOLUMETRIC AND CIRCUMFER'5.NTIAL INDICATIONS
}
i i
NS-8ITU TEST RESULTS TUBE AND EDDY CURRENT NEOftRAATION TUBE N0FORIAATION PLUS POSIT DATA BOSSSL DATA CO M GPRIg GPRIg GPRIgfeOPD MAX Ruw TUBE LOCATION LENGTH VOLTS EST.% OREBfTATIOII VOLTS EJT.%
900PO anna El POST RASLB PRESSURE A.OTSG 93 119 UTS + 1.03 c.24 C 17.13 97%
B SCI 2.3e 93%
UTS Che e
e e
4400 A4TSG '
93 119 UTS + 1.93 0.21 A x 3.25 C 9.2 OGIA ID WOL 2.39 93%
UTS Vetumsetc e
a e
dage A.OTSG 93 119 UTS + e.58 c.24 A x 0.2T C 3.03 IIfA B VOL 3.98 40%
UTS Vetummerte e
e e
4sse
[
A.OTSG 10T 120 ETL-SAs e.2T A x 3.25 C 2.38 10fA BVOL BWO NfA UTSVetumstric e
e e
4400
[
A.OTSG 13T 12e ETL-2.08 e.12 A x 0.19 C 3.31
- NfA ID WOL 1AT 2F%
UTS Volumstric e
e e
440s A.OTSG 1ST 120 ETL - 3.99 0.13 A x e.13 C 2A2 IIfA IDWOL 1A4 13%
UTS Vetuneerte e
e e
dose i
D4TSG Se 13 UTS - 3.17 0.29 C 3.00 SIIA ID SCI 4.28 ST%
UTSF Circ e
0 0
440s l
I D.OTSG 134 19 UTS - e.2e e.14 A x 4.30 C 3.04 lefA IDWOL 2.03 00%
UTSF Volumektc 3
-e e
does f
D-OTSG its 3e UTS + 1.T4 c.51 C 5.11 ItfA ID SCI S.es 43%
UTS Circ e
e e
4400 r
i UTSF = Upper Tuboehoot Secondary Fece l
i KET = IGneSc Exponalen Trenettion t
t
$0G = Non QuenMeetite insEcaMon t
i i
j i
I f
r b
?
I
TABLE 2 PICEP SENSITMTY ANALYSES Case Cracic Size (in)
Load Pressure (pse) Temp Young's Yield
., Surface Leak Rate Leak Rate feORRAAUZED TO Circumferental (Ibf)
Tube / SheE (dog F) teodulus Samos Roighnson (IbnVsec) gpm(579F)
.cas opas 1
0.017 3140 2515/15 500 2.89E+07 4.20E+04 2.00E-C4 3.55E-03 3.60E-02 1.00E+00 2
0.017 1000 2515/15
.80 2.69E+07 4.20E+04 2.00E-04 6.70E-05 6.75E-04 1.88E-02 3
0.017 1000 2515/15 500 2.89E+07 4.20E+04 2.00E-C4 5.40E-05 5.44E-04 1.51E-02 4
0.017 3140 2515/15 80 2.89E+07 4.20E+04 2.00E44 4.30E-03 4.33E-02 1.20E+09 5
0.017 3140 2515/15 500 2.89E+07 4.20E+04 2.00E44 3.40E-03 3.43E-02 9.52E-01 6
0.017 1000 1515/15 80 2.89E+07 4.20E+04 2.00E-04 3.60E-05 3.63E-04 1.01E-02 7
0.017 1000 1515/15 500 2.89E+07 4.20E+04 2.00E-04 3.80E-05 3.63E44 1.01E-02 8
0.017 3140 1515/15 500 2.89E+07 4.20E+04 2.00E-04 1.80E43 1.81E42 -
5.04E-01 9
0.017, 3140 1515/15 80 2.89E+07 4.20E+04 2.00E-04, 2.60E-03 2.62E-02 7.28E-01 10 0.017 1000 2515/15 80 2.00E+07 4.20E+04 2.00E-04
- 1.00E-04 1.01E-03 2.80E-02 11 0.017 3140 2515/15 500 2.00E+07 4.20E+04 2.00E-04 5.10E-03 5.14E42 1.43E+00 -
12 0.017 3140' 1515/15 500 2.00E+07 4.20E+04 2.00E-04 2.70E-03 2.72E-02 7.56E-01 13
'O.017 3140 2515/15 80 2.00E+07 4.20E+04 2.00E-04 '
6.40E-03 6.45E-02 1.79E+00 14 0.017 3140 2515/15 80 3.12E+07 4.20E+04 2.00E-04 4.00E-03 4.03E-02 1.12E+00 15 0.017 3140 2515/15 80
. 2.89E+07 3.50E+04 2.00E-04 6.90E-03 6.96E-02 1.93E+00 16 0.017
~3140 2515/15 80 2.89E+07 4.90E+04 2.00E-04
. 2.90E-03 2.92E 02 8.12E-01 17 0.017 3140 2515/15 80 2.89E+07 4.20E+04 1.50E-04 4.40E-03 4.44E-02 1.23E+00 18
,0.017 3140 2515/15 80 2.89E+07 4.20E+04 2.50E44 4.30E-03 4.33E 02 1.20E+00 l CASE 1 = BASE CASE l
l ATTACilM ENT II Certificate of Service for Additional Information Regarding TMI-I Technical Specification Change Request No. 268
g.
E*-
UNITED STATES OF AMERICA
- NUCLEAR REGULATORY COMMISSION IN THE MATTER OF DOCKET NO. 50 289 GPU NUCLEAR INC LICENSE NO. DPR-50 CERTIFICATE OF SERVICE 1
This is to cenify that a copy of the additional information regarding TMI-l Technical z Specification Change Request No. 268 for Three Mile Island Nuclear Station Unit 1, hat, on the date given below, been filed with executives of Londonderry Township, Dauphin County, Pennsylvania; Dauphin County, Pennsylvania; and the Pennsylvania Department of Environmental Resources, Bureau of Radiation Protection, by deposit in the United States mail, addressed as follows:
Mr. Darryl LeHew, Chairman -
Ms. Sally S. Klein, Chairman j
Board of Supervisors of Board of County Commissioners Londonderry Township of Dauphin County R. D. #1, Geyers Church Road Dauphin County Courthouse Middletown, PA.17057 Harrisburg, PA 17120 Director, Bureau of Radiation Protection
. PA Dept. of Environmental Resources Rachael Carson State Office Building P.O. Box 8469 Harrisburg, PA 17105 8469 Att: Mr. Stan Maingi
=
GPU NUCLEAR INC.
BY:
,m /,'
- 6
~
V e Presideilt anil Di or, TMi
~
/O!/d
'7 DATE:
t
_-