ML20082B952
| ML20082B952 | |
| Person / Time | |
|---|---|
| Site: | Byron, Braidwood  |
| Issue date: | 04/03/1995 |
| From: | Saccomando D COMMONWEALTH EDISON CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9504060104 | |
| Download: ML20082B952 (20) | |
Text
.,.....
,y A
s
.. f"% Commo wealth Edison.
'S
' =
'f
) ' '1400 Opus Placa '
'C, O ~7 Downers Grove, minois 60515 4 g
,/:
a.
F 1 April 3,1995 "
Office of Nuclear Reactor Regulation
.U.S. Nuclear Regulatory Commission -
L Wa'shington, D.C. 20555 Attn: ' Document Control Desk
Subject:
Byron and Braidwood Nuclear Power Stations Additional Information Regarding an Increase of the Interim Plugging Criteria for Steam Generators 7
NRC Docket Nnmbers: 50-454 and 50-456.
L
Reference:
1.
M. Lynch letter to D. Farrar dated March 24,1995, transmitting the Nuclear Regulatory Commission Comments on Commonwealth Edison's Company Proposed Test Program for
.j u
L Steam Generator Tube Indications Restricted From Burst l
2.
D. Saccomando letter to NRC dated March 20,1995 transmitting Information Regarding Questions Pertaining to Increased Interim Plugging Criteria 3.
D. Saccomando letter to the Nuclear Regulatory Commission l
dated March 15,1995, transmitting Commonwealth Edison's l
response to the Nuclear Regulatory Commission Concerns as Documented in Meeting Summary dated March 14,1995
]
4.
M. Lynch letter to D. Farrar dated March 14,1995, J
transmitting a Meeting Summary from' the February 23,1995, NRC/ Comed Meeting '
On February 23,1995, Commonwealth Edison Company (Comed) met with the Nuclear Regulatory Commission (NRC) to discuss the proposed license amendments regarding the increase in the Interim Plugging Criteria for Byron Unit l'and Braidwood Unit 1 steam generators. As a result of that meeting, a
. summary was issued which contained 4 concerns. In Reference 3,' Comed responded to those concerns and included a test plan for obtaining bounding steam
~
line break leak rates for indications restricted from burst (IRB) by the tube l
support plate (TSP). This program was proposed to support an NRC suggestion
~ ;
[
that the Comed WCAP-14273 leak rate calculation be simplified.
l L
.,m. _.,_, J 4 0 0 3 2 3
9504060104 950403 l-PDR ADOCK 05000454 g
{8 p
P, PDR q
g I:
I
E
.NRC Controlled Document April 3,1995 On March 22,1995, a teleconference was held between Comed and the NRC in which the Staff provided some comments regarding the overall conduct and strategy of the test program. A summary of those comments were transmitted in Reference 1. The attached document contains Comed's expanded description of the proposed IRB test program. Specifically, it contains:
Attachment A: History and Test Program Objective Attachment B: Total Test Program Development and Evaluation of Test Variables Attachment C: Proposed Bounding Test Plan J
Attachment D: Response to NRC concerns that were detailed in Reference 1.
It is important to be clear that a full scope, total test program has been developed.
The basis for development of this program is provided in Attachment B. The total i
test program was used to develop a bounding test program which is presented in i
Attachment C. Comed currently plans to conduct tests within the bounding test program only. This testing is currently scheduled to be complete by about July 15, 1995.
Comed understands that the Staff has agreed to meet with us on April 11,1995, to further discuss the elements of the test program and any test results completed prior to the meeting. We look forward to this opportunity and welcome any questions prior to that meeting that we could address on April 11,1995.
If you have any questions concerning this information please contact this office.
i Sincere N
e Denise M. 'accoma\\ndo l
Nuclear Licensing Administrator l
Attachments j
cc:
D. Lynch, Senior Project Manager-NRR R. Assa, Braidwood Project Manager-NRR G. Dick, Byron Project Manager-NRR S. Dupont, Senior Resident Inspector-Braidwood 4
H. Peterson, Senior Resident Inspector-Byron J. Martin, Regional Administrator-RIII Office of Nuclear Safety-IDNS o nia : nye ste.nnecT.st a
n e
Attachment A Historv j
On February 23,1995, Comed and the NRC met to discuss the potential increase of the interim plugging criteria for 3/4" tubes from 1.0 volt to 3.0 volts until steam generator replacement at Braidwood 1 and Byron 1. During that meeting Comed presented a model which quantified the low probability which exists for indicaticns that may potentially " leak" within the confines of the TSP at pressures equal to main steam line break (MSLB) pressure.
I At the St?s request Comed provided the StatTwith data which would quantify 4
the potential benefit from using this analytical leakage model. Additionally, Comed also pursued the use of an alternate leak rate model which did not rely on a bobbin voltage correlation with crack length. A description of this model and results quantifying the potential benefits were provided to the Staffin Reference 2.
To fully support the alternate leak rate model, it is necessary to defino a fixed leak rate for a single indication restricted from burst (IRB) by the presence of the TSP. One approach is to define an IRB bounding leak rate and apply that leak rate to each IRB independent of size. In an effort to obtain a bounding leak rate for IRB, Comed submitted a proposed laboratory test program to the Staffin-Reference 3. This proposed laboratory test program used 7/8" diameter tubing i
specimens with throughwall outside diameter stress corrosion cracking (ODSCC) crack lengths of 0.34",0.40" and 0.50". These specimens would be pressurized to approximately 15% above calculated burst pressure while contained within a simulated support plate. Leak rate measurements would then be performed at both room and elevated temperature.
The development of this phase of the test program was based upon samples immediately available for testing. Specifically, our decision to use 7/8" diameter specimens with 0.34",0.40" and 0.50" ODSCC crack lengths, rather than the Byron and Braidwood size 3/4" diameter tubing was because of specimen availability. Laboratory induced ODSCC specimens take several weeks to l
generate. Furthermore, it is Comed's position that the leak rate is a geometrical function that is net dependent on the tube diameter. Specimens in the 3/4" diameter size will not be available until approximately May 15,1995.
It is Comed's intention that the plan and the preliminary results from this test program will be reviewed by the EPRI SGDSM Ad-Hoc committee, which is scheduled to meet in early April. Modifications to the test plan, as well as the leak
)
rate calculation, will be based upon suggestions from the committee.
E;nla: Bytwd t StgenmCTesT;)
--a.,n-~
., -. ~.,
- On March 22,1995, the StatY and Comed participated in a teleconference to discuss this test program (Reference 3). During that teleconference, Comed recognized that the proposed test program needed to be enhanced to address new concerns.
Based upon these concerns Comed proceeded with the following actions:
1.
_ Redefine objectives of a new and expanded laboratory leak rate test program.
2.
Develop a list of required test variables to meet the leak rate' test program objectives.
3.
Develop a parametric sensitivity study test matrix based upon the list of variables.
4.
Develop a detailed prioritized test program to bound variables of the parametric sensitivity study.
5.
Define a final test plan to meet test objectives.
Test Plan Obiective Develop a bounding MSLB leak rate for large (>0.6") indications which have been restricted from burst (IRB) by the presence of the tube support plate (TSP). This bounding leak rate value will be used in a leak rate calculation which takes into account both freespan and IRB leakage.
Comed does not have an objective to create or test a low voltage (1 volt or less) 3/4" long crack. However, long cracks (>0.6") will be created and tested.
e K:nia:Bytw3:Stoen/IPCTesT:4
Attachment B TOTAL TEST PROGRAM DEVELOPMENT Test Plan Obiective Develop a bounding MSLB leak rate for large (>0.6") indications which have been restricted from burst (IRB) by the presence of the tube support plate (TSP). This bounding leak rate value will be used in a leak rate calculation which takes into account both freespan and IRB leakage.
Comed does not have an objective to create or test a low voltage (1 volt or less) 3/4" long crack. However, long cracks (>0.6") will be created and tested.
Variables Considered in total test nrogram As part of Comed's development of a total laboratory test program from which a bounding test program was derived, the following variables were taken into consideration.
Leak rate sensitivity to:
Tube size Crack length Test temperature Crevice condition TSP gap Crack exposure outside TSP Tube pressurization a' emssures tr - to freespan burst Freespan to IRB leak :.e comparisan Tables B-1 and B-2 define the elements of a total parametric study which would be required if all these variables were tested for each possible combination.
l i
i l
E:nle:Bytwf Stgen/IPeresT:5
~
i
-j
- W, j
Comed's Evaluation'of Test Variables Comed performed an evaluation of all the possible test variables and determined j
which are important to the bounding test program. The following is our evaluation of each variable.
Sensitivity to tube size:
Byron and Braidwood Unit 1 tubing is 3/4" diameter. Due to the immediate availability of 7/8" tubing corrosion test specimens, the test program previously submitted to the NRC in reference 3 will be conducted using 7/8". tubing. These test results will be used to define additional testing requirements 'and test modifications. Subsequent testing will include 3/4" diameter tubing. Preparation of 3/4" specimens with varying ODSCC throughwall crack lengths is in progress.
'A review of the geometry of the hypothetical crack is contained within WCAP-14273. It is important to note that for an IRB, the equations provided to define the crack opening area depend upon the length of the crack, and the gap between the outside diameter of the tube and the inside diameter of the tube support hole.
In addition, the WCAP describes the basis upon which the maximum clearances between hole and tube were assembled. This leads to the fact that the calculation of the crack opening area is based on geometric ec; ations. Review of these equations indicates that there is no dependency of.he crack opening on tube diameter.
To demonstrate that the leak rate is a geometrical function of crack opening and is not dependent on tubing diameter, Comed will perform testing on both 7/8" and-3/4" diameter tubing. Based upon the test results it may not be necessary to do duplicate testing with both 7/8" and 3/4" tubing.
Sensitivity to crack length:
Specimens may be prepared for 3/4" diameter tubing with a goal of achieving throughwall crack lengths m the following ranges; 0.25" - 0.45", 0.45" - 0.60" and 0.60" - 0.75". Use of these ranges will develop a conservative upper bound leakrate based on the following facts:
- No indications below 10 volta have had a throughwall crack >0.37" per the EPRI pulled tube database.
Braidwood 1 and Byron 1 largest indications of 10.3 and 10.9 volts had throughwall crack lengths of 0.20" and 0.27", respectively, with burst pressures
> 4000 psi at operating temperature. This is well within the experience of the EPRI pulled tube database.
M:nle:Sybwd:Stgen/IPCTesTr4
~
l J
Laboratory induced ODSCC specimens with throughwall indications in the upper end of this range (> 0.6") are extremely costly and difficult to produce. The longest laboratory induced axial ODSCC indications are likely to be about 0.6". To achieve i
a 0.75" length, either laboratory cracking from ID and OD exposure to doped steam or, fatiguing the specimen is likely to be required. This may deform the tube and change its mechanical properties and/or result in multiple long throughwall indications. Comed will use the longest ODSCC specimens which are produced in the laboratory and will consider lengthening these cracks by fatigue.
The critical throughwall crack length for a burst at MSLB pressure of 2560 psid for a 3/4" diameter tube at lower tolerance level (95% / 95%) material properties is 0.75". Such an indication would have a bobbin voltage of > 40 volts. Braidwood 1 and Byron 1 largest indications of 10.3 and 10.9 volt had throughwall crack lengths of 0.20" and 0.27", respectively, with burst pressures > 4000 psi at temperature. The EPRI pulled tube data database demonstrates that no i
indication of < 10 volts has had a thorough crack > 0.37". Therefore, it is unrealistic to assume that a 0.75" axial through crack would develop in a plant i
where a 3.0 volt IPC is applied.
j The crack length test variable, with > 0.6" long ODSCC specimens, is expected to bound by a factor of 1.5 the largest indications seen in the EPRI pulled tubo data base for indications below 10 volts.
Sensitivity to test temperature:
l Westinghouse is in the process of modifying a test facility to accommodate high leak rates at operating temperature test conditions. The test facility is expected to have the capability to measure the leak rates up to at least 5 gpm at 615 F.
Testing at elevated temperatures is extremely time consuming, therefore, to expedite testing the test plan will consider development of a temperature correlation factor. A correlation of operating and room temperature conditions will be developed from test data obtained in this program. The existing leak rate adjustment procedures indicate that operating temperature leak rates are about 1/2 room temperature leak rates. This correlation will be confirmed for indications confined within the TSP. Some of the longer crack length tests may be performed l
at room temperature to expedite testing.
E:nla:Bytwd: St gen /IPCTesT:7 l
-~ - -. -.
Sensitivity to crevice condition:
S' me tests may be conducted with both " packed" TSP crevice conditions and o
" clean" TSP crevices. Crevice packing will be performed by packing the crevice I
with magnetite and a binder followed by baking of the specimen. This is typical of the deposits seen on pulled tubes. This test is representative of actual steam generator crevice conditions believed to be present at Byron and Braidwood. If the leak rates are considerably different, test results will be evaluated to determine appropriate application of the data in leakrate calculations.
Sensitivity to TSP gap:
Tests may be conducted within the range of possible maximum TSP clearances for drilled hole TSP steam generators (0.020" - 0.025"). The 0.025" valve was obtained by applying a 95% confidence level to the upper diameter limit of a drilled support plate hole and the 95% confidence level of the minimum diameter of the tubing.
This test at 0.025" clearance will provide a bounding leak rate value for maximum gap.
Sensitivity to crack exposure outside TSP:
The maximum TSP deflection with tube support plate stabilization is shown in WCAP-14273 to be 0.1". This value is extremely conservative with a factor of 2 applied to the Tranflo loads. A review ofinspection data for Byron 1 and Braidwood 1 has not found any axial ODSCC indications extending beyond the edges of the TSP. Therefore, based upon a comparison of the real location of indications within the tube support plates and a review of the conservatisms within the calculation of tube support plate maximum deflection, Comed's assessment is that testing with TSP offset of 0.1" is unnecessary.
Leakrate vs. Tube Pressurization:
Indications contained within the TSP will be pressurized and leak rates measured; at MSLB pressure above calculated burst pressure e
1000 psi increments up to approximately 10,000 psi.
Testing may be repeated with crack offset 0.1" outside TSP.
The objective of these pressurization steps is to determine the maximum leak rate vs crack opening configuration which is postulated to occur.
I l
Comparison of freespan to IRB leak rate:
Certain specimens will be pressurized to valves below MSLB conditions while in freespan so that the freespan leak rate can be measured and extrapolated to MSLB conditions using the EPRI leak rate procedure. This value will then be compared to the IRB leak rate obtained from these specimens while they are confined within the TSP.
These data will be used to show a correlation between the freespan and IRB leak l
rate. The result may be that the freespan leak rate is higher than the IRB leak j
rate for certain size indications.
o nt. w sw.n m er..r a
R.
~
F X X X X X X X X X X X X 5
1 6
g l
g uA M P
X X X X X X X X X X X X EP O M O Er R
=
is P
p X
0 X X X X X X X X X X X X J
A 0
P T
0 T
L A
E 0
M D
1 ist
=
i P
is C
A p
X X X X X X X X X X X X RI T
0 T
L 5
E E
6 M
D 2
A 5
R 7
A P
X X X X X X X X X X X 6
)BR I
Y
(
T l
D V
6 SR M
S U
M 5
X X X X X X X X X X X B
E 4
S M
N O
l E
R 5
F E
4 L
D K
E C
X X X X X X X X X X X T
A 5
C f
2 R
C I
T SER D
E N
K 0
C X
X X
X X
X X X X X X X X X X
A 11A P
C N
S I
E U
D l
B NI N
D 1
E A
B D
K X X X X X X X X X X X X X X X C
E M A L
J P
B l
1 A
T Y
E T
Y Y
is V
I R
I T
D is p m
V U I
T U
p T
A T
0 0
N I
R 0
S S
5 E
C 6
0 S
N E I
T T
2 P P P M" E
P 0
S M 1
P A A A 5
R E
E T
T S
S
= A G G G M4
=
- 6 5
E E
E E
F F
P P G E
7 G T S
F F
M O O A A L
I M M C2 4
6 L
O 5 F
L L
l L
A M T
I M
I l
K5 5
A F
F T
R O
L TL M
A
". 2".
E 5
3 5
O O 1
E A
1 R
P O 0 0 D D 0 1
2 2
C V R 6
-:3_
TABLE B-2 INDICATION RESTRICTED FROM BURST ORB) PARAMETRIC TEST'MATR!X -
e-T 4
PARAMETRIC STUDY l
l l
l l
7/a'otA. msES CPACYLENGTH SENSmVUY DELTA P = ' DELTA P = '-
TEMPEEATURE PACKED UNPACKED
.25.45'
.45
.6'
.6 -.75 ' 2650 psi 10I)00 psi ROOM 615 F I
0' OFFSET X
X 0.1" OFFSET X
X X
X X
X X-X X-0.2' OFFSET X
X X
X X
X X.
X-X DELTA P = 2650 psi X
X X
X X
X-X.
DELTA P = 10000 psi X
X
.X X
X X-
-X-0 MIL GAP X
X X
X X
X X-X X
15 MIL GAP X
X X-X X.
X X
X
- X.
23 MIL GAP X
X X
X X
X X
X
'X 25 Mil GAP X
X X
X X
X-
-X X
X-CRACKIENGTH $ENSMVTTY l
.25.45" X
X X
.X-X X
l
.45
.6' X'
X X
X-X X
j
.6.75 '
X X
X X
X-X l
VOLTAGE SENSITIVITY-X X-X X
X X
~X X-X l
ROOM TEMPERATURE X
X X
X X-X X
615F X
X X-X X
X X-l'
~
a
Attachment C BOUNDING TEST PROGRAM Test Plan Obiective Develop a bounding MSLB leak rate for large (>0.6") indications which have been restricted from burst (IRB) by the presence of the tube support plate (TSP). This bounding leak rate value will be used in a leak rate calculation which takes into account both freespan and IRB leakage.
This test program will be developed to address sensitivity to:
Differences in leak rates between 3/4" and 7/8" diameter tubing Sensitivity of crack length inside the support Differences in leak rate data between room temperature and operating condition tests (validate room temperature to operating condition correlation)
Differences between leak rates in packed versus clean crevices Differences in TSP crevice clearances Length of crack extending outside the TSP Differences in pressurization levels Differences between freespan and IRB leak rates Comed does not have an objective to create or test a low voltage (1 volt or less) 3/4" long crack. However, long cracks (>0.6") will be created and tested.
Proposed Bounding Test Plan Based on Comed's evaluation of the critical variables required to meet the programs objective, a total test program was developed and is provided in Table C-
- 1. The notes for Table C-1 outline the testing sequence and prioritization critena which determine the order in which specimens are tested.
Using Table C-1 the bounding test program is defined as tests 1-1, 1-2, 1-6, 1-7, 2-1,2-4,2-7,2-10 and 4-1. Need for further testing will be based on Comed's assessments of the results of these tests. Each test will involve only one specimen.
In addition, Comed may change the scope, number of specimens and number of tests, depending upon test results obtained.
It is Comed's position that the only testing required to bound the MSLB leak rate l
are tests 2-1,2-4,2-7 and 1-7. Other tests are only planned to improve the i
industries understanding ofleakage from IRB's.
i it :nla:Bytad:Stgen/IPCresT9
-~
Uhd TABLE C-1 Test Matrix for Indications Restrained from Burst (IRBs)
Crevice Crack Length Gap Free Crack to TSP Bladder Test Test Pri.
Tube Temp Span Offset Pres.
Test No.
Seq.
ority Dia.
Leak AP'"
Time.
Open Pack
.25-
.45.6
.6-
.02
.025 Test 0.0" 9.1" 6.2" Offset Days
.45
.75 (inch) 1-1 la 1
7/8 615*
x x
x x
x x
0.0 12 1-2 la 1
72 615*
x x
x x
x x
0.1 12 1-3 la 2
7/8 615*
x x
x x
x x
0.1 12 1-4 lb 4
7/8 615*
x x
x x
x 7
1-5 lb 5
7/8 615*
x x
x x
x x
12 1-6 la 1
3/4 615*
x x
x x
x x
0.0 12 1-7 la 1
3/4 615*
x x
x x
x x
0.1 12 1-8 la 2
3/4 615' x
x x
x x
x 0.1 12 1-9 lb 4
3/4 615*
x x
x x
x 7
1-10 lb 5
3/4 615*
x x
x x
x x
12 1-11 la 3
7/8 615*
x x
x x
x x
0.1 12 1-12 la 3
3/4 615*
x x
x x
x x
0.1 12 2-1 2
1 7/8 615*
x x
x x
x x
0.1 12 2-2 2
2 7/8 615*
x x
x x
x x
0.1 12 2-3 2
3 7/8 615*
x x
x x
x x
0.1 12 2-4 2
1 7/8 615' x
x x
x x
x 0.1 12 l 2-5 2
2
.7/8.
615*
x x
x x
x x
0.1 12
TABLE C-1 Crevice Crack Length Gap Free Crack to TSP Bladder Tcst Test Pri-Tube Terap Span Offset Pres.
Test No.
Seq.
ority Dia.
Leak AP" Time Open Pack
.25-
.45.6
.6-
.02
.025 Test 0.0" 0.1" 0.2" Offset Days
.45
.75 (inch) 2-6 2
3 76 615*
x x
x x
x x
0.1 12 2-7 2
1 3/4 615' x
x x
x x
x 0.1 12 2-8 2
2 3/4 615*
x x
x x
x x
0.1 12 2-9 2
3 3/4 615*
x x
x x
x x
0.1 12 2-10 2
1 3/4 615*
x x
x x
x x
0.1 12 2-11 2
2 3/4 615' x
x x
x x
x 0.1 12 2-12 2
3 3/4 615*
x x
x x
x x
0.1 12 3-1 3
3 7/8 T,RT x
x x
x x
0.1 10 3-2 3
3 7B T,RT x
x x
x x
0.1 10 3-3 3
4 7B T,RT x
x x
x x
0.1 10 l
l 4-1 4
1 7B RT x
x x
x x
0.1 10 4-2 4
2 3/4 RT x
x x
x x
0.1 10 l
Notes:
i
- 1. Test sequence includes pressurizing with a bladder typically to above the free span burst pressure. Some specimens include incremental increases in bladder pressure beyond that equivalent to a free span bunt. For test sequen es 1 and 2, the bladder pressurization and subsequent leak tests follow leak tests that apply primary side pressure and flow to obtain the APs. Thus, for these tests, the bladder pressurization is performed to open the crack beyond that obtained within the pressure capability of the facility.
NOTES FOR TABLE C-1
+
' Test Plan for Indications'with Restrained Burst (IRBs)
Test Sequences Test Sequence 1: 615 F, Crack Lengths > 0.6" Seouence la.
A. Free span leak test at 1500,1700 and 2000 psi AP
'B Leak test with crack centered (0.0 offset)in TSP at 2000,2335 and 2560 psi AP -
C. Measure crack opening diameter and area D. Leak te'st with crack 0.1" offset outside TSP at 2560,2700,2900 psi AP (up to facility limit)
E. Measure crack opening diameter, area and evaluate crack extension F. With crack centered or 0.1" (test matrix specification) offset outside TSP,-
pressurize to above calculated burst pressure with bladder
- Note: Test sequence 1 is applied to the longest cracks which are expected to open up and contact the ID of the TSP hole within the pressure / flow limits of the facility. The bladder pressurization step is applied to assure crack flank to TSP contact for comparisons with the leak rates from step D.
G. Leak test with crack centered or 0.1" offset outside TSP at 2335 and 2560 psi AP H. Measure crack opening diameter, area and evaluate crack extension I. With the same crack position as step F, pressurize with a bladder (with foil if necessary) to about 1000 psi above the prior pressurization step J. Repeat leak test of step G -
K. Repeat steps I and J with increases in bladder pressure in 1000 psi increments until bladder / foil pressurization increase cannot be achieved (approximately burst pressure inside TSP on order of 8000 psi)
- Note: Ifit is demonstrated on one sample in any test sequence, that the leak rates do not significantly increase by the pressurizations of steps I to K, steps I to K may not be repeated on additional specimens and this test sequence will be terminated at step H for the later specimens L. Measure crack opening diameter, area and evaluate crack extension Seauence Ib Perform test with Sequence la steps A to E (i.e., no bladder pressurization steps)
~
t
-c--._
i NOTES FOR TABLE C-1 Test Sequence 2: 6150F, Crack Lengths < 0.6"(Comparison of room temperature to operating conditions)
A. Free span leak test at 1500,1700 and 2000 psi AP B. Leak test with crack tip centered at 2000,2335,2560 psi AP C. Leak test with crack tip 0.1" offset from the edge of the TSP at 2560,2700 psi AP up to facility limit D. Measure crack opening diameter, area and evaluate crack extension E. With crack tip 0.1" offset, pressurize to above calculated burst with bladder Note: Test sequence 2 is applied for shorter cracks which may not open to a
significantly contact the ID of the TSP hole within the pressure / flow limits of the facility. Thus, steps A and B may result in comparable leak rates. The bladder pressurization step is applied to assure that crack flank to TSP contact is achieved to assess the influence on leak rates.
F. Leak test at 2335,2560 psi AP at both room and operating temperature for at least one specimen.
G. Measure crack opening diameter, area and evaluate crack extension H. With the 0.1" crack position, pressurize with a bladder (and foil if necessary) to about 1000 psi above the prior pressurization step L Repeat leak test of step F J. Repeat steps H and I with increases in bladder pressure of 1000 psi increments until bladder / foil pressurization increase cannot be achieved (approximately burst pressure inside TSP on order of 8000 psi)
Note: Ifit is demonstrated on one sample in any test sequence, that the leak rates do not significantly increase by the pressurizations of steps H to J, steps H to J may not be repeated on additional specimens and this test sequence will be terminated at step G for the later specimens K. Measure crack opening diameter, area and evaluate crack extension Test Sequence 3: 615 F and Room Temperature Leak Rate Comparisons A. Pressurize to above calculated burst pressure with bladder B. Room temperature test at 2000,2335,2560 psi AP C. Operating temperature test at same pressure differentials 3
D. Measure crack opening diameter, area and evaluate crack extension E. Move crack to 0.1" outside TSP, pressurize to same pressure with bladder F. Room temperature test at same pressure differentials G. Operating temperature test at same pressure differentials H. Measure crack opening diameter, area and evaluate crack extension
NOTES FOR TABLE C-1 Test Sequence 4:
Room Temperature, Bladder Pressurization to Establish IRB Leak Rate Magnitude and Effects of High " Burst" Pressures on Leak Rates A. Pressurize to above calculated burst pressure with bladder B. Room temperature test at 2335,2560 psi AP C. Measure crack opening diameter, area and evaluate crack extension D. Move crack to 0.1" outside TSP and pressurize to the same pressure E. Room temperature repeat of step B leak rate tests F. Measure crack opening diameter and area G. With the 0.1" crack position, pressurize with a bladder (and foil if necessary) to about 1000 psi above the prior pressurization step H. Repeat leak test of step D I. Repeat steps G and H with increases in bladder pressure of 1000 psi increments until bladder / foil pressurization increase cannot be achieved (approximately burst pressure inside TSP on order of 8000 psi)
Note: Ifit is demonstrated on one sample in this test sequence or by the equivalent pressurizations of Sequences la or 2 tests, that the leak rates do not significantly increase by the pressurizations of steps G to I, steps G to I may not be repeated.
J. Measure crack opening diameter, area and evaluate crack extension Packed Crevice Simulation
- Pack crevice with magnetite and a binder followed by baking of the specimen to obtain a fairly hard deposit typical of that found on pulled tubes at TSP crevices i
7 NOTES FOR TABLE C-1 Test Plan for Indications with Restrained Burst (IRBs)
Test Priorities Priority 1 Recommended test matrix to establish bounding leak rate for large cracks (test
-sequence 1) and the leak rate dependence for more realistic crack lengths (test' sequence 2)
An initial test (sequence 4) of an indication pressurized by a bladder to obtain an early estimate of the magnitude ofleak rates to be expected Priority 2 Repeat of priority 1 tests. Tests performed upon completion of priority 1 tests-only if bad test data are obtained or the initial test results in unacceptable uncertainties on the leak rates relative to adequately defining a leak rate model for IRBs Priority 3 Tests performed to compare room temperature leak rates with operating temperature leak rates. These tests would be performed only if the leak rates of priority 1 tests exceed the facility capability at operating temperatures and some of the priority 1 tests must be run at room temperature.
Tests with packed crevices. Tests to be performed ifin situ leak rates are to be used for reference SLB leak rates rather than free span plus IRB' leak rates.
These tests would provide physical insight into expectations to be obtained.
with in situ leak rate measurements.
Priority 4 Tests of additional variables not considered to be necessary to adequately defm' e IRB leak rates.
Priority 5
. Test variables that are judged not worthy of consideration for testing os l
i
- e.,
~
Attachment D RESPONSE TO NRC COMMENTS DATED MARCH 24,1995
. Comment 1:
The proposed use of only three test specimens appears insufficient to adequately determine the sensitivity of the results with respect to a number of test parameters.
Response
The original proposal for 3 test specimens was based upon the immediate availability of specimens. Other specimens are being prepared and will be available for test within the scope of the attached proposed test program by May 15,1995. The bounding test program in Attachment C contains 9 test specimens and would be completed about July 15,1995.
Comment 2:
The use of only 7/8-inch diameter tubing appears restrictive in that the Westinghouse Model D4 SGs use 3/4-inch diameter tubing.
Response
As noted in WCAP-14273 Comed's understanding of the geometrical calculation to determine possible crack geometries is independent of tube diameter. It depends upon crack length and gap size between the tube OD and the support plate hole ID.
Specimens being prepared in the corrosion test facility, including 3/4" tubing have been added to the proposed test program. These tests will begin after May 15, 1995.
Comment 3:
Limiting the bladder pressure to 15 percent above the pressure at which the test specimens are expected to burst, appears too restrictive. The staff believes that the test specimens should be pressurized in the lab to an extent that ensures the crack faces from the IRBs are opened to the point which would result in a conservative representation of the leakage from an IRB in a SG.
4 Kihle:9ytzd:Stgen/IPCTest:10
Ilesponse:
Comed agrees with this comment. The specimens need to be pressurized to ensure that the crack faces from the IRBs are opened to the point that a conservative representation of the leakage is obtained. Comed has developed a test program in which tubes with cracks are incrementally pressurized with a bladder followed by leak rate measurements. These incremental increases in pressure will continue up to the tube freespan burst pressure which is in the range of 7,000-11,000 pounds. Results from these tests will be used to develop a leakrate versus pressure curve for several specimens, thereby, eliminating confusion as to the maximum leakrate.
Comment 4:
The staff believes that conducting the proposed leak tests at temperatures representative of plant conditions is preferable to conducting these tests at ambient temperatures.
Response
The industry test programs conducted with EPRI and European data indicate that leakage at temperature can be predicted from room temperature tests by the application of an appropriate correlation. However, Comed proposes that additional testing be done at both room temperature and 615 F to validate this correlation for different configurations.
Comment 5:
It is not clear from the formulation of the proposed leakage model whether the leakage flow rates are sensitive to the gap between the outer diameter of the tube specimen and the simulated TSP. If there is a highly sensitive relationship, is a 23 mil gap sufficiently conservative to encompass the as-built configuration of the Byron 1 and Braidwood 1 Model D4 SGs?
Ilesponse:
WCAP-14273 contains a discussion of the methodology to arrive at the 23 mil gap being sufficiently conservative. Comed believes a 23 mils gap is sufficiently conservative. The bounding test program is defined in Attachment C. Attachment B describes the methodology used to develop a bounding TSP gap.
K, nla : Bytwd < $tgen / I PCTest :11
Comment 6:
The test program proposes to use specimens with a maximum induced flaw length of 0.60 inches. Testing flaw lengths up to 0.75 inches should provide an upper bound of the potential leakage flow rates from an IRB.
Response
Laboratory induced ODSCC specimens with throughwall indications > 0.6 are extremely costly and difficult to produce. The longest laboratory induced axial ODSCC indications are likely to be about 0.6". To achieve a 0.75" length, fatiguing the specimen is likely to be required. This may deform the tube and change its mechanical properties. Comed will use the longest ODSCC specimens which are produced in the laboratory and will consider lengthening these cracks by fatigue.
This test with a > 0.6" ODSCC specimen expected to bound by a factor of1.5 the largest indications seen in the EPRI pulled tube data base for indications below 10 volts.
Comment 7:
Assuming that 0.1 inches is the maximum vertical displacement of a TSP under postulated accident conditions, the staff believes that the test collar simulating the restraint of a TSP should be moved to expose an additional 0.1 inches of the induced tube crack face rather than just exposing 0.1 inches of the crack. This consideration would come into play in the event that the tube crack induced by the bladder, extended beyond the 0.75 inch thick collar.
Response
It is important to remember that WCAP-14273 has significant conservatisms built into the tube support plate motion calculation which results in a maximum vertical displacement under a combination of all anticipated conditions of 0.1" In fact, it is believed that displacement in the event of a main steam line break would be substantially less than 0.1", therefore the 0.1" offset will clearly bound this condition.
The overpressure test methodology proposed ofincreasing tube pressures up to specimen freespan burst will evaluate the hypothesis that a tube crack extension is induced by the bladder. Based upon the outcome of this evaluation Comed will consider what additional testing may be required.
Comed has no current plans to test specimens with offset >0.1"
,,,a. _. mm.w
.