ML20153B182
| ML20153B182 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 09/04/1998 |
| From: | MPR ASSOCIATES, INC. |
| To: | |
| Shared Package | |
| ML20153B176 | List: |
| References | |
| 083-248-CBS-01, 083-248-CBS-01-R02, 83-248-CBS-1, 83-248-CBS-1-R2, NUDOCS 9809230004 | |
| Download: ML20153B182 (8) | |
Text
_ _. _ _ _ _ _ _ _ _ _ _ _ _. _.. _ _ _.. _. _. _ _ _ _ _ _. _.. _ _ - - _ -
i a
i MPR Associates,Inc.
. WMPR.
. 320 King Street Alexandria,VA 22314 CALCULATION TITLE PAGE I
6-8 ment mi Pa08 I of PliPg GPU Nuclear
+ Appendices %
3 Project Task No.
Shroud VerticalWeld Evaluation 083-9601-248 0 Title Calculation No.
Shroud Finite Element Evaluation 083-248 CBS-01 Preparer /Date Checker /Date Reviewer //pprover Data Rev. No.
,C.
j B. Lane W. McCurdy h [gsq,
- t bj% ___ Q 0
5h'l" 4
m._
fM sla./w i
i j
1 g
hl.'OL t
syx 7/t%k*g
"/h 'lL l1%
1 dimq?xO-i 2
914hs yy,~
~
wl4/%
T p
i 9909230004 990916 I
PTAt ADOCK 05000219 i
P PDR I
l j
QUALFTY ASSURANCE DOCUMENT TNs doctaient has been prepared, checked, and reviewed in accordance w.th ths1lJuality Assurance requiremerts of 10Cf4SO Appendx B, as specified in the MPR Ouelty Assurance Manual.
cs d' 961# Wtf383818661 'P d35 EBBPHIEBB691EDET6:01
'Un XTts SMUIDOS90 &W:tEPH w
-e y
y-r y-~- - -
r a,-w-T
MPR Associates,Inc.
MMPR
= =
Alexandria,VA 22314 RECORD OF REVISIONS Caledation No.
Prepared By Ch od By 083 248 CBS-OL h (w Page 2 4
~
Revision Des gon 1
0
[nitiallssue 1
Revised seismic loads based on update.d transient dynanfic ana!ytes.
Removed results for 10 wedges because on.*f8 wedges will be installed.
2 Revised Summary of Results $cction (pg 3) to correct a typographical error In the leakap area.
\\
l 4
I l
1 l
J l
l l
1 Es'd 96Tu W022:0T 8661 'P d3S EBEbearRFUSlTE46T6:01
'Un XIN S310100550 &54:m b
MPR Assoclites,Inc.
MMPR 320 King Street Alenndria, VA 22314 Calculation No.
Prepared By Ch ed By 083-248-CBS-01 Page 3 SM
/
1.0 PURPOSE The purpose of this calculation is to determine the flaw tolerance of the vertical welds in the section between circumferential weids H5 and H6A in the Oyster Creek core shroud, i
The verti, cal weld evaluation was originally performed in Reference 1. This calculation considers the effect ofinstalling wedges between the core plate and the shroud wall. A finite eiement model of the shroud section is developed to evaluate the effects.
j 1
The finite element modelis also used to determine the leakage path flow area through the cracked vertical weld during normal operating conditions.
f l
\\
2.0 SUhmIARY OF RESULTS The maximum stresses in the shrcud section between circumferential welds H5 and H6A are summarized for the limiting load cases in Table 2-1. Stress contours for each load case are presented later in this calculation. As shown, these stresses meet the requirements of Subsection NB of the ASME Boiler and Pressure Vessel Code,1989 Etlition. The evaluations are performed with eight core plate wedges installed. All circumferential welds and the vertical welds in the H5/H6A shroud section are assumed to be completely failed.
i The evaluations show that the load through the vertical welds can be reacted by taking credit for compression across the failed circumferential welds due to tie rod preload.
For the MSLB case, if only welds H5 and H6A are failed with all other circumferential welds intact, compression could no longer be maintained across both welds H5 and H6A.
Consequently, some amount of the vertical weld is required to react the hoop load from the differential pressure. Results of the evaluation performed in Appendix A show that if there is ten inches of intact vertical weld, the stresses in the H5 and H6A meet the requirements of the ASME Code.
The maximum leakage path flow area through a fully-cracked vertical weld in the HS/H6A shroud segment during normal operating conditions is 0.495 inz This flow area will be used elsewhere to evaluate the effect of reactor coolant flow that bypasses the core through the cracked vertical weld.
g se u
. - _. _.. -. -. -. -. _ - -. -. ~.
4 a
1
- 5. Provide your basis for stating (MPR.1957, page 4 8) that the 10-wedge configuration bounds the 8-i wedge configuration.
t
^
f
Response
j Several load cases were evaluated using different boundary conditions and wedge arrangements, as l
summarized below-
}
Load Case
. Description
.A 10 Wedge support configuration with conservative boundary conditions j
4 i
2 j
B 10 Wedge support configuration with realistic boundary conditions C
8 Wedge support configuration with tealistic boundary conditions i.
t i
i i
A summary of the calculated reaction loads for each load case is provided below:
{
. Maximum Reaction Loads at Wedges Loading Condition Case A Case B Case C Normal Pressure (AP) 28 kips 0 kips 0 kips i
j OBE + Normal Pressure (AP) 43 kips 36 kips 40 kips
)'
SSE + MSLB 107 kips 67 kips 79 kips j
SSE + RLB 87 kips 64 kips 79 kips f
f i
' Load Case C p5esents the calculated loads that are applicable for the proposed modification (with the 3
installation of 8 wedges around the core plate). For the purpose of adding conservatism to the analysis of E
the core plate structure, GPU Nuclear used the reaction loads from Load Case A, which are higher than l
Load Case C.
j i
Briefly, the higher reactions in Load Case A are the result of conservative boundary conditions on the core l
j
_ plate model, combined with the application of vertical pressure loads. For Load Case A, tb wedges were L
modeled as being rigidly cmstrained in all directions (which ignores the shroud radial stiffness) and zero j
gaps were input betwem the core plate, wedges, and shroud.1 These constraints and modeling conditions, j
. combined with the vertical differential pressure loads, produced secondary type reactions in the core plate 4
that are conservative, but not realistic. For example, as normal vertical pressure loads were applied to the
]
core plate model (in the upward direction), the core plate assembly tried to elastically flex into a convex shape (i.e., the center of the core plate tried to move upward, in the out-o/-plane direction). The core plate l
assembly is about 22 inches deep. The convex shape of the core plate caused the outer t>p cdge of the i
core plate assembly to try to move radially outward (while the bottom edge tried to move radially inward).
j t
i The models for Load Cases B and C included the shroud radial stiffness and initial installation gaps between the wedges and the shroud and core plate.
)
I
~...
Since the wedges were modeled as being completely rigid (with no Gexibility or installation gaps), they restrained the core plate, which generated secondary type reaction loads at the wedge locations. An example of this effect is shown in the table above (for Lead Case A); there is a 28 kip reaction load in the wedges as a resu!: of the normal pressure differential across the core plate. Nete that this reaction load is not realistic and does not appear in Load Cases B or C, since these models included the shroud radial stiffness and initial installation gaps.
With regards to Load Cases B and C, the primary difference is that Load Case B was for a 10-wedge support configuration, while Load Case C for was an 8-wedge configuration. Both of these load cases 1
used the same boundary constraints (which included the modeling of the shroud radial stiffness and initial installation gaps between the wedges and the shroud and core plate). As expected, the reaction loads into the core plate for the eight-wedge con 6guration (Case C) are higher than the ten-wedge conGguration (Case B).110 wever, the bounding (highest) reaction loads are for the ten-wedged con 6guration (Case A),
which used conservative boundary conditions.
In summary, GPU % clear used the reactior, loads from Load Case A since they are conservative and bound Load Case C. This approach is conservative and results in the calculation of higher stresses in the core plate and wedges.
t i
i
i l
i 6.
Provide your basis for using a 10% buoyancy effect (MPR-1957, page 4-7).
)
- Response:
- The effect of buoyancy on the core support plate weight is captured in the analysis model by reducing vertical gravitational acceleration by approximately 10%. Since an equal volume of water is displaced by 1
the core support plate, the buoyant (upward) force on the core support plate is in proportion to the ratio of
- density of water to that of stainless steel. The actual value for this proportion that was used in the analysis model was 9.8%.
P e
i 1
i i
I l
i l
.k 9
~
I
o 7.
Provide the planned level (s) of visual inspection described in MPR-1957; Section 8 (VT-1, VT-3, etc.).
Response
1.
Pre-Installation Inspections Visual exams are to be completed prior to installation of the wedges to confirm that:
Each installation site is free of obstructions and debris, and
+
The core plate (top plate) and shield angles have no signs of degradation that could affect the structural integrity or performance of the wedge installation. Inspections will be completed at each installation site, based on the use of VT-1 procedures.
The inspections of the core plate will be limited to the immediate, accessible area around each installation site (i.e., approximately a 6 to 12 inch circular area on the top plate of the core plate). The core plate will not be examined in its entirety.
Visual inspections (VT-1) of the shield angles will be performed on accessible areas where a wedge is to be installed, including the shield angle itself and the attachment w:Id cf the shield angle to the shroud.
Shroud inspections will be done as part ofother in-Vessel Visual Inspections (IVVis) to confirm the integrity of the shroud vertical welds. No additional structurally related shroud inspections will be performed as part of the wedge installation.
2.
Post-Installation Inspection _s Prior to vessel re-assembly, visual inspections will be performed to verify the installation of each wedge.
The inspections will be performed (VT-1) on accessible areas of the wedges in accordance with approved GPU Nuclear procedures. The inspections will confirm that:
Each wedge is properly located, oriented, and positioned, The retainer springs are properly engaged on thejacking bolt, The fit-up with the shield support angles has been properly established, and All miscellaneous installation tooling and support equipmentihardware have been removed from the vessel (a foreign material exclusion program will be used to monitor materials in the vessel).
E
_- - -.. _..-. ~...-....
..-. - ~ ~.
l-t i
?
l 8.
How does the inspections of MPR-1957, paragraph 8.2.2, compare to BWRVIP-25 (enhanced VT-l?).
Response
I L
Section 8.2.2 of MPR-1957 addresses inspection of the wedges during subsequent refueling outages.
' There are no comreitments in BWRVIP-25 as regards to inspection 'of wedges during subsequent refueling
(.
' outages after wedge installation j
GPU Nuclear is taking steps beyond BWRVIP-25 to insure on-going adequacy of the wedges.
t i
'l i.
l-l f
t.
P 5
f l
r i
I l
i P
j:
s l
i t
i I
i l
t i
o l
6
{
e i..
f
-