ML20210P141
| ML20210P141 | |
| Person / Time | |
|---|---|
| Site: | Yankee Rowe |
| Issue date: | 02/02/1987 |
| From: | Mckenna E Office of Nuclear Reactor Regulation |
| To: | Papanic G YANKEE ATOMIC ELECTRIC CO. |
| References | |
| NUDOCS 8702130309 | |
| Download: ML20210P141 (12) | |
Text
m
~' -
FEB 0 21987 Docket No.50-029 Mr. George Papanic, Jr.
Senior Project Engineer-Licensing Yankee Atomic Electric Company-1671 Worcester Road Framingham, Massachusetts 01701
Dear Mr. Papanic:
SUBJECT:
YANKEE NUCLEAR POWER STATION - ANALYSIS OF THE MAIN STEAM AND FEEDWATER PIPING AND SUPPORT STRUCTURE During the review meeting conducted from January 6-9, 1987, Yankee Atomic provided a discussion of the methods and criteria (Enclosure 1) to be used for analysis of the main steam and feedwater piping and support structure. to this letter provides staff comments regarding your proposal that should be considered as your analysis plans for this system are finalized.
Sincerely,
/s)
Eileen McKenna, Project Manager PWR Project Directorate #1 Division of PWR Licensing, NRR
Enclosures:
As stated cc: See next page-kR 000l05000029 870202 P
PDR Office:
P P/)l$k PD/PADf1 Surname: EMc enna/jb Gleart f7 Date:
(R/ 2/87
/8 i
\\
Mr. George Papanic, Jr.
Yankee Atomic Electric Company Yankee Nuclear Power Station I
cc:
Mr. James E. Tribble, President p
i Yankee Atomic Electric Company 1671 Worcester Road Framingham, Massachusetts 01701 Thomas Dignan, Esquire Ropes and Gray 225 Franklin Street Boston, Massachusetts 02110 Mr. N. N. St. Laurent Plant Superintendent Yankee Atomic Electric Company Star Route Rowe, Massachusetts 01367 r
Chairman Board of Selectmen l
Town of Rowe Rowe, Massachusetts 01367 Resident Inspector l
Yankee Nuclear Power Station i
c/o U.S. NRC Post Office Box 28 Monroe Bridge, Massachusetts 01350 l
Regional Administrator, Region I U.S. Nuclear Regulatory Commission j
631 Park Avenue King of Prussia, Pennsylvania 19406 Robert M. Hallisey, Director Radiation Control Program Massachusetts Department of Public Health 150 Tremont Street, 7th Floor Boston, Massachusetts 02111
ANALYSIS OF THE MAIN STEAM AND FEEDWATER PIPING APO SUPPORT STRUCTLRE OUTSIDE THE VAPCR CONTAINER
1.0 INTRODUCTION
7 The main steam /feedwater piping outside the VC.is supported on a' series of _ light structural frames. The mass and stiffness of the support structure and piping runs are of similar magnitude; therefore, the rigid-support assumption generally used in the majority of the piping analyses is not valid here. Additionally, a num er of pipes may be supported by the same frame, promoting interaction between the pipes. For the above reasons, the main steam /feedwater piping' and the support structure shall be analyzed together in one problem or they can be analyzed separately 1
with interaction explicitly considered.
This document outlines the methods and criteria which shall be used in i
the analysis and qualification of the main steam /feedwater piping outside-the VC. Both seismic and non-seismic portions of the piping shall be analyzed. The extent of the amount of non-seismic piping shall be determined based on its effect on the response of the seismic piping and l
the support structure. Certain modifications may be designed to limit the areas of' interaction between the seismic and non-seismic portions.
L Specific analysis details and evaluation criteria are discussed in the following sections.
- 2. 0 MAIN STEAM /FEEDWATER PIPING ANALYSIS 2.1 Geometry and Computer Modeling The cos ined main steam /feedwater piping and support structure system is large and complex, making it prohibitively costly and time-consuming to analyze a detailed model of the entire system.
To minimize the size of the actual seismic analysis model, substructuring methods shall be used to model the support structure frames to limit the degrees of freedom to a manageable nuser.
Each frame shall first be modeled in detail, considering all structural meters, connections, and other attached piping and equipment.
Consideration shall be given to the supporting scheme on each frame to determine whether the main steam and feedwater pipes will interact on that particular frame.
If the supporting scheme allows interaction, the attachment points of the piping shall be retained, and the detailed frame model shall then be condensed to l
1 YAN: 0599
[
u
retain only the important ment)ers and equivalent masses and stiffnesses of the remaining members. This simplified model shall then be used for the system analysis.
The above procedure shall be used for every support frame.' The analysis model shall consist of the eight main steam and feedwater lines with equivalent support frames connecting them as appropriate. After the system analysis is completed, the reactions at the support points shall be used to analyze the detailed support frames for qualification.
If modifications to the supporting scheme or the support frames are required, the frame models shall be reevaluated to assess the interaction and to determine the necessary degrees of freedom for the revised configuration.
The revised support frame models shall then be regenerated and recondensed.
This iterative process shall be repeated until an acceptable supporting scheme is determined.
The support frames and the main steam and feedwater lines are connected to a number of structures. For the PAB and turbine building, the walls and structural penetrations are stiff compared the support frames and piping, and anchors shall be taken at those attachment points. The piping penetrations for the VC, however, are relatively flexible by comparison. For those penetrations, equivalent stiffnesses shall be calculated.
In all cases, the size and thickness of the reinforcing pad on the VC shell shall be considered when calculating the equivalent stiffnesses. The relative stiffness of the piping section exiting the VC and the reinforcing pad shall be considered in determining the local stiffness of the VC shell.
The connection of the main steam and feedwater piping to the VC resembles a nozzle penetrating a vessel. The most commonly used method to determine the stiffness of the nozzle attachment is WRC Bulletin 297. This method is based on empirical data for cylindrical shells.
In this analysis, the VC is a sphere rather than a cylinder. To determine a compatible stiffness for the penetrations, WRC 297 shall first be used, considering the circumferential stiffness of a nozzle on a large cylinder (comparable to the VC curvature). To verify and bound this stiffness, the longitudinal stiffness data of WC 297 shall be used to determine an upper bound stiffness of the equivalent cylinder, and theoretical means shall be used to determine a lower bound stiffness in a flat plate. These three values shall be used to determine a reasonable value to be used at the VC boundary.
2 YAN: 0599
The remaining modeling aspects of the main steam /feedwater piping and support structure model shall be as specified in the Seismic Retrofit Criteria document, DC-1, Revision 3.
2.2 Loading Conditions 7
The main steam /feedwater piping and support structure shall be analyzed for deadweight, thermal, and site specific spectral loads, including the required anchor motions associated with each of the three load cases. The intent of this evaluation is to ensure the adequacy of the system to function during and after a seismic event to attain a safe shutdown. For this evaluation, the thermal conditions for normal operation shall be used in cod ination with the applicable deadweight and seismic loads for evaluation of the piping and support structures. Thermal anchor motions of the anchor points shall be included as appropriate.
Ground and amplified response spectra shall be generated for each of the anchor points in the system model for the site specific i
spectra. The amplified response spectra shall be generated for appropriate elevations on the VC, PAB, and turbine building.
PVRC damping (ASME Code Case N-411) shall be used in the development of the spectra.
\\;
Seismic response spectrum analyses shall be performed for the system. The seismic load shall be input as independent support motions at the appropriate anchor points (multi-level response j
spectrum method).
As suggested in NlREG-1061, the response between the support levels shall be cosined by. absolute summation, and j
combination within a level shall be performed by square-root-sum-of-the-squares.
The Reg. Guide 1.92 107. grouping method shall be used to cosine modal effects.
i Seismic anchor motions shall be considered for each building as appropriate and as specified in the Seismic Retrofit Criteria j
document DC-1, Revision 3.
In addition to the building motions, i
relative seismic anchor motions of adjacent support frames shall be considered due to seismic wave propagation.
To determine the proper differential motion to be applied to adjacent supports, the seismic wave shall be propagated along the surface.
Using the predicted wave length, frequency, magnitude of displacement, and the spacing of the support frames, a maximum relative displacement between any two anchor points can be calculated, as well as the distribution of the j
anchor motions between a series of supports. A series of maximum
(
3 t
YAN: 0599
\\
i
,mr-
,?
i
^
displacements between supports, to maximize the pipe stresses and b'
-support loads, shall be used in the analysis unless a reduced set of displacements can be justified. Response from the seismic wave load case shall be codined with the standard seismic anchor motion response using the SRSS method.
~
2.3 Acceptance Criteria Acceptance criteria for the main steam /feedwater piping shall be as specified in the Seismic ?.etrofit Criteria document DC-!,
Revision 3.
The basic ASI".E/ ANSI B31.1 equations shall first be used in the qualification.
In the initial evaluation, the piping shall be compared to the SEP criteria as follows:
DW < Sh Eqn 11 DW + Site Specific Spectra < 2.4Sh Eqn 12 1hermal + TAM + SAM < Sa Eqn 13 DW + Thermal + TAM + SAM < Sa + Sh Eqn 14 If Equations 13 and 14 cannot be qualified with the addition of the seismic anchor motion stresses, the SAM stresses can be included in Equation 12.
If the Equation 12 allowables are exceeded, the strain-based criteria and its associated requirements, as specified in the Seismic Retrofit Criteria document DC-1, Revision 3, may be used. These strain criteria and requirements apply to the seismic portions of the main steam and feedwater piping and to the l
non-seismic portions which have an effect on the response of the seismic piping or the frames which support the seismic piping.
- 3. 0 MAIN STEAM /FEEDWATER PIPING SUPPGtT FRAE ANALYSIS l
3.1 Geometry and Computer Modeling The piping shall not be included in the detailed model for the evaluation of the MS/FW support frame.
The frame model shall include all the meders effective in transmitting the seismic load, with other meders considered as lumped masses. The boundary of this frame is connected to the VC, PAB and Turbine Building. All these structures shall be modeled with equivalent stiffness.
4 YAN: 0599
~-
_y
3.2 Loading and Methodology The MS/FW support frames shall be analyzed for dead load, seismic load and pipe reactions.
The wind load does not govern for these L
frames.
The seismic inertia force due to the frame weight shall be considered using the equivalent static method.
It is considered as an accelera-tion equal to the peak of the ground spectra multiplied by the mass of the frame. Since.this frame is made of bolted steel members, and the maximum stress is expected to be close to the yield stress, the 7% damping spectra shall be used.
The load conbinations shall be as follows:
Case 1:
DL + RT + (R X + F X) + (R y + Fy) + (RZ + FZ)
Case 2:
DL + RT + (R X + F X) + (R Y + F y) - (RZ + FZ )
Case 3:
DL + RT + (RX + FX) - (Ry + Fy) + (RZ + FZ)
Case 4:
DL + RT + (R X + FX) - (R y + F y) - (RZ + FZ )
Case 5:
DL + RT - (R X + FX) + (R y + Fy) + (RZ + FZ)
Case 6:
DL + RT - (RX + FX) + (Ry + Fy)
_(RZ + FZ)
Case 7:
DL + RT - (R X + F ) - (Ry + Fy) + (RZ+F) 2 X
Z Case 8:
DL + R '- (R X + FX) - (R y + Fy) - (RZ + FZ)
T Where DL = dead load due to frame weight and pipe weight i
RT = pipe retcW/:s s e to pipe thermal load RX, Ry and RZ " Pipe seismic reactions applied in the X, y and Z directions 1
FX, FY and FZ = frame seismic inertia loads in the X, y and Z directions l'
5 l-YAN: 0599 i
1 s,
if
3.3 Acceptance Criteria Acceptance criteria for the MS/FW support frame shall be as specified in the Seismic Retrofit Criteria document DC-1, Rev. 3, Section
- 5. 4.
The members and connections shall be evaluated for'the most critical load case as discussed in Section 3.3 of this document.
- 4. 0 NON-RETLRN VALVE ENCLOStRE ANALYSIS The enclosure structure around the non-return valves on the main steam lines outside the VC will be analyzed using the Seismic Reevaluation and Retrofit Criteria, DC-1, Rev. 3.
Loads and load combination will be per Sections 5.2 and 5.4.1 of the Criteria. The response spectra to be used as input loading will be calculated using the mode shapes and participation factors from the main steam /feedwater piping analyses.
Analysis methodology and acceptance criteria will be per Sections 5.3.1 and 5.4 of the Criteria.
However, the non-return valve enclosure structure is not part of the Safe Shutdown System, and the sole purpose of the analysis is to demonstrate overall integrity of the enclosure structure such that it does not jeopardize the adequacy of the non-return valves on the main steam piping. Therefore, a nonlinear analysis method such as the one described in Section 6.0 of the Criteria and/or a more liberal acceptance criteria may be used to demonstrate the integrity of the enclosure structure if the allowable stresses of Section 5.4.2 are exceeded.
l 6
YAN: 0599 L
Comments on the Pro)osed Analysis Technique of the Yankee Main Steam and Main ;eedwater (MS/MF) and Support Structures During the January 1987 meeting, the licensee proposed an approach for seismically analyzing the MS/FW piping outside the Vapor Container and the associated supporting steel frame structures. This proposed was reviewed by the staff and its Consultants and the comments are listed below:
4 Main Steam /Feedwater Piping Analysis A.
Geometry and Computer Modeling 1.
Substructuring Method The definition of this method should be provided.
From the discussions held in the previous meetings, it is the staff's understanding that the condensation technique will be applied in order to reduce the degrees of freedom of the entire system.
It is not clear whether dynamic condensation or static condensation is to be used. The details of the theory and application procedures need to be provided for review.
2.
Modeling of Supporting Steel Frame Structures The following concerns should be considered when the model is developed:
a.
Inclusion of mass and stiffness of the horizontal steel members between two frames, e.g. steel beams, cross bracings, etc.
b.
Inclusion of the mass effects of the frames.
c.
Consideration of the flexibility of frame members on which the piping systems are directly supported.
3.
Modeling of Nozzle Stiffness at VC Penetration The use of WRC Bulletin 297 and cylindrical shell theory as I
discussed on page 2, fourth paragraph is not acceptable, because the local stiffness of a spherical shell, e.g. the VC shell, is much stiffer than that of a cylindrical shell with the same radius and thickness. Three P.P. Bijlaard's papers can be used as references for modeling the nozzle stiffness at the VC penetration. They are:
a.
Bijlaard, P.P., " Stress in a Spherical Vessel from Radial Loads l
Acting on a Pipe," WRC Bulletin No. 49, pp. 1-30, April, 1959.
b.
Bijlaard, P.P., " Stresses in a Spherical Vessel from external moments Acting on a Pipe," WRC Bulletin 49, pp. 31-62, April, 1959.
c.
Bijlaard, P.P., " Influence of a Reinforcing Pad on the stresses in a Spherical Vessel Under Local Loadings", WRC Bulletin 49, pp. 63-73, April 1959.
a
_2 4.
Modeling of Piaing Two concerns s1ould be considered:
The four pipe lines should be modeled as one system because they a.
are closely coupled through the support structures.
b.
The motions at the anchor points (translations and rotations) should be defined.
5.
The effect of the crane support structures at the north end of the system can be ignored only if the double "Y" joint can be defined as an anchor point.
6.
All concentrated masses attached to the system should be included in the model.
7.
The non-return valve (NRV) Enclosure should be included in the model.
B.
Loading Conditions 1.
The use of independent support motion analysis technique together with PVRC damping is acceptable provided that the requirments stated in NUREG-1061 are applied.
2.
Since the degrees of freedom (D0F) of the support structures will be condensed in the final dynamic model, proper support motions which include the amplification effects through the support structures should be used as the input for the analysis. Further more, these input response spectra need to be peak-broadened as required by NRC Regulatory Guide (RG) 1.122.
3.
It is not clear whether the PVRC damping will also be applied for the ground response spectra.
4.
The amplified response spectra at all support points (e.g. VC, PAB, TB, etc.) need to be peak-broadened as required by R G 1.122.
Main Steam /Feedwater Piping Support Frame Analysis A.
Geometry and Computer Modeling Not only the stiffness but also the inertia effects of the connected buildings (VC,PAB,TB and Crane Support) should be considered when the model is developed.
B.
Loading and Methodology 1.
Since the systems are located outside the buildings, a certain level of wind load should be combined with the earthquake load.
2.
When a equivalent static method used for the analysis, the requirements specified in Section 3.7.2 of the NRC standard Review Plan (SRP) should be applied,1.5" before calculating the equivalent static forces.i.e. the spectra factor of
~]
3.
The relative motions induced from other buildings (VC, PAB, TB, etc.) should also considered in the analysis.
Non-Return-Valve-Enclosure-Analysis Since no specific nonlinear analysis method and criteria were proposed a detailed description of the method, procedures, and criteria should be submitted for review, if nonlinear technique is chosen for analyzing this structure.
In addition, the licensee should demonstrate that the time histories to be used for the analysis do contain adequate energy content and frequency distribution.
l l
i 4
- y.
. FEB o 21987 3
./
opies:
D NRC PDR Local PDR-PAD #1 r/f'
~-PAD #1 p/f TNovak, Actg. DD NThompson, DHFT OGC-Bethesda EJordan BGrimes JPartlow Glear.
PShuttleworth EMcKenna TCheng CGrimes
- ACRS (10)
LFMB L
l i-
[
I t.