ML19257C273
| ML19257C273 | |
| Person / Time | |
|---|---|
| Site: | Haddam Neck File:Connecticut Yankee Atomic Power Co icon.png |
| Issue date: | 01/31/1980 |
| From: | NORTHEAST UTILITIES |
| To: | |
| Shared Package | |
| ML19257C271 | List: |
| References | |
| TASK-03-06, TASK-3-6, TASK-RR NUDOCS 8001250470 | |
| Download: ML19257C273 (28) | |
Text
.
DOCKET No. 50-213 ATTACHMTNT 1 HADDAM NECK PLANT SEISMIC REEVALUATION PROGRAM CRITERIA DOCUMENT 1809 080' JANUARY, 1980 8001250 hhd
SECTION IV PRIMARY COOLANT IDOP SYSTEM ANALYSIS A.
SCOPE The purpose of this document is to present the analytical methods and stress criteria which will be used for the Connecticut Yankee pri-mary coolant loop system seismic qualification program.
The program will include static analysis of the primary piping / support system for normal operating thermal, pressure, and deadweight loads along with dynamic system analysis for seismic loads.
Stress criteria will be presented for the piping, supports, and primary equipment.
B.
BACKGROUND In the years since the Connecticut Yankee generating station was designed seismic analysis methods have become more rigorous and the ASME Boiler and Pressure Vessel Code,Section III, Nuclear Power Plant Components, has been published reflecting changes in analysis, design, and quality control techniques.
The purpose of this criteria document is to establish requirements for performing the upgrading seismic analyses of the primary coolant loop system with current technology.
The original design criteria used for analysis of this plant's primary piping system is the ANSI B31.1 Code for Pressure Piping.
The reactor pressure vessel, steam generator, and reactor coolant pump were designed and analyzed to the rules of the ASME Code Section VIII.
lor the purposes of this document, the reactor coolant loop piping shall be considered to consist of the hot legs, cold legs, crossover legs, and pressurizer surge line.
The primary equipment i809 081
considered in this document consists of control rod drive mechanism, reactor vessel internals, reactor pressure vessel, steam generator, reactor coolant pump, and pressurizer.
The supports covered by the criteria in this document are those for the reactor pressure vessel, the steam generator, reactor coolant pump, and pressurizer.
C.
LOADING CONDITIONS The reactor coolant loop piping, supports, and components will be analyzed for the following loading conditions:
1.
Normal condition operating pressure, deadweight, and temperature.
2.
SSE Condition Sei.smic - Safe Shutdown Earthquake (SSE) combined with operating pressure and deadweight.
D.
STRESS CRITERIA 1.
Piping The piping analysis that will be performed for the Connecticut Yankee evaluation is based on the rules of the ANSI B31.1-1973 Code, the Summer 1973 Addenda.
The loading combinations and associated stress limits to be used for the piping systems which are part of the seismic qualification program are given in Table 1.
The stress limits used for the SSE condition correspond to faulted condition allowables.
The piping stresses are to be calculated using formulas given in ANSI B31.1-1973, 1973 Summer Addenda.
The maximum loads that the primary coolant loop piping is per-mitted to transmit to the pressurizer, steam generator, reactor coolant pump,and reactor pressure vessel nozzles are listed in Table 2.
1809 087.
Since the loop isolation valves are much thicker and stronger than the attached piping, and since valves of this design have no history of gross failure of their pressure boundaries (as long as the stresses of the piping attached to the valve remain within the limits defined in this document) the valve integrity is assured.
2.
Supports For linear type supports (i.e.,
reactor coolant pump hangers), the basis used for the stress criteria in this section is the AISC steel construction manual.
The other supports in the primary coolant loop system can be classified as plate and shell types.
The stress criteria for the plate and shell supports that is outlined in this document are based on the ASME Code,Section III, Subsection NF.
The load combinations and stress limits for both the lin3ar and plate and shell primary equipment supports are presented in T tble 3.
The information presented in the table will provide allowables for normal operating and seismic conditions.
3.
Components The basis of the stress criteria outlined in this section for the primary equipment is the ASME Code,Section III, Subsection NB.
The load combinations and stress limits to be used with those com-binations are presented in Table 4.
E.
ANALYSIS PROCEDURES 1.
General Procedures The reactor coolant loop piping / support system will be evaluated with three-dimensional static or dynamic models, depending on the load requirements, which include the effects of the equipment supports and equipment.
Static analysis of the piping systems will be performed 1809 083
using displacement techniques with limped parameters and stiffness matrix representations of supports.
It will assume that all com-ponents and piping behave in a linear e.~3stic manner.
The methods to be used for dynamic analysis depend upon which of two techniques is chosen, response spectra or time history.
Details of the two dynamic analysis procedures are presented in the following two sections.
The primary equipment that will be evaluated as part of this pro-gram shall have dynamic analyses performed in accordance with the same procedures as those presented below for piping systems.
In addition to the detailed models that are developed for the evaluations of the individual components, reduced models wi]' be produced for use in the reactor coolant loop system analysis.
Analytical representations of the primary equipment supports shall be produced for inclusion in the reactor coolant loop system
- model, ihe loads that are generated by the reactor coolant loop system mod (1 shall be used to qualify the component supports.
2.
Response Spectrum Analysis procedures If a decision is made to perform a response spectrum seismic analysis, a three-dimensional linear dynamic analytic model of the primary coolant loop system will be developed.
The model will include analytical representations of the components, component supports, and associated piping.
The boundaries of the model will be defined as the component support to containment concrete interface.
The analysis will be performed assuming that the seismic event is initiated with the plant at normal full power condition.
The damping values that will be used are four percent (4%) of critical for the SSE condition.
Since the components are supported a't different t
1809 084
floor elevations within the containment building, the response spectrum in each direction shall be an envelope of the applicable floor spectra.
The analysis shall be performed with a simultaneous input of the two horizontal components and one vertical component of the earth-quake.
The modal response for each item of interest (e.g.,
- force, displacement, stress) shall be obtained by the square root of the sum of the squares method.
3 1/2 2
RT= [ER 3
1 i=1 N
1/2 2
where:
R
[E R
]
1 f
j=1 where:
RT total combined response at a point
=
R 1 value of combined response of direction i
=
R ij absolute value of response for direction i,
=
mode j N
total number of modes considered.
=
For systems having modes with closely spaced frequencies, the above method shall be modified to include the possible effect of these modes.
Combined total response for systems which have such closely spaced modal frequencies will be obtained in accordance with Regulatory Guide 1.92, or as an acceptable alternative, the following metnoa.
Tne groups of closely spaced modes shall be chosen suen that the difference between the frequencies of the first mode and the last mode of the group does not exceed ten percent (10%) of the lower frequency.
Frequency groups are formed starting f un the lowst fraluency 1809 585
and working toward successively higher frequencies.
No frequency should be included in more than one group.
The resultant unidir-ectional response for systems having such closely spaced modal fre-quencies shall be obtained by the square root of the sum of:
(a) the sum of the squares of all modes, and (b) the product of the responses of the modes in various groups of closely spaced modes and associated coupling factors, c.
The mathematical expression for this method (with "R" as the item of interest) is:
S N -1 Nj 2
2 R gR gcgg, for: AfK I
R
+2 I
E Ri
=
i j=1 j=1 K=Mj E=K+1 number of groups of closely spaced modes where:
S
=
lowest model number associated with group j of M
=
3 closely spaced modes highest model number associated with group j of N.
=
U closely spaced modes egg coupling factor with
=
.g g
2_
-1 i
i
~
c KE 1+
=
_(8gwK+0),"t)~
and:
W uj wg [1-(B)2]
=
g OK + (S O
K Kd frequency of closely spaced mode K (rad /sec) w
=
g fraction of critical damping in closely S
=
K spaced mode K duration of the earthquake (seconds) t
=
d i809.086
The analyses performed for piping and supports will not include stresses resulting from SSE induced differential motion.
These stresses are secondary in nature, based on ash 1E' Code rules for piping (NB-3653, NB-3653, F-1360) and component supports (NF-3231).
The
<SE being a very low probability single occurrence event, is treated as a faulted condition.
The analysis of the components subjected to seismic loading will involve several steps that are similar to those outlined above for the system analysis.
A three-dimensional linear elastic analystic repre-sentacion of the component is developed, The component supports and attached primary coolant loop piping shall be represented by stiffness matrices.
The analysis shall be performed with the simultaneous input of three response spectra, two horizontal and one vertical.
Damping values of four percent (4%) for SSE will be used.
The model combina-tion techniques outlined for the system analysis shall also be used for the component analysis.
3.
Time History Seismic Analysis Procedures In the event that time history seismic analysis is required, the following procedures shall be used.
A three-dimensione.1 elastic non-linear model of the reactor coolant loop system shall be used.
The model shall include a simplified representation of the containment interior concrete structure, the components, the component supports and the attached piping.
The effects of the large auxiliary piping systems (e.g., main steam, feedwater) shall be accounted for with stiffness elements in the form of linear springs or stiffness matrices.
Damping for the system model shall be provided using the Rayleigh method based on a computed model energy distribution.
I809 087
If the analysis is performed by applying the three orthogonal earthquake time histories separately, the total response will be obtained by adding the three directional responses by algebraic summation.
If the three time histories are applied simultaneously, direct integration will be used to determine the total response.
F.
MODELING TECIINIQUES The piping system componenta, rtad component supports are to be represented by an ordered set of data which numerically describes the physical system.
The spatial geometric description of the model is to be based upon the as-built isometric piping drawings and equipment drawings.
Node point coordinates and incremental lengths of the members are determined from these drawings.
Node point coordinates are input on network cards.
Incremental member lengths are input on element cards.
The geometrical properties along with the modulus of elasti-city, E, the coefficient of thermal expansion, a, the average tempera-ture changes from the ambient temperature, AT, and the weight per unit are sp'cified for each element.
The supports are repre-s
- length, w,
sented by stiffness matrices which define restraint characteristics of the supports.
A network model is to be made up of a number of sections, each having an overall transfer relationship formed from its group of elements.
The linear elastic prope.~ ties of the section are to be used to define the characteristic stiffness natrix for the section.
Using the transfer relationship for a section, the loads required to suppress all deflections at the ends of the section arising from the 18'09 088
thermal and boundary forces for the section are obtained.
These loads are incorporated into the overall load vector.
After all the sections have been defined in this manner, the overall stiffness matrix (K) and associated load vector to suppress the deflection of all the network points is to be determined.
The flexibility matrix is multiplied by the negative of the load vector to determine the network point deflections due to the thermal and boundary force ef fects.
Using the general transfer relationship, the deflections and internal forces are then determined at all node points in the system.
The support loads (F) are also computed by multiplying the stiffness matrix (K) by the displacement vector (6) at the support point.
The models used in the static analyses are to be modified for use in the dynamic analyses by including the mass characteristics of the piping and equipment.
The lumping of the distributed mass of the piping systems is to be accomplished by locating the total mass at points in the system which will approximately represent the response of the distributed system.
Effectf3 of the equipment motion will be obtained by.aodeling the mass and the stiffness characteristics of the equipment in the overall system model when required.
The supports are again represented by stiffness matrices in the system model for the dynamic analysis.
From the mathematical description of the system, the overall stiffness matrix (K) is to be developed from the individual element stiffness matrices using the transfer matrix (K ) associated with R
mass degrees-of-freedom only.
From the mass matrix and the reduced, stiffness matrix, the natural frequencies and the normal mbdes are to be determined.
1809 089
The effect of eccentric masses, such as valves and extended structures, are considered in the seismic piping analyses.
These eccentric masses are modeled in the system analysis, and the tor-sional effects caused by them are evaluated and included in the total system response.
The total response must meet the limits of the criteria applicable to the safety class of the piping.
l'809 090
TABLE 1 LOADING COMBINATIONS AND STRESS LIMITS FOR PIPING LOADING COMBINATIONS STRESS LIMITS 1.
Normal:
Design Pressure + Deadweight
$S h 2.
SSE:
Op; rating Pressure + Deadweight
+ Maximum Potential Earthquake Loads (SSE) 12.4 Sh where:
Sh = all wable stress from USAS B31.1 Code for Pressure Piping.
TABLE 2 STEAM GENERATOR LOADS INLET AND OUTLET NOZZLES FORCE (kips)
MOMENT (in-kips)
LOAD X
Y Z
X Y
Z Thermal 160 100 25 1200 3000 15000 DW 20
- 20 10 70 100 150 Pressure
+1700 40 15 1500 1500 2000 SSE 250 200 120 6000 6200 7000 REACTOR COOLANT PUMP LOADS INLET NOZZLE FORCE (kips)
MOMENT (in-kips)
LOAD X
Y Z
X Y
Z Thermal 100 30 30 R000 7000 3000 DW
+
20 1
1 50 200 200 Pressure
+1700 30 20 1000 6000 3000 SSE 350 200 275 6000 13000 10000 REACTOR COOLANT PUMP LOADS OUTLET NOZZLE FORCE (kips)
MOMENT (in-kips)
LOAD X
Y Z
X Y
Z Thermal 50 50 40 3000 3000 7000 DW 1
- 10 1
50 20 150 Pressure
+1400 10 10 1000 700 500 SSE 450 150 300 13000 1500 15000 1809 092 NOTE:
1.
All loads are + unless noted.
2.
Coordinate system.
X-Y Plane Vertical Z
By Right Hand Rule
TABLE 2 (Continued)
REACTOR PRESSURE VESSEL LOADS INLET NOZZLE FORCE (kips)
MOMENT (in-kips)
LOAD X
Y 7
X Y
Z Thermal 50 100 30 5000 7000 5000 DW 1
- 20 1
200 60 800 Pressure
+1400 1
10 800 700 200 SSE 300 130 300 9000 13000 10300 REACTOR PRESSURE VESSEL LOADS OUTLET NOZZLE FORCE (kips)
MOMENT (in-kips)
LOAD X
Y Z
X Y
Z Thermal 60 150 30 1000 4000 20000 DW 1
- 20 1
75 100 800 Pressure 1500 5
5 70 000 400 SSE 500 90 160 1600 14000 7000 PRESSURIZER SURGE NOZZLE FORCE (kips)
MOMENT (in-kips)
LOAD X
Y Z
X Y
Z Thermal 3
7 7
1200
.1000 400 DW
+
30 1
1 15 10 35 SSE 3
5 5
250 350' 350 100' NOTES:
1.
All loads are + unless noted.
2.
Coordinate system.
X-Y Plane Vertical Z
By Right Hand Rule
TABLE 3 LOADING COMBINATIONS AND STRESS LIMITS FOR SUPPORTS LOADING LINEAR TYPE PLATE AND SHELL 3
COMBINATION SUPPORTS LIMITS SUPPORTS LIMITS P+D+T Working Stress" Pm i 1.0 S D
m Pm+Pb I 1.5 Sm 1
l P+D+TO+E Within lesser of Pm i 1.2 Fy 2
1.2 F 0.7 S Pm+Pb i 1.8 F y or u
y Y
I t
t times working limits" 1
Not to exceed 0.7 Su-2 Not to exceed 1.05 S u 3
Compressive axial member loads should be kept to less than 0.9 times the critical buckling load.
4 Working stress allowables per Appendix XVII of ASME III.
NOTES:
P pressure
=
deadweight D
=
D thermal-design temperature T
=
TO thermal-operating temperature
=
El SSE
=
material yield strength F
=
y t
allowable tensile stress per ASME Section III, F
=
Appendix XVII 1809 094
TABLE 4 LOADING COMBINATIONS AND STRESS LIMITS FOR COMPONENTS LOADING COMBII;ATION STRESS LIMIT Design Pressure + Deadweight Pm18m Pg (Pm) + PB 1 1.5 Sm Operating Pressure + Deadweight Pm1248m
+ SSE PL (Pm) + PB i 3.6 Sm*
1 Not to exceed 0.7 S u 2
Not to exceed 1.05 Su general primary membrane stress NOTES:
P
=
m L
Primary local membrane stress P
=
B Primary bending stress P
=
all wable stress intensity per ASME,Section III S
=
m ultimate stress at operating temperature G
=
u 1809 095
V.
ANALYSIS AND EVAldATION OF PLANT STRUCTURES 1.
Basic Approach This section outlines criteria that form the basis for the reassess-ment of the structural adequacy of the safety-related structures to resist the SSE loads. The structures that will be included in the reevaluation are:
a)
Containment Shell b)
Containment Internal Structure c)
Screenwall House d)
Primary Auxiliary Building e)
Service-Turbine Building Complex f)
Auxiliary Feedwater Building All of these structures may be classified as seismic Category 1 structures except for some areas of the service-turbine building complex. The new diesel generator building, a recent addition to the plant, has been designed as a seismic Category 1 structure using currently accepted techniques and is not included in this reassessaent.* All structures will be reevaluated using dynamic analyses. Where the preliminary evaluations indicate a considerable margin of safety with respect to the postulated seismic event, a simplified equivalent static procedure may be used.
The following documents establish acceptable methods, stresses and properties and are discussed in detail in the sections.to fol. low:
1809 096 USNRC Standard Review Plan - Sections 3.7.2, 3.8.3, 3.8.4 USNRC Regulatory Guides 1.60, 1.61, 1 92 ACI Codes 318-71, 349-76, 359-77 ASME B&PV Code,Section III Subsections NE, NF AISC Specification for Design, Fabrication and Erection of Structural Steel for Buildings Uniform Building Code - 79 Edition for Unreinforced Brick and Hollow Unit Masonry Although the proposed criteria are essentially the current standard ones, they may subsequently be modified.
If the proposed modifica-tion departs significantly from the present NRC positions, justifications will be documented to support any changes.
- The new diesel generator building was analyzed in 1974. A three dimensional space frame model was constructed and an evaluation made based on a Reg.
Guide 160 input normalized to 0.17 G.
1809' 097
. 2.
Time-History Motions The seismic input has been described in terms of response spectra in Section III.
The seismic input is also needed in terms of time-histories for the computation of floor response spectra as well as in the time-history analyses procedures for structural response computations.
Time-histories that will be developed for such purposes will match the design response spectra of Section III within the limits required by USNRC Standard Review Plan Section 3.7.1.
The overall duration and the rise, strong motion and decay portions of the time-history will be consistent with the hypothesized SSE.
1809 098
. 3.
Material Properties For the determination of the strength and stiffness of the structures under the postulated seismic conditions, the material properties will be taken as either those specified on the contract drawings and docu-ments or the average of actual material properties obtained from tests at the +1me of construction.
In lieu of construction test data, tests a selected cores or samples from existing construction may be performed to obtain actual material properties.
Table 1 lists the specified material properties for concrete, reinforce-ment and structural steel in various structuree.
Damping in reinforced concrete and structural steel shall be taken as 5 percent of critical except 7 percent of critical damping may be used when the stresses induced in the structure by seismic, gravity, and operating loads (see section 6.1) are high (close to allowables, see section 6.2).
180L9099
' TABLE 1 SPECIFIED MATERIAL PROPERTIES I.
Containment & Internals Structure A.
Reinforced Concrete 3000 psi 0 28 days (came concrete - f' = 5000 psi) 0 28 days) 1.
f'
=
2.
Reinforcing Steel (a)
- 14 & #18 (ASTM A408)
(1) Typical fy = 50,000 psi min (2)
Foundation Mat and Exterior Wall Dowels:
fy = 40,000 psi min (b)
- 11 (ASTM A-15 & A-305)
(1) Typical:
fy = 40,000 psi min (2)
Exterior Wall above elev. 31'-6" and dome:
fy = 50,000 psi min (c)
- 10 and smaller: ASTM A-15 and A-305, intermediate grade, fy = 40,000 psi min B.
Structural Steel ASTM A-36, Fy = 36,000 psi min II.
Primary Auxilary Building, Turbine-Service Building Complex and Screenwell House A.
Reinforced Concrete 1.
F' = 3000 psi 0 28 days 2.
Reinforcing Steel: ASTM A-15 & A-305, intermediate grade, 809 000 fy - 40,000 psi min.
. B.
Structural Steel ASTM A-36, Fy - 36,000 psi min C.
Unreinforced Brick and Hollow Unit Masonry:
S' = 1500 psi III. Auxiliary Feedwater Building A.
Structural Steel ASTM A-36, Fy = 36,000 psi min 1805101
. 4.
Analytical Procedures Linear elastic dynamic analyses procedures are intended to be used for all structures.
If nonlinear inelastic procedures are to be used for any structure a separate criterion will be develope) for the non-linear analysis procedures and acceptance criterion.
The USNRC Standard Review Plan Section 3.7.2 shall be followed in those matters not explicitly covered by this document.
The following dynamic analyses procedure may oe used:
o Response spectrum modal superposition o
Time-history modal superposition o
Time-history direct integration Equivalent static procedure may also be used where justified.
4.1 Soil-Structure Interaction Most of the structures at the CY plant are founded on rock (shear wave velocity, Vs > 3500 ft./sec.).
Screenwell house is founded on lean concrete fill of 2 ft. to 20 ft, depth over rock.
A small portion of Turbine Building is also founded on lean concrete fill of small depth over rock. A lightly loaded region of Service Building is founded on select compacted fill (soil) of about 10 ft. depth over rock.
1809 102'
. gn Soil compliance effects in those portions of any structure founded
'on soil backfill will be considered. The seismic input for all structures will be as described in Section 2.
4.2 Structural Modeling Dynamic structural models will be used to calculate the structural responses to the horizontal and vertical components of the ground motion. Material properties used in these models will be as defined in Section 4.3.
In general, the stiffness of reinforced concrete structural members will be calculated using gross cross-sections. Cracked sections will be used when necessary for a realistic assessmert of the stiffness.
Mass calculations shall include the dead weight of the structures as well as the equipment.
The mass of non-structural elements (e.g.,
partitioning) and small pieces of equipment (e.g., electrical cabinets) will be estimated as a uniform weight across the whole floor.
4.3 Coupling Simplified models of the Nuclear Steam Supply System (NSSS) components will be coupled to the dynamic structural model of the Containment Internal :tructure.
Responses at the equipment supports will be cal-culated for later use in NSSS qualification.
1809 103
_9 Structures which are physically connected by structural elements will be analyzed using coupled dynamic models except where it can be shown that coupling does not significantly influence relevant structural responses.
4.4 Torsion Significant eccentricity between mass and stiffness will induce torsional response in a structure subjected to horizontal component of ground motion. Such eccentricity will be taken into account in the modeling of structures.
In addition, to account for variation in location of mass and stiffness in the model and in the structure as well as possible torsional input into the structure, accidental eccentricity or equivalent will be considered.
For structures with rigid diaphragms or equivalent which are modeled by lumped mass models, accidental eccentricity shall be taken equal to 5% of the plan dimension normal to horizontal input component.
Such accidental eccentricity will be additive to geometrical eccen-tricity that may exist at that level.
For other structures where accidental eccentricity cannot be accounted for in a simple manner, the responses to the horizontal input component shall be increased by 5% to account for the effects of accidental eccentricity.
Torsional responses shall be combined with translational responses on an absolute sum basis.
1809 104
. 5.
Floor Response Spectra The peaks in the floor response spectra at structural frequencies are usually broadened to account for the uncertainties in these frequencies due to uncertainties in materi=1 properties, and approxi-mations in modeling techniques and analyses procedures.
When minimum specified properties of structural materials are used in the model, the spectral peaks at structural frequencies will be broadened by 15% on each side of such frequencies.
If actual average structural material properties determined from test data are used in the models, a portion of this uncertainty is accounted for.
The average material properties are usually higher than the minimum specified properties and leads to somewhat higher values for structural frequencies.
In this case the spectral peaks at structural frequencies will be broadened by 5% on the high side and 15% on the low side of the structural frequencies.
Lesser peaks and valleys will be smoothed by free-hand enveloping.
1809 105
. 6.
Acceptance Criteria 6.1 Load Combination The following load combination will be censidered in evaluating the structure:
U = D + L' + 0 + E where U = total load to be resisted L' = actual live load 0 = operating temperature and pressure loads, if any E = SSE load D = dead weight 6.2 Ailowable Stresses The allowable stresses for reinforced concrete portions of structures will be per ACI Code 359-77 for the Containment Exterior and Internal structures and ACI Code 349-76 for other structures.
The stresses for steel portions of structures will be checked per Part 1 of AISC Specification,1979 edition, except thct the allowable stresses will be as delineated in NRC Standard Review Plan Section 3.8.3 and 3.8.4.
1809 106
_12-6.3 Structural Foundations The structural foundations will have a factor of safety 1.1 against slidin) and overturning for the following load combination:
U = D + L' + E where V, D, L' and E are as defined in Section 6.1.
~1809 107