ML20086U062
| ML20086U062 | |
| Person / Time | |
|---|---|
| Site: | Point Beach  |
| Issue date: | 12/23/1991 |
| From: | Zach J WISCONSIN ELECTRIC POWER CO. |
| To: | Samworth R NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20086U063 | List: |
| References | |
| CON-NRC-91-156, RTR-REGGD-01.035, RTR-REGGD-01.035.01, RTR-REGGD-1.035, RTR-REGGD-1.035.01 VPNPD-91-457, NUDOCS 9201070193 | |
| Download: ML20086U062 (41) | |
Text
-!
..$ i.-g Wisconsin.
Electnc POWER COMPANY
- 23t W Michgin, Po Bax 2046. Mhaukee WI 53201 (414)??1 2345 VPNPD 457 NRC-91 156 December-23, 1991 Document Control. Desk U.S. NUCLEAR REGULATORY COMMISSION Mail Station F1-137-Washington, D.C.
20555 Attn:-Dr.-Robert Samworth, Project Manager Project Director III-3 Ocntlemen DOCKET NOS. 50-266 AND 50-301 TENDON SURVEILLANCE REPORTS POINT BEACH NUCLEAR PIANT f'
[
During the week of October 21, 1")91,-a NRC Structural Audit Team i
visited the Point Beach Nuclear Plant to gather information on the
-affects'of age related degradation on plant structures and compo-nents.. At-the exit meeting, the audit team-leader requested that all
-tendon surveillance reports for Point Beach Nuclear Plant be submitted for further review.
The. audit team's: concern was regarding a change made to our Technical l '-
Specifications to conform to Regulatory Guide 1.35 and 1.35.1.
To
.further evaluate this concern, we have contracted-Bechtel to review I
our.ca1cula.tions'and' Technical Specification change.
L Enclosed is the following documentation:
L
- 1.
PBNP FSAR Section 5.1.2.4, pages 5.1-51 through 5.1-88.
2.
PBNP Technical Specification 15.4.4 VII, pages 15.4,4-8 through p
_15.4.4-10.
3.
PBNP' Calculation N-89-094.
n i
4.
.PBNP Unit 1,
" Containment Building Post-Tensioning System One-Year Surveillance."
5.
PBNP Unit 2, " Containment Building Post-Tensioning System One-Year Surveillance."
h 667 2
-es Document Control Desk December 23, 1991 Page 2
-1 6.
PBNP Unit 1, " Containment Building Post-Tensioning System Three-Year Surveillance."
7.
PBNP Unit 2, " Containment Building Post-Tensioning System Three-Year Surveillance."
8.
PBNP Unit 1 and Unit 2, " Containment Building Post-Tensioning System Eight-Year Surveillance."
9.
PBNP Unit 1 and Unit 2, " Containment Building Post-Tensioning System Thirteen-Year Surveillance."
10.
PBNP Unit 1 and Unit 2,
" Containment Building Post-Tensioning System Eighteen-Year Surveillance."
The enclosed material listed above is more than what the audit team leader requested.
We believe the additional information may prove useful in your investigation of this matter.
Very truly yours, Q
4-'
F Wo Jam p J.
Zach Vice President Nuclear Power Enclosure Copy to:
NRC Regional Administrator, Region III (w/o/e)
NRC Resident Inspector (w/ofe) l 4
l l
[
1 1
0 PBNP FSAR SECTION 5.1.2.4 (Pages 5.1-51 through 5.1-88) k i
1 4
/
)
and similar items may be picked up and carried at the maximum wind velocity of 300 mph.
The behavior of heavier, oddly shaped objects such as an automobile, is less predictable.
The design values of 50 mph for a 4000 lb. automobile lifted 25 feet in the air is felt to be representative of what would happen in a 300 mph wind as the automobile was lifted, tumbled along the ground, and ejected from the tornado funnel by centrifugal force.
These tissile velocities are consistent with reported behavior of such objects in previous tornadoes.
Unlift Due to Buovant Forces Uplift forces which are created by the displacement of ground water by the structure are accounted for in the design of the structure.
External Pressure Load External pressure loading with a differential of 2 lbs. per sq. in
, from outside to inside is considered.
i t
The external design pressure is equivalent to having a barometric pressure rise to 31 in, of mercury after the containment is sealed at 29 in
, of mercury at the same time as the containment is cooled to 50"F from an initial maximum operating condition of 105 F.
Therefore, operation of purge valves is not necessary during this shutdown condition. Vacuum breakers are not required.
E.1.2.3 Egismic Desion llassificat_4jm All equipment and structures are classified as Class I, Class II, or Class III as recommended in TID 7024, " Nuclear Reactors and Earthqua August 1963.
es,"
5.1-51
_ _ - ___ -_-_ - - - - -- - ~ ~ ~^
These classifications are defined as follows:
1.
Class 1 Then structures and components including instruments and control f
whose failure might cause or increase the severity of a loss-o -
coolant accident or result in an uncontrolled release of excessive Also, those structures and components amounts of radioactivity.
1 vital to safe shutdown and isolation of the reactor.
4 2.
Class 11 are important to reactor and components which Those structures operation bat not essential to safe shutdown and isolation of the reactor and whose failure could not result in the release of
+
substantial amounts of radioactivity.
y 3.
Class Ill Those structures and components whi 5 are not ielated to reactor operatinr n-containment.
5.1.2.4 Qgt
.. Design General The analysis for the containment structure f alls into two general The analysis and nonaxisymmetric analysis.
axisymmetric categories, axisymmetric analysis is performed through the use of the finite element computer program for the individual loading cases of dead load, live load, temperature, prestr?ss, and pressure using the usual assumptions of the theory of elasticity as described in Section 5.1.2.2.
The finite element approximation of the containment structure does not consider tFa buttresses, and the lateral loads due to seismic or wind are considered in the nonaxisymmetric analysis described later in the section.
5.1-52
This 'ectioh of the fSAR discusses analytical techniques, references, and design philosophy, lie design criterit, analysis, and construction drawings were reviewed by Bechtel's consultants,1. Y. Lin, Kulka, Yang &
Associate.
Axisymmet ric lechni. gym The finite element technique is a general method of structural analysis in which the continuous structure is replaced by a system of elements (members) connected at a finite number of nodal points (joints).
Conventional analysis of frames and trusses can be considered to be examples of the finite element method.
In the application of the method to an axisymmetric solid (e.g., a concrete containment structure), the continuous structure is replated by a system of rings of triangular cross section which are interconnected along circumferential joints.
Based on energy principles, work equilibrium equations are formed in which the ro. dial and axial displacements at the circumferential joints are the unknowns of the system.
The results of the solution cf this tet of equations is the deformation of the structure under the given loading conditions. For t* s output, the stresses are computed knowing the strain and stiffness of each element, The finite element mesh used to describe the structure is shown in figure 5.1-9.
The upper portion and lower portion of the structure are anal / zed independently to permit a greater number of elements to be used for those areas of the structure of major interest such as the ring girder area and the base of the cylinders.
The finite element mesh of the structure base slab is extended down into the foundation material to take into consid nation the elastic nature of the foundation material and its effect upon the behavior of the base slab.
Ihe use of the finite element ccW > program permitted an accurate estimate of the stress pattern at various locations of the structure. The following material properties were used in the program for the various loading conditions:
5.1-53
t i
4 Load Conditions load Condition I
D. F. 1. T.
E 3
E,,,m,,,,,,,,,,..,,,,y 2.7 x 10' 5.O x 10' E... e......, i,,. o 2.5 x 10' 5.0 x 10' v,,,,;,,,
O.17 0,17
"...i...c.......<.......,,,3 0,5 x 1 0
L.. a <,. o 0.13 x 10' O.13 x 10' 30 x 10' 30 x 10'
..e i,. o 30 x 10' 30 x 10'
,i... c,. o fyi..., r.. o 34.000 for definition of Load Conditions. see Section 5.1.2.2.
The3 structure is analyzed assuming an uncracked homogeneous material.
The major benefit of the program is the rapability to predict shears and moments due to internal restraint and the interaction of the foundation slab relative to the soil.
The use of an uncracked section is conservative because the decreased relative stiffness of a cracked section would result in smaller secondary shears and moments, in arriving at the above mentioned values of E,, the effect of creep is included by using the following equation for long-term loads such as thermal load, dead load..vid prestress:
E
- E co ct e,
e, where Ec, Sustained modulus of elasticity of concrete g
E,,
Instantaneous modulus of elasticity of concrete a
5.1-54
t t
Instantaneous strain, in./in. per psi e,
Creep strain, in./in per psi c,
=
The thermal gradients used for design are shown in figure 5.1-17.
The gradients for both the design accident condition and the factored load condition are based on the temperature associated with the factored pressure. The design pressure and temperature of 60 psig and 286'r become 90 psig and 310*f at factored conditions.
For such a small increase in temperature, it was decided to use a single set of thermal gradients to simplify the analysis.
The thermal loads are a result of the temperature differential within the structure.
The design temperature stresses for this finite element analysis were prepared so that wnen temperatures are given at every nodal point, stresses are calculated at the center of each element.
Thus, the liner piste is handled as an integral part of the structure but having different material properties, and not as a mechanism which would act as an outside source to produce loading on the concrete portion of the structure.
Under the design accident condition or factored load condition, cracking of the concrete at the outside face would be expected.
The value of modulus of elasticity of concrete, E,, was used together with the method l
described in ACI Code 505-54 to find the stresses in concrete, reinforcing steel, and liner plate from the predicted design accident thermal loads and factored acciient loads.
1 The isostress plots shown in figures 5.1-10 and 5.1-11 do not consider the concrete cracked.
The thermal stresses are combined in the isostress l
output for the cases of D + f + T and D + F + 1.5P + T.
The first case was critical for concrete stresses and occurs after depressurization of the containment; the second case is critical for the reinforcing stresses I
and it occurs when pressure and thermal loads are combined and cause cracking at the outside face.
5.1-55 LI t
The stresses shown in lable 5.1-1 consider cracking. The general approach of determining stresses in the concrete and reinforcement required the evaluation of the stress blocks of the cross section being analyzed.
The value of stresses was taken from the computer output in case of axisymmetric loading and from analytical solutions is case of nonaxi-symmetric loading. Both computations are based on homogeneous materials, therefore, some adjustment is necessary to evaluate the true stress-strain conditions when cracks develop in the tensile zone of the concrete.
1he procedures used to determine the area of conventional reinforcing required and the stress in the concrete resulting from the loading condition, considering the effects of cracking where required, are presented.
P Basic Assumption: The thermal stresses in the containment are comparable to those developed in a reinforced concrete slab which is restrained from rotation. The temperature varies linearly across the alab. The concrete will crack in tension and the neutral axis will be shifted toward the compressive extreme fiber.
The cracking will reduce the compression at the extreme fiber and increase the tensile stress in reinforcing steel.
The following analysis is based on the equilibrium of normal forces, therefore, any normal force acting on the section must be added to the normal forces resulting from the stress diagram. The effects of Poisson's ratio are considered while the reinforcement is considered to be identical in both directions.
Stress-strain relationship in compressed region of concrete:
E c, = 0, - v o, c
e E c, = - v o, + o, c
c a,= E, '* * ' ' "
2 1 - v c' 5.1-56
)
s
%*tV a,, y 1-V, assuming o, = 0, c o, and e, = c, = c, o, = E, c f
= 1. 2 05 E, c, [if v, = 0.17 )
e The reinforcement is acting in one direction, independently from the reinforcement in the perpendicular direction.
Example:
6 and g,. 3 0 x 10 Jf E, = 3 x 10 6
1.205 x 3 The liner. plate is acting in two directions, similar to the concrete except for the difference caused by the Poissons ratios:
o = E,C, f
= 1. 3 5 E,c, t
If vt = 0.25 and v,= 0.17
- 1. 3 5 x30 n =
= 13. 2 t
1.205 x 3 l
l L
The following is an example of the use of the analytical method derived.
Thermal stress in base slab:
l^
E
- 3 x 10' psi c
E, = 3 0 x 10' p s i V, = 0. 2 '
5.1-57 u
a
4 vt = 0.25 n, = 8. 3 nt= 11.2 Equilibrium of forces considerin;; crack section:
4,42 [293 + Ac ) 8. 3 - [65. 0 + 105.7 + 24. 0) c 1000 + Ao, [12 x 42 + 3 x 11.2) = N = -95,000 lbs ba 156. 5 psi c
o, = [29 3 + 156. 5) 8. 3 = 3,7 31 psi The concrete and reinforcement stresses are calculated by conventional methods from the moment caused by loading other than thermal.
The analyses assume homogeneous concrete sections.
Those concrete and reinforcing steel stresses are then added to the thermal stresses as obtained by the method described.
Notation:
E, Modulus of elasticity of concrete E,
Modulus of elasticity of steel Modular. ratio of liner plate / concrete n
t Hodular ratio of reinforcement / concrete n,
=
A o, = Reduction of concrete compressive stress considering cracking Concrete strain c,
=
Steel strain c,
5.1-E8 e
Concrete strain in X direction c,
9 Concrete strain in Y direction Poisson's ratio of concrete v,
=
Poisson's ratio of liner plate v,
=
Stress in concrete o,
Stress in liner plate a
t Stress in reinforcement c,
Stress in concrete in direction X o,
An equilibrium equation can be written considering the tension force in the reinforcement, the compressive force in the concrete, and the axial force acting on the section, in this manner the neutral axis is shifted from the position defined by the computer analyses into a position which is the f.inction of the amount of reinforcement, the modulus ratio, and the acting axial forces.
Large axial compressive force might prevent the existence of any tension stresses, as in the loading condition, D + F + T, therefore, no self-relieving action is existing; the stresses are taken directly from the computer output, in the case of D + F + 1.5P + T, the development of cracks in the concrete decreases the thermal moment and this effect is considered, but the self-relieving properties of other loadings are not taken into account even in places where they do exist, such as at discontinuities, e.g., the cylinder base slab connection. This means that in analyzing the section, a reduced thermal moment is added to the moment caused by other loadings without any reduction.
5.1-59 l
I NLnEx_j)ymmetric Ana1_yLij The nonaxisymmetric aspects of configuration of loading required various methods of analysis.
The description of the methods used as applied to different parts of the containment are given in the sections below.
Buttresses The? buttresses are analyzed for two effects, nonaxisymmetry and anchorage zone stresses. Both effects are shown in the results of a two dimensional plane strain finite element analysis with loads acting in the plane of the coordinate system (Figure 5.1-12).
At each buttress, the hoop tendons are alternately either+ continuous or spliced by being mutually anchored on the opposite faces of the buttress.
Between the opposite anchorages, the compressive force exerted by the spliced tendon is twice as much as elsewhere, therefore, this increased value added to the ef fects of the tendon which is not spliced will be 1.5 times larger than the prestressing force acting outside of the buttresses.
The cross sectional area of the buttress is about 1.5 times that of the wall so the hoop stress as well as the hoop strains and radial dis-placements can be considered as being nearly constant all around the structure, Isostress plots of the plane strain analysis, figure 5.1-13, confirm this.
The vertical stresses and strains used by the vet tical post-tensioning oecome constant at a short distance away from the anchorages because of the large stiffness of the cylindrical shell.
Since, as stated above, the stresses and strains remain nearly axisyn-metric despite the presence of the buttresses, their effect on the overall analysis is negligible when the structure is loaded with dead load or prestressing loads.
When an increasing internal pressure acts upon the structure, combined with a thermal gradient such as at the design accident condition, the resultant forces being axisymmetric, the stiffness variation caused by the buttresses will be decreased as the concrete develops cracks.
The structure will then tend to shape itself to even more closely f ollow the direction of the a c'. i ng axisymmetric at yield loads, which include 5.1-60
factored pressure, than at design loads including pressure.
This fact, combined with the redundancy of the pressure resisting structural ele-ments, indicates that the buttresses will not reduce the margins of safety available in the structure.
Seismit or Wind lo3Jin_g n
Design requirements dictated by seismic loading of the structure are greater than that of tornado or wind loading.
The seismic analysis is conducted in the following manner.
The loads on the containment structure caused by earthquake are determined by a dynamic analysis of the structure.
The dynamic analysis is made on an idealized structure of lumped masses and weightless elastic columns acting as spring restraints.
The analysis is performed in two stages: the determination of the natural frequencies of the structure and its mode shapes, and the modal response of these modes to the earthquake by the spectrum response method.
The natural frequencies and mode shapes are computed from the equations of motion of the lumped masses established in a virtual displacement method solved by iteration techniques using fully tested (iigital computer program.
The form of the equation is:
( K) x ( A ) = w2 x (M) x (6)
(K)
Matrix of stiffness coef ficient including the combined effects of
- shear, ficxure,
- rotation, and horizontal translation (M)
Matrix of concentrated masses (A)
Matrix of mode shape Angular frequency of vibration u
5.1-61 l
r The results of this computation are the several values of u, and mode shapes An for n = 1, 2, 3 -- m where m is the number of degrees of freedom (i.e., lumped masses) assumed in the idealized structure.
The response of each mode of vibration to the des 4gn earthquake is then computed by the response spectrum technique as follows:
- 1.
- The base shear contribution of the n" mode V = K x S,n (u :b) n n
n where 9,
Angular frequency of the n" mode W,
Effective weight of the structure in the n" mode (E h nW)*
l x
a y",
D,(b,n)*W,.
where the subscript x refers to levels throughout the height of the
- structure and W, is the weight of the lumped mass at level x.
S., (u,;y) - Spectral acceleration of a single degree of freedom system with a damping coefficie6t of y obtained from the response spectrum 2.
The horizontal load distribution for the n" mode was then computed as:
F,
- V ( b W,) / ( E,h,n,)
W n
an The several mode contributions are then combined to give the final resronse of the structure to the design and hypothetical earthquake.
3.
The number of modes to be considered in the analysis is determined to adequately represent the structure being analyzed. Since the spectral response technique yields the maximum value of response for each mode 5.1-62 L
and these maxima do not occur at the same time, the response of the modes of vibration is combined by taking the square root of the sum of the squares of the modal values. The analytical model and results are shown in Figure 5.1-14.
Laroe Openinas (Eculoment Hatch and Personnel Lock Dr,enina)
As stated in the design criteria, the primary loads considered in the design of the equipment hatch and personnel lock opening, as for any of the structure, are dead load, prestress, pressure, earthquake, and thermal loads.
The secondary loads considered, caused by the above primary loads were:
1.
The deflection of tendons around the opening 2.
The curvature of the shell at the ope 'ng 3.
The thickening around the opening The loads described under primary loads are mainly membrane loads with the exception of-the thermal loads.
In addition to membrane loads, accident pressure also produces punching shear around the edge of the opening. The values of these loads for design purposes are the magnitudes of these loads at the center of the opening.
These are fairly simple to establish, knowing the values of hoop and vertical prestress loads, accident pressure loads, and the geometry and location of the opening.
The hoop normal forces caused by either post-tensioning or internal pressure have a very low value right at the base slab and gradually increase at higher elevations, accompanied by varying shear forces.
The effects of the earthquake loading is also a function of the elevation.
The equipment hatch on the Point Beach containment is close to the base slab so that the forces are not constant in the vertical direction.
The analysis considers these forces and the values are obtained from calculations considering a continuous shell.
l 5.1-63
.. shear stress near the edge of the opening. (E), for various components of loading is predicted to be as follows:
Prestress - 19 pri Pressure - 36 psi Earthquake - 3 psi The contribution from temperature and dead load are very small. Under the D4 f + P + T + E case the shear stress is predicted to oe 20 psi.
Secondary loads are redicted by the following methods:
1.
The membrane stress concentration factors and effect of the deflection of the tendons around the equipment hatch are analyzed for a flat plate by the finite element method.
The stresses predicted by conventional stress concentration factors, when compared with those values from the previously mentioned finite element computer program, 7
demonstrated that the deflectior of the tendons does not significantly s
affect the stress concentrations. This is a plane stress analysis and does not include the effect of the curvature of the shell.
- However,
~ it gives an assurance of the correctness of the assumed stress pattern
- caused by the prestressing around the opening.
2.
With the help of Reference 5, stress resultants around the large opening are found for various loading cases.
Comparison of the results found from this reference with the results of a flat plate of uniform thickness with a circular hole show the effect of the cylindrical curvature on stress concentrations around the opening.
Normal shear forces (relative to opening) are modified to account for the effect of twisting moments.
These modified shear forces are called Kirschoff's shear forces. Horizontal wall ties are provided to resist a portion cf these shear forces.
3.
The effect of the thickening on the outside face around the large opening is-considered using a separate axisymmetric finite element computer analysis for a flat plate with anticipated thickening on the 5.1-64 L
l outside ft;e. This particular finite element computer program handles both axisymmetric and nonaxisymmetric loads.
This finite element computer program is also used to predict the effect of concentration of hoop tendons (with respect to the containment) at the top and bottom of the opening.
Various conditions checked by the flat plate plane stress finite element analysis were as follows:
1.
During prestressing with only the hoop tendons stressed 2.
~e:e local effects of hoop tendon curvature under the D + F + 1.5P design load condition 3.
After total prestressing D + F The membrane loads were applied at the flat plate boundary and the tendon loads from curvature in the plane of the model were applied at the tendon locations.
The analysis considered the effects of thickening by assigning increased E values for the elements representing the thickened portion of the shell, but it did not consider the shell curvature effects and the fact that the thickening is not syr. metrical about the opening.
Reference 5 was used to determine the effects of shell curvature on the stress concentrations tround the opening.
For the analysis of the thermal stresses around the opening, the same method is used as for the other loadings.
At the edge of the opening, a uniformly distributed moment equal but opposite to the thermal moment existing on the rest of the shell is applied and evaluated using the methods of Reference 5.
The effects are then superimposed cn the stresses calculated for the other loads and effects.
In the case of accident temperature, af ter the accident pressure has already been decreased, very little or no tension develops on the outside, 5.1-65 l
so thermal strains will exist without the relieving et7ect of the cracks.
However, the liner plate will reach a high strain level, and so will the concrete at the inside corner of the penetration, thereby relieving once again tne very high stresses, but still carrying a high moment in the state of redistribution stresses.
')
In sthe case of 1.5P (prestress fully neutralized) + 1.0T (accident temperature),
the cracked concrete with highly strained tension reinforcement constitutes a shell with stiffness decreased but still essentially constant in all directions.
In order to control the increased hoop moment around the opening, the hoop reinforcement is about twice that of the radial reinforcement (see figure 5.1-15).
The equipment-hatch opening is thickened for the following> reasons:
1.
To reduce the larger than acceptable predicted stresses around the opening; 2.
To accommodate tendon placement; 3.
To accommodate bonded steel reinforcing placement; 4.
To compensate for the redut. tion in the overall shell stiffness due to the opening.
In order to minimize the effect of tensile stresses at the outside face and to distribute the concentration of radial forces exerted by hoop tendons in a more uniform - manner, the inside row of vertical tendons is given a reverse curvature (they are deflected outward as they pass the opening) so at'to reduce the inward acting radial forces (due to hoop tendons) at the top _ and bottom of the opening and to produce inward acting forces on the l-sides of the large opening.
The working stress method (elastic analysis) is applied to both the load combinations for desigr. loads as well as for yield loads for the analytical procedures descr bed above.
The only difference is the higher allowable stresses under yield conditions.
The various factored load combinations 5.1-66 i
(
l a
and capacity Jeduction (actors are specified in Section L.1.2.2 and are used for the 3 'tM toad coTbinatfors using the working stress siesign method.
The design assumption of streight line variation of stresses is maintained under yield conditions.
The governing design condition for the sides of the equipment hatch opening at the outside edge of the opening 1: the e,;cident cordition.
Under-this condition, approximately 60% of the total tiondtt' reinforcing steel needed at the edge of the opening at the outside face is a result of the thermal load.
A breakdown of total 'oading fc' lows:
Stress brgab(p_w_a 1.
From'The.-mal Gradient (Plus 60%)
T.
from Membrane force Ir.t.luding (tiinus 14%)
The lhickering t'ffect "J.
From Mrnent'; Caused by Tliickenini (Plus 60%)
4.
from Hembrane forces and Moments
'(Hinus 6%)
Caused t'y the Etfect of Cylitdrical Cervature
=_
Total (100%)
p Excluding ' thermal load, the remaining stress (eqaivalent to approximately j
40% of the total load, including thermal) at the edge of the outside face is the contribution of the following stress resultants:
1.
Normal stresses resulting from membrane forces, including the effect of thickening, contribute approximately -35% (-14% of total),
l 1
l' 5.1-67
2.
ficxural stresses resulting from the moments caused by thickening on the outside face contribute approximately 150% (60% of total).
3.
Normal and flexural stresses resulting from membranc forces and moments caused by the effect of cylindrical curvature contribute approximately -15% (-6% of total).
1 4
Etnet ratighl n
Analysis of the contcinment penetrations falls into three categories:
1.
The concrete shell; 2.
The liner plate reinforcement and closure to the pipe or electrical canister; 3,
The thermal gradients and protectior, requirements at the high temperature penetraticns.
i lhe three categories will be discussed separately.
The basic computer analyses applied in the design of the containment shell are for axisymmetric solids subjected to axisymmetric loadings; therefore, areas where either the shape er the loadinj is nonaxisymmetric are analyzed by other methods.
The nonaxisymmetric effects are not included in the axisymmetric analyses directly, but the results of two independent calculation methods are combined.
Small penetrations without appreciable accident pressure loads or pipe f ailure loads were analyzed as holes in a flat plate and the stress concentrations from the membrane loads were the main consideration in specifying the reinforcing steel.
fer penetrations which could be sub-jected to external forces and moments, additional reinforcement was added where necessary to resist moments and shear.
5.1-68
1.
Concrete Shell In general special design consideration is given to all opeliing:
the containment structure. Analysis of the various openings ha,
however, indicated that the degree of attention required depends upon the penetration size.
Small penetrations are considered to oe those with a diameter smaller than 1 1/2 times the shell thicknen, i.e.,
approxim tely 8 ft. in diameter or less. Reference 1 indicates iiat' for openings of 8 ft. diemeter or less the curvature etfect of the y
shell is negligible.
In general, the existing concrete wall th d 'ss is found to be capable of taking the imposed stras es using bondel reinforcement and the thickness is increased only as regrire '
' to poimit space requirements for tendon deflection.
Th s inuc.a stresses ~ due to - normal thermal gradients and octtti. ted ruptere conditions distribute rapidly and are of a minor niture corrrared.o th9 numerous loading conditions for which the sheM must be designed.
The penetrations are analyzed as holes in a planc sheet.
Ap. lied.
piping restraint loads due to thermal expansion or accident torces are assumed to distribute in the cylinder at st.ted in Re#erence 3.
Typical details associated with these openings are indica'ei in Figures 5.1-2 and 5.1-3.
2.
Liner plate Closure
'The stress concentrations around openin;s in tht einer plate we calculated using the theory of elasticity lhe stress concentrations are then reduced by the use of a reinforciri plate aroui j tJei opening.
In the case of a penetration with 10 a prhciable exte.nal load, anchor bolts are used to maintain strain compatioility between the liner plate and the concrete.
Inward displacemenc of the-liner plate at the penetration is.also controllel by t.se anchor bolts.
In the case of a ' pipe penetration in which large external g eratiig loads are imNsed upon tne pen!tration, the stress level from the external loads is limited to the design stress intensity values, S.,
given in_ (he ASME Boiler and Pressure Vessel Code, Section 111, 5.3-69 a
Ca i
Article 4.
The stress level in the anchor bolts from external loads is-in accordance with bearing values meeting ACI Code Requirements.
The combining of stresses from all ef fects is done by the methods outlined in the ASME Boiler and Pressure Vessel Code, Section 111, Article 4. Figure 414.
The maximum stress intensity is the value from figure N-415(A) of this code.
Shown in figure 5.1-16 is a typical penetration and the applied loads.
The stresses from the effects of pipe loads, pressure loads, dead load, and earthquake are calculated and the stress intensity kept oelow S,.
-The stresses from the remaining effects are combined with the above calculated stresses and the stress intensity kept below S,.
s 3..
Thermal Gradient i
The only lines penetrating the containment shell normally having high temperatures are the main steam and feedwater.
The steady state temperature gradients are determined for the case with no cooling with maximum insulation using the Generalized Heat Transfer Program.
The'results indicate no ccoling is necessary (see figure 5.1-17).
Liner Plate There are no design conditions under which the liner plate is relied upon to assist the concrete in maintaining the integrity of the structure even though the liner will at times provide such assistance.
Loads are transmitted to the liner plate through the anchorage system and direct contact with the concrete and vice versa.
Loads may be al;o transmitted by bond and/or friction with the concrete.
These loads cause or are caused by liner strain.
The liner is designed to withstand the predicted strains without leaking.
5.1-70
Possible cracking of concrete is considered and reinforcing steel is provided to control the width and spacing of the cracks.
In addition, the design is made such that total structural deformation remains small during the loading conditions and that any cracking will be orders of magnitude less than that sustained in the repeated attempts to f ail the prestressed concrete from overpressure tests of
'Model 2" (both at General Atomic).
(See " Prestressed Concrete Reactor Vessel, Model 1, #GA 7097, HTGR and Laboratory Staff" and
" Concrete Reattor Vessel, Model 2, #GA 7150, Advance HTGR Staff "
Under test condition, the cylinder wall and the dome will be under net membrane compressive stress.
Therefore, there is only a slight possibility of cracking at the outside f ace of the wall and the dome from thermal gradient present during the test across the thickness of the wall and the dome.
The crack width is ca'culated using following reference 6.
Following is the equation as mentioned in the above reference to calculate the maximum size of the crack:
N max. = 0.115 */A x f, x 10 i n,
where W max.
Maximum crack width A = Area ci concrete surrounding each bar, sq. in.
f, Stress in the bar, psi The maximum crack width is predicted to be 0.0055 in.
The corres-ponding spacing of the crack is predicted to be 10 in.
It is expected that the crack pattern will be two dimensional.
However, because of the higher circumferential prestressing compared 5.1-71 u
to the vertical prestressing in the cylinder wall, the size of the vertical crack is predicted to be smaller than th' horizontal crack.
As described, the structural integrity conseque9ces of concrete cracking are limited by the bonded reinforcing and unbonded tendons provided in accordance with the design criteria.
The effect of concrete cracking on the liner plate is also considered. The anchor spacing and other design criteria are such that the liner will sustain, for example, order of magnitude of strain less than did the liner of Model 1 at General Atomic without tensile failure.
Liner Plate Anchors The liner plate anchors are designed to preclude failure when subjected to the worst possible loading combinations.
The anchors are also designed such that, in the event of a missing or f ailed anchor, the total integrity of the anchorage system would not be jeopardized by the failure of adjacent anchors.
l The following loading conditions are considered in the design of the anchorage system:
1.
Prestress; 2.
Internal Pressure; 3.
Shrinkage and Creep of Concrete; 4.
Thermal Gradient (Normal and Design Basis Accident);
5.
Dead Load; 6.
Earthquake; 7.
Vacuum.
5.1-72
l The following factors are considered in the desijn of the anchorage system:
1.
Initial inward curvature of the liner plate between anchors due to fabrication and erection accuracies; 2.
Variation of anchor spacing; I
3.
Misalignment of liner plate seams; 4.
Variation of plate thickness; 5.
Variation of liner plate material yield stress; 6.
Variation of Poisson's iatto for liner material; 7.
Cracking of concrete in anchor zone; 8.
Variation of the anchor stiffness.
The anchorage system satisfies the following conditions:
1.
The anchor has sufficient strength and ductility so that its energy absorbing capability is sufficient to restrain the maximum force and displacement resulting from the condition where.a panel with initial outward curvature is adjacent to a panel with initial inward curvature 2.
The anchor has sufficient flexural strength to resist the bending moment which would result from Condition 1 C
3.
The anchor has sufficient strength to resist radial pull out force When the liner plate moves inward radially as shown in figure 5,i 9, the sections will develop membrane stress due to the f act that the anchors have moved closer together. Due to initial inward curvature, 5.1-73 1
the sect on between 1 and 4 will deflect inward giving a longer length than adjacent sections and some relaxation of membrane strength will occur. it should be noted here that section 1-4 car.not reach an unstable condition due to the manner in which it is loaded.
The first part of the solution for the liner plate and anchorage system is to calculate the amount of relaxation that occurs in section 1-4, since this value is also the force across Anchor 1 if it is infinitely stiff.
This solution is obtained by solving the general differential equation for beams, including the effect of relaxation or the lengthening of section 1-4.
Figure 5.1-18, Sheet 1, shows the symbols for the forces that result from the first step in the solution.
Using the 'ne 'l shewn in figure 5.1-18, Sheet 2, and evaluating the necessary e
- , constants, the anchor is allowed to displace.
The solution yields a force and displacement at Anchor 1, but the force in Section 1-2 is (N)-K,,g,,,,,5 and Anchor 2 is no longer in 3
force equilibrium.
The model shown in figure 5.1-18, Sheet 2, is used to allow Anchor 2 to displace and then to evaluate the effects on Anchor 1.
The displacement of Anchor 1 is 5, + S', and the force an Anchor 1 is K,(5, + S'3).
Then Anchor 3 is not in force equilibrium and the solution is continued to the next anchor.
After the solution is found for displacing Anchor 2 and Anchor 3, the pattern is established with respect to the effect on Anchor 1 and, by inspection, the solution considering an infinite amount of anchors is obtained in the form of a series solution.
The preceding solution yielded all necessary results.
The most important results are the displacement and force on Anchor 1.
5.1-74
l Various patterns of welos attaching the angle anchors to the liner I
plate are tested for ductility and strength when sub.iec t to a transverse shear load such as AN and are shown in figure 5.1-19.
Using the results from these tests together with data from tests made for the fort St. Vrain PSAR, Amendment No. 2 and Oldbury vessels, a range of possible spring constants are evaluated for the Point Beach liner.
By using the solution previously obtained together with a chosen spring constant, the amount of energy required to be absorbed by the anchor is evaluated.
By dividing the amount of energy that the system will absorb by the most probable maximum energy, the result then yielded the factor of safety.
By considering th; worst possible loading condition which resulted from the listed loading conditions and the conditions stated below, the following results are obtained:
Lase 1 Simulates a plate with a yield stress of 32 ksi and no variation in any other parameters
[ase 11 Simulates a 1.25 increase in yield stress and no variation in any other parameurs Case ill Simulates a 1.25 increase in yield stress, a 1.16 increase in plate thickness, and a 1.08 increase for all other parameters Case IV Simulates a 1.88 increase in yield stress with no variation of any other parameters
[ase V is the same as Case 111 except the anchor spacing is doubled to simulate what happens if an anchor is missing or has failed 5.1-75
_._ J
LINER PLATE cat [VLATIONS - RESULTS 1
factor Nominal Initial of Plate inward Anchor Anchor Safety Thickness Displace-Spacing Spacing Against
[ng fin) ment fin)
L (In)
L.Hnl failure
. 1 0.25 0.125 15 15 37.0 11 0.25 0.125 15 15 19.4 111 0.25 0.125 15 15 9.9 IV 0.25 0.125 15 15 6.28 V
0.25 0.25 30 15 4.25 1
Suou:*t!
In designing for structural bracket loads applied perpendicular to the plane of the liner plate or loads transferred through the thick-ness of the liner plate, the following criteria and methods are used:
a.
The liner plate is thickened to reduce the r "dicted stress level in the plane of the liner plate. The thickened plate with the corresponding thicker weld attaching the bracket to the plate will also reduce the probability of the occurrence of a leak at this location, b.
Unf.cr the application of a real tensile load applied perpendic-ular to the plane of the liner plate, no yielding is to occur in thu perpendicular direction.
By limiting the predicted e
s'. rain to 90% of the minimum guaranteed yield value, this criterion is satisfied.
c.
The anowable stress in the perpendicular direction is calcu-lated using the above allowable predicted strain in the perpen-dicular direction together with the predicted stresses in the plane of the liner plate.
5.1-76
d.
In setting the above criteria, the reduced strength and strain ebility of the material perpendicular to the direction of rolling (in plane of plate) is also considered if the bracket did not penetrate the liner reinforcing plate.
In this case, the major stress is nurmal to the plane of the liner plate. The allowable stresses are reduced to 75% of the stress permitted in (c) above.
e.
The necessary plate characteristics are assured by ultrasonic examination of the rcinforcement olates for lamination defects.
Mi s sile.1 The containment structural design considered the following external missiles:
jlq!D Weicht (1b)
Velocity (foil 4 x 12 plank, 12 ft. long 200 440 Automobile 4,000
/4 The depth of penetration of these missiles is analyzed using " Design of Protection Structures," by A. Amirikian, NAV Docks P-51, Bureau of Yards and Docks, Department of the Navy,1950.
None of the above l
missiles would penetrate the containment, l.
1mplementation of Criteria This section documents the manner in which the design criteria are met by the designer.
Various types of documentation are presented.
figures 5.1-10, 5.1-11, and 5.1-13 illustrate isostress plots and tabula-tions of predicted stresses for the various materials.
The isostress plots of the homogeneous uncracked concrete structure indicate the general stress pattern for the structure as a whole under various loading conditions. More specific documentation is made of the predicted stresses for all materials in the structure.
In these tabulations, the predicted 5.1-77
stress is compared with the allowable to permit an easy comparison and evaluation of the adequacy of the design.
Results of Analysis The isostress plots, figures 5.1-10, 5.1-11, and 5.1-13, show the three principal stresses and the direction of the principal stresses normal to the hoop direction.
The principal stresses are the most significant information about the behavior of the structure under the various condi-tions and are a valuable aid for the final design.
The plots are prepared by a cathode-ray tube plotter.
The data for plotting are taken from the stress output of the finite element computer program of the following design load cases:
D+F 0+F4. 15P D + F + 1.5P + T.
D + F + T.
The above axisymmetric loading conditions are found to be governing in the design since they result in highest stresses at various locations of the structure.
The table of predicted stresses, Table 5.1-1, for various materials has been prepared for the presentation of the combined stresses of the axi-symmetric and nonaxisymmetric loading cases. These stresses are computer analyzed considering cracked concrete sections where applicable, in the manner described in 5.1.2.4, No stresses are shown for the tendons due to the almost constant stress level regardless of loeding condition.
The tabulated stresses may be considered the final results of the analysis and
[
design.
f 5.1-78 l
a
~ _ _ _
The upper stress limit for a linear stress-strain relationship was assumed to be 3000 psi (0.6 fi) for use witn analyses made by the use.f the axisymmetric finite element analyticai method.
(The analyses referred to considered the concrete as uncracked and the analytical model is the entire containment.)
However, the maximum predicted compressive stress was about 2600 psi.
The load combination considered was (D + f + T.) and the location for the predicted stress was near the junction of the base slab and cylinder. Therefore, only the linear portion of the stress curve was used in the analyses that used the entire containment structure as a model.
f The compressive stress and strain level is the highest (nf ter the LOCA when temperature is still relatively high 200*f, and pressure is dropping rapidly) at the inside face of the concrete at the edge of openings and also under the liner plate anchors. Neither concentration is a result of what may be considered a real load.
in the case of an opening, the reat stress is a result of prestress, reduced pressure, and dead load.
Applying stress concentration factors to these loads still keeps the concrete ir, essentially the elastic range. When the strain and resulting stress from the thermal gradient are also multiplied by a stress concentration factor, the total strain and resulting stress will be above the linear stress range as determined by a uniaxial compression test. The relatively high stress level is not of real concern due to the following:
1.
The concrete affected is completely surrounded by either other concrete or the penetration nozzle and liner reinforcing plate. This confinement puts the concrete in triaxial compression and gives it the ability to resist fnrces far in excess of that indicated by a uniaxial compression test.
t 2.
The high state of stress and strain exist at a very local area and really have no effect on the overall containment integrity, j
Howevtr, to be conservative, reinforcing steel was placed in these areas and, also, the penetration nozzle will function as compressive reinforce-ment.
5.1-79 l
l The concrete under the liner plate anchors experiences some limited yielding ta order to get the necessary stress distribution required to resist the liner plate self-relieving loads.
Liner Plate Desian Prpvis &qi The1 1iner plate is anchored as shown in figure 5.1-1 with anchorage in both the longitudinal and hoop direction.
The anchor spacing and welds are designed to preclude failure of an individual anchor.
The load deformation tests, referred to above, indicate that tne alternate stitch fillet weld used to secure the anchor to the liner plate would first fail in the weld and not jeopardize the liner plate leaktight integrity.
Erection and fabrication inaccuracies are controlled by specified toler-ances given in Section 5.6.1.5.
Offsets at liner plate seams are controlled in accordance with ASME Section 111 Code which allows 1/16 in, misalignment for 1/4 in. plate.
The flexural strains due to the moment resulting from the misalignment are added to calculate the total strain in the liner plate.
penetration Details Typical penetration details are shown in figures 5.1-2 and 5.1-3.
Horizontal and vertical bonded reinforcement is provided to help resist membrane and flexural loads at the penetrations. This reinforcement will be located on both the inside and outside face of the concrete.
Stirrups are also used to assist in resisting shear loads.
Local crushing of the concrete due to deflection of the reinforcing or tendons is precluded by the following details.
1.
The surface reinforcements either have a very large radius, such as the hoop bars, concentric with the penetration or are practically straight, having only standard hooks as anchorages where necessary.
i 5.1-80
4 2.
The tendans are bent around penetrations at a Mnimum radius of approximately 20 feet. Maximum tendon force at initial prestress is 850 kips, which result in a bearing stress of abrut '.0 psi on the d
- concrete, 11 is also important to note that the deflected ten. dons are continuous past the oper.ings and are isolated from the local effects of stress concentrations by virtue of being unbonded.
In accordance with the ASME Section !!!, all penetration reinforcing plates and the weldment of the pipe closure to it are shop stress relieved as a unit.
This code requirement and the grouptng of penetrations into large shop assemblies permits a minimum of field welding at penetrations.
Butt welds are used between the penetration sleeve and process piping.
Both flued ends and drilled standard weight pipe caps are used for the closure piece between the sleeves and the pipes. The design, fabrication, inspection, and testing of the containnent penetration head fittings is in accordance with ASME Boiler and Pressure Vessel Code, Section 111, Class B,1968 Edition and all add'nda. Inspection procedures used for all closure welis consisted of Pelu oenetrant and local leak pressure testing at the containment design accident presstre.
Open butt welds without ' backing rings were specified prior to June 1970.
All of these welds were radiographed.
Welds efter June 1970 did net have the requirement for backing rings and radiographic inspection. Consequently, most of the Unit 1 penetration closure welds were radiogriphed and the majority of the Unit 2 closure welds were not.
Prestress Losys.n The following categories and values of prestress losses are considered in the design:
5.1-81
l Tvoe of loss Assumed Value i
Seating of Anchorage None Elastic Shortening of Concrete Im
- 5. x 10' Creep of Concrete 0.27 x 10 In/In/ Psi Shrinkage of Concrete 100x 10 In/1,'/ Psi Relaxation of Prestressing Steel 8% of 0.r5f, 4
li'.' Ksi Frictional Loss K - 0.0003,
,,.156 There is no allowance for the seating of the BBRV anchor since no slippage occurs in the anchor during transfer of the tendon lot irto the structure.
Sample lift-off readings will be taken to confirm that any seating loss is negligible.
The loss of tendon stress due to elastic shortening is based on the strain change in the initial tendon relative to the last tendon stressed.
I A concrete properties study using Point Beach samples was conducted at the University of California.
A similar study conducted on a,early identical concrete mix has indicated a
creep value of 0.125 x 10" In/In/ Psi.
Conversion of this unit creep data to hoop, vertical, and dome stress gives these values of stress loss in tendons:
Hoop - 5.5 Ksi Verthal - 2.8 Ksi Dome - 5.5 Ksi A single creep loss figure of 400 x 10-' in/in at 1500 psi (f ) in the concrete is tied throughout the structure.
Tnis results in a prestress loss of 11.8 ksi in the prestressing steel.
The value used for shrinkage loss represents only that shrinkage that could occur after stressing. Since the concrete is, in general, well aged at the time of stressing, little shrinkage is lef t to c' er and add to prestress loss.
5.1-82
- - - _ _ _ - _ = _ - _ _ _ _ - _ - - - - - _ _ - _ _ - - - _ - - - - _ - - - - _ _ - - - - - - - - - - - - - - - - - - - - - - - - - - - -
l
.Q The value of relaxation -loss is based on information furnished by the tendon system vendor, Inland-Ryerson Construction Products Company.
Frictional loss parameters for unintentional curvature (K) and intentional curvature (p) are based on full-scale friction test "ata.
This data
' indicate actual values of K 0.0003 and p - 0.125 versus the design values of K = 0.0003 and p 0.156.
Assuming that the -jacking stress for the tendons is 0.8 f' or 192,000 psi and using the_ assumed prestress loss parameters, the following tabulation shows the magnitude of the design losses and the final effective prestress at end of 40 years for a typical dome, hoop, and vertical tendon.
Dome Hoop Vertical (Ksi)
(Ksi)
(Ksi)
Jacking Stress 192 192 192 Friction loss 18.5
- 20. 8"'
20.0 Seating Loss 0
0 0
Seating Stress 173.5 171.2 172.0
"' Average of crossing tendons Dome Hoop Vertical (Ksi)
(Ksil (Ksi)
Elastic loss 8.8 9.4 4.1 Creep Loss 11.8 11.8 11.8 Shrinkage _ Loss 3.0 3.0 3.0 Relaxation loss
_12J 12.5 12,5 Final Ef fective Stress'"
137.4 134.5 140.6 This force does net include the effect of pressur.zation which H
increases the prestress force 5.1-83 L
To provide assurance of achievement of the desired level of final effec-
)
tive prestress and that ACI 318-63 requirements are met, a written procedure was prepared for guidance of post-tensioning work.
The procedure p.rovided nominal values for end anchor forces in terms of pressure gage readings for calibrated jack-gage combinations.
force measurements were made at the end anchor, of course, since that is the only practical location for such measurements.
Thg procedure required the measured temporary jacking force, fc 1 single tendon, to approach but not exceed 850 kips (0.8f',).
Thus, the limits cet by ACI 318-63, Paragraph 2606(a)l, and of the prestressing system supplier, were observed. Additionally, benefits were obtained by in-place testing of the tendon to provide final assurance that the force capability exceeded that required by design.
During the increase in force, measurements were required of elongation changes and force changes in order to allow documentation of compliance with ACI 318-63, Paragraph 2621(e).
The jacking force of 0.8f', further provided for a means of equalizing the force in individual wires of a tendon to establish g
compliance with ACI 318-63, Paragraph 2621(b),
The procedures required compliance with ACI 318-63 such that if broken wires resulted from the post-tensioning sequence, compliance with Paragraph 2621(d) was docu-mented.
Each of the above procedures contributed to assurance that the desired level of final effective prestress would be achieved.
The requirements of ACI 318-63, Paragraph 2606(a)2 state that f, should not exceed 0.7f( for " post-tensioning tendons immediately af ter anchcring".
Paragraph 2606(a)2 of ACI 318-63 refers to " tendons" rather than to an individual tendon. Further, the paragraph does not refer to the location to be considered for the determination of f, in the manner, for example, of the " temporary jacking force" referred to in Paragraph 2606(a)l.
Two interpretations uere therefore required.
Both interpretations had to consider the effect of the resultant actions on both the prestressing j
system and structure.
The first interpretation was that the location for measurement of the seating force used in calculating f, was at the end anchor and just 5.1-84 j
l
subsequent to the measurement of the " temporary jacking force" referred to in Paragraph 2606(a)l. The advantages of this location are several. One is that it is a practical one and thus the possibility for achieving valid measurements could be made without the added complexity of additional measuring devices.
Another advantage is that measurements at this locat;on provide assurance that the calculated f, does not anywhere exceed the _ maximum f, (0.8f,) to which that tendon has been subjected.
One case considered was that of anchoring each tendon at a measured force of 850 kips (0.8f',).
Although there was no apparent detrimental-effect to the prestressing system or structure, insertion of shims would be almost impossible, further, it was concluded that this case would not establish compliance with ACl 318-63.
The case adopted was to seat each tendon with a measured " pressure" reading for the jack, at " lift-off" of the end anchor, of 775 kips
-(between 0.72 and 0.73 f',).
This procedure had several advantages.
One advantage was that the force on the containment and the tendon was within the bounds of those for which it had been tested and resulted in no known detrimental ef fects.
The second advantage was that the stressing procedure was simplified since the stressing crews did not have to L
accommodate a large number of different anchoring force requirements. The l
third advantage was that, at the compiction of stressing the last tendon, j.
the expected losses were such that the average f, at the end anchors of the tendons would be less than 0.7f',,
thus establishing compliance with l
ACI 318-63, Paragraph 2606_(a)1 and 2.
The fourth advantage was that the percentage loss of prestressing force was less than would be the case if the tendons were anchored-in such a manner that the calculated value of f, nowhere exceeded 0.7f',.
L The latter advantage deserves special mention since it plays a strong role in assuring that the final effective prestress equalled or exceeded the desired value.
For example, -if the f, at anchorage of the _ tendons were 0.lf',,
the final effective prestress, neglecting relaxation for the moment, would be about 86% of the initial prestress.
Clearly, the assurance (that the concrete creep and shrinkage losses have been properly 5.1-85
accounted - for) increases as the f, for the anchored tendons and tendon increases.
However, this design was committed ta meeting the ACI 318-63 requirement and the anchorage force for the tendons was kept at or below 0,7f', in accordance with the interpretation described.
Miscellaneous Considerations In various cases, it is the designer's decision to provide structural c
adequacy in excess of design criteria submitted in the PSAR. Those cases are as follows:
1.
Section b.1.2.2 requires a minimum of 0.15% bonded reinforcing steel in two perpendicular directions on the exterior faces of the wall and dome for proper crack control.
Due to the cold weather exposure, a y minimum of approximately 0.25% is provided.
2.
Section 5.1.2.2 requires a minimum of 0.15% at cross section area bonded steel reinforcing (as stated above) for any location. At the y-base of the cylinder, the controlling design case requires 0.25%
vertical reinforcing. As a result of pursuing the recommendation of the NRC Staff to further investigate current research on shear in concrete, several steps were taken:
c.
The work of Dr. Alan H. Mattock was reviewed and he was retained as a consultant on the implementation of the research being conducted under his direction. The criteria was updated in accordance with his recommendation.
b.
In addition to reviewing Dr. Mattock's work, the firm of T. Y. Lin, Kulka, Yang and Associate was consulted to review the detailed design of the cylinder to slab connection.
Pursuant to their recommendation, j
approximately 0.5% reinforcing was used-rather than the 0.25% reinforcing indicated by the detailed design analysis for the vertical will dowels. This increase insures that there was sufficient flexural steel to place the section within the lower limits of Mattock's test data i
5.1-86 l
l (approximately 0.3%) to prevent flexural cracking from adversely affecting the shear capability of the section.
5.1.2.5 Ouality Control General Quality of materials and construction of structures and systems is assured by a continuous program of quality control and inspection.
Shop inspection of all critical materials and components was undertaken by the Bechtel Power Corporation.
Inspection of construction work was carried out by qualified field supervisory personnel and inspectors.
Project design personnel made frequent visits to the job site to coordinate the construction with the design.
Special measures beyond those normally employed in construction were instituted for the quality control of Class I structures.
Organization Shop inspection of all materials and components for Class I structures was conducted by the Bechtel Power Corporation to insure compliance with applicable specificaticnb drawings, codes, and standards.
A formal quality assurar,;e organization and reporting system was employed to assure that Class I structures were constructed in accordance with applicable specifications and drawings.
A Nuclear Quality Assurance Manual was prepared and distributed to the field to implement this system.
This manual established standard inspection and reporting procedures for Bechtel field personnel.
Field inspection was carried out by the permanent field engineerin. staff located at the job site under the surveillance of a Quality Assurance Engineer who was also located at the job site and permanently assigned to this position.
5.1-87
The Quality Assurance Engineer reported directly to the Project Engineer
)
in the Some office and was completely independent of the construction organization.
The number of field inspectors assigned Lt any one time depended upon the inspection workload.
The inspectors were graduate engineers or had equivalent experience.
They were thoroughly familiar with the design specifications, design drawings, applicable codes, and sampling and testing procedures pertaining to their areas of responsibility.
Independent testing laboratories were used for quality control, testing and reporting of the concrete materials, and for user's testing of reinforcing steel, liner plate, and tendon material.
The independent testing laboratories reported in the Quality Assurance Engineer.
Quality Control Testing Structural Materials
)
1.
Backfill Compacted fill was placed under strict continuous control of field inspectors. Soil Testing Services, who operate a testing laboratory in Green Bay, performed the following tests to verify compaction:
Moisture Content of Fill Material Moisture Density Relations - ASTM D-1557 Field Density Tests Examination of exposed excavation and fill material was made by the inspectors.
2.
Concrete Materials 5.1-88
__-