ML20032A669

Forwards Info Re Structural Stresses Induced by Differential Settlement of Diesel Generator Bldg & Estimates of Relative Density of Granular Fill Matls,Diesel Generator Bldg, in Response to NRC

ML20032A669
Person / Time
Site:	Midland
Issue date:	10/21/1981
From:	Jackie Cook CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:	Harold Denton Office of Nuclear Reactor Regulation
Shared Package
ML20032A670	List: ML20032A669 ML20032A671
References
NUDOCS 8111020144
Download: ML20032A669 (50)

Text

!

i Consumem Power Company 0"'"/,u*"*.,

j,,,o.,,,,,,4.,

e,J Co.structi..

General Offices: 1945 West Pernell Road, Jackson, MI 49201 * (517) 78& O453 October 21, 1981 A

C prn 3 e g gW

.N $

b Ll)

DCT3 0199;w [

yy/ m::e y

@D' liarold R Denton, Director Office of Nuclear Reactor Regulation US Nuclear Regulatory Commission Washington, DC 20555 MIDLAND PROJECT MIDLAND DOCKET NOS 50-329, 50-330 REMAINING NRC SOILS-RELATED CONCERNS FOR DIESEL GENERATOR BUILDING FILE 0485.16, B3.0.3 SERIAL 14316 ENCLOSURES:

(1) STRUCTURAL STRESSES INDUCED BY THE DIFFERENTIAL SETTLEMENT OF THE DIESEL GENERATOR BUILDING (2) SUBGRADE MODULUS AND SPRING CONSTANT VALUES FOR DIESEL GENERATOR BUILDING STRUCTURAL ANALYSIS (3) BEARING CAPACITY EVALUATION OF DIESEL GENERATOR BUILDING FOUNDATION

/ e) LONG-TERM MONITORING OF SETTLEMENT FOR DIESEL GENERATOR BUILDING (5) RELATIVE DENSITY AND SHAKEDOWN SETTLEMENT OF SAND UNDER THE DIESEL GENERATOR BUILDING (6) ESTIMATES OF RELATIVE DENSITY OF GRANULAR FILL MATERIALS, DIESEL GENERATOR BUILDING, MIDLAND PLANT (7) REVIEW AND CONTROL OF FACILITY CHANGES TO THE DIESEL GENERATOR BUILDING (8) DIESEL GENERATOR BUILDING BEARING PRESSURE DUE TO EQUIPMENT AND COMMODITIES On September 24, 1981, a request for additional information relating to the diesel generator building was made by the Staff in a telephone discussion. We are responding to this request by forwarding the enclosures itemized above.

Each enclosure addresses one of the Staff concerns transmitted to us in the September 24, 1981 telecommunication.

sQ

.8111020/oc1081-0462a10,jf%,,

_0 A

t 2

i We believe the enclosed information adequately responds to the request and individual concerns identified for us by the Staff. The discussions and data contained in the enclosures to this correspondence lend further support to our conclusion that the design of the diesel generator building combined with the remedial actions are adequate.and appropriate for this structure.

'V\\ Go tun r- <[3'I/[3 (2 cyc t__

i.

t JWC/RLT/dsb CC Atomic Safety and Licensing Appeal Board, w/o CBechhoefer, ASLB, w/o MMCherry, Esq, w/o FPCowan, ASLB, w/o RJCook, Midland Resident Inspector, w/o RSDecker, ASLB, w/o JHarbour, ASLB, w/o DSHood, NRC, w/a (2)

DFJudd, B&W, w/o JDKane, NRC, w/a FJKelley, Esq, w/o RELandsman, NRC Region III, w/a WHMarshall, Esq, w/o JPMatra, Naval Surface Weapons Centre, w/a W0tto, Army Corps of Engineers, w/a WDPaton, Esq, w/o i

FRinaldi, NRC, w/a HSingh, Army Corps of Engineers, w/a BStamiris, w/o i

l l

i l

l l

l i

t

[

l

[

oc1081-0462a100 l

I

+

e

.g

/

, f a

/

r f

J d

TECHN*lCAL REPORT STRUCTURAL STRESSES INDUCED BY DIFFERENTIAL SETTLEMENT OF THE 4

DIESEL GENERATOR BUILDING e'

-s y

I

/

s' 0

r, di 4

CONSUMERS POWER COMPANY MIDLAND PLANT UNITS 1 AND 2 1

S

/, "

' /

9 I

.-m.,4

=.-

.... -,.,.,., ~.... -

,,, _. ~,.. _,.....,.., _,,,,,,,,,,.,,. _, -.,, ~... -,. _ _ _ ~.,,..,

..:y..

~

a MIDLAND PLANT UNITS 1 AND 2 7

TECHNICAL REPORT CTURAL STRESSES INDUCED BY

/

DIFFERENTIAL SETTLEMENT OF THE DIESEL GENERATOR BUILDING TABLE OF CONTENTS Page 1.0 STRUCTURAL REANALYSIS 1

1.1 STRUCTURAL ACCEPTANCE CRITERIA 1

1.1.1 Load Cases 2

1.1.2 Load Combinations 2

1.1.3 Allowable Material Limits 4

2.0 DIESEL GENERATOR BUILDING ANALYTICAL MODEL 4

2.1 APPLICATION OF LOADS TO THE BUILDING MODEL 4

-2.1.1 Dead Loads 5

g

~

2,1. 2' Settlement Loads 5

e 2.1.3 /

Live Loads 6

2[k.4' Wi?.d Tmads 7

f 2.1.5 Tornado Loads 7

2.1.6 Seismic Loads 8

2.1.7 Thermal Loads 9

3.0 STRUCTURAL ADEQUACY COMPUTATIONS 10

4.0 CONCLUSION

S 11 REFERENCES 12 APPENDIX l

A OPTCON' 9

a TABLES Q>

f I-l Loads and Load Combinations for Concrete Structures Other than the Containment Building From the FSAR

- /

and Question 15 of Responses to NRC Requests Regarding Plant Fill

~ )/ '

/'

il s's-

~

Midland Plcnt Unitz 1 a 2

Structural Stresses Induced by Differential Settlemant in the Diesel Generator Building Table of Contents (Continued)

I-2 Loads and Load Combinations for Compar_ son Analysis Requested in Question 26 of NRC Requests R.egarding Plant Fill I-3 Soil Properties Used in the Scismic Analysis I-4 Steel and Concrete Stresses for Diesel Generator Building Structural Memo ~ers (According to FSAR and the Responses to NRC Requests Regarding Plant Fill, Question 15)

FIGURES I-l Didsel Generator Building Dynamic Lumped Mass Model for Seismic Analysis I-2 Diesel Generai 1r Building Finite Element Model I-3 Es'timated Secondary Compression Settlements from August 15, 1979, to December 31, 2005, Assuming Sur-

• charge Remains I-4.

Comparison of Estimated Secondary Compression Values with' Settlement' Values Resulting from a Finite Element Analysis of the Diesel Generator Building I-5, Basis for Calculation of Equivalent Shear Wave Velocity iii

MIDLAND PLANT UNITS 1 AND 2 TECHNICAL REPORT STRUCTURAL STRESSES INDUCED BY DIFFERENTIAL SETTLI 'ENT OF THE DIESEL GENERATOk 'UILDING l

1.0 STRUCTURAL REANALYSIS To account for the effect of the observed and predicted settlement on the diesel generator building, a structural reanalysis was performed.

This reanalysis proceeded.by defining the acceptance criteria for the structure (see subsection 1.1).

These acceptance criteria differ frou the acceptance criteria used in the original design and analysis of the structure and set forth in the Final Safety Analysis Report (FSAR) only in the additf.on of four load combinations that include the effect of settlemant.

These additional load combinations are described in Subsection 1.1.2.

'Jo investigate the effects of the load combinations on the structure, the structural reanalysis uses two different mathematical models of the diesel generator building:

a dynamic, lumped mass model and a static, finite element model.

The dynamic, lumped mass model (described in Subsection 2.1.6 and illustrated in Figure I-1) is used to generate seismic forces in the building, given the input ground motion from the operating 4

Ibasis earthquake and safe shutdown earthquake (SSE)- specified

.in the FSAR.

The finite element model illustrated in Figure I-2 is a more complex mathematical model that reduces the diesel generator

~

building to an interrelated system of plate, beam, and boundary elements representing the walls, slabs, foundation, and soil.

The finite element model is used to assess the effect on individual elements of various load combinations applied to the structure as a whole.

(These load combinations include seismic forces generated with the dynamic, lumped mass model.)

The finLte element model.thereby allows the identification of those sections of the diesel generator building that will experience the greatest forces due to the postulated load combinations.

The allowable stress is then calculated and compared to the actual stress level in these sections based on the forces derived from the finite element model.

This comparison shows that even those i

sections of the building experiencing the highest forces meet the accejtance criteria.

1.1 STRUCTURAL ACCEPTANCE CRITERIA Because of the settlement problem, a structural reanalysis of the diesel generator building was performed in accordance with the structural acceptance criteric which are consistent with FSAR Subsection 3.8.6.3, with settlement effects included as outlined in the response to NRC Requests Regarding Plant Fill, Question 15 (Revision 3, September 1979).

1

Midland Pltnt Units 1 cnd 2 Structural Strescos Inducsd by Difforcntial Settlement in the Diesel Generator Building 1.1.1 Load Cases I

The following loads are considered in the reanalysis:

a.

Dead loads (D) b.

Effects of settlement combined with creep, shrinkage, and temperature (T) 1 i

c.

Live loads (L) d.

Wind loads (W) e.

Tornado loads (W')

f.

OBE loads (E) g.

SSE loads (E')

h.

Thermal effects (To)

Thermal effects appear twice in this list (Items b and h).

For load combinations committed to in the response to Question 15 of the NRC Requests Regarding Plant Fill, thermal effects are i

contained within the settlement effects term, T.

For load combinations committed to in FSAR Subsection 3.8.6.3, thermal effects are contained in the thermal term, To (Refer to Table I-1).

All other load cases appearing in the load combinations for Seismic Category I structures listed in' FSAR Subsection 3.8.6.3 (e.g.,

rupture of pipe lines) do not occur in the diesel generator building and are not addressed.

1.1.2 Load Combinations The load combinations employed for the original analysis and design of the diesel generator building are provided in FSAR Subsection 3.8.6.3.

The original FSAR load combinations did not contain a settlement effects term (T).

For the structural i

reanalysis performed in response to Question 15 of the NRC L

Requests Regarding Plant Fill (September 1979), four additional i

load combinations were established and committed to be considered.

These additional combinations

sider the effects of differential settlement in combination wl long-term i

operating conditions and with either wind load or OBE.

Table I-l provides the load combinations listed in FSAR Subsection 3.8.6.3 and the four additional load combinations.

These load combinations comprise the acceptance criteria for the diesel generator building and are hereinafter referred to as the Midland

=

acceptance criteria.

i 2

l i

_,_.,,,_...m___...,_,..

,,..., _., ~,. _, _, _.,..

.,,__,_,__.-._..____,..____.__.,...,__...._c.,..

Midland Plcnt Units 1 cnd 2 Structurcl Strc:ges Inducsd by Differential Settlement in the Diesel Generator Building By requiring combination of differential settlement with wind loads and OBE, the Midland acceptance criteria are more stringent than the requirements of American Concrete Institute (ACI) 318.

ACI 318 only requires combining the effects of differential settlement with the dead loads and live loads.

The Midland acceptance criteria are less stringent than ACI 349, because ACI 349 (as supplemented by Regulatory Guide 1.142) includes load combinations that combine the effects of differential settlement with extreme loads such as tornados and SSEs.

In the response to Question 26 of NRC Requests Regarding Plant Fill, a commitment was made to do a separate structural reanalysis of the diesel generator building in accordance with ACI 349, as supplemented by Regulatory Guide 1.142, for comparative purposes only.

Table I-2 provides the load combinations of ACI 349 as supplemented by Regulatory Guide 1.142.

It is unnecessary to use all Table I-1 load combinations in the structural reanalysis.

A number of combinations can be eliminated from the analysis after comparison with more severe loads or load equations.

For example, Equations 6 and 10 from Table I-1 are:

a.

U = 1.25 (D + L + H + E) + 1.0T (6) o b.

U = 1.4 (D + L + E) + 1.0To + 1.25H (10)

Because there are no significant forces on the structure due to thermal expansion of pipes (H ),

these two expressions can be o

rewritten in simpler forms:

U = 1.25 (D + L + E) + 1.0To (6) a.

b.

U = 1.4 (D + L + E) + 1.0T (10)

,o The second expression is more critical than the first.

Therefore, Equation 10 is used in the analysis and is considered to envelop the lower force components resulting from an analysis using Equation 6.

Utilizing this approach with the entire set of load cvmbinations eliminates the less critical equations and condenses the list to nine load combinations.

I l

l Table I-1 Load Combinations Equation No.

t a.

1.05D + 1.28L + 1.05T (1) b.

1.4D + 1.4T (2) c.

1.0D + 1.0L + 1.0W + 1.0T (3) o d.

1.0D + 1.0L + 1.0E + 1.0T (4) o 3

l L

Midland Plant Unita 1 and 2 Structural Strasses Inducad by Differential Settlement in the Diesel Generator Building e.

1.4D + 1.7L (5) f.

1.25 (D + L + W) + 1.0To (7) g.

1.4 (D + L + E) + 1.0T (10) o

{

h.

1.0 (D + L + E') + 1.0To (15) 1.

1.0 (D + L + W') + 1.0T (18) o 1.1.3 Allowable Material Limits In accordance with regulatory requirements and the i

recommendations of the American Concrete Institute.(ACI 318 and ACI 349), the maximum rebar tensile stress allowed in the diesel generator building rebar equals 0.90 f (where f equals yield y

y stress) for computation of section capacities.

Because the diesel generator buildirg rebar has an f value of 60 kai, the y

maximum allowable tensile rebar stress due to flexural and axial loads is 54.0 ksi.

In similar fashion, the ultimate compressive strength of concrete is based on a strain of 0.003 in./in.

Rebar stress values subsequently calculated for critical, reinforced concrete sections of the diesel generator building were based on this maximum allowable rebar stress value (54 ksi)' and a maximum

~

allowable concrete strain level of 0.003.

2.0 DIESEL GENERATOR BUILDING ANALYTICAL MODEL The$ structural reanalysis of the diesel generator building uses a finite element model.

The required load combinations were i

applied to this model and the resulting forces were investigated for compliance with the structural acceptance criteria.

The diesel generator building was modeled as an assemblage of plate, beam, and boundary elements.

The structure is defined by a set of 853 nodal points and 1,294 elements.

Of these elements, 901 are plate elements representing walls and slabs, 141 are beam elements, and 252 are boundary elements (translational springs, in both the vertical and horizontal directions) representing varying soil pressures.

Certain items, such as steel platforms and lightly reinforced interior secondary structural walls, have not 1

been included in the model for the reasons listed in subsequent sections.

Figure I-2 illustrates an isometric view of the finite element model.

I 2.1 APPLICATION OF LOADS TO THE BUILDING MODEL The following loads have been applied to the model in the manner i

noted.

t

]

I i

4 I

.~.,m_..--..__..

___...._,__.__.-.._.~,_.<...~_______._._,,_._m.

_ -. _ _ _.,. _ _. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _. -. _ _ _ _. -. - - - _ _ ~.

Midltnd Pltnt Units 1 and 2 Structural Strecces Inducsd by Differential Settlement in the Diesel Generator Building 2.1.1 Dead Loads The dead load of the structure was simulated by specifying a mass acceleration value equaling that of gravity (32.2 ft/s2 ),

Secondary structural walls and platforms were not included in the model because their contribution to the gross weight of the structure is minimal (less than an estimated 5 percent) relative to the sum of the other loads considered.

Their exclusion does not significantly affect the magnitude or distribution of stresses.

The louvers on both the north wall and routh wall, along with the doors on the north and south walls of the building, were moceled simply as penetrations, with dimensions equivalent to those of the doors and louvers.

This is acceptable because the doors and louvers contribute insignificant 1y to the building stiffness and total building weight.

The diesel generator pedestals and the ground floor slabs were omitted from the finite element model because they were not constructed monolithically with the remainder of the structure.

Consequently, they do not add stiffness to the structure.

2.1.2 Settlement Loads The civil engineering group modeled settlement effects into the structure by representing varying soil conditions as boundary elements comprised of translational (vertical and horizontal) springs.

At 84 locations along the building footing, a set of various spring values (one vertical spring and at least one horizontal spring) was applied to represent the nonhomogenous nature of soil conditions existing beneath the diesel generator building.

Spring values were developed for two general cases:

those springs calculated for long-term loading and those springs calculated for short-term loading, e.g.,

tornados and earthquakes.

For long-term loading, a set of springs was calculated for the determination of structural stresses caused by

(

the settlement of the diesel generator building after 40 years.

1 These springs were calculated at each nodal point along the foundation by dividing the total load represented at the selected point by the predicted settlement at that point, so that the spring constant was expressed in terms of force / unit displacement.

The estimated secondary compression settlement values from j

August 15, 1979, to December 31, 2025, are shown in Figure I-3 and are explained in Dr. Peck's testimony (Figure I-3 is the sum I

of settlement from August 15, 1979, to December 31, 1981, and I

from December 31, 1981, to December 31, 2025, as shown in l

Figures 27-12 and Figure 27-13, respectively, of the Responses to l

NRC Requests Regarding Plant Fill.)

These estimates are based on the conservative assumption that the surcharge remains in place i

5

Midland Plcnt Units 1 cnd 2 Structural Stressac Inductd by Differential Settlement in the Diesel Generator Building over the 40-year life of the plant, thus exceeding actual settlement predictions.

Figure I-4 compares these settlement values with those settlement values resulting from the finite element analysis of the diesel generator building model.

The comparison shows a closte correlation between values resulting from the finite element model and estimated settlement values i

l generated by Dr. Peck an$ Bechtel soil engineers.

Because thc estimates of the soils engineers are based on the conservative assumption of the surchnrge remaining in place over the 40-year life of the plant, the model overestimates the settlement loads on the structure considered in the structural reanalysis and is therefore conservative.

Figure I-4 also indicates the settlement and differential settlement occurring in the building subsequent to August 1979 (when the surcharge material was removed).

As Figure I-4 shows, the settlement and differential settlement which has occurred since the removal of the surcharge are very small compared with the settlement and differential settlement conservatively estimated for the purpose of the structural reanalysis.

The other set of springs was developed for short-term loading, in which it was assumed that the structural movement was small enough to assume the soil was linearly elastic.

The modulus of elasticity was estimated using the results of laboratory and field investigations.

Springs were developed for the vertical and horizontal modes.

These springs were calculated by determining the amount of force required to produce a unit displacement in the direction indicated by the particular mode.

The footings of the diesel generator building were assumed to be resting on a large mass of elastic soil for the vertical mode and embedded within the mass of soil for the horizontal mode.

The settlement due to seismic shakedown was also identified as a possible occurrence during a seismic event.

The maximum dif ferential settlement due to seismic shakedown, as stated in H,

Question 15 of the NRC Requests Regarding Plant Fill is approximately one-half inch.

The effects of seismic shakedown settlement will act to reduce the effects of differential l

settlement and tor this reason was not the governing case in the structural reanalysis of the diesel generator building.

2.1.3 Live Loads I

Live loads were applied to the modeled structure by applying pressure loads on the plate elements which represent the floor slab at el 664'-0" and the roof at el 680'-0".

During the plant life, a maximum live load of 100 psf is predicted to occur on the roof slab, whereas for the floor at el 664'-0", a maximum live l

load of 250 psf is postulated.

One hundred percent of the live load was used in the design of individual structural members, 6

L

Midland Plant Units 1 and 2 Structural Stresses Induced by Differential Settlement in the Diesel Generator Building such as floor slab at el 664'-0" and roof slab at el 680'-0".

For overall building response, however, the live loads considered were limited to 25 percent of the above maximum loads.

This 25 percent value represents the live load expected to be present when the plant is in operation, i.e.,

100 percent of the live load will not act simultaneously on every square foot of the floor space.

2.1.4 Wind Loads Loads resulting from the design wind (100-year recurrence with a velocity of 85 mph) were applied to the modeled structure as a pressure load on the plate elements that represent the exposed walls.

Wind loads on the roof and south wall hatch covers were determined assuming the hatch covers were in place.

These loads were then distributed to the nodal points which define the perimeter of the respective hatches.

2.1.5 Tornado Loads As specified in BC-TOP-3-A (Reference 1), various combinations of velocity wind pressure, differential pressure, and local pressures were applied to the modeled structure.

The maximum wind velocity of the tornado was 360 1ph.

The original structural analysis performed in accordance with the FSAR considered various tornado-generated missiles.

The analysis j

considered missiles equivalent to a 4" by 12" by 12' wooden plank (108 pounds) traveling end-on at 300 mph at any height; a 4,000 pound automobile with a velocity of 72 mph no higher than 30 feet above the ground with a contact area of 20 square feet; a 1-inch diameter, 3-foot long, 8-pound steel bar traveling at 216 mph at any height in any direction, and a 35-foot long utility pole, 13-1/2 inches in diameter, weighing 1,490 pounds, traveling at 144 mph, and striking the structure not more than 30 feet above the ground.

For tornado-generated missile loads, the structure was allowed to locally exceed the yield strain.

The results of the original tornado-generated missile load analysis showed the diesel generator building was acceptable.

Results of missile impact tests conducted over the last 6 years indicate that reinforced concrete walls, thinner than the exterior walls of the diesel generator building, have a considerable margin against local damage.

The tests indicate that a wall thickness of 12 inches would sufficiently preclude unacceptable local damage (spalling) from these missiles.

(The thinnest exterior wall of the diesel generator building is 30 inches thick.)

7

l Midland Plcnt Unita 1 and 2 Structural Strecaco Induced by Differential Settlemant in the Diesel Generator Building 2.1.6 Seismic Loads The seismic r,ponse ot a structure depends on the stiffness properties au mass of the structure, the input seismic motion at the structure _acation, and the soil properties of the foundation medium.

Of these parameters, only soil properties are affected by insufficient compaction of backfill.

The following paragraphs describe how the effects of insufficient compaction and eventual surcharging were accounted for in the revised diesel generator building seismic analysis.

The analytical models used for the original seismic analysis and for the seismic reanalyses described in this report are one-dimensional, stick-type, lumped mass models using beam elements to represent the structural stiffness and impedence functions of the foundation medium (see Figure I-1).

The models were analyzed by the modal superposition method, a conventional method used in dynamic analysis.

Design responses are calculated by modal superposition in conjunction with the site ground spectra.

The site ground spectra are those associated with ground acceleration set forth in FSAR Section 3.7 and approved by the NRC at the construction permit stage.

The floor response spectra in the building are calculated by the modal superposition method using the design time history.

The design time history is a modification of the N21E (north, 21' east) ground motion component recorded during the July 21, 1952, earthquake at Taft, California.

The effect of soil-structural interaction is accounted for by coupling the structural model with the foundation media.

The foundation media are represented by impedance functions which represent the equivalent spring stif fness and radiation damping coefficients as specified in BC-TOP-4-A (Reference 2).

The structural portion of the lumped mass model was not revised in the new dynamic analysis.

The difference in the new model was confined to the treatment of the soil-structural interface.

The revised analysis developed the impedance functions based on the building's foundation dimensions and the modification in the soil properties described below.

_n addition, the weight of the soil and the concrete pedestals and diesel generator pedestals within the building were included in this revised model.

The original (presettlement) diesel generator building seismic analysis was based on the underlying till material, which has a shear wave velocity value of 1,359 ft/s (see Table I-2).

This value was not adjusted for the 30 feet of plant fill between the l

till and building foundation elevation.

The seismic reanalysis I

accounted for the soil properties of the fill by averaging the low strain shear wave velocity of the fill and underlying till (Figure I-5) over a depth or 75 feet, which is the smallest I

I B

MidlEnd Plant Unite 1 cnd 2 Structural Stroscos Inductd by Differential Settlement in the Diesel Generator Building dimension of the building.

This resulted in the value of 796 ft/s, which was used in the seismic reanalysis.

However, the effect of decreasing shear wave velocity to.a lower bound estimate of 500 ft/s was also analyzed.

Both the low strain shear wave velocity value of 796 ft/s and the lower bound shear wave velocity value of 500 ft/s were supplied by soil consultants.

The floor spectra at all elevations ot the diesel generator building were generated using a shear wave velccity value of 796 ft/s.

The resulting floor response spectra were combined in an enveloping fashion with the spectra developed in the original analysis which used a shear wave velocity value of 1,359 ft/s.

The floor response spectra were further broadened to account for a lower bound shear wave velocity of 500 ft/s.

Thus, conservative floor response spectra were generated.

The results of the seismic reanalysis indicated that the seismic forces at all elevations of the diesel generator building were somewhat higher than the forces determined in the original analysis.

This increased seismic load was conservatively simulated by applying the maximum structural acceleration occurring in the dynamic model to the finite element model in north-south, east-west, and vertical directions.

The combined effect of the three directional responses was assessed using the square-root-of-the-sum-of-the-squares method recommended in NRC Regulatory Guide 1.92.

The ability of the st ucture to withstand these increased seismic forces in combinatio-with the other loads is described in Section 3.0.

2.1.7 Thermal Loads Thermal effects were included in the analysis as a linear variation of temperature across the thickness of an element.

The thermal effect due to linear variation of temperature across the thickness of an element (also called gradient) results in bending moments being applied to the element.

In general, the temperature gradient which is of most concern for the diesel generator building is that anticipated to occur in the winter.

In accordance with the Handbook of Concrete Engineering (Reference 3) and FSAR meteorological data, the equivalent steady-state exterior winter temperature of 14.6F was calculated.

The corresponding maximum interior ambient air temperature was 75F.

For information on how thermal effects were applied to the model, see Section 3.0.

9

Midland Plant Units 1 cnd 2 Structurcl Stresses Inducad by Differential Settlement in the Diesel Generator Building 3.0 STRUCTURAL ADEQUACY COMPUTATIONS The' computations necessary to verify structural adequacy were performed using a computer analysis program (OPTCON) capable of analyzing reinforced concrete sections.

This reinforced concrete analysis program models a portion of the diesel generator l

building and analyzes it for forces that resulted from the BSAP finite element model analysis.

Refer to Appendix A for additional information concerning OPTCON.

To determine the structural adequacy of the diesel generator building, the modeled structure was partitioned into structural' categories (i.e., north wall, center wall, roof, etc).

Critical elements from each category were then selected for further investigation based on.their axial force, moment, and in-plane shear force.

Using OPTCON, rebar stress values cere then calculated in these critical elements to verify'that the allowable rebar tensile stress value was not exceeded.

To facilitate-the calculation process, a computer program was specifically written for selecting critical elements that would r

undergo OPTCON investigation.

This program was written so that its selection of critical elements.was based on a comparison of the axial force, bending moment, and in-plane shear force of each separate element within a structural category with all other elements of the same structural category.

Once these critical elements were selected, a thermal gradient was assigned to.cach element based on the location of that element within the building.

The' gradient is assigned on a temperature basis, and is converted by OPTCON into a thermal moment.

Based upon the procedure discussad above, all structural i

categories of the diesel generator building were investigated and found to meet the structural acceptance criteria.

Table I-4 shows.the results of the analysis.

The left-hand column of Table I-4 shows-the element with the highest rebar stress value for each structural category.

The second column shows the load combination which produces the highest stress.

In other words, this is the load combination which is critical for this category.

The next three columns show the axial, flexural, and in-plane shear force calculated by BSAP for this element's critical load combination.

The sixth column presents the rebar stress value i

computed by-OPTCON for each critical element within each structural category.

The highest rebar stress value (reflecting the combined effects of flexural, axial, and in-plane shear l-loads) exists in the south wall where the rebar stress value is 42.5 ksi.

The last three columns compare maximum separate force component allowables in all structural categories (axial, flexural, and in-plane) against the corresponding critical loads genersted by BSAP.

This comparison of separate force components 10

[.

i

Midlcnd Plent Unite 1 and 2 Structural Strecsas Induced by Differential Settlement in the Diesel Generator Building is provided for information only.

The interactive method used by OPTCON to calculite actual rebar stress values more accurately depicts how close an element is to the maximum allowaole stress value of 54 ksi as it considers the combined effect af flexural, axial, and in-plane loads.

The final structural reanalysis of the diesel generator building showed that the critical load combinations (Table I-1) are those which include either the tornado load case (W') or the SSE load case (E), specifically:

a.

1.0D + 1.0L + 1.0W' + 1.0T, (18) b.

1.0D + 1.0L + 1.0E' + 1.0T (10)

In approximately 70 percent of the diesel generator building, the c

tornado load combinations produce the stress levels.

4.0 CONCLUSION

S The diesel generator building is a massive, reinfor ced concrete structure with extensive reserve strength.

The structural reanalysis performed on the diesel generator building verifies that the integrity of the structure will not be violated even under the most critical load combinations.

Based on the analysis performed, it can be stated that the settlement has had minimal effect on the structure, and there is reasonable assurance that the diesel generator building will safely perform its intended function over the operating life of the Midland plant.

n 11

Midland Plant Units 1 and 2 Structural Stracroc Inducsd by Differential Sattlemsnt in the Diesel Generator Building REFERENCES 1.

Bechtel Power Corporation, Tornado and Extreme Wind Design Criteria for Nuclear Power Plants, Revision 3, August 1974 (BC-TOP-3-A) 2.

Bechtel Power Corporation, Seismic Analyses of Structures and Equipment for Nuclear Power Plants, Revision 3, November 1974 (BC-TOP-4-A) 3.

M.

Fintel, Handbook of Concrete Engineering, Van Nostrand Reinhold Company, Fsptember 1974 I

l l

l l

12 I.

Midicnd Plcnt Unita 1 cnd 2 Structural Strascas Induced by j

Differential Settlemsnt in the Diecal Generator Building TABLE I-1 LOADS AND LOAD COMBINATIONS FOR CONCRETE STRUCTURES OTHER THAN THE CONTAINMENT BUILDING FROM THE FSAR AND QUESTION 15 OF RESPONSES TO NRC REQUESTS REGARDING PLANT FILL Responses to NRC Requests Regarding Plant Fill, Question 15 a.

Service Load Condition U = 1.05b 1.28L + 1.05T (1)

U = 1.4D + 1.4T (2) b.

Severe Environmental Condition U = 1.0D + 1.0L + 1.0W + 1.0T (3)

[

U = 1.0D + 1.0L + 1.0E + 1.0T (4)

FSAR Subsection 3.8.6.3 a.

Normal Load Condition U = 1.4D + 1.7L (5) b.

Severe Environmental Condition U = 1.25 (D + L + Ho + E) + 1.0T (6) o U = 1.25 (D + L + Ho + W) + 1.0To (7)

U = 0.9D + 1.25 (Ho + E) + 1.0To (8) l U = 0.9D + 1.25 (Ho + W) + 1.0To (9) i I

c.

Shear Walls and Moment Resisting Frames U = 1.4 (D + L + E) + 1.0T + 1.25H (10) o o

l U = 0.9D + 1.25E + 1.0T + 1.25H (11) o o

d.

Structural elements carrying mainly earthquake forces, such as equipment supports j

U = 1.0D + 1.0L + 1.8E + 1.0To + 1.25Ho (12) l 1

Midland Plcnt Units 1 cnd 2 Structural Stresses Inducrd by Differ &ntial Setticmant in the Diccol Ganorator Building Table I-1 (continued) e.

Extreme Environmental and Accident Conditions U = 1.05D + 1.00L + 1.25E + 1.0T + 1.0H + 1.OR (13) g 3

U = 0.95D + 1.25E + 1.0Ti + 1.0Ha + 1.0R (14)

U = 1.0D + 1.0L + 1.0E' + 1.0To + 1.25Ho + 1.OR (15)

U = 1.0D + 1.0L + 1.0E' + 1.0Ti + 1.0Ha + 1.OR (16)

U = 1.0D + 1.0L + 1.0B + 1.0To + 1.25Ho (17)

U = 1.0D + 1.0L + 1.0To + 1.25Ho + 1.OW' (18) where B = hydrostatic forces due to the postulated maximum flood D = dead loads of structures and equipment and other permanent lead contributing stress E = operating basis earthquake (OBE)

E' = safe shutdown earthquake load (SSE)

Jo = force on structure caused by thermal expansion of pipes under operating conditions H = force on structure caused by thermal expansion of a

pipes under accident conditions L = conventional floor and roof live loads (includes moveable equipment loads or other loads which very in intensity)

R = local force, pressure on structure, or penetration l

caused by rupture of pipe T = effects of differential settlement, creep, shrinkage, and temperature T

= thermal effects during normal operating conditions, o

including linear expansion of equipment and tempera-ture gradients t

l T,

= total thermal effects which may occur during a design accident U = required strength to resist design loacr or their related internal moments and forces l

l l

2

Midlcnd Plant Units 1 cnd 2 Structural Strs:sta Induced by Differential Settlemant in the Diesel Generator Building Table I-1 (continued)

W = design wind load W' = tornado wind loads, excluding missile effects, if applicable (refer to Subsection 2.2.3.5)

)

s e

t e

l l

\\

i l

l i

l t

3 l

1

Midicnd Plcnt Unita 1 cnd 2 Structural Stronces Induced by Differential Sattitmant in tbs Diesel Generator Building TABLE I-2' LOADS AND LOAD COMBINATIONS FOR COMPARISON ANALYSIS REQUESTED IN QUESTION 26 OF NRC REQUESTS REGARDING PLANT FILL ACI 349 as Supplemented by Regulatory Guide 1.142 a.

Normal Load Condition:

U = 1.4 (D + T) + 1.7L + 1.7R o

U = 0.75 [1. 4 (D + T) + 1.7L + 1.7Tf+ 1.7Ro]

b.

Severe Environmental Condition:

U = 1.4 (D + T) + 1.4F + 1.7L + 1.7H + 1.9Eo + 1.7Ro U = 1.4 (D + T) + 1.4F + 1.7L + 1.7H + 1.7W + 1.7Ro U = 0.75 [1.4 (D + T) + 1.4F + 1.7L + 1.7H + 1.9Eo + 1.7Tb

+ 1.7Roj U = 0.75 [1.4 (D + T) + 1.4F + 1.7L + 1.7H + 1.7W + 1.7T o

+ 1.7R ]

o c.

Extreme Environmental Conditions:

U= ( D + T) + F + L + H + To + Ro+Wt U= (D + T) +F+L+H+To+Ro + E.,

d.

Abnormal Load Conditions:

U= (D + T) +F+L+H+T.

+ R, + 1.5P.

U=

( D + T ) + F + L + H + T.

+ R, + 1.25P, + 1. 0 ( Yr + Yi

+ Ym) + 1.25Eo U= (D + T) +F+L+H+T.

+ R, + 1.0P, +.l. 0 ( Y,

+ Yi

+ Ym) + 1. 0 U.,

1

Midlcnd Plant Unita 1 End 2 Structural Stregacs Induced by DiffGrcntial SGttlem:nt in the Diesel Generator Building Table I-2 ( Contin ued )

where Normal loads are those loads encountered during normal plant operatior. and shutdown, and include:

T = settlement loads D = dead loads or their related internal moments and forces L = applicable live loads or their related internal moments and forces F = lateral and vertical pressure of liquids or their rela-ted internal moments and forces H = lateral earth pressure or its related internal moments and forces To = thermal effects and loads during normal operating or shutdown conditions, based on the most critical transient or steady-state condition Ro = maximum pipe and equipment reactions if not included in the above loads Severe environmental loads are those loads that could infre-quently be encountered during the plant life and include:

En = loads generated by the operating basis earthquake (BOE)

W = loads generated by the operating basis wind (OBU) speci-fled for the plant Extreme environmental loads are those loads which are credible but highly improbable, and include:

E.,= loads generated by the safe shutdown earthquake (SSE) l We = loads generated by the design tornado specified for the l

Plant Abnormal loads are those loads generated by a postulated high-energy pipe break accident and include:

P. = maximum differential pressure load generated by a i

postulated break l

T. = thermal loads under accident conditions generated by a postulated break and including To 2

i L

Midlcnd Plant Unita 1 End 2 Structural Stresecs Induced by Differential Settlement in the Diesel Generator Building Table I-2 ( Continued)

R. = pipe and equipment reactions under accident conditions generated by a postulated break and including Ro U

= required strength to resist design loads or their related internal moments and forces Y, = loads on the structure generated by the reaction on the broken high-energy pipe during a postulated break Yi = jet impingement load on a structure generated by a postulated break Ym = missile impact load on a structure generated by or during a postulated break, such as pipe whipping 6

]

3

Midland Plant Units 1 Cnd 2 Structural Strsacco Induend by Differential Settlement in the Diesel Generator Building TABLE I-3 SOIL PROPERTIES USED IN THE SEISMIC ANALYSIS First Second Original Revised (1)

Revised 03 Analysis Analysis Analysis Modulus of Elasticity (E) 22,000 ksf 6,598 ksf 2,609 ksf Poisson's Ratio 0.42 0.45 0.40 Unit Weight (w) 135 pcf 116 pcf 120 pc/s Shear Wave Velocity (Vs) 1,359 ft/s 796 ft/s 500 ft/s Shear Modulus 7,746 ksf 2,275 ksf 971 ksf UI Note different shear wave velocity values.

I 1

f

m s

. ~ _ _ _ _ - _ _. _

Midland Plant Units 1 and 2 Structural Stresses Inducsd by Differential Settlement in the Diesel Generator Building TABLE I-4 REBAR STRESS VALUES -

FOR Tile DIESEL GENERATOR BUILDING STRUCTURAL NENBERS

( ACCORDING TO FSAR AND THE RESPONSES TO NRC REQUESTS REGARDING PLANT FILL, QUESTION 15)

Maximum Calculated Loads Rebar Nazimum

( B.c AP)

Stress Value Allowable Loads ConcreteHI Axial In-Plane (ksi)

Axial In-Plane Stress (ksi)

Description of Load'U Tension Flexural Shearsal. (Allowable Tension Flexural Sheartas Gradient TAllowable Nembers/ Location combination (k/ft)

(k-ft/ft)

(k/ft)

= 54 kai)

(k/ft)

(k-ft/ft)

(k/ft)

(*F)

= 3.400 ksi)

Exterior - West 2'-6" thick wall Tornado 9.73 27.17 5.36 22.17 85.3 95.7 34.2 0

0.354 horizontal reinforce-ment, plate element 44 Exterior - South 2'-6" thick wall _

Seismic 27.30.

1.33 67.58 42.46 85.3 95.7 34.2 60.4 0.000""

horizontal reinforce-ment, plate element 287 Elevation - 664'-O" 2'-0" floor slab Tornado 17.70 13.67 5.34 39.15 47.5 44.7 26.7 24 0.068 E-W reinforcement,

. plate element 167 Elevation - 680'-0" l'-9" floor slab Tornado 3.51 26.62 1.77 36.06 85.3 63.7 22.7 24 0.834 N-S reinforcement, plate element 788-South 2'-0" missile' shield Seismic 15.b1 12.50 14.34 32.84 64.8 55.5 26.7 60.4 0.372 wall south, horizontal reinforcement, plate element 631 Interior 2'-0" interior missile Tornado 18.99 3.78 1.35 28.06 47.5 55.5 26.7 24 0.000Hb 8 shield wall, vertical reinforcement, plate element 824 1

Midland Plant Unita 1 and 2 Structural Strasses Induced by Differential Settlement in the Diesel Generator Building Tabla I-4 (continued)

Maximum Calculated Loads Rebar Maximum (BSAP)

Stress Value Allowable Loads Concreteidi Axial In-Plane (ksi)

Axial In-Pla5e Stress (ksi)

Description of LoadM8 Tension Flexural Shea r t al ' (Allowable Tension Flexural Shear (U Gradient TX11owable Members / Location Combination (k/ft) ik-ft/ft) (k/ft)

= 54 ksi)

(k/ft)

(k-ft/ft) _jk/ft)

(*F)

= 3.400 ksi)

North 2'-0" missile shield Tornado 23.46 9.14 7.76 13.85 137.2 114.7 26.7 0

0.000'*3 wall north, horizontal reinforcement, plate element 839 Exterior - North 2'-6" thick wall Tornado 11.17 20.27 5.19 21.90 85.3 95.7 34.2 24.n 0.313 horizontal reinforce-ment, plate element 767 Exterior - East 2'-6" thick wall

. Torn _Jo 9.15 25.44 8.11 23.64 85.3 95.7 34.2 24.0 0.403 horizontal reinforce-ment, plate element 896 Interior l'-6" thick wall Tornado 17.0 0.65 0.31 16.66 64.8 41.2 20.6 24.0 0.000

horizontal reinforce-ment, plate element 683 South 0.00

8 2'-0" thick box Tornado 4.93 1.62 1.45 8.02 48.6 39.2' 26.7 0

missile shield / south, horizontal reinforce-ment, plate element 732 2

-+

Midland Plant Unita 1 and 2 Structural Stresses Induced by Differential Settlement in the Diesel Generator Building Table I-4 (continued)

Maximum Calculated Loads Rebar Maximum (BSAP)

Stress Value Allowable Loads Concrete 84)

Axial In-Plane (kai)

Axial In-Plane Stress (ksi)-~

l3' Gradient (Allowable Description of boad 8 8 8 Tension Flexural Shear:23 (Allowable Tension Flexural Shear 3.400 ksi) i Members /Locat ion Combination (k/ft)

(k-ft/ft) (k/ft)

= 54 ksi)

(k/ft)

(k-ft/ft) -(k/ft)

(*F)

=

e i

Footing 46.0d N/A 92.14 34.20 75.0 2'-6" thick footing, Seismic beam element 87 NOTES:

4 8'8The tornado load combination is 1.0 (D + L) + 1.0 (Wr) + 1.0 (T,).

The seismic load combination is 1.0 (D + 7) + 1.0 (E') + 1.0 (T ).

4 i

s280ut-of-plane shear loads were investigated independently from axial, flexural, and in-plane loads. This investigation indicated that the maximum allowable out-of-plane shear force was never exceeded.

83' Shear capacity of concrete only with no tension load on the section 8'IStresses are in compressive sense. Concrete stresses shown are associated with maximum rebar tensile stresses shown in this table, 868Section is cracked.

l*lThis value is conservatively high and will be reduced.

1 4

5 t

i 3

4

Midland Plant Units 1 and 2 Structural Stresses Induced by Differential Settlement in the Diesel Generator Building El. 680'-0" M SumW Man PoinQ 1

1 (Member Number)

El. 664'-0" M

2 2

El. 647'-0" M

3 Rotational Spring 3

Horizontal Translational Spring K

M K

/

y 4

r x

El. 630'-0"

/

-=

j yl Horizontal Damper i

Rotational Damper

,C C,

j g xx r i i t / /// ///trii r/rr /

/

Vertical Damper Vertical Translational Spring

{

FIGURE l-1 DIESEL GENERATOR BUILDING DYNAMIC LUMPED MASS MODEL FOR SEISMIC ANALYSIS l

1 L

Midland Plant Units 1 and 2' 4

i Structural Stresses Induced by Differential Settlernent in the Diesel Generator Building i

l 67%'

/

centerline to centerline f

152*-6" centerline to centerline i

l

/

h ' its l

/

/

/

/

i L

h,/

y s,/

"?

/

s s

g, g[ /

gi:

/

h)l -

l

/

/:/, s l x, %

l N s

/

N !

h 49'-10%"

\\

ff (ctnterline of roof

\\

g.

/

/

/

i slab to centerline N

h N

/

/)%

r **

y f

s p

N h

[./

of footing)

/

/

NORT11 s

- ~ -

N N

prf

/

N s y-j p% d$I

/

f

\\r\\

/

}x.m ~,

f N /

N N /

FIGURE l-2 DIESEL GENERATOR BUILDING FINITE ELEMENT MODEL typical vertical translational spring g

Midland Plant Units 1 and 2 Structural Stresses induced by Differential Settlemsnt of the Diesel Generator Building DIESEL GENERATOR BUILDING I*

1.13 1.13 1 45 i.32 9dio O Fra RO O FF 0

y s.-

... a

s.

n.

m:

.*,;.v q*....

n..

..a:

s 101

r "Q.Q Cl

[

l Q Ol 4

e{

j t

i; 1.70 O BAY 1 BAY 2 BAY 3 BAY 4 b

i i

i e

1.94 0 1.94 o 1.53 O 2.41 0 2.36 o NORTH LEGEND O

BUILDING SETTLEMENT M ARKER 2.36 SETTLEMENTC 'NCHES (THIS DRAWING AND T.sl INFORMATION CONTAINED ON IT WERE OBTAINED FROM FIGURES 27-12 AND 27-13 OF THE RESPONSES TO NRC QUESTIONS REGARDING PLANT FILL)

FIGUKE l-3 ESTIMATED SECONDARY COMPRESSION SETTLEMENTS FROM 8-15-79 to 12-31-202S ASSUMING SURCHARGE REMAINS

Midland Plant Units 1 and 2' Structural Stresses Induced by Differential Settlement in The Diesel Generator Building NORTH

/*"*""*"""""

y 0.24

/ 0.25 0.07 0.07 0.24

/

/

/

/

SURFACE A

/

0.%

1.03 1 09 1.13 I

/ JO J: '..

C -

d---

i 1.29 0.12 / /

1.13 1.16 1.28 j

f 1.45 SURFACEB j

/j f 0.16 (Greatest differential I

/

// BAYI BAY 2 BAY 3 BAY 4

/

j settlement = 0.21 inches)

/(j/

I

/

lf.39

/

f.....g.............................................................................,/

/ 1.68 I 1.70 0.11 f/ 1.70 0.08 0.12 0.13 0.20 /

(Greatest differential

//

i ll

/

settlement = 1.17 inches) 1.90 SURFACE D

'***W'"*

u - - - ---

f (Greatest differential

% ~ ~ _ _ _- - < / 2.13 l 94 1.94 1.93

__a 2.36 settlement = 1.28 inches) 2.41 SURFACE A REFERENCE (AS OF AUGUST 15, 1979)

SURFACE B ~~.=~~...~. ACTUAL SETTLEMENT VALUES FROM SEPTEMBER 14,1979 TO JULY 9,1981 SURFACE C ------ SECONDARY COMPRESSION SETTLEMENT VALUES CALCULATED BY FINITE ELEMENT ANALYSIS SURFACE D ----- ESTIMATED SECONDARY COMPRESSION SETTLEMENT VALUES FROM AUGUST 15,1979 TO DECEMBER 31, 2025 ASSUMING SURCilARGE REMAINS IN PLACE.

FIGUREl-4 i

COMPARISON OF ESTIMATED SECONDARY COMPRESSION SETTLEMENT VALUES WITil

. SETTLEMENT VALUES RESULTING FROM A FINITE ELEMENT ANALYSIS OF Tile DIESEL GENERATOR BUILDING Y

[

'O

,)

t g

I Midland Plant Units 1 and 2 Structural Strcsses Induced by Differential Settlement of the Diesel Generator Building

</

/

M

^

i Elevation 664 s e 111 s /

s e ii i <

1i

[ Final Plant Grad Elevation 634'-0" A

Ill lil lill lil

-(/'

/ -

Elevation 628'-0"

',[

- - V = 500 feet /second s-e.

r?.,;

,[-

Elevation 615 APP,75FT, V,= 850 feet /second

t t

j..

s Elevation 600

~

[(original gra,de) y A

[

,, /,_ :, > e- -

~<wNL;i.n;

.m :: :

C* >~ I~ b<< A s~, a ;m,

v~ e.

• ss.L f

g l

V W 850 feet /second j

s

.. i i

/

ee Elevation 550 (depth of V = 2300 feet /second eff. soil) s

' FIGURE I 5 BASIS FOR CALCULATION OF EQOlVALENT SHEAR WAVE VELOCITY VALUES (V,)

(Shaded region represents the area over which low-strain shear wave velocity values (V,) were averaged, resulting in a V, value of 796 ft[sec.)

m gliwt

• w

+

eu--

-e--e--

i.-er---e-#

pew wy, y--i, y,

,yw--

y.

=,g,,e

,www wi,

-ww ww,,

w,,

e-am -

-r o.--

w

Midicnd Plant Units 1 End 2 Structurtl Stroacco Inducad by Differential Settlement in the Diesel Generator Building APPENDIX A OPTCON The OPTCON computer code is a versatile and complete design and analysis program for reinforced concrete structures.

The program may be used for the investigation of an existing reinforced concrete section where the reinforcing steel area is predetermined.

Alternatively, it can be used for obtaining an optimum design by allowing the program to determine the minimum reinforcement required.

The computer program operates on the axial force / moment interaction diagram (IAD) of a section, where an IAD is a plot of the maximum allowable resistance of a section for given stress and strain limitations.

Combinations of moment (M) and axial load (P) falling within this area are acceptable.

Figure IA-1 depicts the appropriate IAD for a symmetrically reinforced, symmetrically shaped section subjected to a combination of flexural and axial loads.

The OPTCON program handles loads consisting of axial forces and corresponding bending moments due to different types of loads.

Special subroutines are provided to incorporate the thermal effects into the design and/or investigation.

The cracking effect of the concrete and the yielding effects of the reinforcement (as allowed by the appropriate stress / strain yielding criteria) are considered in the calculation of the thermal loads and moments computed by the program.

A-1

Midland Plant Units 1 and 2 Structural Stresses Induced by the Differential Settlement of the Diesel Generator Building 1

Compression C(-)

Satifies Design Criteria l

Interaction Diagram Compression Failure Zone a

's

'i::it?:

~

?

' b:k

,siiiiiji/ e,Q s

.: j:: 9 +

%i. :....:..

-:[$$ $$.fii[

asna:!:fW

, gl;g:;.g, ~ " "

T'

!!.y:::.

f[h}i

ggg,

\\ RafaDCed Failure

...!::$s

. s.::$: S

• 7 Tension Failure Zone N

Moment f'

(+)M l

s Tension T(+)

Figure IA 1 TYPICAL INTERACTION DIAGRAM (for single axis bending on a section with symmetrical reinforcement) i l

l

ENCLOSURE 2, SUBGRADE MODULUS AND SPRING CONSTANT VALUES FOR DIESEL GENERATOR BUILDING STRUCTURAL ANALYSIS 1

Issue:

Basis and actual numerical values of adopted spring constants - both 1

static and dynamic (reconcile with settlement predictions and observed l

behavior)

Response

Introduction This describes the methods used to develop the subgrade modulus and spring constant values intended for use in the structural and seismic analysis of the diesc1 generator building. These values are indicated in Appendix A.

Springs for Dynamic (Short-Term Static) Condition Springs were developed for movements which were assumed small enough to cause a negligible decrease in the low-strain shear modulus which was determined from field geophysical testing. Vertical translation, hori-zontal translation, and rocking mode springs were developed. The shear J

wave velocity profile assumed for use in the spring development consis-ted of a shear wave velocity of 500 fps from elevation 634 to 615 feet and 850 fps from elevation 615 to 550 feet. The diesel generator building foundation elevation is 628 feet. In all cases, it was assumed i

that the structure foundation and walls were rigid. The spring constant values were developed for the best estimate of the existing soil shear wave velocity. However, as discussed later, a parametric study was performed and the effect of varying shear wave ve.", city profiles on the spring constants was determine (,

5 The spring constant for the vertical translation mode is 9.6 x 10 k/ft (subgrade modulus, 163 kcf). This value was developed using the fol-lowing formula developed by Timoshenko and Goodier which is described in Reference 2, Pages 347 and 350:

o 14cd K=1_

r,=y,

g This formula represents the ratio of force to deflection for a rigid circular plate resting on an elastic half-space with shear modulus, G.

The shear modulus used was a weighted average of the shear moduli pro-file for a depth below the foundation equal to the width (78 feet) of the building. The subgrade modulus indicated,163 kcf, was determined

.by dividing the spring constant by the area of the diesel generator building wall footings.

1 l

The spring constant for the horizontal translation mode was developed as separata components considering the effects of:

1)

The horizontal shear force of the base of the building and soil within the building perimeter on the elastic half-space 2)

The horizontal shear force of the buried portion of the exterior side walls on the elastic half-space 3)

The horizontal force of the walls perpendicular to the translation directiog on the half-space.

The base shear component, 3.1 x 10 k/f, was developed using the following formula developed by Bycroft which is described in Richart, Hall, and Woods (1970),

Pages 347 and 350:

32 (1 -u ) Gr o " J4cd o

K=x 7 - 8u Yr This formula represents the ratio of horizontal force to horizontal deflection for a rigid circular plate resting on an elastic half-space with shear modulus, G.

The shear modulus used was based on the shear wave velocity of 500 fps.

5 5

The side shear components, 1.48 x 10 and 2.97 x 10 k/f, were developed using the following formula developed by Groth and Chapman which is described in Reference 1, Pages 99 through 102:

p,qaI E

This formula represents the horizontal deflection of the top or bottom corner (depending on the value of I) of a flexible rectangle buried in an elastic half-space with Young's modulus, E.

The spring constant was decennined by rearranging terms and multiplying by the area of the rectangle (wall). Because the diesel generator building walls are assumed rigid, the spring constants developed for the top and bottom corners were averaged. The Young's modulus used was based on a shear wave velocity of 500 fps.

The components represented by the wall pushing against the soil, 3.56 x 10 and 2.0 x 10 k/f were developed using the following formulas i

developed by Douglas and Davis which are described in Reference 1, Pages 97 and 98:

2

(3 - 4u) F3+F4 + 4 (1 - 2p)(1 -u)F3 (Upper Corners) o

=

x 32 r G (

-p)

(3 - Au) Ft+F2 + 4 (1 - 2 )(1 -u)F3 (Lower Corners) o 32 r G

-p)

=

9 These formulas represent the horizontal deflection of the upper and lower corners of a flexible rectangle buried in an elastic half-space with shear modulus, G.

The spring constant was determined by rearran-ging terms and multiplying by the area of the rectangle (wall). Because the diesel generator building walls are assumed rigid, the spring con-stants developed for the top and bottom corners were averaged. The Young's modulus used was based on a shear wave velocity of 500 fps.

9 The spring constants for the rocking mode are 1.85 x 10 and 4.65 x 10 k-f/ rad. These values were developed using the following formula developed by Gorbunov-Possadov which is described in Reference 2, Page 350:

.2 G

<cc

0 9=

1,,,

This formula represents the ratio of moment to angular rotation for a rigid rectangular plate resting on an elastic half-space with shear modulus, G.

The shear modulus used was the same weighted average value that was used for the vertical translation mode.

.nully, it,as necessary to datermine the vari: tion in the vertical translation and rocking mode spring constants for a variation of back-fill properties consisting of 1) fill below foundation level (el 628 to 600 feet) with a constant shear wave velocity of 500 fps and, 2) fill below foundation level with a constant shear wave velocity of 1,350 fps.

This was done by substituting the weighted average shear moduli for these cases for the weighted averages used earlier 'in the calculations for a shear wav

• velocity profile for fill (el 628 to 600 feet) varying f r om 500 f ps to 850 f ps.

Because the vertical translations and rocking mode spring constants, K, and K,4, respectively, are linearly propor-tional to the shear modulus, G',

the above can also be accomplished by multiplying the original spring constants by 0.85 and 1.8 for fill shear wave velocities of 500 fps and 1,359 fps, respectively.

Sprines for Long-Term Static Condition The subgrade moduli for tne long-term settlement condition of the diesel generator building were developed from the settlement of the structure

~

af ter the surcharge was removed neglecting the immediate.. eave which occurred following load removal, September 14, 1979, to December 31, 2025. Figure 2 contains the contact pressures used to determine these j

-subgrade moduli.

3 i

The vertical subgrade moduli were determined by dividing the contact pressures by the corresponding measured and estimated settlements.

The settlement used is the sum of 1) the measured settlement which occurred frem September 14, 1979, to January 16, 1980, neglectin3 the imnediate heave occurring af ter surcharge removal on August 15, 1979,

2) the estimated settlement from January 16, 1980, to December 31, 2025, extrapolated from settlement versus log (time) plots of the building se t tlement =arkers which were plotted for the time period during sur-charge loading, and 3) the esti=ated dewatering settlement which had an estimated range of 0 inch to 0.25 inch. The estimated values were then proportioned within this range according to the settlement predicted f rom August 15, 1979, to December 31, 1981, by extrapolation of settle-ment versus log (time) plots described in (2) above. The effect of seismic shakedown settlements has been omitted as discussed in Subsee-tion 2.1.2 of Enclosure 1.

The horizontal spring constant value to be used is the same as the value computed for the short-term static case.

3: inas fo r 3eismic c.nalys is The seismic analysis for the diesel generator building was done by using the half-space lumped spring and mass representation approach presented in BC-TOP-4-A, Revision 3.

Dif ferences may be noted between the spring values used for the seismic d ysis a.u.

in spri gs used for t..c cina

i: (.:.: c:-t a r e. :: c) c.i.:. -

sis. This is primarily due to the consideration of only local ef fects for the short-term static analysis', whereas the seismic analysis must consider global ef fects due to ground motion. Minor differences arise also because of the use of alternate formulas to calculate equivalent area, variations in Poisson's ratio, and graphical interpolation.

9 4

4 i

REFERENCES 1.

Poulos, H.G., and Davis, E.H.,

Elastic Solutions for Soil and Rock

!!echanics, John Uiley and Sons, Inc.,1974, (New York),.411 pp.

2.

Richart, F.E., Jr., Hall, J.R., Jr., and Woods, R.D., Vibrations of Soils and Foundations, Prentice-Hall, Inc. 1970, (Englewood Cliffs, N.J.), 414 pp.

e e

b 4

5

APPE:,' DIX A SU5 GRADE ".00C*l.*S AND S?RI'iG CO::5?A:;!

VALUES FOR STRUCTURAL AND SEISMIC ANALYSIS OF TriE DIESEL CENERATOR BUILDING 9

4 r-,

w.

-,-s

APPENDIX A I.

Dynamic (Short-Term Static) Condition The vertical subgrade modulus values and the horizontal acd rocking spring constant values for the dynamic (short-term static) analyses of the diesel generator building are listed below.

Spring Constaats N-S Direction E-W Direction Mode Subgrade Modulus (k/ft)

(k/ft)

Ve rtical 163 kef Horizontal 5

5 1.

Base shear 3.1 x 10 3.1 x 10 5

2.

Stue frictinn be-1.48 x 10 2.97 x 10 tween soil and wall 5

3.

Wall pushing against 3.56 x 10 2.0 x 10 soil 5

5 TO AL OF 1, 2, and 3 8.14 x 10 g,g7 x yg Rocking Rocking Axis Rocking Axis E-U ( k-f t/ rad) N-S (k-f t/ rid) 9 9

1.85 x 10 4.65 x 10 II.

Long-Term Static Condition The vertical subgrade modulus values and the horizontal spring constants values for the long-term static analysis of the diesel generator buil-ding are given below.

1.

Vertical subgrade modulus - The subgrade moduli of the selected points along the exterior wall footings are shown in Figure 1.

2.

Horizontal spring constants - The horizontal spring constants given previously for the short-term static or dynamic case can also be used for the long-term, static case. This is because horizontal displacements are small and in the elastic range.

III. Seismic Analysis Soil springs for the seismic analysis of the diesel generator building were determined for two conditions to envelope the anticipated range of soil properties beneath the building. These springs are used for both the safe shutdown egrthquake and the operating basis earthquake.

Springs

. for the E = 22 x 10 kaf are as follows:

A-1 e

y,,

y ~+

w r+-m--

-y e

v-v w

y w

I 1

North-South East-West Vertical 6

6 K (K/ft) 1.5 x 10 1.5 x 10 x

0 2.0 x 10 K (K/ft) 9 9

K9 (K-f t/ rad) 3.30 x 10 8 10 x 10 Soil springs based on revised soil properties with E = 6.6 x 10 ksf and V = 796 f t/s as follows:

s North-South East-West Vertical 5

5 K (k/ft) 6.9 x 10 7.0 x 10 x

5 9.3 x 10 K (k/ft) 9 9

K9 (k-f t/ rad) 1.70 x 10 3 50 x 10

=Sen-vs e velocity of 796 f t/s is a leu strain value.

Its deriva-tion is discussed in Subsection 2.1.6 of Enclosure 1.

Io account fo r strain ef fects on the modulus of elasticity, this low strain value is degraded to 660 f t/s. The modulus of elasticity corresponding to this reduced value is then varied by +50% to account for variations in soil prope rt ies. The ef fect of lowering the shear wave velocity to 500 f t/s was also analyzed to ensure that the cost conservative values were used ftr the s:ructural analysis.

l.

l J

O e

A-2

FIGURE 1 LONC-TEFJ! STATIC SUBGRADE MODULI IN KCF DIESEL GENEPATOR EUILDI:'G s

hl' 32.0 33.3 31.4 35.9 26.5 l

I I

l l

25.5 24.7 1

I I

i l

24.7 24.1 23.8 18.8 18.8 i

I 1

l l

e e

l I

L

~n.-

FIGUPI 2

SUMMARY

OF BEARING STRESS (KSF) DUE TO DEAD LOAD s

i I

3.17 3.14 3.11 3.08 3.05' I

i

~

l 1

g I

1 I

t 3.80 3.76 3.73 3.70 3.67 e

e

'o

~

ENCLOSURE 3 BEARING CAPACITY EVALUATION OF DIESEL GENERATOR BUILDING FOUNDATION INTRODUCTION Analyses were carried out to ef31uate the factor of safety against bearing capacity failure of the diesel generator building footings when subjected to static and combined static and earthquake loading.

Factors of Safety Factors of safety against bearing capacity failure were calculated for two different conditions:

1.

A long time af4er the permanent structural load h s been applied and the induced pore watet pressures below the footing have dissipated. Effective shear strength paramet-rs of the soil were used to calculate the factors of safety under these conditions.

Laboratory testing of the fill materials indicated the effective strength parameters to be c' = 0 psf and $' = 30*.

Factors of safety under drained conditions were calculated for a high water level at elevation 628' (at the bottom of the footing) and a low water table below elevation 603'.

Factors of safety equal to 5.3 and 7.2 were calculated for these two conditions, respectively. These values exceed the value of 3 required in the Midland FSAR for dead and live load conditions.

2.

Factor of safety against bearing capacity failure were calculated for a second condition corresponding to an earthquake occurrence sometime after the structure has been built.

In the analysis it was assumed that during the time span before the earthquake occurs the excess pore water pressure in the ground, induced by the static loads on the footing, have dissi-pated. The undrained shear strength of the soil corresponding to consol-idation under the 6 feet of surcharge plus the static footing load was used in estimating the factor of safety. The effective confining stresses along the failure surface was estimated by utilizing the method of slices to determine the mobilized friction angle required to maintain static equilibrium.

The effective normal stress and shear stress at the base of each slice were then used to estimate the absolute values of the major and minor principle stresses II and17 at the base of each slice due to the long 1

3 term static loads. These initial consolidation stresses were used by Woodward-Clyde Consultants for determining the range of consolidation pressures and the appropriate anisotropic consolidation stress ratio, 3 /6, t u8e in anis tr P cally consolidated undrained triaxial tests.

i Theresultsofthesetriaxialtestsweresummarizedbyplottingthe 3

10/19/81 mi1081-0833a100

2 undrained shear strength, tg, along the failure surface for each sample asdefinedbyLoeweandKarakiath(1959),andtheinitialeffective confining stress, 3, on that surface for both anisotropically and isotropically consofidated samples.

It was found that the anisotropically consolidated samples gave slightly higher undrained strengths than the isotropically consolidated samples.

The factors of safety for the combined earthquake and static loading were obtained by using the method of slices. The undrained shear strength, gg, at the base of each slice was ' determined from the plots of I I

versus 77 referencedabove,byenteringtheplotsatthe.valueoftheINtial sea, tic confining stress, U, at the base.of oach slice.

Factors of safety were calculated for a high water table' at elevation 628' for the anisotropic and isotropic shear strength relationships which yielded factors of safety of 2.6 and 2.4, respectively.

For the dewatered case when the water table is below the failure surfaces considered, a factor of safety of 3.1 was calculated from the anisotropic shear strength relationship.

In current engineering practice which is consistent with the Midland FSAR, a factor of safety of 2 is considered adequate when considering combined static and earthquake loads; under these conditions it is concluded that the present diesel generator building footings are adequate for the static and earthquake loads considered above.

10/19/81 mi1081-0833a100

. _ - ~

l i

r ENCLOSURE 4 i

LONG-TERM MONITORING OF SETTLEMENT l

FOR DIESEL GENERATOR BUILDING

, Issue:

E.ans for Long-Term Monitoring of Settlement for Diesel Generator Building (Includes Technical Specifications, etc)

Response

The settlement monitoring program for the diesel generator building has been established and is being implemented. Settlement monitoring points have been located around the building and on the machine pedestals to identify any tilting or warpage.

These points are surveyed every 60 days during construction, and every 90 days during the first year of plant operation. It is currently l

planned to evaluate the settlement data during the first year of plant operation and develop an appropriate monitoring interval for the re-maining plant operating life. As a minimum, the building would be monitored annually for the next 5 years of operation and then at 5-year intervals thereafter. At least 6 points on the building will be monitored for the operating life of the plant: one point at each l

building corner and a point at the center of each east-west wall. Each corner of each machine pedestal will also be monitored as discussed above.

If the rate of settlement increases at any tire during the monitoring program to a value greater than predicted for shat monitoring point (see the response to NRC 10 CFR 50.54(f) Question 27), the monitoring interval will be increased to every 60 days to per_it evaluation of the change in settlemer.t. The allowable limit of absciute settlement af any point and relative. settlement between points will be provided as part of the technical sprcifications in the Final Safety Analysis Report. These values for the yedestals are given in the response to NRC 10 CFR 50.54(f)

Question 8.

Tae building values due to fill consolidation are given below.

I SW Corner South SE Corner NE Corner North NW Corner DG 1 DG 21 DG 3 DG 28 DG 26 DG 24 Allowable limit 1.85 1.89 2.34 1.38 1.19 1.18 of settlement After August 15, 1979 (inches) 10/19/81

..o ENCLOSURE 5 RELATIVE DENSITY AND SHAKEDOWN SETTLEMENT OF SAND UNDER THE DIESEL GENERATOR BUILDING Comparison of Relative Densities The relative densities of the sand fill in the area of the diesel generator building were compared based on information from the DG series borings conducted before surcharge and the COE series borings conducted after surcharge. The relative _ densities from the DG series borings were determined from standard penetration blow count data using Gibbs and Ho'Itz (1957) relationships. The relative densities from the COE series borings were determined by Woodward-Clyde Consultants (WCC) from in-situ densities of tube samples and grain size data as described in the report entitled "Estir;tes of Relative Density of Granular Fill McLerials, Diesel Generator BuildiGg, Midland Plauc, Units 1 & 2, Midland, Michigan" dated July 24, 1981.

For the purpose of presentation of relative density data the building was divided into four quadrants as shown in Figure 1.

The same Figure 1 also shows the location of DG series borings and COE series borings. The relative densities are compared below.

Relative Density Comparison (%)

DG Series Borings COE Series Borings Quadrant Range Average Range Average Northwest 15-100 62 42-100 82 Northeast 15-100 72 28-100 74 Southwest 45 45 57-100 88 Southeast 10-100 69 56-100 88 It is seen from the above comparison that the relative density obtai.ned from the COE series borings after surcharge are higher than the relative densities obtained from the DG borings before surcharge.

Makedown Settlement The settlement of the granular fill materials under the diesel generator building due to ground shaking (SSE = 0.12g) caused by earthquakes was calculated based on tests performed on soil samples obtained after the surcharge progara during the recent (1981) soil investigation program by Woodward-Clyde Consultants (WCC). The details of the tests and results obtained are presented in the WCC report entitled, " Estimates of Relative Density of Granular Fill Materials, Diesel Generator Building, Midland Plant Units 1 and 2, Midland, Michigan" dated July 24, 1981.

The settlements were estimated based on the approach described by Seed and Silver (1969) and the recommendations on multidirectional shaking by Pike, Seed and Chen (1975) at mi1081-0459a100

e i

2 4

I the location of Borings COE 8, COE 10, COE 11, COE 12 and COE 13.

For the COE borings the relative density used in the analyses was the average relative density reported by WCC in the report referenced above. The settlement values based on DS series borings prior to surcharge were also calculated using the above approach. However, the relative density values were obtained using the standard penetration blow count and the Gibbs and Holtz relationship. The settlement estimate based on borings prior to and after surcharge program are compared below.

Location DG Series Borings COE Series Borings (Inch)

(Inch)

Northwest Quarter 0.02-0.36 0.07 Northeast Quarter 0.00-0.25 0.07 Southwest Quarter 0.00-0.02 0.02-0.11 Southeast Quarter 0.00-0.14 0.02-0.11 It can be seen from the above comparison that the maximum settlement calculated based on the COE series borings (after surcharge) are lower than that calculated based on the DG series borings (before surcharge). Therefore, the design values of settlement and differential settlement of 1/2 inch provided in the response to.10CFR50.54(f) Question 27 are conservative.

10/15/81 1

t mi1081-0459a100

DIESEL GENEf1 ATOl1 DUILDING Cos 5 A NE QUADRANT d COE-10 NW QUADRANT eDG-28

1. DG-31 eDG-32 R

Wl T

Gi FJ0~DG-19I r 13 i

t_;

i-l

%a',e 0

r l

DG-14 a

H-1T Cit-14 T

ll-15 DG-10 DG-9 O

DG-27 DG-20.e DC-71e e DG-8 DG-13 S a

r- - - - - j g- - - -'- 1 r-- ~ ~ -- I DG-2 r- - - - - -)

l 1

1 I

8

,3 8

1 L

e i

i I

i l

l 3

I I

I l

I

'DG-22 8

I 8

i e

\\

b h

j i

IDC-15 I

I I

(DG-24 I

(

l iS I

1 1

Ie I

i I

i l

I DC-12 j.IDG-26 i

1 8

i DC-17 S

DG-16 l

G N-11 I

go a Cil-17 egCll-la I

e I

I I

Cil-16 8

i u---J L----

L.._-J L_ _ _ - _J J

L_J L

1 L__J L__

d COE-ll SW QUADRANT 1 COE-12 SE QUADRANT DG-1 e gCOE-13 e PREVIOUS (1979) BECHTEL DIESEL t;ENERATOR BORING dRECENT(1981)UCCCOEBORING FIGURE 1 BORING LOCATION PLAN IN AREA 0F DIESEL GENERATOR BUILDING

I ENCLOSURE 7 REVIEW AND CONTROL OF FACILITY CIMNGES TO THE DIESEL GENERATOR BUILDING Facility changes, ie, changes in structures or changes and additions in equipment and bulk commodities, to the Midland Plant safety-related structures are subjected to the re.eiews of the organizations identified in the Administrative Controls Section of the Midland.echnical Specifications, FSAR Subsection 16.6.5.

The two major organizations responsible for the reviews of facility changes are the Plant Review Committee (PRC), composed of plant staff members, and the independent Safety and Audit Review Board (SARB), which reports directly to the Vice President of Nuclear Operations. The functions, responsibilities and authority of these two review groups are identified in the Midland FSAR Subsections 16.6.5.1 and 16.6.5.2.

Facility change reviews will be conducted in accordance with the requirements of 10CFR50.59. Facility change reviews will include a review of the Midland FSAR to determine whether the change affects the safety of the structure or system and whether an unreviewed safety question is involved. This review is performed by a Consumers Power employee or a consultant. The subsequent evaluation is reviewed by the onsite Plant Review Committee prior to implementation of the change. The SARB also conducts an independent review of facility changes.

10/15/81 mi1081-0459a100

4 ENCLOSURE 8 DIESEL GENERATOR BUILDING BEARING PRESSURE DUE TO EQUIPMENT AND COMMODITIES A detailed weight summary has been made for equipment and commodities (piping, cable tray, wire, etc) which were included as live loads in the bearing pressure calculations for the diesel generator building.

The total weight of equipment is estimated to be 1,474 kips of which i

1,217 kips is on the pedestals and 257 kips is distributed throughout the I

building. The weight of commodities distributed throughout.the building is estimated to be 614 kips.

The weight of equipment was determined from vendor drawings for each piece of equipment. The majority of equipment is directly mounted on the diesel generator pedestal and contributes to the bearing pressure below the pedestals. The remaining pieces of equipment are mounted on walls and elevated slabs and contribute to the bearing pressure below the building spread footings.

The weight of commodities was determined by performing a take-off of the lineal footage or square footage of the various commodities. These values are then multiplied by an appropriate unit weight for each commodity to determine the total weight of the commodity.

The commodities are attached to the walls and elevated slabs of the building and contribute to the bearing pressure below the building spread footings.

The contact area of each pedestal is approximately 745 square feet and the contact area of the building spread footings is approximately 6,425 square l

feet. Thus, the load intensity due to equipment and commodities is 408 psf under each pedestal and 136 psf under the building footings.

Because the building is normally unoccupied, occupancy loads contributing to settlement will be negligible. The specified live load for the building floors represents the maximum estimated load on the floor during construction and maintenance. This load is used for design of the floor slab. Hence, the assumption of a 5 psf allowance for the occupancy load at the building footings is considered conservative. A 5 psf allowance for occupancy live loads on the grade slab would also be conservative.

10/15/81 mi1081-0459a100

____ ___-__ -___. _