ML19345D635

Interim Deficiency Rept Re Reactor Vessel Anchor Bolt Failure,Completing Design Package of Proposed Mods.Prestress Levels of Studs Have Been Lowered.Next Rept Will Be Sent by 810331

ML19345D635
Person / Time
Site:	Midland
Issue date:	12/10/1980
From:	Bailey K, Jackie Cook BECHTEL GROUP, INC., BECHTEL POWER CORP., CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:	James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
References
10CFR-050.55E, 10CFR-50.55E, 9787, MCAR-37, NUDOCS 8012160353
Download: ML19345D635 (54)

Text

- . - . _ . . _. -

(r.50

% CODSilm8f5

.I 'd -!

i POW 8f James W Cook

(

\ / 0 00f Vice Presalent - Projects. Engsnersvrg Q ond Construction General Offn es: 1945 West Pernall Row, Jackson. MI 49201 e (517) 78&o453 December 10, 1980

~

Mr J G Keppler, Regional Director Office of Inspection and Enforcement US Nuclear Regulatory Commission Region III 799 Roosevelt Road Glen Ellyn, IL 60137 MIDL*dD PROJECT -

UNI'I NO 1, DOCKET NO 50-329 UNIT NO 2, DOCKET NO 50-330 UNIT NO 1, REACTOR VESSEL BROKEN ANCHOR BOLT -

FILE 0.4.9.35 UFI 73*10*01, 02111(S), 21114(E) SERI?.L 9787

References:

1. S H Iiowell Letters to J G Keppler; Mid..ad Nuclear Plant; Unit No 1, Docket No 50-329, Unit No 2, Docket No 50-330; Unit No 1 Reactor Vessel Broken Anchct Lolt;

a. Howe-311-79; dated December 14, 1979

b. Howe-267-79; dated October 12, 1979 .

c. Howe-51-80; dated March 3, 1980  ;

d. Howe-80-80; dated April 30, 1980 7

M s2 J V Cook letter to J G Keppler; Midland Nuclear Plant;

, .'" Serial 8971; dated May 16, 1980

.~

~

3. J W Cook letter to J G Keppler; Midland Nuclear Plant; Serial 9330; dated July 24, 1980

(( 4. J W Cook letter to J G Keppler; Midland Nuclear Plant; O Serial 8809; dated August 1, 1980 e

I 5. NRC (D S Hood) letter to CP Co, dated July 7, 1980,  %)

Subject:

Summary of b y 23, 1980 Meeting on Preservice 6, '

Failure of Three Reactor Vessel Hold-Down Studs

6. J G Keppler letter to S H Howell, Docket No 50-329 and 50-330, dated August 18, 1980 oc1280-0081a112 8012160 M c

SERIAL 9787 2

References 1, 2 and 4 ( re interim 50.55(e) reports, as is this letter, concerning broken anchor bolts in the Unit I reactor vessel support skirt.

Reference 3 r- ie d interim technical information concerning the reactor pressure ve.. 1 support modification and the schedule for the accomplishment of that modification. In Reference 5, the NRC requested a detailed description of the analytical techniques being used to assess the modified NSSS support system. Enclosure 1 to this report provides the requested' information.

Enclosure 2 provides the status of actions taken to resolve this condition.

Another 50.5. (e) report, either interim er final, will be sent on or before March 31, 1901.

I Reference 6 transmitted the NRC investigation report regarding the reactor vessel anchor bolt failures. Further, Reference 6 specified that "... actual plant modifications to compensate for the defective bolts will not be started

- on Unit I until approval of the design concept is received from NRR."

Reference 4 to this letter generally stated NRR staff concurrence win the design concept, and also alluded to the schedule and type of further information submittals. The attached report and our previous submittals comprise the complete package of materials describing the design concept.

Based on the current procurement and fabrication schedule underway, we request that the Staff complete their review of the attached report by the middle of January. Immediately following the review, it is the Company's intent to meet with NRR to resolve any staff conce:ns, and thereby obtain formal recognition that the condit L specified in your R tter of August 18 (Reference 6) has been met. The final NRR approval is required by February 1, 1981 in order to support our construction schedule. If the Staff has any concerns from our previous report (Reference 4), we would appreciate being notified as soon as possible so that they can be resolved.

& n 4. CwL 1 WRB/cl l

Enclosures:

(1) Report entitled, " Reactor Pressure Vessel Support I

Modification for Midland Nuclear Power Plant, Midland, l Michigan, Report No 2," dated October 1980 (2) MCAR-37, Interim Report //4, dated November 5,1980, l

entitled, " Broken Reactor Vessel Anchor Studs in r Unit 1" CC: Director. Of fice of Inspection & Enforcement Att hr Victor Stello,.USNPC (38)

Director, Office of Management Information & Program Control, USNRC (!)

RCook, USNRC Resident Inspector oc1280-0081a112

- - - . ---w . - - y.m--7..g .-ev , * , - -,- -.c ~ ..-.=,,c w-. -

c - , - - - ,

. , w 4

SERIAL 9787 3 GAlinenberger, ASLB Panel FPCowan AS&L Appeal Panel MMCherry, Esq MSinclair CRStephens, USNRC WDPaton, Esq, USNRC ,

FJKelly, Esq, Attorne., '..aeral SHFreeman, Esq, Asst Attorney General GTTaylor, Esq, Asst Attorney General VHMarshall GJMerritt, Esq, TNK&J l

l B

l i

l l

l l

l i

oc1280-0081a112-e

• e ., --my-,-eawr -

w 9 -gew w w **-e - e , -w--,te we

REACTOR PRESSURE VESSEL SUPPORT MODIFICATION FOR MIDLAND NUCLEAR P0kIR PLANT MIDLAND, MICHIGAN REPORT NO 2 DECEMBER 1980 l CONSUMERS POWER COMPANY JACKSON, MICHIGAN

^7 rp1280-0036a112 t

TABLE OF CONTENTS (CONT'D)

Page 3.2 Analysis of the RV Support for the 14 Final Loads 3.2.1 Analysis of the Anchor Studs 14 3.2.2 Analysis of Upper Lateral Supports 16 4.0 STATUS OF ANALYSIS AND DESIGN 17 4.1 Final Support Load Generation 17 4.2 ULS Design land Fabrication 17 5.0 STUD DETENSIONING STATUS . 18

6.0 CONCLUSION

19

7.0 REFERENCES

20 TABLES 1 Detensioning Data 21 FIGURES 1 Lateral Support Concept 26 2 Detensioning Data 27 3 Walls and Slab Plan at El 636'-0" 28

! 4 Upper Lateral Support Plan 29 5 Upper Lateral Support Shim Detai.' 30 6 Upper Lateral Support Detail 37 7 Upper Lateral Support Embedment De tail 32 l 8 Plan View cf Lower RV Support Deta il 33 9 Concrete Pedestal Detail 34 j 10 Anchor Stud Detail 35 rp1280-0036a112 l

l

_ = _ _ _ _ _ _ ._= . _ . .. - .. - . . - . ._ _ ..

TABLE OF CONTENTS (CONT'D) f8Le 12 Reactor Coolant System Boundaries 37 13 RV Isolated Model, Reactor Internals and SSS 38 14 RV Isolated Model, Skirt-Supported Plant, Plan View 39 15 RV Iselated Model, Elevation View A-A, Hot Leg 40 16 Reactor Internals and Service Support Structure 41

.; 17 Internal Wall Structures Model 42 18 Utilization of Computer Programs 43

't i

.I

[

l l

rp1280-0036a112

___ _ _ _ = . __ - ..

I REACTOR PRESSURZ VESSEL SUPPORT MODIFICATION FOR MIDLAND NUCLEAR POWER PIANT .

TABLE OF CONTENTC

[ age,

1.0 INTRODUCTION

1

2.0 DESCRIPTION

OF THE EXISTING REACTOR VESSEL 1 SUPPORT DESIGN CRITERIA 3.0 ANALYTICAL PROCEDURES 1 3.1 Generation of Support Loads 1 3.1.1 Technical Basis 1 3.1.2 Mathematical Model 2 3.1.2.1 NSSS Model 3 3.1.2.2 Intern'al Wall Structures 4 3.1.2.3 NSSS Supports 5 3.1.3 Load Cases Analysed 5 3.1.4 Method of Analysis 6 3.1.4.1 Seismic Forcing Functions 6 3.1.4.2 LOCA Forcing Functions 7 3.1.4.3 Computer Codes Used 9 3.1.5 Seismic Analysis 11 3.1.6 LOCA Analysis 12 3.1.7 Preliminary Design Loads 13 rp1280-0036a112

I

1.0 INTRODUCTION

This report provides a description of the anal'rtical techniquea that will be used in the analyses of the Midland Unit 1 Re:ctor Vessei 'odified support system. This report is a continuation of the report submitted to the NRC in July, 1980 entitled, " Reactor Pressure Vessel Support Modification for Midland Nuclear Power Plant, Midland, Michigan, Preliminary Report No 1."

I 2.0 REACTOR VESSEL SUPPORT DESIGN CRITERIA The design criteria for the reactor vessel support system are those stated in the previous report, therefore please refer to Section 2.0 of the July 1980 Report for the discussion on this topic.

3.0 ANALYTICAL PROCEDURES 3.1 GENERATION OT. SUPPORT LOADS 3.1.1 TECHNICAL BASIS j The methodology used to generate the design loads for the modified Nuclear Steam Supply System (NSSS) supports will utilize the same l

l analytical techniques and computer codes as used in developing.the l 3&W's Owners Group Report entitled, Effects of Asymmetric LOCA l

I i Loadings, BAW 1621 B&W 177-FA, (Reference 2) which has been i

submitted to the NRC for review in July 1980.

i Modifications will be made to the existing mathematical models of the NSSS and its supports to incorporate the upper lateral support spring rates, reactor vessel anchor stud spring rates, internal rp1280-0036a112 l

L

i 2

wall structures, and boundary conditions at the reactor coolant pumps and steam generators rpecific to the Midland Plant. The seismic forcing functions are Midland specific, however the LOCA forcing functions (ie, cavity pressurization, and reactor internal differential pressures) used i' determine the support loadings are based on larger breaks than those specifically applicable to Midland.

The analyses will incorporate techniques (described herein) which insure that all components supporting, and attached to, the reactor vessel will receive a full review for structural integrity under the modified support design.

3.1.2 MATHEMATICAL MODEL It is assumed that the initial loads to which the Reactor Vessel (RV) and its supports are subjected will not produce component yielding. Therefore, model construction and subsequent analyses are based on linear analytical techniques. The validity of thes2 assumptions is assured by comparing the linearly derived dynamic stresses to allowable stresses for a linear analysis.

In describing the mathematical model which will produce the final loads on the NSSS supports it is convenient to discuss the model as three integrated components; the NSSS, the internal wall structures, and the NSSS supports attached to the internal wall structures.

rp1280-0036a112

3 3.1.2.1 NSSS MODEL Because of the complexity of the RV loading conditions and the number of attachments to the vessel, a detailed isolated model of this component is constructed. This model is a complete representation of the reactor vessel and its appendages (eg, control rod drive mechanisms, service support structure, and reactor internals). It also incudes the hot legs extending to the steam generators and the cold legs extending to the pumps for loops A and'B. Boundary conditions are imposed at the ends of the pipes where they connect to the components to simulate the remainder of the NSSS. The isolated model is shown in Figures 12 through 15. .

The isolated portion of the NSSS is modeled utilizing finite beam-element and lumped mass representations of each component. Finite element methods are used where necessary to define the structural characteristics of components such as the fuel and plenum assecblies. Once determined by finite element techniques, the structural characteristics of components are used to generate the equivalent finite-beam element and lumped mass representations: The criteria for developing the equivalent structural representation is that component stiffness and frequency must be retained.

rp1280-0036a112

4 The various components that make up the total RV and its internals are identified in Figure 16. By comparing Figure 16 with the lumped-mass model shown in Figure 13, the correlation between the components and the model elements representing them can be seen.

In addition to the structural representation of the components, the NSSS mathematical model incorporates the effects of fluid coupling between components into the overall structural response of the syste n. This is accomplished by develping a mass matrix using the height of concentric cylinders, the distance between the cylinders, and various parameters describing the fluid between the cylinders. The mass matrix which is generated is combined with the diagonal mass matrix terms defining component mass distribution to generate a full system mass matrix l 3.1.2.2 INTERNAL WALL S'.-  ; J.

The internal wall structural model properties included, are tht rea, shear area, area moments of inertia, 1

l modulus of elasticity, and Poisson's ratio for different l

elevations in the wall. Lumped mase-s at different elevations define the mass distribution and mass resistance of the wall structure. The internal wall structure is mode.ed to the center of the concrete 1

basemat and the boundary conditions at that point are rp1280-0036a112

5 fixed such that no relative rotation or translation is allowed. The internal wall structure model is shown in Figure 17.

3.1.2.3 NSSS SUPPORTS For the isolated RV model, the NSSS supports can be described as the boundary conditions imposed on the cold leg piping at the pumps and the hot leg piping at the steam generators, the reactor vessel skirt support, and the upper lateral supports near the RV flange.

The boundary conditions imposed on the reactor coolant piping at the pumps and steam generators consist of stiffness matrices that represent the characteristics of the structures to which the pipes are attached. They are obtained from a full system model by disconnecting the pipes at the component nozzles and computing a stiffness

, matrix of the remaining component with its supporting l

structures and other attached piping.

The reactor vessel skirt support is modeled as a boundary l

condition at the base of the RV skirt support in the form l

of a set of springs. The boundary conditions reflect the flexibility of the anchor studs, localized concrete flexibility, and overall flexibility of the RV pedestal from the RV skirt support to the center of the basemat.

rp1280-0036a112

6 The Upper Lateral Support (ULS) tie the RV to the internal wall structures. ULS structural properties are incorporated into equivalent beams with end conditions reflecting the axial load carrying ability of the supports and appropriate cross sections properties to reflect the support flexibility.

Localized concrete deformation is included in the considerations of the support flexibility. The ULS equivalent beams are shown in Figures 12, 13, and 17 as i

they connect the RV with the internal wall structures.

3.1.3 LOAD CASES ANALYZED The isolated model will be subjected to four load cases in the process of determining the design loads on the supports. Two sets of seismic analyses will be performed; one for the Operating Basis Earthquake (OBE) and the other for the Safe Shutdown Earthquake (SSE). Two Loss of Coolant Accidents (LOCA) cases will be considered; a guillotine at the hot leg outlet of the RV and a guillotine at the cold leg inlet to the RV. The support system is designed such that the ULS receive no deadweight or thermal loads from the RV. Deadweight and thermal loads for the RV lower support have been prev!ausly computed and will not be affected by the support modifications.

3.1.4 METHOD OF ANALYSIS 3.1.4.1 SEISMIC FORCING FUNCTIONS rp1280-0036a112

7 The seismic forcing functions that will be applied to the mathematical model consist of response spectra curves for SSE at damping values from 1% to 5%. Response spectra is supplied for earthquakes in five directions, North-South, East-West, vertical, rotation about North-South, and rotation about East-West. The rotation is applied as occurring about the geometric center of the RV at the elevation of the basemat.

3.1.4.2 LOCA FORCING FUNCTIONS LOCA forcing functions are composed of three sets of time histories which are applied simultaneously to individual degrees of freedom. The forcing functions are the result of blowdown into the cavity between the RV and the primary shield wall, and pressure wavs propagation inside the RV due to the break in the reactor coolant pressure bounda ry.

Core Bounce The vertical response of the reactor internals and Fuel i Assemblies (FA) result in a time varying force composed of the structural response to differential pressures.

Core bounce is the terminology given to this response phenomina. The nonlinear structural response reflecting holddown springs and vertical gaps is calculated in a decoupled analysis. The FA core and reactor internals l rp1280-3036a 112 ,

l l

1

, - . - - - - . , - . . . _ . - . . , . , - . , . . - , . . . . - - . , _ . _ _ . _ , . _ , , . _ . . .. ., , ..r ,

8 are simulated with a planar model const. 'ing of beam elements, nonlinear axial springs, and it Ted marses.

The ANSYS code is used to calculate the vertic:41 reactions of the core, which are then used as applied force time histories on the reactor vessel in the system f

dynamic analysis. The core bounce LOCA forcing functions are the result of the worst case possible double end guillotine pipe breaks at the RV nozzel'.

Thermal Hydraulics and Linear Dynamic Response The pressure waves through the RV produce several r

l reactions that are not considered in the core bounce forcing functions and which can be applied directly to a linear dynamic system.

For the reactor vessel, the horizontal pressure gradient results in horizontal forces on the RV, core support cylinder, thermal shield, and the plenum cylinder. The vertical gradient results in vertical forces on the RV.

l The integration of the pressure-time history defines time history forces which are applied to discrete mass joints of the mathematical model.

The thermal hydraulic loadings applied directly to the linear dynamic model are the result of a hot leg pipe rupture and a cold leg rupture.

. rp1280-0036a112 1

9 Asymmetric Cavity Pressures Pipe ruptures which occur in the cavity between the RV and the wall result in differential pressures across the RV in a time varying manner. The differential pressures, when integrated across the area of the RV, produce time varying forces which are applied to discrete mass joints on the RV. The cavity pressure loadings on the RV for these analyses are produced by the Architect-Engineer end are the result of mass and energy data from single ended pipe guillotine ruptures.

3.1.4.3 COMPUTER CODES USED FOR NSS^. ANALYSIS The two analytical computer programs and the four data reduction codes used in the seismic and/or LOCA analyses for the support design loads are described herein.

Structural Analysis Codes l 1. HYDROE - A comput.er code used in calculating the l

hydrodynamic mass coupling of concentric cylinders.

2. STALUM - A coaputer program for analyzing three-dimensional, finite segment systems consisting of uniform or nonuniform bar/ piping segments, closed-j loop arrangements, and supporting elements. STALUM l

performs both static and dynamic structural analyses undergoing small linear, elastic deformacions. The l

l rp1280-0036a112

10 static analysis is based on the matrix displacement method. The static loadings are static mechanical forces, thermal, and/or support displacement loadings. The dynamic analysis is based on lumped-mass and normal-mode extraction techniques. The dynamic input loadings can be response spectra or time history forcing functions.

The essential input to the program consists of the physical properties of the system, the boundtry conditions, and/or the loading inforsation; the essential output consists of the resultant joint displacements, rotations, forces, momenos at both ends of each segment, and stresses at various locations in each segment.

Data Reduction Codes

1. FTRAN - A computer code used for Fourier analysis of t forcing functions to determine the frequency content of the forcing function.

2. S1235 - A post-processor program used to tabulate forces, moments, displacements, and rotations in a specification format.

3. INTFCE - A prograc used to convert pressure-loading data to force-loading data acceptable for use by the structural analysis codes.

rp1280-0036a112

11

4. LOPL - A post-processor program used to provide time history tabulations and plots of spring forces and resulting loads and displacements.

3.1.5 SEISMIC ANALYSIS Utilizing the geometric and structural properties of the mathematical model shown in Figures 13 thru 15, and 17, the STALUM code is used to determine the structural frequencies and mode shapes of the isolated NSSS, the internal wall structures,and the NSSS supports as a coupled system. Each degree of freedom (DOF) in the model is assigned a damping value based on the location and type of component the DOF represents. Strain energy damping is used to determine a composite damping for each mode. The modal accelerations are applied to the model dynamically to reflect the structural amplification. Equivalent static forces for each mode

! are determined and applied to each DOF to give resulting modal displacements and member forces. The modal responses for each l individual earthquake will be combined, and the individual member responses will be corbined by taking the square root of the sum of the squares (SRSS) results of all six components. Figure 18 shows i

l the flow diagram for the seismic analysis.

RV_ Support Anchor Loads The seismic loads on the RV support are taken directly from the seismic analyses and are the farces and moments from the combined five earthquakes at the base of the RV skirt. These centerline rp1280-0036a112 l

l

l 1

12 a

leads are resolved int' support loads for the stress evaluation described in Section 3.2.

ULS Loads The combined five earthquake ULS load is distributed in a worst case manner to obtain a maximum load for an individual support member for which each is designed. The e <bined earthquake dynamic load on the equivalent beams representing the ULS in the mathematical model is given as the total horizontal primary shield wall load.

3.1.6 LOCA ANALYSIS 4 The geometric and structural properties of the mathematical model are used to determine the mode shapes and frequencies of the structure in the same manner as in the seismic analysis. The four sets of LOCA forcing functions are applied simultaneously to individual DOF's to represent the structural loadings to the romponents during the LOCA event. Modal displacement and member force responses are determined for each mode and the modal results are combined by direct algebraic summation. The resulting displacements and member forces and moments are stored such that time for time or peak results are available for any member or joints.

rp1280-0036a112

13 RV Support Loads The peak forces and moments, regardless of their time of occurrance, will be obtained from the time history LOCA analysis once ut, and used as the total centerline load imposed by the RV on the support.

ULS Loads The LOCA loads are determined in a fashion similar to the seismic loads. The peak LOCA horizontal dynamic load is distributed in a worst case manner to determine the peak individual ULS load for which each will be designed. The totsi horizontal force on the equivalent beams representing the ULS will be given as the maximum load on the primary shield wall.

3.1.7 PRELIMINARY DESIGN LOADS B&W has performed preliminary analysis using the upper lateral support along with a conservatively assumed zero pretension loaded anchor studs. The load cases analyzed were SSE and a B&W l identified worst case LOCA involving a het leg guillotine at the l

l E7. The analysis was done assuming the upper lateral supports in contact with the reactor pressure vessel. The loads transmitted from the RV to the support system at the RV skirt and the upper lateral support are given below.

rp1280-0036a112 l

14 RV SKIRT F F M M (kips)(1) (kips) (ft-kips)(2) (ft-kips)(1) -

SSE 114 233 147 1,646 LOCA 1,003 3,347 3,529 1,113 UPPER LATERAL SUPPORT (RADIAL LOADS)

Total Wall Load Maximum Individual ULS Load (kips) (kips)

SSE 166 55 LOCA 3,377 1,126 (1) Treated as a shearing load on the shear pins and keys provided in the RV skirt to pedestal connection.

(2) M is in effect, the overturning moment.

3.2 ANALYSIS OF THE RV SUPPORTS FOR THE FINAL LOADS 3.2.1 ANALYSIS OF ANCHOR STUDS The RV anchor stud stress analysis has assumed that the studs would resist the tensile forces in the base that result from vertical uplift forces and from overturning moments. Horizontal rp1280-0036a112

15 shears and the torsional moments are transferred from the RV skirt flange to the 5-1/2 inch thick sole plate by 48 shear pins, and then transferred by shear lugs to the concrete pedestal (See Figures 8 thru 11).

The determination of the stud stresses for the final loads will be performed by means of a finite element analysis. The finite

! element model will include the RV skirt and flange represented by shell elements, along with boundary spring elements to simulate the anchor stud tensile stiffness, compressive stiffness of the concrete, and the shear pins embeded in the sole plate. The broken studs in Unit I will be accounted for by omitting the cension boundary springs at their corresponding node point locations. The reactions from the vertical uplift forces, overturning moments, and horizontal shears will be resolved into discrete nodal loads at the top of the RV skirt model. The applied reaction forces will be oriented such that the naximum tensile stresses in the studs will occur in the neighborhood of the broken studs. The stud prestress forces will be simulated by equivalent compressive forces applied to the base nodes representing the stud locations.

The finite element program being used to assess the stud stresses produces only linear solutions. The analysis will require a number of iterations to achieve a balanced solution. The analysis will be initiated with the neutral axis coinciding with the geometric center of the RV skirt flange. After the. loads are rp1280-0036a112

16 applied, the ~ ..y node point stresses will be checked to verify that neither the studs have exceeded their prestress force nor the concrete bearing stress has exceeded its allowable. If either of these conditions are not true, then adjustments will be made to the position of the neutral axis by either declaring more studs with tension loads above the prestress, and/or smaller areas of concrete capable of resisting bearing loads. This iterative process will be continued until the boundary stresses are balanced.

3.2.2 ANALYSIS OF UPPER LATERAL SUPPORTS The ULS bracket analysis assumes that the bracket would resist both the compressive loads from seismic and LOCA forces on the RPV and the bending loads from upwaro pressurization of the shield plugs. The preliminary analysis indicates that the maximum anticipated pressurization load applied to the brackets in addition to the preliminary axial load produce stresses well within the range of allowable stresses. The trapezoidal ULS bracket will be assessed by taking sectional properties at several locations along the length. The allowable yield stress for the steel will be reduced at each section to account for the higher temperature according to the AISC 1971 code edition, that till occur from having the bumper in contact with the RPV. The thermal analysis assumes an RPV surface temperature of 530 F and 16,000 cfm airflow at 130 F. The results of the temperature analysis of the ULS indicate that the exposed edge of the primary deeld wall rp1280-0036a112

17 (Point A on Figure 6) will be 248 F, and the concrete behind the ULS embedment (Point B on Figure 6) will be heated to 159 F.

The material used to strengthen the bracket, according to the current preliminary design, will be the same material used to fabricate the bracket, which is ASTM A516 Grade 70 steel. The shim material will be ASTM A240 Type XM-19 stainless steel.

4.0 STATUS OF ANALYSIS AND DESIGN 4.1 FINAL SUPPORT LOAD GENERATION The analysis by B&W incorporating the final mathematical representation of the modified boundary conditions to simulate the ULS and the reduced stud prestressing is ic progress. Results verifying the design will be submitted to the NRL upon completion of the analyses.

4.2 ULS DESIGN AND FABRICATION The preliminary upper lateral support design has been completed and the structural drawings are being prepared to proe;re the material and proceed with fabrication.

l The final design of the ULS has not started but will begin shortly after the loads have been developed. It has been anticipated that the final loads will be less than the capacity of the bracket since their design is based upon a conservatively estimated set of preliminary loads.

i The existing brackets, which will be a part of the ULS design, were originally designed to support the cavity annular shield plug at El 632'.

The layout and details of the ULS brackets are shown in Figures 3 thru 7.

rp1280-0036a112

18 As shown in Figures 5 and 6, the additional stiffness required by the ULS will be obtained by adding steel plates to the bottom flange and to either side of the top flenge. The clear distance between the brackets and the RV varies between 1-1/4 and 6-1/2 inches. This gap will be shimmed tight with both the RV and the ULS in the hot operating condition. A shimming procedure is currently under development to measure the thermal displacements of the ULS and RV in order to establish the required shimming distance. A method of measuring the change in the gap between the Reactor Pressure Vessel (RPV) and the bracket end that will work in the extreme environmental conditions of the hot functional test is being developed for use.

5.0 STUD DETENSIONING STATUS The Unit 1 studs were detensioned in order to preclude further failures and are currently at a nominal stress level of about 6 ksi as recommended in TES Report TR-3887-2, Rev 1 ( Reference 1). The detensioning procedure is also being evaluated to ensure that the limits of accuracy of the measured stud stress levels are compatible with the criteria of Reference 1.

The stud detensioning procedure that was used required that the liftoff values be recorded. These values are shown in Table 1 and Figure 2, and exhibit a certain amount of scatter. A consultant specializing _n the field of tensioning behavior and tensioning systems is being retained to establish the possible reasons for this scatter as well as to comment on the procedure used to tension and detension the studs to assure that the 6 ksi prestress design allowable will not be exceeded. The recommended rp1280-0036a112

}

19 criteria for establishing an allowable short-term stress was established in TR-3887-2, Rev 1 (Reference 1) and included in Section 3.2.2 of the first report of this subject. With this criteria in mind, the studs that had a recorded liftoff stress of less than 75 ksi were proof-test tensioned to 75 ksi for several minutes so that a value of half of the indicated tensile load, 37.5 ksi, could be used as an allowable short-term stress.

6.0 CONCLUSION

This report has described in detail the modeling techniques being used in the analyses of the modified reactor vessel support system for the Midland Nuclear Power Station. These methods represent the standard techniques r*ilized by the NSS suppliers for primary system analysis under the ia fous design conditions. The design modification is mandatory for Unit 1 because of the anchor stud failures experienced.

Based on the investigations conducted, the Company originally recommended using the Unit 2 reactor vessel support design in its original condition although this matter is still under review with the NRC staff. However, it is the Company's intent if practicable to modify the Unit 2 design with upper lateral supports to be similar to the Unit I support design.

Analyses for Unit 2 wit; aer lateral supports will also be carried out using the techniques described in this report with the appropriate changes being made to the input data to properly represent the Unit 2 configuration.

This report provides information regarding the detailed analytical techniques which fulfill the Company's understanding of the material rp1280-0036a112

19a necessary for final NRC review and concurrence of the reactor vessel support design modification concept. The design of the upper lateral supports has proceeded using preliminary design loads as described in the report. The supports are conservatively designed with respect to these preliminary loads and will be able to wittstand loads in excess of those anticipated from tse final analyses. The confirmation of the adequacy of the design will be made upon receipt of the final support loads.

Appropriate status reports and final analytical results will be submitted in the future to document the completion of the detailed design.

T

..-l-rp1280-0036a112

20

7.0 REFERENCES

1. Teledyne Engineering Services Report, TR-3887-2, Rev 1,

" Acceptability for Service of Midland RPV Anchor Studs," May 20, 1980.

2. BAW 1621 B&W 177-FA Owners Group, " Effects of Asymmetric LOCA Loadings", Phase II Analysis, July 1980.

3. Consumers Power Company, " Reactor Pressure Vessel Support Modification for Midland Nuclear Power Plant, Midland, Michigan, Preliminary Report No 1," July 1980.

i i

l l

rp1280-0036a112

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

Midland Plant Units 1 and 2 21 RPV Support Modification TABLE 1 DETENSIONING DATA UNIT 1 REACTOR VESSEL ANCHOR STUDS Hydraulic Stud Number (2) Pressure Bolt Stress Sequence B&W ~ Teledyne Date (psig) I to Nea rest ksi 1 01 in 37 in 4-03 13,000 88 2 02 in 13 in 4-23 11,900 81 3 03 in 01 in 4-25 13,400 91 4 04 in 25 in 5-19** 9,300 63*

5 01 out 37 out 8,000 54*

6 02 out 13 out 12,500 85 7 03 out 01 out 10,800 73*

8 04 out 25 out 5-12 8,400 57

+

9 05 out 43 out 5-13 12,500 85 i 10 06 out 19 out 5-13 12,500 85 11 07 out 07 out 5-13 13,400 91 12 08 out 31 out 5-14 13,800 94 l 13 05 in 43 in 5-14 12,300 83 f

I

'14 06 in 19 in 5-14 11,500 78 I

15 07 in 07 in 5-15 12,000 81 j 16 08 in 31 in 5-15 11,400- 77 f 17 09 in 40 in 5-16 12,300 83  !

18 10 in 16 in 5-16 11,700 79 f 19 11 in 04 in 5-19 13,700 93 20 12 in 28 in 5-19 12,400 84 21 09 out 40 out 5-20 12,200 83 22 10 out 16 out 5-20 12,500 85 rp1280-0036a112

. _ = . .

Midland Plant Units 1 and 2 22 RPV Support Modification TABLE _1 (Continued)

Hydraulic Stud Number (2) Pressure Colt Stress Seguence B&W Teledyne Date (psig) I to Nearest ksi 23 11 out 04 out 5-20 13,000 88

.4 12 out 28 out 5-21 12,300 83 1

25 13 out 46 out 5-21 12,800 87 26 14 out 22 out 5-21 11,500 78 27 15 out 10 out 5-21 12,300 83 28 16 out 34 out 5-22 12,600 85 29 13 in 46.in 5-22 11,100 75 30 14 in 22 in 5-22 12,100 82 31 15 in 10 in 5-23 9,300 63*

32 16 in 34 in 5-23 13,100 89 33 17 in 38 in 5-23 11,600 79 34 18 in 14 in 5-27 9,500 64*

35 19 in 02 in 5-27 13,300 90 36 20 in 26 in 5-27 9,600 65*

37 17 out 38 out 5-28 12,500 85 38 18 out 14 out 5-28 12,300 83 39 19 out 02 out 5-29 14,000 95 1

40 20 out 26 out 5-29 12,100 82 41 21 out 44 out 5-30 12,200 83 42 22 out 20 out 5-30 12,300 83 43 23 out 08 out 6-17 12,300 83 rp1280-0036a112 '

l 1

l

Midland Plant Units 1 and 2 23 RPV Support Modification TABLE 1 (Continued)

Hydraulic Stud Number (2) Pressure Bolt Stress Sequence B&W Teledyne Date (psig) I to Nearest ksi 44 24 out 32 out 6-18 12,300 83 45 21 in 44 in 6-18 12,800 87 46 22 in 2G in 6-18 10,900 74*

47 23 in 08 in 6-19 12,300 83 48 24 in 32 in 6-19 12,400 84 49 25 in 41 in 6-20 12,200 83 50 26 in 17 in 6-20 11,800 80 51 27 in 05 in 6-20 13,000 88 1

52 28 in 29 in 6-23 12,800 87 53 25 out 41 out 6-23 . 12,500 85 54 26 out 17 out 6-24 12,700 86 55 27 out 05 out 6-24 8,900 60*

56 28 out 29 out 6-25 12,500 85 57 29 out 47 out 6-25 10,200 69 j 58 30 out 23 out 6-25 12,200 83 59 31 out 11 out 6-26 12,200 83 60 32 out 35 out BR0 KEN 61 29 in 47 in 6-26 11,900 81 62 30 in 23 in 6-27 12,400 84 63 31 oin 11 in 6-27 11,800 80 64 32 in 35 in 6-27 11,600 79

< 65 33 in 39 in 7-02 11,700 79 l

rp1280-0036a112 1

a 4 Midland Plant Un'its 1 and 2 24 RPV Support Modification TABLE 1 (Continued)

Hydraulic Stud Number (2) Pressure Bolt Stress Sequence B&W Teledyne Date (psig) I to Nearest ksi 66 34 in 15 in 7-02 11,700 79 67 35 in 03 in BR0 KEN 68 36 in 27 in 7-03 12,300 83 69 33 out 39 out 7-03 12,100 82 70 34 out 15 out 7-03 12,300 83 71 35 out 03 out 7-07 12,000 81 72 36 out 27 out 7-07 10,300 70*

73 37 out 45 out 7-G7 12,600 85 74 38 out 21 out 7-08 12,500 85 75 39 out 09 out 7-08 12,200 83 76 40 out 33 out 7-08 13,600 92 77 37 in 45 in 7-09 13,000 88 78 38 in 21 in 7-09 11,500 78 79 39 in 09 in 7-09 12,200 83 80 40 in 33 in 7-10 13,200 90 81 41 in 42 in 7-10 11,800 80 82 42 in 18 in 7-10 12,500 85 83 43 in 06 in 7-11 10,200 69*

84 44 in 30 in 7-11 12,300 83 85 41 out 42 out 7-11 12,200 83 86 42 out 18 out 7-14 10,400 71*

87 43 out 06 out 7-14 11,800 80 i

rp1280-0036a112 I

Midland Plant Units 1 and 2 25 RPV Support Modification TABLE 1 (Continued)

Hydraulic-Stud Number (2) _

Pressure Bolt Stress Sequence B&W Teledyne Date (psig) I to Nearest ksi 88 44 out 30 out 7-14 11,700 79 i 89 45 out 48 out 7-15 13,100 89 90 46 out 24 out 7-15 10,400 71*

4 31 47 out 12 out 7-15 11,700 79 92 48 out 36 out BR0 KEN 93 45 in 48 in 7-16 12,500 85 94 46 in 24 in 7-16 11,900 81

, 95 47 in 12 in 7-16 12,100 82 96 48 in 36 in 7-17 11,700 79 NOTES:

1

1) Ram area of tensioner = 27.134 sq in, bolt area = 4.00'sq in.

! 2) Refer to Figure 1 of Reference 3 for the locations of the studs.

• ) Proof loaded to 75 ksi after detensioning.

• • ) Tensioner run up to 14,200 psig/96 ksi on initial attempt without being able to rotate nut. Lift-off data shown are results of detensioning attempt after 20th in sequence.

l i

i

! rp1280-0036a112

SEE FIGURE 6 C

i-

,[Tl ip

b@,y@-

JJ ( .g

- ==

r- , LATERAL 'a *i , l 2

sSueeORT  :

[,N.

b - ,

'I

~

O! fdM.

X) T - \ - -

% b Q

i>~ -

yh . .ty C.G.

yp (n

C A Py% e9

. y. i.-

. ~

L m .

~0 h 0 - E *i n.: ' -

JD .a

~~I .i.?$f  ?'

If.9?/

O ROTATIONAL ROTATIONAL I/5 E t l

M 4

SPRING RATE K2 SEE FIGtIRE 8 o.,

SPRING RATE K1 df rn -- - - - - - ( .

m (qq w_ .,3 w

7 _ .. g

~0  ? RQ r l'

d  :. $h?il 0, ru f b's 9

Ei ORIGINAL NEW

1. • g:/.N' O8

g

m:

PROPOSED Yi3;?i g

\

M '

LSFE FIGURE 9 K1 >> K2 a 10T T 04

~

, MDLAND JOB 7220 UNIT-1 i LIFT-OFF VALUES DURING DETENSIONING APRIL 81980 TO JULY 171980 TOTAL: 93 STUDS i

a _ .

1 i C b I ,1

, b x $ U

am - m l v3

g e - -

Pn u -

o O e o -

e o o

2 , -e B k g o o b o o o 8 8 8 5 m

O G .

= Bs

i n N I

, G.)

W U CL REACTOR ELDG CONTAINMENT I SHELL WALL

>p 4 !O"( TYP) RREACTOR -

~

Zr -

VESSEL I . r,

~

D > FIG. 6 PRIMARY SHIELI) '

/ -

Z .

A ( WALL, _

mO I g

.g.

g _.,

\f

/

, .j3,w.

.1:P. ~ N 'W e i I

_: ,/,$ ~

R -~ I 4 b  : / ,e L\ \ l

/

b>so g i e

em /

'6 s I

W FIG. 7 [ [S e.+ ) [

_ _4  ! 4 "l _

! . - -i _

O

' i i i j

FL EL 63/JO" M FL E 634'- 0" l \

\-

i

/ FL EL 619'6* )

% / f f

\ bl '

/ I .

.d 'i..':

h{

r\ ~ .. .a-O, l

. . .. '!)-{

\ '

f ^ SECONDARY -REFUELING g _.

S-IlELD WALL C ANAL WALL

f.

m p 3% O'

4. UPPER LATERAL SUPPORT x '

s 17*

(TYP) y

\ ~

,/'~

~~

"N v

,'! ;~ r

^

270i V7 I k'90

,d

+ -

3 i

,W R.V EDGE ,'

' i

(% ~,.Q

^

l N&

A

/

\ ,

l CONCRETE /'

SHIELD PLUG ,

- FOR DETAll l

PRIMARY SHIELD _

/ SEE FIGURE S l

WALL igo l

UPPER LATERAL SUPPORT PLAN l

FIGURE 4 l

29 l

t

REACTOR PRESSURE VESSEL UPPER LATERAL SUPPORT FIGURE 5 I

E /2' x 5' x 2'- 1' 1/4' STIFE E's.

( A 516 GRADE 70)

I E 2 /2' x 9'x 2'-1' Q(C_7{.)

%7

+ \/ 7%

/\ 4

/\f  % l

( A516 GRADE 70)S 4

1I/16 6 HOLES FOR

% v17s///s y xxxxNQ + ,

4 5/8 % BOLTS (A-325) D"M > S I

E II2'x 6 /2'x 5' E 1/4'x 1

12' y TYR BOTH SIDES j

( A 516 GRADE 70) g TYP.

• NOTE: PLATES NOT CALLED OUT ARE EXIST. STEEL

[

SECT!ON /M aspO/ ,,

W

< m i .

STAIN 6ESS STEEL E E l'x .9'x i-laTYP) y _9 [ 12" x 5 ( A 240 )

STAINLESS STEEL S <

( A2.40) p p 9 ,y "

8 I a -

A gj~ g

a g W T mim- "'"

l8 "

SHIM MATERIAL STAINLESS STEEL D $ i-

,t' "N'

II II (A 240) 2 .

i lI

< Il

' QG. '6 TYP\ y '

7 PLAN AT EL. 63 2'-3" 1 .- - -

30

l l

l l l

l

~. l

4 '

s.i

• EDGE .OF REACTOR

9. f-8 . VESSEL I

I 3'- 6' VARIES : 1!'4'-2/2' I 't!;.6-

~

l r, ~, EL. 632'- 3'  ;

I r--- - = i o i g i

!.f-- -

u-- = i i

!l ll ll ll ll 11 ll1 il

,--- 7k'

~. 9 6

_i I

ll si i p a -

,, ii i 5 I

p. a g_ y .e = --

a

,_ g ., , l

, C- -3 pq , ~ ~L f i+ W z"j  !

p % /;;

l 8

[,

. CONCRETE NOT POURED YET

@p_______-_t_

, r - ' ,

q 8i Jie. .,< 4 .

= =.

VARIES : 5"- 6 l/4" (EXIST. STEEL EMBED (SEE FIG. 7 )

.SECTION (D ,

QiG.'6 -

UPPER LATERAL SUPPORT DETAll:

FIGURE 6 I

l l l .NEW ADDITIONAL STEEL- 1k4'fP l

( TYP) A 516 GRADE 70 l _

l'- 2' __

SECTION @

D 31

FACE OF CONC. ~

TO. REFU ELING i C ANA I

EL.6L34-OFLP . '//

" 1 o .- - 14~

.. a a I

l,

" ~'

BARldxI- 4 -

[4] 5 BARS 1/2 x h x 2'-1'

-I' (TYP.) .e -

64

h. 7%' W my me C -

x .g;- -

a a \ =~jf EL. 63 2'-3's 1d

~~,

c ']

> 3v fk.1I4

~

>, x ij - -

a s -

f I .

o w

o

/  % -

x  : o y-_- 'f*1 1 .

.1 o to

, a-

[

MN w o

y a

,  ;- to - e .

Q *3 %, / v a-Ik.7/8 '? i. ,gl .

'd

. . .# g.

\ \

TYP*/ pg /  ; 3f4'p xs' NELSON

1. W{

~

4 4 STUDS OR EQUAL l

9 ,' e.i, a 1 74 _ NOTE 1_Tl , , , ,

1: k. 2P4 x 25x 3-0

.0-bs f

~

y l ,,

a

\- a ,m 0=__ _g%/ % .- - a o

," m,,g -

0- --

2 r r_ _.=_ - - -

j v" 3 -

4 m

w 0:==:k J, a.

,sy o

, ,- i -

y ...

1 _ .. .

, 5h. .'

L*

FACE OF C ONC.

NOTE 1 THIS DIMENSION l'-O'FOR

~

TOP ANCHOF l'-6 FOR BO T. ANCHOR GURE 7 32

dtzscrse 16:5EL o 0 .

asa w4s g 6 HEAR PMS dEE F/4 11 is' I

/#*

EEAcror Sncr -

t kc.7 m, . .

7 X,

:. 93 270 + o

= = - -

~

, n  ; ,;*' r ,

g, f,0 +

1. 6, Ik

%Y Q Q% C g* k %l9,. .': .o'

}

g; j

- e s ..

2  %;~t p

1 1 s MTE2AL A *l l go'E 4 ArtEAL A "2 .

I8 O*

~

PLAN AT EL. 603'-l" F GURE 8

. pj@@@$$$0Ab

C REAC. TOR SKIRT I

i q .,

M t 18 ADD L n 6 s g i s Jh

'(mmN Di ie ArssuD

g/,/f//ff, . ,T m ur o w .s r n s,- e .. a w.

p . .* . 2 ._.5 , i :

- tm_ . . _

l s

1- -

n s/

l;l .t l

l,li o e ,, i, , m

' jL f I l i ! I i \9 ii If Ii .

b I i i ;

i t s i u

l<; :l Se =

q*' i I' ,

',e

.! o-

l

'E o/

~

ll! r > ie

__ , I ee D 4mp c2

, 6d = 5K9D ll ll  :-

3 l  : ' ~

' ,l - - - - - _ _ _ __

~!

p . .. . . 1. e- .

3

'g . . . . _ , ,2 O, ,

, i l i e- a

\b

+._

• g

- S~ g

<~ g S

e r

ss CONCRETE PEDESTAL DET.

m.m.

' - 3h

~

~ ~

R = 7 '-4 34 " REACTOR SEMI-FIN HEAW "Q_ SKIRT HEX JAM 8 NUT HEAVY HEX NUT PLAIN WASHER

_ ( HARDENED )

108 6 , = WASHER 1" THK I

2%Vj6 ID x 5' OD L

im i r

-- es-. 3" LATE _ P._ '

m ASTM

~i C (-"l

+-+

_ A 36

}

l

= s 1 L ..

y I l \\\ / STN 3 m xu = e ij Z '

=o x

\\

% =

z x ll 11 n

T s

~i

/ W A \ l

/ 7 - _

\_ - _

(-

9yga -

4 9

(SOLER. ) '

s, 8 = ASTM A 36 t)

N $

\ 4 i

4 -

i l*

4x14;2 3 1 6" 2Mp ANCHOR STUD SHEAR LUG ASTM A 354 GR.BD ASTM A 36 , _ , _

9 # ~

9 I P 3Yx 17

~

g ASTM A 36

'LE ss ~I T

3 Yb

.=--

IIN -

_ HEAW HEX NUT l

SEMI- FIN HEAVY HEX J AM B NUT FIG.10 ANCHOR STUD DETAIL 35

DRILL 1Mt HOLE AND

- R= 7' -4 743" -

M TO 22 M "4 IN SOLE & .015 E. REACTOR SKIRT

'D 6" .

"h ,

1-l

, 2"4 7.. r.:i .5* ,*,#.'-

lM\\h/I'i'. d 2i ' i.

TO.C. EL /

/' -

/nj >(

b'l.8 a-*

f j @ .

),3@, i.'

• M t

602'-31 /2"g  ?.:

v..w-u n.sa a.s-x s.e c,:...n.m T

@ s a

a

>.'

3ft 1/8- *, . ..,

.' en . > .

[, , ; _ >* , ? l 1

.. .~ -

.? o. .v[ * > > _ . .

DRLL 1% HOLE FOR REAMER P! LOT BAR

2"4

+O \

-0.01/

'1 L

e

,5 o M o o 45 CHAMFER

~

U [. "

l  !

1/16" 1 3

.6

\(2 & x O.48-REQ'D) d6 ASTM A 354 GRADE BD SHEAR PIN DETAIL FIGURE 11 36

FIGURE 12 Reactor Coolant System Boundaries

[K] -

[K]

e 1

I i

l

[K D w Y

@ i

'$ U

[K)

( i l [K] vg 1

w D-I

[K] Nt.s I

I

~

[KJ l

l l

[K] = Stiffness matrix 37

,-,--e , m , , p- y n - -- ,.y -emy ym e- ,m ---n,---g ,,-y-- e,,- ,-~ , - . , ,w---n--- - --, w - - ~ -

FIGURE 13 RV Isolated Model, Reactor Internals and SSS e

M _Q 05 3 r--- :m m L] ?43xl IS m2 T r-- 7 N

- ', >3

• F' I l e

1 I

e ;- - - - . -

c.wem - i N l em - Ik b l t tem i

! S I fD '. SR j ,,

y' e s N i

set. De W L A e- / _. m

'N""

an.coe esee.- l @ .g

, m_- -m 8  % -

=.- ceu a "W

_%sg=ee W e.! 3.a ___IS. a% g E %* g A ==

-J.9 "2

_, c

, o-.

wgk*

E I ib - e[s:

pa g" 1,

- u. ,g ,.,

g ..

x ,fg "w,e Y P "2 l i w ., j

- .%,W.n%,T* - ,l sm

-- e. ,~f- ,

3 3 l =

K)F  %,-

M b

%o /

Ne m '

.r_

&x c-_.

c --' ' -

sib c_.~,

@ uts no 4

3 es ,

• l 0

[L ,

s Nr Ir eI

/ewNh

~ (nA u

~/

.A I l s* );j . \/ '

\\ I

\\i Ngi "

'/

I'll /N2 \ \._ / )

i .

8

1. i

@ :N/ f Wg*j j i

a l n

dm

=

%ri-z - r . ;e E'=,.id g i 3 e

o :_

s.. @J rir, 2 ,

Ie m.a .

o.

sa 1 - ,_ :P j

d; ' N%f% 155:.9f..xg)'c  %

m

- @2ch. / /, f-=~. mm -?v

mm;)./m *.)

v \ J, c.

O s pg - k c.

s - =e}- y = -

s ' -

k

(p_, 1g 8 9= gisq =3-2

- Q E 2:c 09, , &

/ $ [ ?k~ /, $ 2_

• 5

• \ .

. a: E g5 bys '

km

- z ...(.} -

,# 1. e __

D ' V. >> ]

g'$ o lif.) de $?> [

b h me r;e,a cgr% e

,a 2

g 73 j/m ,,y g ((y_s=l  ; y 1

(\ "/V f e /a

\\

~

/)f 2 (C._ M // \'._@

Y el@

[\ :': \ - / --

\

c00 e =

' -- ' / -

is si .

!s1s!

5 :

5 l k ;i b E:

E i

c.

d 39

FIGURE 15 RV Isolated Model, Elevation View A-A, Hot Leg

-% c s=

\ {

< si i h-a i p i M, i i e s

/-

\lit '

i l li[- -

\

i , ,

! 7 .: . # 1 I

I I View A-A 1  % i I l 1 I I I I l 1 I l

l N i l

i l I ofl a s ii I i A i 1 ,

I I i l' I I

I I C

[ ____K -

l i

i i i

• T-il y e H- -

ses 3'I M y$ $ _.

a yb__ug~%=]

i 1

I 7- ==q xs/

I O_

I c=

3 ---.,_. , _.,

l -'

I

. Ld 1

FIGURE 16 Reactor Internals and Service Support Structure d

I i

CONTROL ROD

[ GRIVE it

[5 L UE

. O d' c,

/

[r \

- n i

CONTROL R00

' - GUIDE TyBE l

' (CCLUMN WELCMENT)

-l ./

i PLENUM COVER p,

Q

/ -

f PLENUM ASSEMBLY Q[ '}

1 h/ o .

UPPER / lf h' CORE SUPPORT GRIO [; i CYLINDER FUEL l, REACTOR CORE

[ (~w u

L h iHE W L LOWER GRID \ {

!i j SH.IELO 1

s( '

2.q GUIDE LUG

- /r SUPP0 T FORGING _

DISTRI UTOR

, ( N SKIRT SUPPORT

_N i

kl 1

aos g<

FIGURE 17 D TERNAL WALL STRUCn mES 1X i & O - u ses '-o* /-  % s

\

/

/ N E [ + nra

\

/ \

I \

@ D - a s a g '- o " m

\ g S =

e i z

i g

/

~

@ 3 - a *

• o '- o \ ,, m ,I

\ /

E \ /

s CiR o - u eai .sj ~(av auwea)N

, /

L&C REACTOR VESS EL. B UMPER R D - E ' 6 2 6 '- o " ELE VAT /ON m

i W n - s = s o s '- o '

N

>777 - EL .sa7-o "

l /fo 75 i

l 42

--W

FIGURE 18 Utili:ation of Computer Programs

~

DEVELOPMENT OF HYOROMASS HYORGE MASS MATRIX STALUM DEVELOPMENT OF GE0M SilFFNESS &

MODULE FLEXI91LITY MATR!CS e

STALUM FREQUENCIES &

LUMP MODE SHAPES MODULE i

i STALUM LUMP EQUIVALENT STATIC FORCES NODULE STALU'd RESULTANT LOADS RSTA (DETERMINE 0 MODULE STAIlCALLY) l l

l h3

. . . . . . . - . . . - . . - - . - - .. . . . . ~ .. ..-. ~.

. . . . . ~

Enclosure 2

~~

Serial 9787 Bechtel Power Corporation

SUBJECT:

MCAR 37 (issued 12/28/79)

Broken Reactor Vessel Anchor Stud in Unit 1 INTERIM REPORT 4 DATE: November 5, 1980 PROJECT: Consumers Power Company Midland Plant Units 1 and 2 Bec~ntel Job 7220 Introduction The discrepancies discussed in this report concern the failed reactor vessel (RV) anchor studs in Unit 1.

Background

The anchor studs in question are 2-1/2 inches in diameter and 7 feet, 4 inches long, embedded in the reinforced concrete RV pedestal. The anchor studs were purchased from Mississippi Valley Structural Steel of St. Louis, Missouri; fabricated by Southern Bolt and Fastener of

, Shreveport, Louisiana; and neat-treated by J.W. Rex of Lansdale, Pennsyl-vania. These studs were received on site by Bechtel in early 1976; embedded in concrete by Bechtel in April 1977; and tensioned by Babcock

& Wilcox Construction Company in late July 1979. The first stud failure was discovered on September 14, 1979. The second and third stud failures were reported on December 20, 1979, and February 5,1980, respectively.

Investigative Action Teledyne Engineering Services' (TES) investigation for Bechtel is complete. The resulting reports discuss the stud failure investigation and the use of the present studs for service. Consumers Power Company and TES are currently investigating the root cause for the excessive hardness of the studs. Bechtel and Consumers Power Company are in the process of retaining a consultant in bolt tensioning to evaluate the tensioning procedure and explain the scatter of lif t-off values that occurred during detensioning of the Unit I studs.

Bechtel has calculated stresses in the studs and upper lateral support brackets based on conservative preliminary loads provided by B&W for the accident condition of a combined seismic and loss-of-coolant accident event.

Bechtel has found those stresses.co be within the allowable range.

, c. .

Bechtel Power Corporation MCAR 37 Interim Report 4 015p.g Page 2 Corrective Action The prestress levels of the Unit I studs have been lowered to 6 ksi.

The lif t-off values, recorded for these studs during detensioning, are shown in Figure 1. The studs that lifted off at a stress of less than 75 ksi were proof-test tensioned to 75 kai so .* minimum value of 37.5 ksi could be used as an allowable short-term stress.

Reactor Pressure Vessel Support Modification for Midland Nuclear Power Plant, Midland, Michigan, Preliminary Report No. 1, July 1980, was transmitted to Region III by Serial 9330 on July 24, 1980. Report No. 2, which provides the analytical techniques for design, is currently being

, prepared and will be transmitted by the end of October 1980.

Safety Implications If uncorrected, this deficiency could adversely affect the safety of operation of the Midland plant at any time throughout the plant's expected life.

Reportability This condition was reported to the NRC by Consumers Power Company under 10 CFR 50.55(e) on September 14, 1979.

Submitted by: O$ N $**1a k OT Approved by:,.[# M 7k -  % I-.u.cu aru Concurrence by: '

. BD/CB/sg /

/

Attachment:

Figure 1 f

1 l

l l

, 9- -- -- -

~~

GAUGE TENSILE t PRESSJRE STRESS o FIGORE ,I (psig) (ksi) .

o a . .

3-3-15,000 o

>]

a>

<-100 O

4 m .

N D D b C-14,000 W

hO c

-95 83 0

zOM D a

C -

g I.L G D- --90 Z h$jj ~~

13,000

' :'u_fg E o_ $

. . .y: -

Q '

--85 1 J o <C 7 _

.,L

• t~..

o - ..

-12,000

' _~

i ,

- - , . ~80

. i.

[ -

L_

11,000 -75 L -

, p. .99

-10,000 .

[ -65

-9,000 i r.D

r

8,000 -55

-50 7 000 16 '14 12 10 8 6 4 2 O NUMBER OF STUDS *

- x st.us.s 4de *T GEEN V E m Gt E D W"I E b C#N

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

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