ML20205N857

Masonry Wall Design, Technical Evaluation Rept for Pilgrim Nuclear Generating Station 1

ML20205N857
Person / Time
Site:	Pilgrim
Issue date:	05/14/1986
From:	Con V, Pandey S CALSPAN CORP.
To:	NRC
Shared Package
ML20205N862	List: ML20205N857 ML20206D122 ML20206D174
References
CON-NRC-03-81-130, CON-NRC-3-81-130 IEB-80-11, TAC-42885, TER-C5506-158, NUDOCS 8605160275
Download: ML20205N857 (57)

Text

. _ __ _ _ _ _ _ _ _ _

TECHNICAL EVALUATION REPORT NRC DOCKET NO. 50-293 FRC PROJECT C5506 NRCTAC NO 42885 FRC ASSIGNMENT 6 NRC CONTRACT NO. NRC 03-81-130 FRCTASK 158 MASONRY WALL DESIGN BOSTON EDISON COMPANY PILGRIM NUCLEAR GENERATING STATION UNIT 1 TER-C5506-158 Preparedfor Nuclear Regulatory Commission FRC Group Leader: V. N. Con

, Washington, D.C. 20555 NRC Lead Engineer: N. C. Chokshi May 14, 1986 This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Govemment nor any agency thereof, or any of their employees, makes any wettanty, expressed or implied, or assumes any legaf liability or responsibility for any third party's use, or the results of such use, of any information, appa-retus, product or process disclosed in this report, or represents that its use by such third party would not infringe privately owned rights.

Prepared by: Reviewed by: Approved by:

Va Utet W Principalkuthor

.4 A. S L A iL der l V -

p for DepartmentDirector I Date: I ~IN -Ib Date: EN4[0h Date- f//W8s l

PRANKLIN RESEARCH CENTER 1 ONISION OF AmRN/CALSpAN '

som a nacs steem maaommaa.pn ses

?h GS Kc{/#7 ~a yk l

l TER-C5506-158 j 1

l CONTENTS Section Title Page 1 INTRODUCTION . . . . . . . . . . . . . I 1.1 Purpose of Review . . . . . . . . . . . I 1.2 Generic Issue Background . . . . . . . . . 1 1.3 Plant-Specific Background . . . . . . . . . 1 2 EVALUATION CRITERIA. . . . . . . . . . . . 4

^

3 TECHNICAL EVALUATION . .. . . . . . . . . . 5 3.1 Evaluation of Licensee's Criteria . . . . . . . 5 3.2 Evaluation of Licensee's Approach to Wall Modifications . 33 4 CONCLUSIONS. . . . . . '. . . . . . . . 34 5 REFERENCES . . . . . . . . . . . . . . 36 APPENDIX A - SGEB CRITERIA FOR SAFETY-REIATED MASONRY NALL EVALUATION (DEVELOPED BY THE STRUCTURAL AND GEOTECHNICAL ENGINEERING

• s BRANCH (SGEB] OF THE NRC)

APPENDIX B - REVIEW OF THE ANALYSIS OF MASONRY WALLS IN PILGRIM STATION, BY A. A. HAMID iii

TER-C5506-158 FOREWORD This Technical Evaluation Report was prepared by Franklin Research Center under a contract with the U.S. Nuclear Regulatory Ceaunission (Office of Nuclear Reactor Regulation, Division of Operating Reactors) for technical assistance in support of NRC operating reactor licensing actions. The technical evaluation was conducted in accordance with criteria established by the NRC.

fe l

V ,

TER-C5506-158

1. INTRODUCTION 1.1 PURPOSE OF REVIEW The purpose of this review is to provide technical evaluations of licensee responses to IE Bulletin 80-11 [1]* with respect to compliance with the Nuclear Regul6 tory Commission (NRC) masonry wall criteria. In addition, if a licensee has planned repair work on masonry walls, the planned methods and procedures are to be reviewed for acceptability.

i l

1.2 GENERIC ISSUE BACKGROUND In the course of conducting inspections at the Trojan Nuclear Plant, l Portland General Electric Company determined that some concrete masonry walls did not have adequate structural strength. Further investigation indicated that the problem resulted from errors in engineering judgment, a lack of established procedures and procedural details', and inadequate design criteria. Because of the implication of similar deficiencies at other operating plants, the NRC issued IE Bulletin 80-11 on May 8, 1980.

IE Bulletin 80-11 required licensees to identify plant masonry walls and their Intended functions. Licensees were also required to present reevaluation criteria for the masonry walls with the analyses to justify those criteria.

If modifications were proposed, licensees wara to state the methods and schedules for the modifications.

1.3 PLANT-SPECIFIC BACKGROUND ,

I In response to IE Bulletin 80-11, Boston Edison Company provided the NRC with documents (July 14, 1980; November 5,1980; March 18,1981) describing the status of masonry walls at the Pilgrim Nuclear Power Station. These documents were reviewed, and a request for additional information was sent on January 7, 1982, to which the Licensee responded (March 22, 1982). On June 16 and 17, 1982, the NRC, its consultants,. and the Licensee participated in a l meeting and tour of the Pilgrim site. Several action items resulted from this

• Numbers in brackets indicate references, which are cited in Section 5.

i l

TER-C5506-158 meeting and have been addressed in the Licensee's subsequent submittal dated Septemoer 29, 1982. This submittal was reviewed, and another request for additional information was sent to the Licensee. ,

I Another meeting and site visit was held on June 6 and 7, 1984, at which the additional requests were discussed. The result of this meeting was a list of action items. The Licensee provided its responses to these items at a third meeting, which was held on July 18, 1984. Included in Licensee's July 1 18 presentation were sampls calculations for several walls and information on its statistical analysis method to determine the boundary strength. However, l all issues could not be resolved at this meeting; therefore, a final set of action items was created. Two of these items required the NRC to review the i Licensee's statistical analysis method and sample wall calculations; the other three required Licensee action. As a result of its evaluation, the NRC requested, in the letter of January 10, 1985, clarification of several items relating to the statistical analysis method and the orthotropic plate analysis l

used by the Licensee. The Licensee respond d to this request (February 21, 1985) as well as to the remaining action items (July 26, 1985). On November 21, 1985, another meeting was held with the Licensee to discuss the plate analysis method used in the analysis. Subsequent to this meeting, the Licens'ee provided additional information dated December 31, 1985 to fulfill the requirements of this meeting. 1 It is worth noting that a number of meetings and, site visits were conducted by the NRC, FRC, and FRC's consultant to discuss and resolve a number of technical issues associated with the qualification of the safety-related masonry walls in the Pilgrim plant. An extensive audit review, including a site inspection and review of calculations and design drawings, was part of these meetings.

The Licensee has reported on 242 walls at the Pilgrim plant. These walls function as shielding barriers, pressure containments, enclosures, and supports for small amounts of piping, cable trays, conduit, instrumentation, and control panels. All walls are reinforced and are either single- or multi-wythe. The materials used in construction are as follows:

l

TER-C5506-158 Concrete Block Hollow Block ASTM C90 i

Grade U-1 Heavyweight ,

i i

Solid Block ASTM C145 Grade U-1 '

Heavyweight i

Masonry Reinforcement Bars ASTM A615 Grade 40 l "DUR-O-WAL" ASTM A82 )

Heavyweight, truss type -

~

Mortar ASTM C476 Type PL Compressive strength at 28 days = 2000 psi Grout ASTM C476 g Coarse l Compressive strength at 28 days = 2000 psi j i

As a result of its reevaluation of masonry walls, the Licensee determined l that 8( walls would require structural modifications. The modifications included the addition of structural steel members across the wall face to act as intermediate supports and at the edges to assure proper boundary conditions.

i In the submittal dated December 31, 1985, the Licensee identified three {

walls in the cable spreading room (194.17, 194.21, and 194.22) as being

' unqualified for tornado depressurization loads without considerable modifica- 1 i

tions. The modifications will require a station outage. The Licensee claims j a hardship exemption on completing such modifications. Also, the Licensee {

will not qualify eight walls in the radwaste corridor (191.29, 191.37, 193.11, l 193.12, 193.5, 193.6, 193.7, and 193.8) because during the next refueling outage (RFO 7) two safety-related cables will be removed from the area of i influence of these walls, thus causing the walls to be reclassified as non-safety related. These two cables are the only safety-related components in the vicinity of these walls. The NRC staff is in the process of reviewing these issues and will issue its evaluation on this subject.

1 l

TER-C5506-158 l

2. EVALUATION CRITERIA i The basic documents used for guidance in this review were the criteria developed by the Structural and Geotechnical Engineering Branch (SGEB) of the NRC (attached as Appendix A to this report), the Uniform Building Code [2],

and ACI 531-79 [3).

The materials, testing, analysis, design, construction, and inspection of safety-related concrete masonry structures should conform to the SGEB criteria.

For operating plants, the loads and load combinations for qualifying the masonry walls should conform to the appropriate specifications in the Final Safety Analysis Report (FSAR) for the plant. Allowable stresses are specified in Reference 3, and the appropriate increase factors for abnormal and extreme environmental loads are given in the SGEB criteria (Appendix A).

'l l

TER-C3506-158 f

i 3. TECHNICAL EVALUATION 3.1 LICENSEE'S CRITERIA The Licensee's criteria are summarised below:

o The loads and load combinations are in accordance with the FSAR o Allowable stresses follow the ACI 531-79 code o Damping values used in the analysis are in accordance with Regulatory Guide 1.61 o The working stress design method was used to qualified the walls.

Two levels of analysis (plate analysis) were employed:  %

- Level 1: The natural frequency of the wall was determined using fully cracked section properties. An equivalent static analysis was performed, and a factor of 1.3 was used to account for multi-mode effects.

Level 2: If walls could not be, qualified by Level 1, Level 2 was performed in which walls were initially assumed uncracked. If cracks developed, cracked section properties would be monitored and determined as they propagated on the finite element model.

Modal analysis was performad and the results were based on the first 12 modes.

N - Level 3: A nonlinear analysis may be performed to resolve local overstresses. However, level 3 was never used.

o Shear capacity of the wall boundary was considered based on the results of a statistical analysis of a field test program in which boundary anchorages were checked for their existence. It is noted that the shear capacity of the wall boundary also relied on the shear strength (assemblage of mortar block, and grout) of the wall itself.

In this section, the Licensee's responses to requests for additional information will be given in chronological order. The requests for additional information were either based on the review of Licensee's submittals or resulted from meetings with the Licensee. Since a response to a particular question may not be complete and/or satisfactory, additional inforination was j subsequently submitted for additional review. In these cases, responses to the same question have been combined to facilitate the review and reporting process.

The following responses have been reviewed (in their chronological order):

l

TER-C5506-158 o March 22, 1982 (4) o September 29, 1982 [5]

o July 18, 1984 (6].

o February 21, 1985 [7]

o July 26, 1985 [8]

o December 31, 1985 [9].

Response Dated March 22, 1982 Following a review of the Licensee's initial responses [10, 11, 12], to IE Bulletin 80-11, the NRC requested additional information in a letter dated January 7, 1982. The following four responses were contained in the March 22, 1983 submittal (4].

Question 1 With respect to the equivalent static analysis, the Licensee should provide justification for the use of the amplification factor of 1.3 to account for multi-mode effects.

.,~

Response 1 In this response, the Licensee presented an analysis for 17 walls at Pilgrim Unit 1 in which responses based on the fundamental mode were compared with those obtained from the first 12 modes. The results illustrated that the fundamental mode of vibration contributes to at least 99.5% of the total response.

In view of the above, the Licensee's response is adequate and is consis-tent with the SGEB criteria.

l Question 2 i With regard to the availability of the amplified response spectra, the Licensee stated that "if no ARS is available, earthquake loads may be evaluated based on a rigid range acceleration value as determined by the original building seismic analysis. In this case the wall must be shown to have a fundamental frequency greater than the rigid range cutoff frequency value." The Licensee should indicate how the rigid range acceleration and cutoff frequency could be determined.

1

, l l

l

TER-C5506-158 Response 2 The Licensee provided clarification that amplified response spectra are available for all rasonry walls evaluated at the Pilgrim plant. Therefore, this response has resolved the concerns associated with seismic loading used in the analysis.

Question 3 For the permissible strain of non-shear walls, the Licensee should provide the technical basis for using 0.1% and 0.01% for confined and unconfined walls, respectively.

Response 3 The Licensee's response is sununarized below.

For Confined Walls The Licensee refers to a series of shear tests on concrete masonry piers performed at California State Polytechnic College at Pomona (13). These tests were performed on masonry piers constructed using 8-inch ASTM type C-90 block type vs mortar. Another set of test data (14] obtained from an actual concrete masonry structure were subjected to a major earthquake ground motion. The structure is North Hall located on the campus of the Ohiversity of California at Santa Barbara. These test results demonstrated that a permissible strain of 0.1% is adequate.

For Unconfined Walls

! Test results by Becica and Fishburn (15, 16] were cited. These tests showed a permissible strain of 0.01% is justifiable.

The review of information provided by other plants indicated that the permissible strains of non-shear walls presented here are adequate and satisfactory.

Question 4-With respect to the special inspection category of stress values, the Licensee should indicate if the construction practice for the masonry structures at Pilgrim Unit I was in conformance with the provisions specified for the special inspection category in ACI 531-79.

l TER-C5506-158 Response 4 The Licensee stated that Bechtel Power Corporation was both Architect /

Engineer and Constructor for the Pilgrim plant and provided continuous surveillance during construction. A number of tests were performed to determine the following:

i o compressive strength o absorption o weight o dimensions o material certificates.

The Licensee also submitted documents showing the instructions used for testing of masonry walls, grouting masonry walls, and specifications for furnishing, delivery, and installation of concrete unit masonry.

Prior to starting the IE Bulletin 80-11 reevaluation, a plant walkdown by the Licensee's consultants indicated that the concrete block walls appeared to be in good shape (based on visual examination). In addition, a test program was conducted to verify the construction details. Details to be verified were i

reinforcement, grouted cells, and anchorage. Further details on this subject are provided in on page 14.

A The Licensee's response is considered adequate and acceptable.

Response Dated September 29, 1982 On June 16 and 17, 1982, the NRC, its consultants, and the Licensee participated in a meeting and tour of the Pilgrim site. Several unresolved I issued were raised at the meeting, and the Licensee was requested to provide additional information. The fc11owing five responses, provided in the September 29, 1982 submittal (5), addressed this request for information.

Question 1 Document the shear stresses for the safety-related shear walls.

Response 1 1 1

The Licensee provided a table of actual and allowable shear stresses for l l

each wall at the Pilgrim plant. The minimum SSE allowable stress was 50 psi, l i

TER-C5506-158 based on the ACI 531-79 criteria for shear walls for vM/Vd y 1 1.

M is the maximum bending moment occurring simultanaously with the shear load V at the section under consideration. yd is the length of the wall in the direction of shear. The maximum SSE allowable was 110 psi, based on the ACI 531-79 criteria for M/Vd < 1. The actual SSE shear stresses ranged from 4.99 psi y

(wall 68.3) to 49.10 psi (wall 66.12). The minimum OBE allowable stress was l 33 psi and the maximum was 73 psi (M/Vd y 1 1 and M/Vd y < 1, respectively). l The actual OBE shear stresses ranged from 2.50 psi (wall 68.3) to 24.60 psi (wall 66.12).

This response is adequate.

Question 2  ;

Check the effect of block pullout if only the horizontal joints are used  ;

I

. to resist shear.

i Response 2 The Licensee provided the allowable block pullout shears assuming only horizontal joints resist shear. These were compared to the original pullout valuesa. For shield walls, the allowables were reduced 33%. For partition walls, the maximum reduction was 22%. The Licensee reviewed the original pullout calculations and found no actual pullout values exceeding the reduced  ;

allowables.

This response is adequate and complies with the SGEB criteria.

Question 3 Verify the use of the fourth order Branson formula for computing effective stiffness.

Response 3 In this response, the Licensee indicated that hand and computer analyses were compared with test results to verify the use of the fourth order Branson formula for computing effective stiffness. Displacements.were calculated using both the third and fourth order Branson formulas as recommended by ACI Committee 435, Articles ACI 435.6R-74 and ACI 435.2R-66. The test results

- j TER-C5506-158 l l

were taken from the report Results of Variation of b or Effective Width in Flexure in Concrete Block Panels (Masonry Institute of America, reprinted 1971). It was found that the calculated displacements using the Branson formulas were comparable to the test displacement results. As the moment approached the allowable limit, the Branson formulas actually produced larger displacements than the test results. Therefore, the Branson formula is i considered adequate to predict effective stiffness.

Question 4 '

l l Masonry wall 64.4 was observed to have significant cracks. Verify that these cracks do not have any significant effect on the results of the '

analysis.

Response 4 In this response, the Licensee indicated,that wall 64.4 has been j reanalyzed considering the effect of the cracking. The wall is reinforced vertically with #5 bars at 16 inches and horizontally with a bond beam at 8 feet, O inches from the base. In the updated analysis, the strength of the horisontal span of the wall was neglected to account for cracking. Only the bond beam that was modeled as a strip of finite elements was considered to have stiffness in the horizontal span. The results of this reanalysis showed that, even after modifying the horizontal properties of the wall to reflect l cracking, wall 64.4 still qualified under the SGEB criteria. The reanalysis ,

indicated a wall moment in the vertical direction of 1876 in-lb/in, compared ,

with an allowable of 3620 in-lb/in. The bond beam moment in the horisontal direction was 2720 in-lb/in, compared with an allowable of 5372 in-lb/in.

This response is satisfactory.

Question 5 Justify the increase from 1.3 to 1.5 for allowable tension (SGEB allows 1.3).

Response 5 The Licensee responded that the 1.5 factor was not actually used in any of the wall qualifications since it applies only to unreinforced walls. The walls considered in this evaluation were reinforced.

TER-C5506-150 It is obvious that even though provided in the Licensee's design criteria,  ;

this factor was not actually used.

Responses to Action Items Presented in July 18, 1984 Meeting During the course of on-going evaluation of the walls, the Licensee decided to rely on the connections along the wall boundaries for load transfer (instead of ignoring them altogether in the original analysis). A test program was conducted at the plant site to verify the existence of the connections along the boundary, and a statistical analysis was performed using the test results. The primary objective of this analysis was to establish a load boundary by statistical approach. On June 7, 1984, the NRC, it consultants, and the Licensee met at Boston Edison Company's Engineering Offices to discuss the reevaluation criteria for masonry walls, in particular the use of statistici analysis to determine boundary loads. Several action

_ items r3sult'ed from this meeting. The following three responses to these actions items were presented and discussed in a meeting on July 18, 1984 [6]

at the NRC offices in Bethesda.

Action Item No. 1 B$COwillprovideasummarizedstatusofallthewalls.

Response No. 1 In this response, the Licensee provided a sumanary of the status of all safety-related masonry walls, which is shown in Table 3-1.

Action Ites No. 2 BECO will provide the following information regarding the test program to verify the construction details of the walls:

a. Total number of walls and subtotals by building 4

b. Total number of walls sampled and subtotals by building

c. Type of sampling (i.e., top, side, and base)

d. Correlation (if any) between length of sample versus total length of wall

e. Additional information about the capacity of boundary at the base.

l TER-C5506-158 Action Item No. 3 BECO will provide a copy of the summary report of statistical analysis of test data regarding the boundary line load.

Response Nos. 2 and 3 The Licensee provided the total number of walls tested in each building; this information is given in the table below:

Unique No. Walls Building Total SR Walls Tested

• Auxiliary 29 8 ,

Diesel Generator 5 2 Intake 5 1 Radwaste 61 16 Reactor 125 20 Turbine 17 3 242 50 Shear espacity at the bottom boundary With regard to the shear capacity of the bottom boundary, the Licensee relied on test data [17] for shear capacity. The shear strength is expressed t'

as:

T = to + pen where To = shear bond strength On = normal stress W = coefficient of friction.

A regression analysis of test data produced the following relationships for type S mortar:

Ungrouted T= 76 + 1.07 on  !

Weak grout i = 114 + 1.08 on Strong grout t = 156 + 1.54 On

• Many walls have multiple data points.

i TER-C5506-158 Table 3-1. Summary of Status of Safety-Related Masonry Walls f r

- The blockwall effort consisted of 220 safety-related masonry walls in I February 1982 when the NRC last inspected Boston Edison's program, 24 I o Qualified by analysis (not using statistical application of test data) l

[

o Qualified by modification 43 l o Modification deferred until 1983-84 refueling outage 1  !

(wall 64.4) (now completed)  !

o Walls to be considered using more rigorous analysis 152 f Total 220 l l

1

- BECo has completed the reconsideration of 152 walls with the following

(

results:

o Reclassified non-safety-related

~

7 j l

o Qualified by reliance on statistical 50 I application of test data ]

P o Qualified by use of deterministically 51 {

established structural mechanisms j o Scheduled for qualification by modification 44 Total 152 i J

I

- Scope added in 1984 New safety-related walls identified 10  !

o o New safety-related blockouts identified 16 Sub Total 26 o Walls previously designated non-safety-related 3 which are now safety-related due to recent modifications o Walls previously designated safety-related which -7 are now non-safety-related Adjustment to total scope +22 Total safety-related walls /blockouts = 242

TER-C5506-158 The Licensee indicated that Pilgrim walls are composed of material similar to those found in the tests [17]. The purpose of these tests was to l i

investigate the effects of the mortar type, grout strength, and bed-joint i reinforcement on the shear-strength characteristics of concrete masonry bed joints. Since the walls are not load bearing (in most cases), the contribution of the normal force is neglected. l 1

Figure 3-1 sununarizes the bottom shear boundary stress of Pilgrim walls as a percentage of the shear strength for ungrouted masonry.

Statistical analysis of boundary strength i

Field inspection of the anchorage conditions at the boundaries of the l masonry walls was undertaken by the Licensee in the latter part of 1981 and early 1982. The information given in this section is the result of a l statistical analysis of the test data obtained from the field inspection. The results of this analysis have been factored into the reevaluation of the walls. Basically, these results provided a set of boundary allowables against which the calculated stresses were compared.

Boundary types Existing boundary types are grouped as follows:

I Top boundary o Metal Q-decking o Structural steel WF section o Concrete Side boundary o Structural steel WF section o Concrete o Intersecting masonry L l

o Intersecting masonry T Test procedures Too boundary Three consecutive blocks were cut. The cut into the block was to be a -

minimum depth of 50% of the block thickness (if anchorage is found) to a

I TER-C5506-158 1

I 25 - ,

l l

l 20  !

de

, ,I j 15 g I

l Z ~/' '

s 11 , l .

to 10 E .

8 hl .

1 -

Ol 4 i l ll:1 l:

0 .

0-- g g g  ;  ; g i g i  ; ; g g i g g i g i g o 5 to 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

% of Sheer Strength i

l l

- l Figure 3-1. Boundary Shear Stress as a Percentage of Shear Strength I l

4 TER-C5506-158 maximum of the entire depth of grout in the cells. If only one dowel was found in the section of the cut, an additional amount of block was cut so that a minimum exposure of 24 inches on either side of the dowel was be achieved.

The result was considered satisfactory if a dowel was found in each block that was cut away because this corresponds to the required 16-inch spacing.

Side boundary Four consecutive blocks were cut away, and the remaining procedures are similar to the above case. The result was considered satisfactory if at least 50% of the exposed blocks contained dowels. This corresponds to a 16-inch spacing as required.

Test results A total of 51 tests were performed on the top boundaries and 37 on the side boundaries. Tables 3-2 and 3-3 present the results for the top boundary and side boundary, respectively.

Statistical analysis The statistical methodology for each boundary type is based on the number of anchorages per unit of length, and the analysis is performed on this set of numbers. The confidence level chosen for all of the analysis is 95%.

To simplify the analysis, it is assumed that the spacing for anchorages on all boundary types is 16 inches, which is not necessarily the design condition for each boundary.

The results of the analysis are summarized below:

4 o Top boundary to Q-deck 0%

o Top boundary to structural steel 0%

o Top boundary to concrete 19.3%

o Side boundary to structural steel 33%

o Side boundary to concrete 20%

1

TER-C5506-158 Table 3-2. Results irom Top Boundary Tests imALL 800seO4MY SXPOSED seuadSER OF seuestem TYPE LahoTM oneness ApeCHOm40Es as.2 concfme se .

2

~~ I s2.2 Concree 4s ' 2 63 5 0-Deca ss 0 64 4 Steet as 2 64 4 Steet M 2 64 6 0-Dect 51 1 64 12 . Concrete se 2 45.18 Steet 43 1 85.21 Steet 60

• 1 64 0 Steet 42 1 44 2 0-Dect 72.5 0 64.5 Q-Deck 20 0 64.5 0-Deca 16 0

, 64 4 Steet 13 1 64.8 Steet 51 2 64 6 Concrete 30 2 84,7 O-Oeca de 2 04 11 Concrete 44 1

, 44.12 0-Dect 47 0 64 18 Steet 50 0 47.1 Steet SS 1 SS.10 Steet 44 0 111.7 O-Dect 40 2 ,

184.2 Steel 38 0 i 164 4 Steet 44 1 164.7 Concrete de 3 164 5 Concrete 49 3 168.2 Steet 44 0 4 168.3 0-Deck Se 3 100.9 Steel 47 0 188.10 Concrete 49 3 191.26 Steet 73 1 191.35 Steel 81 1 191.49 O-Dect 64 0 191.55 Steet 48 1 194.21 0-Deck 89 0 194.21 0-Deck 47 0 194.22 0-Dect 49 0 194.23 Steet 49 1 194.25 0-Dect 32 0 195.14 Steet 48 2 195 18 O-Deca . 41 0 195.22 Steel 68 1 195.23 0-Dect 6 0 195.23 0-Deca SS 0 195.23 0-Oeca 12 0 194 4 Steet 57.5 3 194.7 Steet 10 0 196 7 Steet 80.5 3 Concrete ** 1

, 2090 ,

212 1 Concrete de 1

i TER-C5506-158 l

l l

Table 3-3. Results from Side Boundary Tests  ;

umu. ar= === sNPOSED NuesSEA OF BIUteSEM TYPE LENGTH enensea *Afe ;MOAAGES 45.2 Concrete 33 1 42.2 Concrete as 2 83.5 Concrete 30 0 64 4. Concrete as 0

- 64.13- Steet 37 2 65.21 Concrete 33 1 68.0 Concrete 33 s SS 2- Concrete 44 2 68.5 Concrete 40 - 2 M.8 Steet 33 2 ,

68.8 Steet 33 2

-87.1 Steet 45 2 87.1- Concrete 43 2 48R- Concrete 82 2 88.10 Concrete 38 2 184.4 Steet 32 1 184.9 Concrete 32 2 1e8.2 Steet 42 2 188.2 Steet 37 2 188.3 Steel 43 2 188.9 Concrete 34 1 <

188.10 Steet M 2 191J8 Stees 33 1 191.35 Steel 54 3 191.35 Steel SS S g 181.55 Steel 69 3 194.3 Steel 82 1 194.2 Concrete 32 0 194.22 Concrete 84 3 194.25 Concrete 32 1 1 195.9 Steel SS.S 3 195.14 Concrete 44 2 198.6 Steel 39 3 198.0 Concrete SI' 1 19t.3 Concrete SS 2 209.0 Concrete 38 1 212.0 Steet 33 0 i

I i

I

l l

f

' TER-C5506-158 -

~l l l Table 3-4 illustrates the resua.;s for the last three categories mentioned above. The allowables shcars were determined by applying the above percentages to the ideal boundary strength (assuming anchors were specified on the design drawings). It can be seen in this table that the allowable shear ' '

has been arrived at for different wall thicknesses. .,

It is noted that besides the anchorage along the boundary, the assemblage of the mannry (block, mortar, and grout) also provides shear resistance.

i Response Dated February 21, 1985 -

- As a result of the information presented in tria July 18,' 1984 meeting (above),.several questions arose concerning the statistical analysis'and l orthotropic plate analysis. These questions were sent'to the Licensee on January 10, 1985. The following ten responses address the statistical

'- analysis (Responses 1 and 2) and the orthotropic plate analysis (Responses 3 ,

l to 10). s l Ouestion 1 With respect to the sampling t'echnique used in the test verification program, please provide a technical assessment.of the use of unequal

.. exposed lengths for anchor verification in different walls. Also provide a technical assessment of the fact that the exposed ler.gth war. not

.rt: lated to the length of'the wall (i.e., total ofl44 inches of exposure ves applied noti only for a short wall, say an 8-f t wall, but' also for a long wall, say a 20-ft wall or longer). i Response.1 The etstlistical analysis of the test data has 'been based on a single number (anchorage per unit length), calculated as the number of anchors observed divided by the exposed length. An alternative approach would be to consider each sample point as two measurements (length exposed and number of anchors observed; this method represents two variables: length and number of anchorages).

For the thrae populations for which some anchors can be relied upon (side-steel, side-concrets, top-concrete), the sample statistics were recalculated based on the above two variables and results given in Table 3-5. The results l indicated that in all cases, the previously calculated means and standard deviations are approximately equal'to those presented in Table 3-5.

f y ~ - ~ - -

, ,, e. , . . - , - - . , , , - ,,.,,,.,,--,-q,, , ,,,-.,.,y, , , , . , ._.a. ,e,- ,, , , . ,,_w,n,--.m_m~,-,e.,, _. w w w-.,n,

TER-C5506-158 Table 3-4. Allowable Boundary Line Loads WALL POSITION SOUNDAfW Aa.LOmastg SMSAA COMasENTS j THICatM SS TVPES v sh/W l 4' TOP O 120 See Note 3 W 0

• j C 90 See Note 4

~

SIDE W 137 See Note 5 C 93 See Note 6 M 100 See Note 7 12* TOP Q 120 Sea Note 3 W ,0 C 90 See Note 4 SICE W 127 See Note 5 C S3 See Note 6 M 160 See Note 7 1 *-4* TCP Q 240 See Note 3 W 0 C 100 See Note 4 SIDE W 355 See Note 5 C 185 See Note 6 M = See Note 9 2 -0* TOP Q , 340 See Note 3 to 2 -2* W 0 C 180 See Note 4 SICE W -

355 See Note 5 C 185 See Note 6 M

• See Note B 3 -s* TOP Q 440 See Note 3 to 3'-0* Q 000 3'4' Well W 0 C 100 See Note 4 SIDE W 355 See Note 5

% C 185 See Note 6 M

• See Note a 3 -4' TOP Q 720 See Note 3 W 0 C 180 See Note 4 SIDE W 302 See Note 5 C 278 See Note 5 M

• See Ncte 8 NOTES:

1. Tacutates allowaaie sneer force va is la pounes per enca lengtn of souneary

2. Souncery types are as follows: Q: C Deca ssee W: Structures Sieet C: Concrete M: Intersecting masonry west.

3. Strengtn apotles to walls pareHet to me nas of ane 0-ooca. Por weils perpenoicular to tne Q-cock rtes. anowsoie sness to essen as aero.

~

4. Allowsoie sneer sootles eney a meets wenn longen greater man 8 -

8*. For shorter wens. the aseewesse saast to temen as aero.

5. Allowoote snear acoues eney a meses seen steegnt greater man 4'-

0*. For snorter wans. tne asieweese snais se amen as aero.

8. Allowooie sneer appfles only to weHe witn fleegnt gegeter gnan $*-

8*. For snorter wells. the asteweeie snait to temen as aero.

7. Values for nevneartes termos et masonry wesis apoty ter mens as wnicn latertocking of teocks from me two wells is esparent. For wens unere no interseeming leant emise teaseen me two wells. me snowsoie soun eary force sna*8 to when as sere.

8. For multi-wytne wells wenn Intertecting masonry else beuneartes, allowsole sneers shalt to summes ever asi wymes ter each of the intersecting wens. The newest such sescusasse esteer snais appty to an saae boungertes at me Intersecuen scese 7 eD980es to seen wytne. The alloweele sneer ser a e' mism eyene is 60 It/In.

TER-C5506-15@

Table 3-5. Sample Statistics of Test Data TOP CONCRETE l POPULA110N $10E STEEL $10E CONCRETE l

STAfl5 TIC We Length Variatten (Previous Results) l 0.0449 0.0354' O.0441 l e mean 1 0.0145 0.0103 0.0100 ,

e 5td. Sev. .

Length Verlation 0.0451 0.0353 0.0439 e mean 0.0116 0.0110 0.0176 e 5td. Dev. ,

so I

v

TER-C5506-158 Exposed Length vs. Length of Wall The exposed length requirements in the test program were not related to l ahe length of the wall. The data were treated on a "per unit length" basis.

The Licensee also stated that tests were prepared on walls of varying length l l

from 4 feet or 5 feet to 30 feet or higher. The sample is made up from walls '

having a wide range of lengths.

The Licensee's response is adequate and satisfactory.

Question 2 In a few cases, no anchors were found with a predetermined exposed length. The Licensee should extend this exposed length to locate the anchors. The results will help to reinforce the adequacy of the statistical analysis method.

Response 2 s'

There are two boundary types which rely on anchors and have at least one

" sero" data point: side-steel (1 sero data point in 17 tests) and side-concrete (3 zero data points in 20 tests). The Licensee noted that in each case the zero data points are from tests with short exposed lengths (30 in, 32 in, 33 'in, and 36 in) .

These four cases were also re-examined, and the following results were given:

o Wall 212: (Side-steel)

The actual sample location was limited due to physical obstructions.

o Wall 63.5: (Side-concrete)

This sample was interrupted by physical obstructions attached to and adjacent to the wall.

o Wall 64.4: (Side-concrete)

This particular wall had some boundary modifications that cover the '

area of the sample in question, and the wall was qualified relying on

]

modifications.  ;

TER-C5506-158 o Wall 194.20: (Side-concrete)

The sample was interrupted by some attachments. The Licensee stated that this wall requires modifications to resist seismic loads. In the analysis, the side boundary was considered to be pinned with shear capacity of 3 lb/in, which is only 3% of the Licensee's acceptance criteria (93 lb/in). In addition, this wall is prod.minantly vertical spanned. An alternative calculation has been done to show the wall with planned modifications will be qualified without relying on the side boundary in question.

The Licensee's response is considered adequate and acceptable.

The Licensee employed a two-way cracked analysis to qualify the walls in the plant. This method has not been normally used in practice. Therefore, a number of questions were raised regarding the validity of this method. The '

following are the Licensee's responses.

, Question 3 J CYGNA's methodology calls for two-way c~r'acked analysis of block walls (level I and level II). There are no acceptable methods available in the literature for the bending analysis of block masonry walls in the post-cracking stage. This is primarily because of the complexity of the problem due to material anisotropy, the existence of planes of weakness which affect crack progagation, discontinuity due to partial grouting, aqd the uncertainty about the contribution of joint reinforcement in the lateral load resistance. In light of the above comments, justify the two-way cracked analysis.

Response 3 The masonry walls at the Pilgrim plant are vertically reinforced at every other cell, with those cells fully grouted. Some walls have no horizontal reinforcement except joint wire reinforcement. A number of walls have horizontal bond beams. For walls without bond beams, finite element orthotropic plate models were used for analysis.

At low levels of load, the entire wall remains uncracked, and since most of the stress occurs in the face shell, the wall acts like an isotropic slab.

The vertical grouted cores do not significantly affect the wall response because they are near the neutral axis.

l

TER-C5506-158 At higher loads, cracks will develop in sections of the wall. In the vertical direction, the grouted cores provide sufficient continuity to ensure beam action. In the horizontal direction, when the moment exceeds the allowables (considering only the face shell area), it is assumed that cracking occurs. In this case, the stiffness of the finite element model in the horizontal direction was set to zero, and the wall becomes in effect a vertical beam.

r In walls containing horizontal bond beams, the load carrying capacity of the bond beam was exploited. This was accomplished by considering an average j steel area (horizontal) over the wall hcight and using a two-way analysis.

This approach was shown by comparative analysis with wall 64.4/65.8 to be l

conservative relative to neglecting the horizontal strength except for the narrow bond beam strip.

This response is satisfactory and is consistent with the SGEB criteria.

e Question 4 Equations developed for adequately reinforced concrete slabs have been used in the analysis to account for the orthotropic properties resulting f4cm differing steel reinforcement details in the vertical and horizontal directions. The applicability of these equations to block masonry walls is questionable because of the notable differences between a reinforced concrete slab and a block masonry wall. First, concrete is a globally homogeneous material, whereas masonry is not. This is particularly true for partially grouted walls. Secondly, the percentage of reinforcement and detailing in the two directions are quite different in the two '

cases. Steel orthotropy for which these equations were developed is not ,

applicable for the Pilgrim walls which have no horizontal steel. Thirdly, l because masonry is a jointed medium, one espects crack patterns, and consequently the steel contribution, to be different from those in j reinforced concrete. In light of these comments, justify the use of equations developed for the reinforced concrete slab to qualify the masonry walls.

Response 4 In this response, the Licensee defended application of equations used for concrete slabs to reinforced masonry walls to account for different steel details in the horizontal and vertical directions. It stated that in the 1

i l

i TER-C5506-158 i

i elastic theory of working stress design, most of the formulas for masonry are [

t similar to concrete except for the material properties and ultimate strength  ;

of masonry, i

The differences in reinforcement between the horizontal and vertical i i

directions were taken into account by the approach der,cribed in Response 1. }

That is, horizontal stiffness was considered to be t.oro for cracked sections  ;

l (no horizontal steel) and the problem reduces to one-way vertical bending.  !

-Because the principal directions are parallel or normal to the bed joints l (mortar joints initiate cracking), the special orthotropic assumptions used in the Pilgrim analysis are appropriate.

This response is adequate. Further discussion on this subject is ]

provided in Appendix B. I I

I Question 5 j l

Modulus of elasticity of the walls is assumed to be equal in the two orthogonal directions. This is not true for masonry which is a composite material. Assessment of the accuracy of this assumption and its impact  !

on the outcome of the analysis needs to be investigated. l I

i Response 5  !

{

The Licensee indicated that it used the value of modulus of elasticity l given in ACI-531-79. For uniaxial direct stress tests, grouted cores will I affect the composite modulus. For 1.ateral bending, however, the stiffness in the horizontal and vertical directions is dominated by the stiffness of the ,

t l

face shall regions of the block. Therefore, horizontal and vertical moduli j should be close for lateral bending.  ;

This response is satisfactory. r 1

i Question 6 .{

The Branson equation has been used in Level II analysis to determine the 1 effective moment of inertias of different elements. This empirical (

equation was originally developed for reinforced concrete members under l l uniaxial bending. Its applicability to two-way bending of block masonry j walls needs to be demonstrated. It must be noted that the Branson i equation has been used to express effective moment of inertia in the I horizontal direction where there is no reinforcing steel.  !

1 i

i l

l

__. _ _ _ . _ _ _ . . _ _ _ . . _ _ . I,

l TER-C5506-158 I l

Response 6

~

The Licensee indicated that Jofreit and McNeice (19] have demonstrated the applicability of Branson's equations to one-way and two-way slabs. In Reference 19, Jofreit and McNeice show how the finite element method can be used to analyze reinforced concrete slabs under progressive cracking. To demonstrate the capability of the method for reinforced concrete members, they compared experimental load-deflection data for one-way and two-way slabs with results calculated using Branson's equation. The experimental results compared well with the analytical results. Also, Branson's equations were'not used in 1 the analysis for the horizontal direction if the walls were not reinforced horizontally with bond beams.

This response is adequate.

)

. Question 7 Higher damping values have been used in evel II analysis. What is the basis for choosing higher damping values? Are these values changing with the level of loading?

Response 7 The Licensee indicated that higher damping values were used in the Level II analysis than in the Level I analysis because Level I was performed prior to the issuance of the SGEB criteria. Conservative values were used in the absence of guidelines for damping in masonry. Level II analysis was performed after the SGEB criteria were issued, and the damping values were changed to reflect the higher SGEB recommendations. (Note: For the Level I analysis, the damping values for OBE and SSE were 2% and 5%. For the Level II analysis, these values were 4% and 7%.)

This response is satisfactory and consistent with the SGEB criteria.

Question 8 Review of calculations of wall 64.4/65.8 revealed a significant difference in element moments from Level I and Level II analyses. For example, moments in the critical element of the bond beam were reduced by 88%

shifting from Level I to Level II analysis. How could this reduction be justified and what are the main reasons for such a large change?

TER-C5506-158 Response 8 The Licensee indicated that moments calculated in the Level II analysis were greatly reduced from the Level I analysis because of the conservatisms  ;

I ' inherent in the Level I analysis. In the Level I analysis, walls were assumed cracked over the full height and width of the wall. The inertial load was taken as the peak response at 5% damping and increased by 30% to account for  ;

higher modes. Level II analysis used a response at 7% damping (non-peak) that included the effects of higher modes which typically increase the response by  !

less than 5%.

The Licensee's response has resolved this concern satisfactorily.

Question 9 Wall 188.10 has an aspect ratio greater than 3, which calls for almost a single curvature with bending primarily in the shorter direction. The crack pattern (parallel to the shorter direction), which is predicted from the computer analysis, does not ssem to be consistent with the one-way bending action of the wall. This inconsistency does not provide confidence in the capability of the proposed analytical model to predict l

i actual behavior.

! Response 9 The Licensee indicated that in the initial computer run for wall 188.10, cracking moments due to vertical bending were exceeded, causing horisontal cracking. After full length horisontal cracking, load redistribution causes vertical cracks (parallel to the shorter direction) to occur near the ,

mid-height of the wall; however, the wall is still spanning vertically.

Original calculations show that the top and bottom boundaries of the wall carry 92% of the load. Thus, vertical cracking, as predicted by the computer analysis, is not inconsistent with the espected one-way vertical action of the I

wall.

i Question 10 It is not clear how existing cracks (e.g., in wall 64.4) have been accounted for in the analysis.

i . _ _ _ _ __ _ ,. - _ __ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . _ . _ _ _ _ _ _ , . . _ . _ . . , _ _ _ . _ _ _ _ .

TER-C5506-158 Response 10 The Licensee indicated that wall 64.4/65.8 was assumed to be cracked over the entire height and width of the wall, not just where the cracks are visible.

The horizontal stiffness was assumed to be sero (except for the narrow bond beam strip). Branson's equation was used to calculate the effective moment of inertia for the bond beam and for the vertical span.

This responsa has confirmed that the existing cracks were factored in the analysis.

Response Dated July 26, 1985 Sev.eral action items resulted from the meeting on July 18, 1984 (6).

Three items required action by the Licensee, two items required action by the NRC. The following three responses represent the Licensee's action on these items, and were presented in the July 26, 1985 submittal [8).

l Question 1 The NRC staff accepted the blockout criteria proposed by the Licensee, provided the following conditions were met:

a'., The Licensee was to survey all blockouts (except 116.3) for cracks on boundaries - acceptance of criteria based on no evidence of through cracks;

b. provide the results of survey for the staff's review and acceptance, and,

c. for blockout 116.3, provide modifications (on one face) to boundaries to resist peak load resulting from the load combination involving PBOC load components. Notify the staff if tornado differential pressures.are greater than 1.5 psi and not acting in the same I

direction as the PBOC load.

Response 1 The Licensee responded that all blockouts were surveyed for cracks at the I

boundaries. Out of 20 blockouts, 17 set the acceptance criteria. Two of the remaining three were closely inspected by cutting into the be?. joint with a masonry saw. The hairline cracks observed were found to be tite effects of surface shrinkage. The third wall was inaccescible for this type of test, but it was concluded on the strength of examination of the other two walls that

b i

TER-C5506-158 F

the hairline crack observed on one face of the third wall was also the result of surface shrinkage. Therefore, all 20 blockouts were considered j acceptable. Also, the Licensee intends to schedule a modification to blockout t 116.3 for PBOC loads only. This load is undirectional, so the modification j will not consider reverse direction loads. Wall 116.3 is not subject to tornado depressurization loads, and seismic forces are negligible compared with the PBOC load.

This response is adequate.

i Question 2 l t

Provide representative calculations to show differences between the prior i Cygna analysis and the subsequent refined analysis for walls qualified j without reliance on the statistically determined line loads. j i

4 Response 2 )

- J The Licensee indicated in this response that nine walls rely on test data l to determine the allowable boundary line loeds. Although this differs from l t

the prior Cygna analysis, the same acceptance criteria and analytical methods previously used by Cygna are still applied. The Licensee has provided Cygna

! calcul$tions for walls 188.10 and 63.4 for comparison with Bechtel/Computech calculations, which were submitted at the July 18, 1984 meeting. These l calculations have been reviewed and it was concluded that the analysis is adequate. t i

Question 3 i i

Provide the alternate qualification scheme for walls 209.13 and 209.14. l Response 3 r i

The equipment that would be endangered by the failure of walls 209.13 and

! 209.14 are safety-related switches mounted on blockout 929.3. The Licensee  !

investigated walls 209.13 and 209.14 and determined that wall 209.14 will not  ;

impact the switches. Wall 209.13 will be modified to resolve this issue.  !

l

_ . _ . . _j

TER-C5506-158 i This response is adequate.

Response Dated December 31, 1985 1

Based on the information provided by the Licensee regarding the two-way cracked analysis, a number of technical issues still needed further explanation and/or clarification. In an effort to resolve this topic, a meeting was held on November 1, 1985 between the NRC and the Licensee to

discuss the Licensee's two-way cracked analysis. The following five responses are to questions discussed in the November 21, 1985 meeting, and were presented in the submittal dated December 31, 1985 [9).

Question la Explain how the cracking moment for a particular element was determined.

Response la 4

In this response, the Licensee indicated that the cracking moments in the vertical and horizontal directions (M, and M, ) were constant values in each d.irection depending on the reinforcement and thickness of the wall. The cracki5gmomentwascalculatedaccordingtotheformula.

Me r = FtgI /Y where F is the modulus of rupture, I,is the moment of inertia of the t

uncracked section, and y is the distance from the neutral plane to the tension face.

4 I

j Question Ib i

1

Indicate how the effects of the directional variation of principal stresses were detemined and how they affect the wall stiffness.

1 Response Ib i

In this response, the Licensee indicated that principal stresses occur in two orthogonal directions, which, in this case, are parallel and normal to the

. bed joints since cracking initiates along the mortar joints. The mortar joints are aligned with the finite element x and y directions. This allows the 3

1 i

TER-C5506-158 directional stiffnesses to be specified independently in the ana2ysis. The program used in the analysis uses Branson's formula to calculate the inertial section properties.

j l

Responses la and lb are judged to be adequate.

Question 2a ,

Since the Branson equation was developed for structural members, justify its applicability to an element of the finite element model.

)

Response 2a I

The Licensee responded that the Branson equation is a method for detemining the effective moment of inertia as a function of moment, section l l properties, and concrete strength. This equation was originally developed for l beams, but Jofreit and McNeice (20] demonstrated its applicability to tne finite element analysis of one-way and two-way slabs. The Branson's equation j is permitted under the 1971 ACI code for two-way slabs.

This response is satisfactory.

l Questton 2b Based on an element stiffness, explain how the inertial forces were obtained taking into account the stiffness changes (i.e., frequency shift). Explain how the stiffness of each element was combined in the evaluation of the inertial forces.

i

?

Response 2b ,

l The Licensee responded that stiffness was varied on an element-by-element [

i basis to take into account cracking. As the applied inertial load causes

! moments to exceed the cracking moment, the element stiffness properties must t ,

i be adjusted. In the vertical direction, this is accomplished using Branson's  !

I formula. In the horizontal direction, the stiffness is set to zero once l cracking (based on the face shell area only) begins, unless a bond beam exists.

{

For walls with bond beams, the horizontal stiffness is based on an average j steel area over the height of the wall (this was shown to be conservative when l compared to considering only the narrow bond beam strip). The adjusted [

l stiffness properties are reinserted into the model, which is then reanalyzed.

h t

E

l .. l l

TER-C5506-158 l 1

The change in the element stiffness properties causes a decrease in the wall l frequency, hence a frequency shift. The change in frequency causes a change in inertial loads, which in turn causes the stiffness properties to be readjusted. This iterative process continues with the stiffness properties changing until convergence is achieved.

This response is adequate and consistent with SGEB criteria.

Question 3 I

I Provide the total number of walls qualified by the two-way cracked )

analysis and indicate how many of these walls have horizontal bond beams.

Response 3 The Licensee indicated that approximately 10% of the total of 200 walls at the Pilgrim plant have bond beams. j 1

l This response is adequate.

Question 4 For walls without horizontal bond beams, explain how the stiffness was i edaluated for the case in which the moment in the horizontal direction

! exceeded the unreinforced allowable (i.e., it was not clear in the

! Licensee's response (1) how cracking was represented in the modal).

Response 4 In this response, the Licensee indicated that if the moment exceeds the l unreinforced cracking moment in the horizontal Jirection (based on the face l sheel area only), the horizontal stiffness in the finite model was set to zero (i.e., no load carrying capacity in the unreinforced horisontal direction).

Branson's equation was not used in the horizontal direction.

This response is satisfactory.

I l Ouestion 5 j For walls with horizontal bond beams, explain how the stiffness along the horizontal direction was determined.

l

)

i TER-C5506-158 i

l Response 5 The Licensee indicated that for walls with horizontal bond beams, the initial properties in the horizontal direction are the same as uncracked walls with no bond beam. Once the cracking moment is exceeded, the cracked moment of inertia, using Branson's equation, is calculated based on the assumption that the bond beam steel area is distributed over the wall height. Thus, walls with bond beams are evaluated using a two-way action analysis.

This response is satisfactory.

Review of the Licensee's analysis indicated that some conservatisms were also built into the analysis, which include the following: .

o Joint reinforcement was not included in the anaiysis.

o The cracking moment in the horizontal direction was based on

' faceshell area only; the contribution of grout core was not accounted for.

3.2 EVALUATION OF LICENSEE'S APPROACH TO WALL MODIFICATIONS As a result of its reevaluation of masonry walls, the Licensee determined that 88 walls could not meet the acceptance criteria without structural 1 ,

modiffeation. In general, the modifications consist of structural steel members that act as intermediate supports or boundary reinforcements. All modi *ications were scheduled for completion by the 1986 refueling outage.

The Licensee's approach to the modification of masonry walls has been reviewed and found to be adequate.

1

~

TER-C5506-158 l

4. CONCLUSIONS A detailed study was performed to provide a technical evaluation of the 1 masonry walls at Pilgrim Nuclear Power Station. Review of the Licensee's criteria and additional information provided by the Licensee led to the conclusions given below:

o The Licensee's criteria have been evaluated and judged to be adequate and meet the SGEB requirements.

o With respect to the allowable line loads for the masonry walls at Pilgrim Nuclear Power Station Unit 1, it is judged that while the l

results of the field test program indicated that the boundary anchorages were not in accordance with the design drawings, detailed statistical analysis of the test data illustrated that certain

percentages of the specified anchors can be relied upon (with 95%

i confidence). These percentages have been used to factor the ideal strength (assuming anchors as per design) to arrive at a set of allowable boundary line loads. This approach is considered adequate. Further discussion on this subject is provided in Appendix A.

o Regarding the two-way cracked analysis, it was shown that the Pilgrim walls are reinforced vertically at every other cell, with the cell fully grouted, joint reinforcement is also installed along the horizontal direction (a number of walls also have horizontal bond

\ beams with reinforcing steel). The analysis employed finite element  !

orthotropic plate bending models. The reduction in wall rigidity was  !

accounted for by using Branson's equation. In the horizontal direction if the moment exceeds the unreinforced allowable, the section is assumed to be cracked and unable to transmit any load (i.e., the element stiffness in the horizontal direction is set to zero, contribution from grout cores or joint reinforcement was i

neglected). The Licensee's approach is judged to be adequate and satisfactory.

Regarding wall modifications, the Licensee indicated that wall modifications consist of structural steel members that act as intermediate

I

! supports or boundary reinforcements. A total of 88 walls require .

1 modifications and all modifications were scheduled for completion by the 1986 refueling outage. The Licensee's approach to wall modifications has been reviewed and found to be adequate.

As discussed in Section 1.3, three walls in the cable spreading room (194.17, 194.21, and 194.22) are not qualified for the tornado depressurization I

. . - , , _,y_-.,...__,__._.,__m ,___m.___y _____y_ .__,,_.,,,7,_,...-___ ,,.,_._,__m,.___..__,._.-_c.__,__m-, - , . ,.

TER-C5506-3,58 i

4 loads without considerable modifications. These modifications will require a [

t statior. outage. The Licensee claims a hardship exemption on completing such modifications. An additional eight walls in the radwaste corridor (191.29, I

191.37, 193.11, 193.12, 193.5, 193.6, 193.7, and 193.8) will not be qualified because the scheduled relocation of safety-related equipment will cause these walls to be reclassified as non-safety-related. The NRC staff is currently in  ;

the process of reviewing these issues and will issue its evaluation on this [

subject. l 1

l

\

l i

i I

I P

I i

l r

I i

(

I i

i l

l l

[

i I

i t

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

TER-C9506-3,58

5. REFERENCES I

l

. 1. IE Bulletin 80-11 l

! Masonry Wall Design NRC, May 8, 1981  ;

2. Uniform Building Code International Conference of Building Officials, 1979

l

3. Building Code Requirements of Concrete Masonry Structures Detroit: American Concrete Institute, 1979 ACI 531-79 and ACI 531-R-79 i

4. W. H. Deacon Letter with Attachments to D. B. Vassallo (NRC)

Subject:

Additional Information on Block Walls Boston Edison Company, March 22, 1982

5. A. 'V. Morisi ,

Letter with Attachments to D. B. Vassallo (NRC)

Subject:

Additional Information on Block Walls Boston Edison Company, September 29, 1982 i

6. Meeting Between Boston Edison Company and NRC at NRC (Bethesda)

July 18, 1984

7. W. D. Harrington Letter with Attachments to D. B. Vassallo (NRC) i

Subject:

Additional Information, Masonry Wall Design Boston Edison Company  ;

February 21, 1985

8. W. D. Harrington Letter with Attachments to D. B. Vassallo (NRC)

Subject:

Response to NRC Action Items i Boston Edison Company July 26, 1985 I

9. W. D. Harrington Letter with Attachments to John A. Zwolinski (NRC)

Subject:

Request for Additional Information, IE Bulletin 80-11 Boston Edison Company December 31, 1985

10. G. C. Andognini Letter to B. H. Grier (NRC)

Subject:

Response to IE Bulletin 80-11 Boston Edison Company July 14, 1980

- - _ - . = _ -_

Y TER-C5506-158 i

11. A. V. Morisi Letter to B. H. Grier (NRC)

Subject:

180 Day Response to IE Bulletin 80-11 Boston Edison Company November 15, 1980 l

12. A. V. Morisi Letter to B. H. Grier (NRC)

Subject:

180 Day Response to IE Bulletin 80-11. " Masonry Wall Design" Boston Edison Company March 18, 1981  ;

i 13. Schneider, R. R. " Shear in Concrete Masonry Priers," California State  ;

i Polytechnic College, Pomona, CA, 1959

14. Englekirk, R. E. and G. C. Hart, " Seismic Design of Concrete Masonry .

Shear Walls," ASCE National Convention, Hollywood, Florida, October 1980

, 15. Becica I. J. and H. G. Harris, " Evaluation of Techniques in the Direct Modeling of Concrete Masonry Structures," Drexel University Structural Models Laboratory Report No. M77-1, June 1977 j

16. Fishburn, C. C., "Effect of Mortar Properties on Strength of Masonry,"

National Bureau of Standards Honograph No. 36, U.S. Government Printing Office, November 1961.

17. Hamid, A., Drysdale, and Heidebrecht, " Shear Strength of Concrete Masonry Joints," Journal of the Structural Division, ASCE, July 1979

18. d'rysdale and Hamid, " Tension Failure Criteria for Plain Concrete 4

Masonry," Journal of the Structural Division, ASCE, February 1984

19. Jofreit and McNeice, "Finits Element Analysis of Reinforced Concrete Slabs," Journal of the Structural Division, ASCE, March 1971

20. Jofreit and McNeice, " Finite Element Analysis of Reinforced Concrete Slabs," Journal of the Structural Division, ASCE, July 1979 i  :

• l i

l 1

. - - _ _ . _ . . _ _ . _ . _ _ . . - . ~ _ _ . , __-___

l i

APPENDIX A l

scEB cnITERIA ron sarETv-nELATED nasoNay watt EvAtuarIoM i (DEVELOPED By THE STRUCTURAL AMD GEC7FECHNICAL ENGINEERING BRANCH

[SGEB] Or THE NRC) l i

l l

I s.

s.

l l

l 1

1 I

FRANKLIN RESEARCH CENTER OtVISION OF ARVIN/CALSPAN 20th & RACE STREET 5. PHILADELPHIA.PA 19103 I

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

. TER-C5506-158 CONTENTS Section Title Page 1 GENERAL REQUIREMENTS . . . . . . . . . . . A-1 2 LOADS AND LOAD COMBINATIONS. . . . . . . . . . A-1

a. Service Load Combinations . . . . . . . . . A-1

b. Extreme Environmental, Abnormal, Abnormal / Severe Environmental, and Abnormal / Extreme Environmental Conditions . . . . . . . . . . . . . A-2 3 ALLOWABLE STRESSES . . . . . . . . . . . A-2 4 DESIGN AND ANALYSIS CONSIDERATIONS . . . . . . . . A-3 5 REFERDJCES . . . . . . . . . . . . . . A-4

%'n l

iii

TER-C5506-15 8

1. General Requirements The materials, testing, analysis, design, construction, and inspection related to the design and construction of safety-related concrete masonry walls should conform to the applicable requirements contained in Uniform Building Code - 1979, unless specified otherwise, by the provisions in this criteria.

The use of other standards or codes, such as ACI-531, ATC-3, or NCMA, is also acceptable. However, when the provisions of these codes are less conservative than the corresponding provisions of the criteria, their use should be justified on a case-by-case basis.

In new construction, no unreinforced masonry walls will be permitted. For operating plants, existing unreinforced walls will be evaluated by the provisions of these criteria. Plants which are applying for an operating license and which have already built unreinforced masonry walls will be evaluated on a case-by-case basis.

2. Loads and Load Combinations The loads and load combinations shall include consideration of norwal loads, severe environmental loads, extreme environmental loads, and abnormal loads. Specifically, for operating plants, the load combinations provided in the plant's FSAR shall govern. For operating license applications, the following load combinations shall apply (for definition of load terms, see SRP Section 3.3.4II-3).

(a) Service Load Conditions (1) D+L (2) D+L+E (3) D + L + W If thernal stresses due to T o and Ro are present, they should be included in the above combinations as follows:

(la) D+L+To + Ro ,

(2a) D+L+To+Ro+E (34) D+L+To + Ro + W Check load combination for controlling condition for maximum 'L' and for no 'L'.

I

• l l

A-1 l

e . ll TER-C5506-198 ;

(b) Extreme Environmental, Abnormal, Abnormal / Severe Environmental. and [

Abnormal / Extreme Environmental Conditions ,

(4) D+L+To+Ro+E (5) D+L+To+Ro+Wt j (6) D+L+Ta+Ra + 1.5 Pa l (7) D+L+Ta+Ra + 1.25 Pa + 1.0 (Yr + Yj + Ym) + 1.25 E i

(8) D+L+Ta+Ra + 1.0 Pa + 1.0 (Yr + Yj + Ym) + 1.0 E' j In combinations (6), (7), and (8) the maximum values of P a , Ta '  !

Ra , Yj, Yr, and Y m, including an appropriate dynamic load factor, should be used unless a time-history analysis is performed to justify otherwise. Combinations (5), (7), and (8) and the j corresponding structural acceptance criteria should be satisfied first without the tornado missile load in (5) and without Yr

• Yj' ,

and Ym in (7) and (8). When considering these loads, local section strength capacities may be exceeded under these concentrated loads, provided there will be no loss of function of any safety-related j system. l t

Both cases of L having its full va16e or being completely absent should be checked. l

3. Allowable Stresses Allowable stresses provided in ACI-531-79, as supplemented by the fqflowingmodifications/ exceptions,shallapply.

(a) When wind or seismic loads (OBE) are considered in the loading combinations, no increase in the allowable stresses is permitted.  !

(b) Use of allowable stresses corresponding to special inspection  ;

category shall be substantiated by demonstration of compliance with the inspection requirements of the SEB criteria, j (c) When tension perpendicular to bed joints is used in qualifying the  !

unreinforced masonry walls, the allowable value will be justified by test program or other means pertinent to the plant and loading conditions. For reinforced masonry walls, all the tensile stresses  :

will be resisted by reinforcement.

(d) For load conditions which represent extreme environmental, abnormal, I abnormal / severe environmental, and abnormal / extreme environmental  ;

conditions, the allowable working stress may be multiplied by the  !

factors shown in the following table:

i 1

A-2 l

l

i TER-C3506-15 8 i l

l 1 Type of Stress Factor l

2  !

{ Axial or Flexural Compression ~ 2.5

) Bearing 2.5 i

Reinforcement stress except shear 2.0 but not to exceed 0.9 fy l

Shear reinforcement and/or bolts 1.5 Masonry tension parallel to bed joint 1.5 I Shear carried by masonry 1.3 Masonry tension perpendicular to bed joint for reinforced masonry 0 for unreinforced masonry2 1,3 Notes i .

i (1) When anchor bolts are used, design should prevent facial

{ spalling of masonry unit. 4

+

! (2) See 3(c).

}

4. Desion and Analysis Considerations j (a) The analysis should follow established principles of engineering

mechanics and take into account sound engineering practices. i (b) Assumptions and modeling techniques used shall give proper considerations to boundary conditions, cracking of sections, if any, and the dynamic behavior of masonry walls.  ;

l (c) Damping values to be used for dynamic analysis shall be those for I reinforced concrete given in Regulatory Guide 1.61.

! (d) In general, for operating plants, the seismic analysis and Category I l structural requirements of FSAR shall apply. For other plants, corresponding SRP requirements shall apply. The seismic analysis shall account for the variations and uncertainties in mass, materials, and other pertinent parameters used.

{

! (e) The analysis should consider both in-plane and out-of-plana loads.

(f) Interstory drift effects should be considered. l ll 1

l i 1 4

A-3 i t

. o TER-C5506-158 *

(g) In new construction, grout in concrete masonry walls, whenever used, shall be compacted by vibration.

(h) For masonry shear walls, the minimum reinforcement requirements of ACI-531 shall apply.

(i) Special constructions (e.g. , multiwythe, composite) or other items not covered by the code shall be reviewed on a case-by-case basis for their acceptance.

(j) Licensees or applicants shall submit QA/QC information, if available, for staff's review.

i

^

In the event QA/QC information is not available, a field survey and a test program reviewed and approved by the staff shall be implemented to ascertain the conformance of masonry construction to design drawings and specifications (e.g., rebar and grouting).

(k) ~ For masonry walls requiring protection from spalling and scabbing due to accident pipe reaction (Y r), jet impingement (Y 3 ), and missile i

impact (Ym), the requirements similar to those of SRP 3.5.3 shall apply. However, actual review will be conducted on a case-by-case

~

basis.

l 1 5. References (a) Uniform Building Code - 1979 Edition.

j (b) Building Code Requirements for Concrete Masonry Structures ACI-531-79 and Commentary ACI-531R-79.

s (c) Tentative Provisions for the Development of Seismic Regulations for Buildings - Applied Technology Council ATC 3-06.

(d) Specification for the Design and Construction of Load-Bearing Concrete Masonry - NCMA August, 1979, i (e) Trojan Nuclear Plant Concrete Masonry Design Criteria Safety Evaluation Report Supplement - November, 1980.

i A-4

I l

l l

I i

APPENDIX B I I

t REVIEW OF THE ANALYSIS OF MASONRY WALLS IN PILGRIM STATION, BY A. A. HAMID i

• 5, i

l l

'l ,

i l l

l

l 1

-l l

t FRANKLIN RESEARCH CENTER DIVISION OF ARVIN/ CAL 5 PAN 20tn & # ACE STREETS. PHILADELPHIA.PA 19103 I

_,------,,,-,,.,.,_-w_w,

-- - - - - --,, _ . - ,,,.. ,.__,, ., -, _ _ _ _ , . , , _ - . - . _ _ . _ , , , . _ , _ . . , , . , _ _ _ _ , , , , , - - _ , , _ . _ , _ _ , , , g,,.,--,, gn._

1 1

1 Technical Report on REVIEW OF THE ANALYSIS OF MASONRY WALLS IN PILGRIM STATION submitted to Dr. Vu Cori Nuclear Engineering Department Franklin Research Center

"- Philadelphia, PA by Dr. Ahmad A Hamid, P. E.

Associate Professor of Civil Engineering Drexel University, Philadelphia, PA April 1986

l Table of Contents

1. INTRODUCTION 3

2. EVALUATION OF BOUNDARY STRENGTHS 2.1 Boundary Types 3

- 2.2 Boundary Strengths Based on Interlock 2.2.1 Block Interlock 4

.. 2.2.2 Grout Interlock 4 2.3 Boundary Strengths Based on Anchors 2.3.1 Background '4 2.3.2 Review of Test Procedures 5 s 2.3.3 Review of Statistical Analysis 5

3. PLATE ANALYSIS 3.1 Methodology B 3.2 Cracking Moments 9 3.3 Flexural Stiffness 9 3.4 Review of Sample Calculations 10

4. CONCLUSIONS 11

5. REFERENCES 12 2

l. INTRODUCTION  !

Block masonry walls at Pilgrith Station were originally analyzed by Cygna Enerr;f Services using cc5servative assumptions and simplified

analytical techniques which have been reviewed by the TRC and NRC staff and consultants in 1982. Two years later, Boston Edison Company, in an attempt to redu'ce the amount of modifications required, decided to refine their analytical methodology.~ Two refinements have been incorporated; i) reassessment of bounday strengths and 2) two-way bending analysis i using orthotropic plate properties. It is the objective of this report to I evaluate the new methodologies developed by Computech Engineering Services (CES) t'cr Pilgrim walls. " l i

2. EVALUATION OF BOUNDARY STRENGTHS 2.1 Boundarv lygs Boundary strength of masonry walls is required to adequately transfer the out-of-plane shear to the supporting structure. There are three types of boundarles for shear transfer mechanism at Pilgrim; l) shear friction provided by steel anchores,2) interlocking of blocks at L side boundary and 3) Interlocking of grout at the top boundary of 0-deck.

j The following section presents evaluation of boundary strengths of

] Pilgrim walls which have been performed by Computech Engineerlag Services (1).  !

i 3

l l

2.2 Boundary Strengths Based gn Interlock  !

l 2.2.1 Block Interlock For L side boundary between two masonry walls boundary line loads can be developed due to shear across the plane of interlock. Under out-of-plane loads, this shear is in the mortar bed joints. An allowable value of 34 pst is used for Type S-mortar. Interlocking due to grouting and the higher shear resistance due to friction from the self-weight were ignored which resulted in a conservative estimate of bed joint allowable shear (3).

. A masonry T side boundary was assigned zero strength since detailing does not call for intarlocking of blocks and bond of mortar collar joints can not be relled on for shear transfer.

. 2.2.2 Grout Interlock

(

For top boundary where ribs of the 0-deck run parallel to the wall, mechanical interlock exists since design drawings indicate that grout is forced between the ribs and the top of the wall. An allowable shear stress of 40 pst is used which is a conservative estimate for 2000 psi strength grout. An adequate margin of safety of at least 3 does exist for this type of boundary.

2.3 Boundary Strengths Bgsgion Anchors 2.3.1 Rackground Field inspection of the anchorage conditions at the boundaries of the masonry walls at Pilgrim was conducted in 1982 under the direction l 4

sub groups which is essential for meaningful results.

Because of the inconsistency of exposed lengths, data was converted to number of anchorages per unit length. This procedure is acceptable based on the reasonable assumption that anchors are uniformly distributed.

The analysis has been based on the assumption that underlying ,

distribution is a normal one. Goodness of fit test was performed which indicated that at both the 5R and IOR significance levels the hypothesis of a normal distribution sh,ould not be rejected.

A confidence level of 955 was chosen for Computech statistical analyses which is consistent with other design provisions used in

"" engineering analysis of masonry walls in nuclear power plants.

2.3.3.2 Results Computech Statistical analysis resulted in the following percentage of the " ideal anchors" for each boundary type :

Top boundary to 0-deck 02 Top boundary to structural steel 0%

Top boundary to concrete 19.32 Side boundary to structural steel 332 Side boundary to concrete 202 l

The results revealed unreliable anchores for top boundary to 0-deck  !

6

l and structural steel. l l

2.3.3.3 IntalBoundmev Strenaths l The ideal strength was based on strength criteria ( shear friction transfer ) which is consistent with the extreme conditions considered i

for SSE loading. Assumptions used for the development of anchorage force are conservative.

2.3.3A .Zaca. Data PEnts A number of exposed lengths with no anchores were found. However, the majority of these zero data points were for top boundaries of 0-deck and structural steel for which the statistical analysis revealed zero percentage of the ideal anchores. For side boundaries to structural steel only one exposed length out of 17 has no anchores whereas three exposed lengths out of 20 have no anchores for side boundaries to concrete. No zero data points were reported for top boundary to concrete. This indicated that a small percentage of exposed lengths ( 5% to 15% ) has no anchores. A detailed study of the effect of the zero data points was presented in Appendix A of Ref.1.

The analysis shows tnat the probability of occurrence of the zero data points is 275 for 75 in. exposed length. Therefore, it is apparent that zero data points are expected and acceptable within the context of the overall acceptance criteria of the statistical methodology.

Therefore, no extra data was deemed necessary (2).

7

i i

2.3.3.5 conservatism g_statistien1 Annivsis i Different levels of conservatism are included in Computech statistical analyses;

- Population standard deviation was taken equal to the sample i

standard deviation. l l

Boundarles shorter than the minimum length shall have a zero  !

allowable load All :tero data points were considered in the analysis despite of the fact that there is a very strong correlation between length  ;

exposed and number of anchorages found which Indicates that little weight should be placed on any sample point with relatively short exposed length and no anchorages.

l l

l

3. PLATE ANALYSIS l

\  !

i 3.1 Methodolony l 1

Stresses in the walls are calculated using the working stress method l l

of analysis. Loading on the walls was analyzed at three levels. The l results at each level were compound to the acceptance criteria before proceeding to the next level. The three levels are:

l l- level 1 Analvsis - The natural frequency of the walls was based on fully cracked section properties in the two directions. E value of

)

600f'm was used.

11- level 2 Analvsis - An interative analysis taking into account the status of wall cracking and using effective moment of inertia and E

= 1000 f'm 8 i i  :

111- level 3 Analvsis - Performed to resolve local stresses. It is not critical for wall qualification.

3.2 cracking Moments '

Cracking is assumed to occur when flexural stresses in the vertical direction exceed 68 psi. This is based on twice the allowable tensile stress for factored loading condition. The assumed nodulus of rupture is a low bound estimate (4) and would cause the walls to be stiffer, the .

freg'uency to be higher and the moments to be lower. Use of higher value would result in higher number of walls that can be q'tallfled based on untracked elastic behavior.

^,

1 Flexural tension in the horizontal direction is assumed equal to twice the modulus of rupture in the vertical direction. For partially grouted wills this assumption may not be conservative. However, because low moduli of rupture are used, this assumption would not lead to non-conservative results 3.3 Flevural stiffness l

The fundemental frequency of the walls was based on lower bound values of moment of inertia ( cracked sections ) and modulus of elasticity

( 600 f*m ) and as such is underestimated. However, peak acceleration was used if the wall frequency is less than the frequency of the peak.

i Branson's equation that has been commonly used to calculate 9

i

l effective moment of inertia for reinforced concrete beams was used to predict the effective stiffness of reinforced block masonry walls at Pilgrim ( level 2 Analysis ). Jofreit and McNelce (5) Cmonstrated the applicability of the equation to finite element analysis of two-way reinforced concrete slabs. Since no contribution is given to joint reinforcement Branson equation was not used in the horizontal direction.

Equal modull of elasticity are assumed in the two orthogonal directions; parallel and normal to the bed joints. This assumption is not always valid for block masonry which is a non-Isotropic material (6). It is to be noted, however, that Pilgrim walls are partially grouted in which grouting provides adequate continuity and additional stiffness across the weak bed '

., l joint planes. Therefore, the two moduli should not be significantly l

different for Pilgrim walls and a representative value of 1000 T'm ( for I level 2 Analysis ) in the two orthogonal directions would be adequate for stif finess calculations.

3.4 Review of Samnle Calculations Sample calculations for 5 walls were reviewed at November 21, meeting. Some walls passed by level I analysis whereas other passed by  ;

level 2 analysis. Loading due to differential pressure was, in most cases, l the governing load. Using plate analysis result in higher frequency ( in the )

range of 7-16 Hz ) of the walls and in few cases,a shift to the peak of the response spectra was obtained.

l l

Review of the deflected shapes of the walls at critical sections l

Indicates that the analysis is reasonably capable of predicting the overall  !

10

r .

I wall behavior. It is apparent that conservative assumptions are bu the methodology; namely 1- low bound values for modulus of rupture are assumed.  ;

11- No contribution of Dur-O-Wal joint reinforcement was cons  !

and therefore The walls are assumed to exhibit two-way action only when uncracked in the horizontal direction.

111- cracking moment in the horizontal direction was based o area only, i.e. Ignore contribution of grouting.

4 iv - Multiwythe walls are assumed uncoupled, i.e. they act sepa In addition, maximum stresses are well below the allowables and deflection levels are very low. Thus, it can be concluded that the methodology is conservative.

4. CO'NCLUSIONS The plate analysis performed by CES for Pilgrim walls has been to provide reasonable prediction of behavior in the elastic range.

Conservative assumptions are built into the methodology and displacements and stresses are well below the allowables.

Statistical analysis of the anchorage strengths of Pilgrim walls soundly based on engineering principles and documented m strengths. The proposed allowable line loads for the walls are conservative and, therefore, acceptable.

11

_ - _ _ . . - - .