ML20107A186
| ML20107A186 | |
| Person / Time | |
|---|---|
| Site: | Pilgrim |
| Issue date: | 09/30/1984 |
| From: | COMPUTECH ENGINEERING SERVICES, INC. |
| To: | |
| Shared Package | |
| ML20107A189 | List: |
| References | |
| 560-02, 560-02-R01, 560-2, 560-2-R1, PI2180, NUDOCS 8411010478 | |
| Download: ML20107A186 (72) | |
Text
--
~ e '. "., .. ,
I)
ATTACHMENT 6 ' ( N"hoML h W f
- 4 7 hd4 '[ 4 H*d*'PI2320 i .
PILGRIM NUCLEAR GENERATING STATION - UNIT 1 i
4 i I l
1 !
STATISTICAL ANALYSIS OF BOUNDARY STRENGTHS
- . FOR MASONRY WALLS -
FROM FIELD TEST DATA
^
l .
Prepared For
, BECHTEL POWER CORPORATION San Francisco. California Prepared By COMPUTECH ENGINEERING SERVICES. INC.
I 2855 Telegraph Avenue Berkeley, California 94705 September 1983
. Report 560-02 Revision 1 -
- 4 y e
2 e
. .- - PI2180.
$ ,F""" '"
%g .
t
e 5 .,
- O U .
TABLE OF CONTENTS 1 INTFtODUCTION '
1
.........................I.....
,. t 2 BOUNDARY TYPES ............................ 2 2.1 Top Boundary ................... ....... 2 2.2 Side Boundary ..... ..................... 2 3 TEST PROCEDURES ........................... 3 .
3.1 Specified Procedures . . . . . . . . . . ............. 3
~
3.2 Compilance with Procedures . . . . . . . . . . . . . . . . . . . . 3 3.3 - General Comments on Procedures ...... ......... 4 4 TEST FINDINGS ............................. 6 5 STATISTICAL ANALYSIS OF TEST FINDINGS . . . . . . . . . . . . . . . 9 5.1 Statistical Methodology . . . . . . . . . . . . . . . . . . . . . . . 9 5.2 Summary of Results .......................,. ,
9 6 IDEAL BOUNDARY STRENGTHS .....................11 6.1 Strength Criteria ......................... 11-6.1.1 Steel Yield Strength ....................12 6.1.2 Weld Strength - . . . . . . . . . . . . . . . . . . . . . . . . 12 6.1.3 Bon d Stress . . . . . . . . . . . . . . . . . . . . . . . . . 12 6.1.4 Insert Capacity .......................13 6.1.5 Shear Strength- of Grout ..................13 6.1.6 Shear Strength of Mortar . . . . . . . . . . . . . . . . . . 13 6.2 Ideal Strength (Specified Anchors) . . . . . . . . . - . . . . . . 16 6.3 Strengths Based on Interlock . . . . . . . . . . . . . . . . . . .'17 ,
6.3.1 Masonry L Side Boundary . . . . . . . . . . . . . . . . . 17 6.3.2 Q-Deck (Parallel) Top Boundary ..............18 7 ALLOWABLE BOUNDARY LOADS . . . . . . . . . . . . . . . . . . . . . 19 7.1 Reliab'e Strength (Actual Anchorsr . . . . . . . . . . . . . . . . 19 .
7.2 Exceptions to Generic Results ..................19_
8 CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 9 REFERENCES ..............................25 APPENDIX A
~I e wt
e t
, 1 INTRODUCTION l
This report describes a series of statistical analyses performed
- by Computech Engineering Services. Inc. (CES) for Bechtel Power Corporation (BP ). The purpose i
e of the analyses is to evaluate the boundary strengths of the m&sonry walls at '
Pilgrim Nuclear Generating Station. Unit 1.
In the latter part of 1981 and early 1982. field Inspection of the anchorage conditions at the boundarles of the masonry walls at Pilgrim was undertaken.
This work was performed under the direction of Cygna Energy Services. The
, work presented herein is the statistical analysis of the existing test data, and the incorporation of the results of this analysis into strength calculations for the bounderles of the masonry walls at Pilgrim Station. Unit 1. The end result Is a. set of boundary allowables for use in the evaluation of the walls.
S,ection 2 of the report describes the boundary types under consideration. Sections
- 3 and 4 describe ti e test procedures and test findings respectively. The statistical analysis of the test results is briefly described in Section 5. and the calculation of boundary strengths is described in Section 6. Section 7 presents a summary of boundary strengths for use in the evaluation of the walls, and the report closes with conclusions in Section 8. In Appendix A a detailed description of the statistical analyses is presented.
I 1
- l
[
l l
e*
J
,, .g CES fi a e = ^ -
2 BOUNDARY TYPES in this section the various types of boundaries found in the walls at Pilgrim are described. The top boundarles are described first. followej by the sides.
- 2.1 Top Boundary \
There are essentially three types of top boundaries, although for strength purposes. one of these types is further sub-divided.
A.
Metal 0-Deckino. This is a standard ribbed metal deck over which is poured a concrete stab' forming the floor. Design drawings call for self drilling anchors at 16" centers and for grout to be forced into the space between the top of the wall and the ribs of the ,
, Q-deck. The ribs of the metal deck generally run either perpendicular or parallel to the wall. It is because of this directional distinction that this top boundary category will later be further ,
divided into two sub-categories.
B. Structural Steel WF Section. For this top boundary condition. the wall butts up against a structural steel section. Anchors are specified to be welded to the flange of the steel section at 16*
centers.
C.
Concrete. In this condition, the top of the wall butts up against structural concrete (typically a concrete slab). Again, anchors are specified at 16* centers.
2.2 Side Boundary There are. four distinct conditions found at the side bounderles of the walls.
A.
Structural Steel WF Section. In this condition the side of the well l is in. contact with a structural steel section. Anchors are srecified j -
(welded to the flange) at bond beam locations only with a minimum of one anchor for every two horizontal reinforcing bars.
B.
Concrete. In this case the side of the well is in -contact with '
structural concrete ftypically a column or another well). Anchors -
are specified at bond beam locations only, with a minimum of one -
anchor for every two horizontal reinforcing -bars. .
C.
Intersectina Masonry L. For this boundary condition. the side of t
the wall meets another masonry wall (at right angles). Joint or -
horizontal reinforcement in the wall is specif'ed to be continuous ~
. around the L-joint. Interlocking 'of blocks -is ' also required.
D.
- -intersectino Masonry T. In this case. the side of the masonry wall - ~
' forms -a T junction with another ' masonry ~ wall. Jo!nf (Dur-o-wall) -
l reinforcement is specified 'as continuous throc,- the T-joint. : For , .
l
' this boundary type. Interlocking of blocks will depend on construction -
l sequence. and cannot be generally guaranteed.
CES 2
l i . . _ . - - -
_, , 9-22-89'. -
e 3 TEST PROCEDURES 1
The test procedures used to examine the walls for anchorages are now describec.
I The procedures as required by Cygna are first presented. followed by an interpretation of the compliance with the procedures as deduced frc$1 the reporting g of test results. %
3.1 Spectfled Procedures r -
The following procedures are taken from Cygna document 80034. Wi-6. Revision
. 3. January 1982. ' Work Instruction for Testing of Masonry Walls".
Along the top edge of the wall. three consecutive blocks were required to be ,'
cut away. The cut into the block was to be to a minimum depth of one half the bipck thickness (if anchorage is found). and to a maximum of the entire depth pf grout in the cells, care being taken not to break through the far face shell of the Ofock. Furthermore. If only one dowel was observed in the section of trie wall that was cut away. an additional amount of block was to be cut away along the boundary, to expose a' minimum of 24' on either side of the observed dowel. The wall was considered _ to be in accordance with the design drawings if a dowel was observed in each block that was cut away. (This corresponds to the required 16* spacing.)
Along the side boundaries. four consecutive blocks were to be' cut away. The requirements for the cuts were the same as for the top cuts. including the 24*
exposed length. The well was considered to be in compilance with the design drawings if at least half of the exposed blocks contained dowels positively anchored to the adjacent structure. This corresponds to a 16* spacing as .
e required.
After inspection of the cut-away blocks for the existence of dowels and the
, documentation of their size. number and location. along with the quality of grout and welds. all portions of the walls disturbed or damaged by the testing were
. to be restored with grout.
3.2 Compliance with Procedures
- j. This section evaluates the compilance of the anchorage tests with the appropriate Instruction in
- Work Instruction for Testing of Masonry Walls *. In general, the Individual tests have not been carried out in a consistent manner and in- '
accordance witn tne Instruction. -
- 1. In testing for top . anchorages. the work Instruction ~ requires 3 L consecutive b:ocks to be inspected for dowels, and the well is:
j ~ deemed satisfactory only if all 3 blocks do have dowels. Also,
' number and location . of dowels were to be reported in the test j ,
results. This has not been done in all cases. The well would, be deemed in " fall
- the test if 0.1 or 2 dowe:. . sere observed. As a well would " fail
- the anchorage test as soon as a block is found
~
with no anchor. It 6ppears that many ~ tests have stopped short of i
t .
l -
CES 3- ,
9-22-03' -
j .
t . ,
I the 3-block requirement, although some have gone more' than the 3-block requirement, perhaps looking for anchorages. If the data ,
l had been collected as specified. then the additional Information j
would have been statistically useful in the context of a revised l
analysis of the data. ,
! In testing for side anchorages. the work Instruction requl@s 4 f 2.
l consecutive blocks to be
- chipped.* and the wall is deemed l
satisfactory if at least 2 anchorages are found. This requirement l is reasonable for the wider multiwythe walls where 6 bars are j
i specified every 8 feet. However. for the 8' and 12* walls only 1 or 2. bars are specified every 8 feet where called for. .Thus, even l if these walls were built in accordance with the drawings, they would
! be very likely to " fall
- the side anchora'ge tests.
l l
Again, as for the top anchorages, some violation of the 4-block l
requirement is apparent, although the side tests conform better to the work instructions than do the tests for anchorage on the l .
tops of the walls.
Anomalies such as those mentioned above tend to give a higher scatter in the test data than the inherent verlability due to variations in wall anchorages alone.
I i
i 3.3 General Comments on Procedures j
The statistical procedure used as the basis of the field testing although
- sophisticated. Is not suited for the walls at Pilgrim. Also, application of the procedure in the field has not followed the basis of the method. Nonetheless.
on examination of the test data and the criteria developed for evaluating the test data by Cygna. the conclusion that the anchorages do not conform to design drawings is correct. However, the same conclusion could have been reached with substantially less testing. Also, given the amount of data collected. the final conclusion made by Cygna that no anchorages can be relled upon is unfairly harsh on the evaluation of the wall anchorages. since not all the specified anchorages are required for the structural integrity of the walls.-
t it is the position of CES that the purpose of the test program was incorrectly
- defined. The objective should have been to determine what percentage of the anchorages could be relled upon (with a given confidence)' rather than a yes or no on all of the anchorages being present. Although the test data does
- indicate that the anchorages are not in accordance with the design drawings.
taking absolutely no account of boundary strengths G.e.. presuming that there
- are zero anchors preseno is the most conservative stance possible.
I It is,' however. Nssible to use the eulsting data as the basis . for a statistical-reassessment .of the state of the wall anchorages at Pilgrim Station Unit 1.' to .
j determine the percentage of anchorages present (w' a given confidence) for
,each of -the. boundary types - present at the piant. 'The original test data has l ,
- therefore been re-examined with the aim of extracting a percentage of enchors
. , D
.O _.
~~'
CES '4 e ,, .
1 s ..
which can be relled upon in any given wall. This is briefly described in Section 5 of this report, while the details of the statistical analyses are presented in l Appendix A. The resulting boundary strengths are summarized in Section 7.
I
? I l l
e
.l l
4 -
4 CES .S s-gg-es-
4 TEST FINDINGS A summary of the results from the relevant field tests is giv91 in Tables 4.1 (top boundary) and Table 4.2 (side boundary). A total of 51 $eful tests were f performed on the top boundaries and 37 on the side boundsdies. There were other tests performed to examine the boundary anchorages but insufficient
- nformation was recorded during those tests to includ.e them In the data space.
it should be noted that the top boundary table (Table 4.1) does not distinguish between Q-deck (parallel) and Q-deck (perpendicular). This distinction is made later and described more fully in Section 6.2.
It should also be noted that the side boun'dary table (Table 4.2) has no data on either masonry L or masonry T boundaries. Cygna's conclusions regarding
- the masonry L and T side boundary types were sufficient for the purposes of ,
this work., and were accepted without further study. The relevant conclusions ~
from Cygna work are that at masonry L side bounderles interlocking between l , the. blocks from the two incoming walls can be relled upon, whereas at masonry i
T boundaries. no interlocking of blocks Detween the two sections of the T is apparent. These conclusions are used directly to develop allowable strengths l
for these boundary types in section s.3 of inis report.
l t
t 4
i I
O 6
s O
O e
ces *
- ->>-a=
_ .m .
i i
! WALL BOUNDARY EXPOSED NUMBER OF
. NUMBER TYPE LENGTH (Inches) i ANCHORAGES 1 45.2 Concrete 66 1
62.2 2 Concrete 48 ,'2 63.5 0-Deck 53
'64.4 O Steel 38 64.4 .2 Steel 36 64.6 2
{ O-Deck 51 1 64.12 Concrete 38 2 65.16 Steel ' 43 1 65.21 Steel '
60 1 66.0 Steel 42 1
i 66.2 0-Deck 72.5 66.5 0
- 0-Deck 28 0 66.5 O-Deck 16 66.6 0 Steel 13 1 66.6 Steel 51 66.6 2 Concrete 30 2
- 66.7 O-Deck 48 2
66.11 Concrete 48 1 66.12 0-Deck 47 0 66.18 Steel 50 0 67.1 Steel 58 1 4
68.10 Steel 48 . 0 111.7 O-Deck 40 ~
2 184.2 Steel 38 0 184.4 Steel 48 1 184.7 Concrete 48
,84.8 3
Concrete 49 3 188.2 Steel 48 0 188.3 0-Deck 56 3 -
188.9 Steel 47 0 188.10 Concrete 49 3 191.26 Steel 73 .1 191.35 Steel 81 1 191.49 O-Deck- 64 0 191.55 Steel 48 1 194.21 Q-Deck 59 0 194.21 0-Deck 47 0 194.22 0-Deck 49 0 194.23 Steet 49 1 194.25 0-Deck 32 0 195.14 Steel 48 2
195.18 O-Deck 41 -0 195.22 Steet 68 1 195.23 O-Deck 6 0 195.23 0-Deck 35 _ 0-195.23 0-Deck 12 0 196.6 Steel - 57.5 2.
196.7 Steel 10 0
- 196.7 Steel _60.5 '2
~
, , 209.0 Concrete 44 :1-212.1 Concrete ' 48
- 1 i .
T.Ama a 4.1. purRULTS FROM TOP ROUMMMY TESTS CES 7_ 9-22-83
~
j . .*
j j WALL BOUNDARY EXPOSED NUMBER OF
- NUMBER TYPE LENGTH (Inches) ANCHORAGES 45.2 Concrete 33 1 I 62.2 Concrete 38 2 63.5 64.4.
Concrete
~
Concrete 30 36 f0 10 l l ;
- 64.13- Steel 37 '2 I 65.21 Concrete 33 1 l 66.0 Concrete 33 1 l t 66.2~ Concrete 48 2 i j 66.5 Concrete 40 -
2 '
66.6 Steel 33 2 l 66.6 Steel 33 2 i -67.1 Steel 45 2 67.1- Concrete 43 2 l
,68.10- Concrete 52 2
. 68.10 Concrete 36 2 ll *
. 184.4 Steel 32 1
~ 184.9 Concrete 32 2 188.2 Steet 42 2
- 188.2 Steel 37 2 i
188.3 Steel 43 2 188.9 Concrete 36 1 i 188.10 Steel 38 2 i -
191.26 Steel 33 -
1 .
191.35 Steel 54 3 191.35 Stesi 65 3 191.55 Steel 69 3 194.2 Steel 32 1 194.2 Concrete 32 0 194.22 Concrete 54 3 194.25 Concrete 32 1 195.9 Steel 55.5 3 195.14 Concrete 48 2 196.6 Steel 59 3 198.0 Concrete 31 1' 198.3 Concrete 36 2 209.0 Concrete 38 1 212.0 Steel 33 0
$l TABLE 4.2 : f4ESULTS_FROM SIDE BOUNDARY TESTS I' .
^
. -CES 8 9-22-83^ -
- - - --- - -- - ^ -^--- - - - - -
i .. .
1 5 STATISTICAL ANALYSIS OF TEST FINDINGS in this Section, the statistical methodology used in developing the boundary strengths for the walls at Pligrim Station Unit 1 is briefly described. The results ,
of the statistical analyses are then summarized for use in Section 7. A detailed l
presentation of the statistical analyses is made In Appendix A. j l ;
} - 5.1 Statistical Methodology -
The test data presented in Tables 4.1 (top data) and 4.2 (side data) consists of data taken from inherently different populations. The top data includes tests l' on three different types of boundarles: Q-deck. structural steel and concrete.
- The side data includes tests on concrete and steel bounderles. Because of ,
I differing construction conditions and requireme'nts for each of these boundary types. there is no reason to believe that ecen- population will have the same f l
underlylr!g distribution. Therefore it is imperative that the data be separated into l
sub.-groups, with each sub-group of data containing results from anchorage tests i
- on.similar boundary types. This leads to five sample spaces from five underlying 1
- populations .
4 2 The statistical methodology adopted for each boundary type is very similar. Data from each boundary type is firstly converted to number of anchorages per unit i
length, and the analysis is performed on this set of numbers. In some cases it is convenient to perform an
- Inverse" analysis whereby the data is converted to length per unit anchorage Instead. In either case. a percentage of anchorages l
which can be rolled upon with a certain confidence is calculated. This is 'further described in Appendix A. The confidence level chosen for all analyses is 95%.
This level is consistent with previously adopted confidence levels for interpretation of test data for use in the nuclear Industry.
< For consistency of analysis. It is assumed that the spacing for anchorages on all boundary types is 16". This is agl Intended to represent the design conditon for each boundary, but rather is a convenience for analysis. *ldeal strengths" are subsequently calculated on this basis. making it a simple matter to factor the results of the statistical analyses into the ideal strength to arrive at a reliable
. strength.
5.2 Summary of Results The following results give the percentage of the *ldeal anchors
- for each boundary type which can be relled upon with 95% confidence. These results are drawn ,
from Appendix A.
- 1. Top boundary to Q-deck - 0%
- e. ' Top boundary to structural steet .- 0%
F
- 3. Top bcundary to concrete - 19.3% (80 inc.. i -
- 4. $1de boundary to structural stee! - 33%
(48 inches)-
e 9
g ces e e-en-es
Side boundary to concrete - 20% (80 Inchss) 5.
- deal anchors
- are spaced at l'6* for allanchors" boundary types, which canasbe discussed in Section relled upon 5.1. Inhererit in the percentage of *idealBoundarles shorter than this minimum length is a minimum boundary length.
shall have a zero allowable. load. This minimum length is calcula ideal spacing (16*)
boundarles with zero allowables. this minimum length has no meanin'g. However, for the remaining boundaries. the minimum lengths are given in parentheses above.
O e
W e
D e
e t .*
l .' .
4 ,
10 9-22-83 CES
- l. .
o .
t
l l
S IDEAL BOUNDARY STRENGTHS This Section describes the calculation of the allowable boundary line loads for '
use in the wall evaluation. Section 6.1 discusses the strength criteria and shear transfer mechanisms appropriate at the various bounderles. Sec; ion 6.2 uses l the results from Section 6.1 together with the anchorage prope$les specified l
- on the design drawings, to arrive at ideal boundary strengths. Thoh boundaries which rely on mechanical Interlock for shear transfer (Q-deck parallel and masonry L boundarles) are assigned strengths in Section 6.3.
6.1 ,
Strength Cettesia ,
Chapter 26 of the UBC states that shear friction provisions are appropriate at an Interface between dissimilar materials or an Interface between concrete cast
, at' different times. It is our bellef that this applies to the boundary conditions
, of the 'walls at the Pilgrim Station. Unit 1.
.An ultimate strength formulation is adopted for the boundary allowables. This is consistent with the extreme nature of the loading conditions considered in the wall and boundary evaluation. Selsmic, tornado and PSOC loads are applied
, to the walls. These loads (SSE. tornado and PBOC) are considered ultimate I conditions, and allowable stresses under these loads are generally increased l
from the usual working stress allowables.
Allowable stressed based on shear friction are derived Nom the-total normal force applied across the interface by the reinforcing and the coefficient of friction between the masonry and the boundary material. For reinforced. concrete the total anchorage force is the area of steel times its yleid strength. In the case of the masonry walls at Pilgrim, this force is taken as the feast of the forces 6 j calculated from:
l l 1. the area of. steel anchorage times its yleid strength.
l . ,
- 2. the product of the allowable bond stress around the anchorage times the length of the anchorage times its perimeter.
- 3. the capacity of the weld or Insert -of the anchorage bar into the boundary.
l The lowest of the three so calculated forces is taken as the 'llmiting 'normai force across the boundary.
The ccefficient of friction is assumed to ' be 1.0 for concrete bounderles -and 0.7 for steel bounderles.
l The values for material strengths (such as ' steel yleid strength) have been' based .
on the requirements for -the original design. and experimental data -has been used for bond and insert . capacities as discussed below.
s-t . .
9 CES- ,
11 's-22-83 .
J .
i . .,
i 6.1.1 Steel Yield Strength
) In all strength calculations requiring a yield strengtn for steel. the i following values have been used:
(a) mild steel .rebar: 40 ksi. j
- t (b) steel flat plates and structural sections: 36 ksi.
l S.1.2 Weld Strength l
The usual permissable weld stress of 21 kal has been increased by l i
en ultimate strength factor of 1.67. giving an allowable stress of 35 ksi.
1/4' fillet welds all round have been used as the basis for the tensile strength calculation for the welds. .
. 6.1.3 Bond Stress As a first step in arriving at an allowable bond stress. the values for 4
bond stress inherent in Section 2612 of the UBC were examined. For deformed bars, this section of the code gives the following " development lengths *:
Ld b 0.04 A Ib y / / I'c Ld v 0.0004 d iby Assuming that the development length is based upon ,
Le = A fsy/IObb where OD is the bar perimeter. then titese two equations reduce to:
fb = 7.96 / f*c / db 4 625 psi These equations -Imply the following bond stresses -(as a function of bar diameter) for 3000 psi concrete: for a #5 bar a bond stress of 625.
psi. and for 'a #6 bar, a bond stress 'of 561 psi.
~
From Figure 6.1 ' (reproduced from Park and Paulay [1] Figure 9.11).
for a cover of approximately 5 bar diameters. (typical of . cover to' a ; '
central #5 bar in on 6* block) the allowable bond stress is approximately .
- .1000 psl. To be on the conservative side, a bond stress of 700 psi
- will be used in all strength calculations for deformed bars. This is entirely
~
con:/. tent with the impiled bond stresses in the USC code for the bar
' sizes in question (as calculated above). It should be noted . that the .
Implied code values. should be expected to be conservative. This is again consistent with the values calculated above. Fr:.. Park and Paulay Figures 9.6 and 9.7 (reproduced herein as Figures 6.2 and 6.3), it can be seen that a value of 300 psi. Is a reasonable value for bond stress in plain CES 12 9-22-63 .
[ ,
bars.
These two values (700 pst and 300 psi) are used in subsequent strength calculations involving deformed bars and plain bars or flat plates l respectively.
i<
6.1.4 Insert Capacity The allo'.vable forces for embedded concrete inserts are taken from the Clemson tests on such Inserts (21. From the results of this extensive test program. It is reasonable (and on, the conservative side) to assign a value of 10 times the USC value for the ultimate load on these inserts.
This provides for the following allowable concrete Insert forces: 7500 lbs for #5 bars and 11000 lbs for #6 bars.
~
'6.1.h Shear Strength of Grout In the case where the ribs of the Q-deck run parallel to the wall, mechanical interlock exists at the top boundary since the, design drawings indicate that grout is forced between the top of the wal! and the ribs of the Q-deck. A boundary shear can. be developed as a result of this Interlock.
The grout at Pilgrim Station Unit 1 la specified to hillve a minimum strength of 2000 psl. and it is treated ,as plain concrete..Section 2622(d) of the USC Ccde app!les. which gives an allowable shear stress in plain concrete of 0.02f*e. With a unit strength of 2000 psl. this implies an -
allowable shear stress of 40 pst in the grout. This allowable stress is l used in Section 6.3.
t i .
I 6.1.6 Shear Strength of Mortar .
At side bounderles to other masonry walls where lhterlocking of blocks
~
is demonstrated, boundary line loads can be developed due to shear across the plane of interlock. This is a shear in the mortar between the blocks in question. An allowable - shear stress -is required for this .
Situation.
The mortar at Pilgrim Station . Unit l' is'specified to be type PL of '
standard STMC-476 (1964). with a minimum 20 day strength. of 2000 psi. This standard is now outdated, but type PL corresponds vary closely to type S mortar in the latest standard. Type 8 mortar has a specified minimum strength of 1800 psi. .
Colville (3) Indicates an average shear bond in type-~ M . mortar of 47i psi. It is reasonable to assume in the absence of other data..that the average shear bond for type 8 mortar ccr. -J scaled from the' value- -
j
' for type ' M by the ratio of the minimum compressive ' strengths for the two mortar typt.. This gives ,a value. for. the average . shear bond in type -
l
- CES .
13 9-22-63. .
! s : ,.
e
+
I i !
se, s.,.
4 en mm X g 95 e
- d 17 5 4 1 25 4
- Too nors Sortom Dars 1160 8.0 80 ,
g A " .
i 6.4 6.4 g\
i l i i 928
- \ ! ! l [%, t
- 48 1 996 48 l-
. i. ON I
- i iNW .
8 I .
- i .
. , l 444 .
3.2 32 I e ! ! ;
i l 236
-l 16 16 , ,
- 1 4
I ! il l l I I i 0 0 0 4 4 2 0 4 4 2 0 Cover s., e e.
4
~
Figure 6.1 : The Effect of Cover on Bond Strength O %
W 8 e O
- 14 :9~22*S3 '.
- CES.
- 9 e s e
s.
I
- 1 1 l l \
0 0001 0 00:
0 12 Oct 0 s e. :
0 11 f II !' I . '8
0 10 II i I' I[
- Mi l i l 0 09 '! '
l l Na IMl IMI I h i Y D"'
i c og . I l l h 007 - ! 'I lltI
- 3 } 0 06 WI I I- i
- 8 Il e l *! ! !!I I
- f l o ,00$
j- r i :; e ,
er : ,
!il l
,,, j l
0 03 I ' I' I l- @
1 0 02 ' '. - -
6 4, - p Ho'rescatas 0 01 88f f IbI I 'II i I ll l l ll! I ll} l Omt 0.00s ooi 0 04 0.1 04 toaimmi
$ leo Figure 6.2 :
The Load-Slip Relationships for 85 Plain Round Bar (oses 1
300 0.2 0.3 smme i s geto* **' 'l
-d 40 '
/ Deevee eaac.ag too ter - -
30 j ,
f
/ , ,,,- o.c 20 2
,co t . ~ es _
/ i 8 Too es, ""
30 100 l'
l -
l .
0 /
0.001 0 25 0 0.01 Imp pro.= o I
Figure 6.3 :
'Sond Stress-Silp Relationship' for Plain Round Bars
. W Affected by the Settleme... 4 Fresh Concress)
- t. t ,
l e
.CES
- l15 -
' 9-22-83
i j . ,.
! S (and hence PL) mortar of l
4 47 psi x 1800/2500 = 34 psi.
It is reasonable to use an average strength since along any side boundary there ar,e a large number ofJed joints,where thg mortar is j i in shear. There is also a margin of safety in the cafculatlog since the mortar at the plant was specified with a minimum strength of 2000 psi.
and the scaling was done using 1800 psl. Also. due to long term aging.
4 the mortar is likely to have a current strength considerably in excess of either of these two numbers. This value of 34 psi is consistent with l
the allowable shear for the grout (40 psD developed in Section C.1.5 i and Is used in Section 6.3. .
T 6.2 Ideal Strength (Specified Anchors)
The anchorages specified on the design drawings vary with wall thickness and ,,
, boundary type. Anchorages along the top edge of the walls are specified at ~
16* centers for all wall thicknesses. Anchorages along the side edges are specified at bond beam locations only, with a minimum of one anchor for every centers.
l i
two horizontal reinforcing bars. Bond beams are specified at 8'-0*
1 i
For the case of the top of a wall against a metal deck or a concrete slab.
specified anchors are #5 bars at 16* centers for 8* and 12* walls, and two
- 5 bars at 16* centers for multi-wythe walls. For the case of the. top of a wall against structural steel, specified anchors are #5 bars at 16' centers for2'- ~
8' and 12* walls, two #5 bars at 16' centers for multi-wythe walls less than 6* thick, and two #6 bars. at 16* centers for thicker multi-wyhte walls.
E l
For the case of side boundaries one bar every 8'-0* la specified for 8' walls.
l
' two bars for 12" walls, three bars for 18* walls. four bars for walls between f
2'-0* and 3'-0* thick, and six bars every 8'-0" for S'-6* walls. From - the fleid test program, these anchors were generally #5 bars. While not specified on any of the design drawings. In many instances side anchorages to structural steel sections .were flat plates approximately l' x 1/4" In section.
The actual anchorage properties used for the ideal strength _ calculations are thus generally #5 bars (with 1*.x 1/4" plates for side bounderles to structural steed. with _ #6 bars used where specified on the top boundaries. For the side boundary to concrete. #5 bars are used as the ' basis for. boundary strength, as this dlameter of bar was most often- found . at these boundaries (and 'Its capacity is smaller than the #6 bars specified). Lengths of anchorages are taken . ~
from the standard details applicable to these walls. ,as no test data is 'available on . anchorage lengths. ,
In order to more readily incorporate the findings from the stat!stical analyses. ~
(which related percentages to an assumed ideal spacing of 16*). side boundary Ideal strengths are based on 1 anchorage per wythe at 16* spacing. This allows.
l I direct scaling of the strength by the percentags of c. srs wh6ch can be relied upon from the statistical' analyses, and' normalizes the varied spacing requirements-
! ' for the side anchorages.
\ .
e l , , ,
~
16 e-22-es
,. ces .
l l * .
l On this basis. Table 6.1 is prepared. giving *ldeal* boundary strengths under f the above assumptions. These strengths are then factored by the results from l 7.1.
the statistical analysis of the test data. This is discussed in Section l
ALLOWAELE UNE LOAD Ob/in)
TOP BOUNDARY SIDE BOUNDARY WALL STEEL l CONCRETE THICKNESS Q-DECK l STEEL \ CONCRETE 480 468 390 468 8* 328 480 468 390 468 12* 328 960 937 '
780 937 1 '-6
- 656 A 960 937 780 937 2'-0* 656 960 937 I 780 937 2'-p* 656 1156 937 780 937 2 '- 6
- 656 1156 937 780 937 3*-0
- 656 1156 937 1170 1404 3'-6* 656 TABLE 6.1 IDEAL BOUNDARY STRENGTHS (SHEAR FRICTION FORMULATION) 6.3 Strengths Based on Interlock Two boundary strengths are considered in this section. The first is the case of a masonry L side boundary where field tests and Inspection confirmed that interlocking of blocks at such boundarles is present. The second is the case of the Q-deck top boundary in the situation where the ribs of the metal deck run parallel to the wall in each case there is also another boundary f jeggingly simil,ar type, namely a masonry T side boundary, and a Q-dec ciploifndtry. However. for the masonry T side boundary, fleid inspection and testing Indicated that Interlocking of blocks could not be assumed at this boundary. Also, the presence of norliontal Dur-o-wall reinforcement was doubtful (conclusion from field tests on this type of reinforcement). Thus a masonry T side boundary is assigned zero strength.
In the case of Q-decking running perpendicular to the waii. no strength can ~
be assumed at this boundary for out-of-plane wall loading.
6.3.1 Masonry L Side Boundary Shear is assumed to be transferred across half the area of the block (excluding the central web). at each bed joint. ene allowable shear stress Is 34 psi as developed in Section- 6.1.6. This formulation gives the 17 9-22-83 CES L . .
following allowable boundary line loads:
6* block - 60 lb/in 8* block - 100 lb/in 12' block - 160 lb/in 1
I 6.3.2 Q-Deck (Parallel) Top Boundary For this boundary type. shear is transferred through the grout extending up into the troughs of the metal decking. The basic shear stress is 40 psi (Section 6.1.5). For each wall thickness. the minimum number of complete troughs across the thickness of the wall is determined from
- the dimensions of the Q-decking. This gives a shear area per inch of .
wall, which is then converted to a line load via the allowable stress.
The following allowable line loads result:
WALL ALLOWABLE UNE LOAD d WIDTH Gh/In.)
8' 120 12* 120 .
. 1 *-6* 240 2 *-0
- 360 2*-2* 360 2'-6* 400 .
S*-0* 600 3'-6* 720
.3,gLE 6.2
- ALLOWABLE LOADS - O-DECM (PARALLED a
. CES 16 9-22-83 ,
4- - - , -, ,, . , , - _ . . - , , - -
i 1
1 .
7 ALLOWABLE BOUNDARY LOADS 4
In Section 7.1. the ideal strengths from Section 6 are factored, to account for i
the actual number of anchors which can be relied upon (with g5% confidence)
I at any particular boundary. Exceptions to the
- generic
- strengths based on actual anchors found in a particular wall are presented in Section 7.2.
{ A summary of allowable boundary line loads is presented in Table 7.2. The j
'. particular boundary allowable is categorized by the wall thickness at the boundary.
i and the type and position of the boundary, 4
j 7.1 Reliable Strength (Actual Anchors)
The allowable strength values for boundarles relying on mechanical anchorages
] for shear transfer are calculated from the product of the appropriate ideal strength value from Section 6.2 and the reliable percentage of anchorages (95%
tonfidence) for that particular boundary type from the statistical analyses (Section
{ 5.2).
The appropriate percentage of the ideal strength for each boundary type is summarized in Table 7.1. The final boundary allowables incorporating these percentages, as well as allowables for boundaries relying on mechanical Interlock i
for shear transfer (see Section 6.3) are presented in Table 7.2. .
i i'
?
, SOUNDARY SOUNDARY fELIABLE PERCENTAGE
, TYPE LOCATION OF IDEAL STRENGTH Q-Deck . Top 0%
Steel Top 0%
Concrete Top 19.3 %
Steel Side ' 33.0% '
- Concrete Side 20.0 %
TABLE 7.1 *
SUMMARY
OF RELIAM F PERCENTAGE OF IDEAL STRENGTH 4
7.2 Exceptions to Generic Raoutts
' Nine walls (45.2.184.7.191.35,191.55.194.22.194.23. "95.22.196.6.' and ~ 196.7) -
have one,. boundary which is .' stronger than the generic criteria 'value. This is :
based on the actual anchors 'found along that boundary when the.particular wall was tested. -
i.
For example.: suppose a particular wall is.12 feet ... sength and was._ included ;
j in the test program.' Suppose also that its top boundary was examined and 3
- anchors were found when 60 inches
- of the boundary were ' chipped away.1This.
, f. - -
- . CES- 19-9-22-_83] 7
l
)
i data is then included as a sample point for the generic category of walls with t 4
this particular type of top boundary. Suppose also that upon Completion of the statistical analysis of this data, the conclusion is reached that with 95%
Therefore onone confidence. a 12anchor is present every 6 feet on this kl{d of boundary.
a foot boundary. one would expect to find *2 anchors. The
- generic strengtn is based on this number. But for this particular wall, it is known for a fact that at least
! three anchorages are present on the top boundary.
Therefore value.
this particular wall has a strengtn greater than the generic criteria it is on. this basis that the nine walls me'ntioned above have one boundary strength which is an exception to the generic criteria..A summary of the particular i
boundary strengths for these walls is given in Table 7.3.
f
- e
]
4 B
b 1
O e
O e e CES
& 20. . m
~
I WALL POSITION BOUNDARY ALLOWABLE SHEAR COMMENTS THICKNESS TYPES va ob/IM 8* TOP Q 120 See Note 3 W 0 C 90 See Note 4 l SIDE W 127 See Note 5 i
C 93 See Note 6 M 100 See Note 7
,12' TOP Q 120 See Note 3 W ,0 C 90 See Note 4 t -
SIDE W 127 See Note 5 l C 93 See Note 6 M 160 See Note 7 1 *- 6
- TOP Q 240 See Note 3 W 0 C 180 See Note 4 SIDE W 255 See Note 5 C 185 See Note 6 M =
See Note 8 2'-0* TOP Q 360 to 2*-2* See Note 3 W 0 -
C 180 See Note 4
! SIDE W 25'5 See Note 5 C 185 See Note 6 M
- See Note. 8 >
2 '-6
- TOP Q 480 to 3'-0*
See Note 3 Q 600 3*-0* Wall W 0 C 180 See Note .4 SIDE W 255 See Note 5 l C
' 185 See Note 6 M =
- See Note 8 3*-6* TOP Q 720 See Note 3 W 0 C 180 See Note 4 SIDE W 382 See Note 5 C 278 See Note 6 M =
See Note 8 For Notes see following page.
~
' TABLE 7.2
- An i nwAaLE BOUNDARY LINE E MADS .
. e
\ .
~ .
, 'CES 21 .
9-22-83
j .
1 j
}; NOTES:
i j 1.
i Tabulated allowable shear forceav is in pounds per inch length of boundary.
i .o i 2.
4 Boundary types are as follows: l
- 1 t Q
- Q Deck slab s
W: Structural Steel C: Concrete M: Intersecting masonry wall.
l
- 3.
Strength' applies to walls parallel to the ribs of the Q-deck. For walls perpendicular to the Q-deck ribs, allowable shall be taken l as zero.
- i .
4.
1, Allowable shear applies only to walls with length greater than 6*-
$". For shorter walls, the allowable shall be taken as zero.
} . .
5:
Allowable shear appIles only to walls with height greater than 4'-
l 0".
For shorter walls. the allowable shall be taken as zero.
- 6.
Allowable shear applies only to walls with height greater than 6'-
8".
l For shorter walls. the allowable shall be taken as zero.
7.
i, Values for bounderles formed of masonry walls apply to' walls at t I which Interlocking of blocks from the two walls is apparent. For walls where no interlocking joint exists between the two walls, the allo' * :le boundary force shall be taken as zero.
8.
i For multi-wythe walls with Interlocking masonry side bounderles, allowable shears shall be summed over all wythes for each of the '
intersecting walls. The lowest such calculated shear shall apply to all side bounderles at the Intersection. Note 7 applies to each wythe. The allowable shear for a 6' thick wythe is 60 lb/in.
! 9.
For boundaries at which the fleid Inspection has revealed the
' presence of visible cracks, the allowable boundary stress shall be taken as zero.
l 10.
Excentions based on actual anchors found _ during field survey are listed in Table 7.3.
l TABLE 7.2 HUTEn ,
l l
l.
y i .
e CES -22 9-22-83 -
0 WALL BOUNDARY ALLOWABLE STRESS Ob/In)
I.D. POSITION TYPE ACTUAL CRITERIA , GOVERNING ANCHORS VALUE l VALUE 45.2 Top Concrete 187 90 187 184.7 Top Concrete 134 90 134 191.35 Top Steel 68 0 68 191.55 Top Steel 37 0 37 194.23 Top Steel 52 O 52 195.22 Top Steel 37 0 37 196.6 . Top Steels . 174 0 174 196.7 Top Steel ?' 142 0 142
- 194.22 South Concrete 171 93 171 NOTES:
i .
1.
This table includes only values for which the actual anchors found in the wall tests provided a higher allowable boundary stress than the criteria values.
A JLBLE 7.3 - AlIrMAfARI F BOUNGARY UNE LOADS GAfAl i M INCLUDED IN TEST PROGRAM) l G
O 1 CES M "
8 CONCLUSIONS This report has presented a set of allowable boundary line loads for the masonry walls at Pilgrim Nuclear Generating Station Unit 1. The allowables are soundly based on engineering principles and well documented mate llal. strengths.
- 1 Factored into these allowables are the results from the field' test program performed at the plant. While the results of these tests indicated that the boundary anchorages were not in accordance with the design drawings. detailed statistical analysis of the available test data has indicated that certain percentages of the specified anchors can be relled upon (with 95% confidence). These percentages have been used to factor the ideal strength (assuming anchors as specified on the drawings) to arrive at the set of allowable boundary line loads for use in the masonry wall re-evaluation at Pilgrim Unit 1. '
For ce'rtain . walls, exceptions to the allowable boundary strengths have been presented. These exceptions are for those walls on which boundary strength tests were performed. and for which the actual number of anchorages found In the inspected region provided a greater strength than the statistically dependable (95% confidence) number of anchorages along the entire boundary.
e l
e t
.e CES 24
, 9 22-03 L ,
4 9 REFERENCES i
1 Park. R. and Paulay. T.,
Wiley and Sons.1975.
- Reinforced Concrete Structur
- John .
2 Brown. R. H. and Whitlock. A. R.. " Strength of Anchor Bolts in Grouted Concrete Masonry.' ASCE Journal of Structural Engineering.
Vol.109. No. 6. June 1983.
3 Colville. J.,
- Response to Comments on Appendix.B of Dr. Colville's Report of 2/13/80 on Trojan Masonry Walls'." WRC Submittal. April 1
- 8. 1980.
4 ,UBC 1979 Edition. '
5' ' Benjamin. J.R., and Cornell. C.A.. " Probability. Statistics, and Decision for Civil Engineers.* McGraw-Hill. 1970.
i l
l l
i t
[ .
l i *
- i CES
- 25 m
O e
- 6 9e 4
0 O
e e
O O
e 4 O
O APPENDIX A G
4 b
D t
=
e 9 e . .
22-83 CES
- e
This Appendix presents thn details of the statistical analyses of the test data
- for each of five boundary types -- side boundaries to steel and concrete. and top boundaries to concrete. steel and metal Q-decking. ,
l
. A.1 SIDE BOUNDARY TO STRUCTURAL STEEL t The sample space for side anchorages to structural steel is summarized in Table l A.1. A total of 17 tests were performed on this boundary type. The histogram of this data is presented in Figure A.1. Each data point is plotted as a ratio of theofspecified inch wall). anchors (based on a spacing of 16*, or 0.0625 anchors per TEST WALL EXPOSED NUMBER OF NUMBER ANCHORAGES LENGTH
- NUMBER LENGTH ANCHORAGES UNIT LENGTH
, UNIT ANCHORAGE SS-1 64.13 37.0 -2 0.05405 18.5 SS-2 66.6 33.0 2 0.06061 16.5 SS-3 66.6 33.0 2 0.06061 16.5 i
SS-4 67.1 45.0 2 0.04444 22.5 SS-5 184.4 32.0 1 0.03125 32.0 SS-6 188.2 42.0 2 SS-7 0.04762 -
31.0 188.2 37.0 2 0.05405 SS-8 188.3 18.5
' 43.0 2 0.04651 SS-9 188.10 21.5 38.0 2 0.05263 SS-10 19.0 I
-199.26 33.0 "1 0.03030 33.0 SS-11 191.35- 54.0 3 0.05556 18.0 l SS-12 191.35 65.0 3
! 0.04615 21.7 SS-13 191.55 69.0 3 0.04348 23.0 SS-14 194.2 32.0 1 0.03125 32.0 SS-15 195.9 55.5 3 0.05405 18.5 SS-16 196.6 59.0 3 0.05085 19.7 SS-17 212.0 33.0 0 'O.00000 00 TABLE A.1
- SIDE ANCHORAGER TO STRUCTURAL STEEL Flest of all the ' anchorage per unit length
- data is examined. The . sample statistics for this data are as follows:
- 1. the mean number of anchorages per unit length is- 0.0449, or 72%
of the specified anchors. ,
2.
the standard deviation of number of anchors per unit length is
. 0.0145, .or 23% of the specified enchors. This implies a coefncient of variation of 32%. *
- 3. the correlation coefficient between the exposou length and the number of anchors found is 0.82. Thus the data supports the CES A-1
. 9-22-83
1 ,
I '
- l statement that as longer lengths are exposed. more anchorages .
are likely to be found.
i .
The fact that there is a very strong correlation between lengtte exposed and number of anchorages, found Indicates that little weignt should {be 'placed on any sample point with a relatively short exposed length and no anchorages.
- Using the full . set of data the following analysis is performed. It is asstimed j
that the underlying distribution is normal. Firstly, tne distribution for the population j mean is examined. With the sample statistics:
{ X = 0.72. SX = 0.23. N = 17 -
the standard deviation of the mean is: ,
- X
[Si = 7 = 0.056
. and hence (assuming that the distribution of the mean is normal) It can be
- stated with 95% confidence that the population mean is at least .
0.72 - 1.64 x 0.056 = 0.63 it is assumed that the population standard deviation is equal to the sample standard deviation. This is a reasonable (and probably conservative) assumption in light of the comments made in Section 3.2 regarding the contribution to scatter in the data due to anomalies In the test procedures. Thus the population has
'a mean of 0.63 and a standard deviation of 0.23. Assuming a normal distribution.
these figures imply that over 99.5% of the population lies above the zero point, which is considered satisfactory. Using these population parameters, it can be concluded with 95% confidence that there are at least the following ratio of anchors present ..
0.63 - 1.64 x 0.23 = 0.25 That is. 25% of the specified anchorages are present. This corresponds to one anchorage every 64 inches.
A similar analysis performed on those sample points with an exposed length greater than or equal to 40* indicates that 62% of the anchors could be relled upon with 95% confidence. 62% of the specified anchors' amounts to 1 anchor approximately every 26 Inches. However, there is one data point with a 33 inch -
length and no anchors. This data point ' suggests that examining only, the '40 inch
- data would lead to erroneous conclusions. and the 95% strength is
~s ignificantly less than 62% of. the specified anchors. '
- A detailed study of the effect of various assumptions .r'egarding the *zero', data point (test SS-17) is made. The one data point in question does not effect the sample mean a great deal. but it does have. a'significant effect on the sample (and hence by assumption, the population) standard deviation. "
l ,
. The following assump ans are made regarding the zero data point. Firstly, it
,is ignored completely: 4 6condly, it is replaced with a test finding one anchorage
. ' ~ , , '
[ ., .CES A-2 9-22-83
- in an exposed length of 64* (the conclusion from the first analysis of the data including the zero point), and finally It is replaced with a data point finding 1 one anchorage in an exposed length of 80* (a more conservative assumption).
It is reasonable to replace the *zero* point with another data point having a greater exposed length and one anchorage due to the high corr l elation be number of anchorages found and exposed length'. The .effect on the sample mean and standard deviation is as follows:
Database Sample Mean Standard Deviation i
j All data 0.72' O.23 Ignore test SS-17 0.76 0.15 SS-17 (1 anch. In 64*)
0.73 0.19 SS-17 (1 anch. In 80*) 0.73 0.20 l .
An analysis similar to those previosly described was performed on the third data
'. base (*zero* test replaced witn 1 anchorage in 64*) and the result is that with 95% confidence. 33% of the specified anchors can be relied upon. This can alternatively be interpreted as finding one anchorage every 48.*. Indeed out of the eight tests with exposed lengths greater than 40*. there was 'at . Ioast one anchorage found in each case, which is consistent with the conclusion from this analysis.
As a final substantiation for replacing the *zero* point. an analysis la performed '
l assuming there is a un'rorm anchorage spacing of 48' (the last conclusion).
There are six tests with exposed lengths in the range of 30* to 36* one of l which found zero anchors. The probabilliy of one test out of six (all assumed to have 33" lengths) where no anchors are found is calculated. The binomial distribution is assumed. with the occurrence probability for a single event calculated from the geometry of 48' spacing and 33' exposed length. This gives a probability of 0.29 of not finding any anchors in 33*. The binomial distribution gives a probability of finding one test out of six (33' length) .with no anchors as 0.31. There is a probability of 0.12 of finding zero tests out of six with no
- anchors. Thus the observed occurrence of one test out of six with no anchors in a length of approximately 33' would be a relatively common occurrence for
~
an anchorage spacing of 48' (probability of 31%).
On this basis, the hypothesis that there is one anchor every 48' (33% of the.
specified anchors) is accepted with 95% confidence.
These analyses asye been based on the assumption that the undertying distribution is a normal one. A Kolmogorov-Smirnov goodness of fit test has been performed on the complete set of data points (17 points). At both the' 8% and ~1 0%
significance levels it is concluded that the hypothesis of a normal distribution should not .De rejected.
I As a final check on the above analyses. the " inverse
- set lof data is analysed
. cgs- A-3
, 22 - --
._ - . . . _ _ _ - _ _ _ - -. - . - .- . .- - _ _ . _ _- .. - . - _ ~ _
- l ..
l.
i in a similar manner to that described abo ve. The zero point is again replaced with a test finding one,ancnorage in an exposed length of 64*. and it is concluded with 95% conficence that there is one anchor (;very 47*. This conclusion is extremely consistent with that from the anchorage ,per unit length data, specified and therefore it can be concluded with 95% confidence, that 33% of the ancnors are present on the side boundaries to structural steel.
j A.2 SIDE SOUNDARY TO CONCRETE j
l The sam'ple space for side anchorages to concrete is summarized in Table A.2.
} A total of 20 tests were performed on this boundary type. The histogram of j this data is presented in Figure A.2. Each data point is plotted as a ratio of .
j the oer ideal inch of anchors wall. which are based on a spacing of 16*. or 0.0625 anchors . l
, TEST WALL EXPOSED j NUMBER OF ANCHORAGES LENGTH NUMBER NUMBER LENGTH ANCHORAGES UNIT LENGTH UNIT ANCHORAGE SC-1 45.2 33.0 SC-2 1 0.03030 33.0 62.2 38.0 2 i
SC-3 0.05263 19.0 63.5 30.0 0
{ SC-4 0.00000 00 64.4 36.0 0 I SC-5 0.00000 eO 65.21 33.0 SC-6 1 0.03030 33.0 66.0 33.0
( SC-7 1 0.03030 33.0 66.2 48.0 2 ,
l SC-8 0.04167 24.0 66.5 40.0 2 -0.05000 SC-9 67.1 43.0 20.0 2 0.04651 SC-10 68.10 52.0 21.5 2 0.03846 26.0 i
SC-11 68.10 36.0 2 SC-12 0.05556 18.0 184.9 32.0 2 SC-13 0.06250 16.0 168.9 36.0 SC-14 1 0.02778 36.0 194.2 32.0 0 0.00000 SC-15 '194.22 54.0 00 3 0.05556 SC-16 194.25 10.0 32.0 1 0.03125 SC-17 195.14 48.0 32.0 2 0.04167 SC-18 198.0 24.0 31.0 1 0.03226 SC-19 198.3 31.0 36.0 2 0.05556 SC-20 209.0 16.0 38.0 1 0.02632 38,0 TABLE A.2
- RIDE ANCHOP.^.SFR 70 CONcRFTE i
- .First of all the
- anchorage per unit length' data is examined. The sample '
[ statistics for this data are as follows:
l
- 1.
the mean numtHir of anchorages per unit lengtn is 0.0354. or 57%
, of the specified anchors.
<_ A-4 W
l ..
1 lg 2.
the standard deviation of number of anchors per unit length is O.0183. or 29% of the specified anchors. This implies a coefficient of variation of 52%.
4
- 3. the correlation . coefficient between the exposed length nd ' the
' number of anchors found is 0.70. Thus the data supp6rts the statement that as longer lengths are exposed, more anchorages are likely to be found. -
The fact that there is a strong correlation between length exposed and number of anchorages found indicates that relatively little weight should be placed on any sample point with a relatively short exposed length and no anchorages.
t i ..
Using the full set of data. an analysis similar to that described in Section A.1 4
i
- was performed. However, when a normal distribution was fitted to the data, approximately 5% of the distribution was below zero. This is a physically unrealizable situation, and this analysis was carried no further.
- A strnitar analysis performed on those sample points with an exposed length greater than or equal to 40' Indicates that 52% of the anchors could be relied i
upon with 95% confidence. 52% of the specified anchors amounts to 1 anchor approximately every 31 inches. However there are three data points with lengths of 30". 32* and 36*, all with no anchors. These data points suggest that examining only the '40 incl)* data leads to erroneous conclusion,s. and the 95%
strength is significantly less than 52% of the specified anchors. -
i A detailed study of the effect of various assumptions regarding the *zero" data points (tests SC-3.4 and 14) is now made. The three data points in question do not effect the sample mean a great deal. but do have a significant effect (
! on the sample (and hence by assumption, the population) standard deviation.
The following assumptions are made regarding the zero data points. Firstly they .
are ignored completely. Secondly, they are each replaced with a test finding i
' one anchorage in a length 32" greater than the actual test length -(one anchor in 32' was the conclusion from the 40' data), and finally they are replaced with data points finding one anchorage in an exposed length of 80* (a more 1
conservative assumption).' It is reasonable to replace the *zero* points with an equal number of data points having a greater exposed length and one anchorage.
due to the high correlation between number of anchorages found and exposed length. The effect on the sample rneen and standard deviation is as follows:
Datanese Sample Mean Standere Devlamon All data 0.57 0.29 Ignore tests SC-3.4 and 14 0.67 ~ 0.18 SC-3.4 and 14 (add 1 anch. In 32*) 0.60 0.23 SC-3.4 and 14 (1 anch. In 80*) 0.60 0.24 CES A-5 9-22-83
i An analysis similar to those previosly described was performed on the third data base (*zero* tests replaced with an equal number of tests finding 1 anchorage in a length 32* greater than the actual test length) and the result is that with 95% confidence.14% of the specified anchors can be relled bpon. This can ,
l
. alternatively be interpreted as finding one anchorage every 112*.
t Again, as a substantiation for replacing the *zero* points, an analysis is performed assuming there is a uniform anchorage spacing of 112" (the last conclusion).
There are eleven. tests with exposed lengths in the range of 30* to 36* three of which found zero anchors. The probability of three tests out of eleven (all assumed to have 36' lengths) where no arichors are found is calculated. The j binomial distribution is assumed, with the occurrence probability for a single event calculated from the geometry of 112" spacing and 36* exposed length.
This gives a probability of not finding any anchors in 36' of 0.67. The binomial -
distribution gives a probability of finding three tests out of eleven (36' length)
! with no anchors as less than 1%. This is a very rare event. It being most likely
- Jo find seven or eight tests out of eleven with no anchorages in 36' given a i
uniform spacing of 112* for the anchorages and random positioning of the 36*
length. This analysis indicates that the 112" spacing may be quite conservative.
In an attempt to clarify the situation the
- Inverse
- set of data is analysed in a similar manner to that described above. The zero points are again replaced as described above and it is concluded with 95% confidence that there -la one anchor every 75*. In light of this conclusion the probability of occurrence of the zero data points is re-examined. It is concluded that there is a 7% chance of finding three tests out of eleven (of 36' length) with no anchorages. Thus the zero points are a much more likely occurrence under the 75* spacing than under the 112* spacing. This 75* spacing is increased to 80* (an even multiple of the 8" blocit heigh 0.
These analyses have been based on the assumption that the underlying distribution '
is a normal one. A Molmogorov-Smirnov goodness of fit test has been performed on the complete set of data points - (20 points). At both the 5% and 10 %
significanc'e levels it is concluded that the hypothesis of a normal distribution should not be rejected.
On this basis, the hypothesis that there is one anchor every 80* (20% of the specified anchors). Is accepted as the 95% strength.
A.3 TOP 900NOARY TO NmONTE i
- The sample space for top anchorages to concrete is summarized in Table A.S.
A total of 10 tests were performed on this boundary type. The histogram _of -
this data is presented ' In Figure A.S. Each data ] nt is plotted es a ratio of the specified anchors (based on a spacing of 16*, or 0.0625 anchors per inch of wall).
1s 4
CES O-O -
-N-
i
.' i TEST WALL EXPOSED NUMBEH OF ANCHORAGES ENGTH NUMBER NUMBER ENGTH ANCHORAGES UNIT ENGTH UNIT ANCHORAGE f
I 1 TC-1 45.2 66.0 2 0.03030 33.0 TC-2 62.2 48.0 2 0.04167 24.0 4
TC-3 64.12 38.0 2 0.05263
' 19.0 TC-4 66.6 30.0 2 0.06667 15.0 TC-5 66.11 48.0 1 0.02003 48.0 TC-6 184.7 48.0 3 0.06250 16.0 j TC-7 184.8 49.0 3 0.06122 16.3 +
i TC-8 188.10 49.0 3 0.06122 16.3 TC-9 209.0 44.0 1 0.02273 44.0 TC-10 212.1 48.0 1 0.02083 48.0
)
~
TABE A.3
- TOP ANCHORAGES TO CONCRETE First of all the
- anchorage per unit length
- data is examined. The sample
, statistics for this data are as follows: ,
!. 1. the mean number of anchorages per unit length is 0.0441.' or 71W
(, of the specified anchors.
)
- 2. the standard deviation of number of anchors per unit length is ,
0.0180, or 29% of the specified anchors. This impiles a coefficient of variation of 41%.
From a visual examination of the histogram of this data (Figure A.3). there is I
no reason to fit any distribution other than a uniform distribution. The question arises as to what parameters should be given to the uniform distribution, f.e..
what are its lower and upper limits?
Using the sample statistics above to fit a uniform distribution, the lower bound is calculated as 0.21. and the upper bound as 1.21. This gives some guidelines for estabilshing the lower bound at around 0.2. It also gives weight to the logical choice of 1.0 for the upper bound on the distribution. However, the 95% strength is very sensitive to these assumed bounds for the distribution, and these were recalculated based on the 7 tests with lengths ranging from 44* to 49*. (This range is very close to the test requirement of 44*.) With this data base, the ,
sample mean is 0.0416 anchorages per unit length (87% of those specified),
with a sample standard deviation of 0.0186 (30 %). When the above analysis is repeated using these nuenbers, the lower bound for the uniform distribution is 0.15 and the upper bound is 1.19.
In view of these two analyses, it is reasonable to asMon the lower end of the uniform distribution at 0.15 and the upper bound at 1.u. This assumed distribution
!I -
. Is plotted ln' Figure A.3. 95% of this distribution lies above.
- 4
,. CES A-7 9-22-83
~ . _ _ _ ___. , ___ - _ __ _. _ . . _. _ _ ._ ._ _ .
i i
s j
(1.0 - 0.15) x 0.05 + 0.15 r 0.193 d
or 19.3% of the specified anchors. ,
l Thus 19.3% of the specified anchors at the top boundary to concrete can be relled upon with 95% confidence. This is equivalent to 1 anchorage every 80 1
i inches. !
~
A.4 TOP BOUNDARY TO SMUCTURAL STEEL The sample space for the top anchorages to structural steel is summarized in Table A.4. A total of 23 tests were performed on this boundary type. The ,
histogram cf this data is presented in Figure A.4. Each data point is plotted as a .ratto of the specified anchors (based on a spacing of 16* or 0.0625 anchors per inch of boundary).
TEST WALL EXPOSED NUMBER OF NUMBER ANCHORAGES NUMBER LENGTH ANCHORAGES UNIT LENGTH TS-1 64.4 38.0 2 0.05263 TS-2 64.4 36.0 2 O.05556 TS-3 65.16 43.0 1 0.02326 TS-4 65.21 60.0 1 C.01667 TS-5 66.0 42.0
' 1 0.02381 TS-6 66.6 13.0 1 0.07692
- TS-7 66.6 51.0 2 0.03922 TS-8 66.18 50.0 0 0.00000 TS-9 67.1 58.0
' 1 0.01724 TS-10 88.10 48.0 0 0.00000 TS-11 184.2 38.0 0 0.00000 TS-12 184.4 48.0 1 0.02083 TS-13 188.2 48.0 0 0.00000 l TS-14 188.9 47.0 0 0.00000 TS-15 191.26 73.0 1 0.01370 TS-16 191.35 81.0
- l. 1 0.01235 TS-17 191.55 48.0 1 0.02083 l TS-18 194.23 49.0 1 0.02041 TS-19 195.14 48.0 2 0.04167 TS-20 195.22 68.0 1 0.01471 TS-21 196.6 57.5 2 0.03478 -
TS-22 196.7 10.0 0 0.00000 TS-23 196.7 80.5 2 0.03306 TABLE A.4
- TOP ANCHORAGES TO STRUCTURAL STEEL s-4 -
CES A-8 9-22-83
f l
l It should be noted that more than 25% of the data'!6 of 23 tests) is' lumped at the zero point. Four of these six data points come from tests with considerable 4 length (47 inches or more). It is thus apparent that substitution of data points
, for this zero data is a rather suspect course of action In this cpse. Also. with ,
, many data points having large exposed lengths and zero anchors.!the correlation
- coefficient between exposed length and number of anchors fourid will be low.
l Again this warns against substution of' data.
With these considerations in mind. the only possible value for the 95% strength l
. ~ for this boundary type is zero. i 4
! A.5 TOP BOUNDARY .TO O-DECK
, The sample space for the top anchorages to metal Q-deck is summarized in Table. A.S. 'A total of 18 tests were performed on this boundary type. The
. histogram of this data is presented in Figure A.S. Each data point is plotted l
as a ~ ratio of the specified anchors (based on a spacing of 16*. or 0.0625 anchors per inch of boundary).
4 i TEST WALL EXPOSED NUMBER OF ANCHORAGES NUMBER NUMBER LENGTH ANCHORAGES UNIT LENGTH i .
' TQ-1 63.5- 53.0 0 0.00000
'TQ-2 64.6 51.0 1 0.01961 i TQ-3 66.2 72.5 0 0.00000 TQ-4 66.5 28.0 0 0.00000 TQ-5 66.5 16.0 0 0.00000
( TQ-6 66.7 48.0 2 0.04167 i TQ-7 66.12 47.0 0 0.00000 l TQ-8 111.7 40.0 2 0.05000 TQ-9 '188.3 56.0 3 0.05357 TQ-10 191.49 64.0 0 0.00000 TQ-11 .194.21 59.0 0 0.00000 TQ-12 194.21 47.0 0 0.00000 TQ-13 194.22 49.0 0 0.00000 TQ-14 194.25 32.0 0 0.00000 1 TQ-15 195.18 41.0 0- 0.00000 TQ-16 195.23 6.0 0 0.00000 TQ-17 195.23 35.0 0 0.00000 TQ-18 195.23 -12.0 0 0.00000 ,
~
. TABLE A.5 : TOP ANQHORAGES TO O-DECK .
lt is readily apparent from a visual inspection of the histogram that no anchorages can be relied upon with any confidence for this bouncery type. This is due to i.
( , the entremely large (70%) proportion of the sample points for which no
. anchorages were found. .
. CES A-9 9-22 '
( :
, Number of Tests 1
1 4-5 N 3- N N N N
' . N N 5 .
N N\ 5
' \ N \\ \
iQ '
TN g NN N s .N> D N N\ \
=
O.1. O.A O.Co 06 l.0 Ratio of Specified Anchorages
. Figure A.1 : Side Anchorages to Steel Boundary Number of Tests d i 4' 7 N
3 5 N X 2
N \
(s sN T 5\
I's s\ T N N\
\
s NN N\ h (N T Nh) s\
N\
(\ '
0,Z 04 06 06 l'o Ratio of Specified Anchorages Figure A.2 : Side Anchorages to Concrete woundary G
4 g CES 9-22-83
, A-10 6
Number of Tests b
3" Assumed Uniform Distribudon Funcuon
\
2' K \
K N i N 1
S: _ _ _ )__g _ ___ % _ ._ _A N, s_.y ds > ,
\ N .
N\ _
- 07. o+ o C. OS to Rado of Specified Anchorages Figure A.3 : Top Anchorages to Concrete Boundary Number of Tests 4,
4 8"7
\
5-A-
N i
N 5 \
i N 5 1-N 3 gi N\ \
N I'
N ' \ \ ,
s
\ s \ s N g I' g N s \N 5 N ,
\ \ ( \ _
O.2. O'4 04 ot 80 i.1 Ret?
Specified Anchorages
(
- Figure A.4 : Top Anchorages to Seset Boundary
- O A.]i 9-22-83
l
.* ?..
~,,
l Number of Tests ' 1 L
M 7
\ . .
N gg . N .
\
- N ,
\ '
fo-N N
\
N 8-N
\ '
N .
(, - N N
\ -
\
4-N N *
\
- g. \
N N -
m
\ \
\ \
\
\
O.2 6'$ 64 06 -
l.C Rateo ut Specified Anchorages Figure A.5 : Top . Anchorages to Mt &kk e A-12 ,
W-
ATTACHMENT 7
( R ,'pa % r b W w s 1'M 4 v{IDP4lioc*d1)
PILGRAM NUCLEAR GENERATING STATION UNIT 1 STATISTICAL ANALYSIS OF BOUNDARY STRENGTHS FOR MASONRY WALLS
- FROM FIELD TEST DATA O
s
- 9 INTRODUCTION i
9 RESULTS S STATISTICS REVIEW G DETAILS OF ANALYSES i
e i
s 4
1
INTRDDUCTION G BOUNDARIES OF MASONRY WALLS AT PILORIM WERE
- CHIPPED AWAY" OVER LIMITED LENGTHS TO CHECK ANCHORAGE i
' CONFORMANCE WITH CONSTRUCTI'ON DRAWINGS.
4 ANCHORAGES D6D NOT CONFORM TO SPECIFICATIONS: HOWEVER.
THERE WERE A SUBSTANTIAL NUMBER OF ANCHORS PRESENT.
O STATISTICAL ANALYSES OF THE FIELD TEST DATA WERE PERFORMED TO DETERMINE WHAT PERCENTAGE OF THE ANCHORS SPECIFIED COULD BE REUED UPON WITH 95% CONFIDENCE.
- i l
l l
r t
i l
a l . .
m _--
RESULTS 9 DRAWINGS SPECIFIED ONE ANCHOR EVERY 16' FOR ALL BOUNDARIES.
O THE FOLLOWING ARE CONCLUSIONS FROM THE STATISTICAL ANALYSES OF
. THE TEST DATA:
O SIDE - STEEL 48" SIDE - CONCIETE 80*
TOP - CONCfETE 80*
TOP - STEEL NO ANCHORS TOP - Q-DECK NO ANCHORS .
O THESE* RESULTS ARE AT THE 95% CONFIDENCE LEVEL 3 .
I 4
1 i
BASIC STATISTICS N
e SAMPE MEAN a X, j
tal
, 9 SAMPE STANDARD DEMATION N i
{ ~
i
- s. p ,tal2 (Xt-X)t j
e POPULATION MEAN IS A RANDOM VARIABL$ WITH:
~ .
MEAN =X STANDARD DEMATION =.E.
i I
O POPUl.ATION STANDARD DLVIATION TAKE 7=3 l
l I
I l
1 4
4 i
1 I
i i
i i
i t POPULATION MEAN d
- s W
.I Area m o.os i
fk w
=R X
i e
WITH 95% CONFIDENCE.
4 .
POPULATION MEAN =
h X - l.64 WE USE THIS VALUE FOR THE POPULATK)N MEAN.
G
.5
4 i
j l
4 POPULATION STANDARD DEVIATION I
WE TAKE THE POPUl.ATION STANDARD DEMATION (e) TO BE EQUAL TO THE i'
, SAMPLE STANDARD DEVIATION (3 ).
1 .
0" = s 1
THIS IS CONSERVATIVE BECAUSE SOME SCATTER IN THE SAMPl.E IS A RESULT 1 .
OF THE FIELD TEST PROCEDURES.
1 NOT ALL TESTS WERE PERFORMED OVER THE SAME LENGTH j
RANGE IS FROM 13" TO 81*
6
i i l i
i i .
i I
, e l
i I i
1 POPULATION STATISTICS AND DISTF4BUTION l
THE POPULATION IS CHARACTERIZED BY:
. MEAN =
STANDARD DEVIATION = 7 THE DISTRIB,UTION IS ASSUMED NORMAL AND A " GOODNESS OF FIT
- TEST IS PERFORMED TO CONFIRM THIS ASSUMPTION.
t
'IN ONE CASE. THE POPULATION IS ASSUMED TO HAVE A UNIFORM DISTRIBUTION.
7
k 1
i i
l j RELIABLE ANCHORAGES 1
4
?
l . ,
FOR EACH BOUNDARY TYPE. USING THE POPULATION ~ STATISTICS AND THE DISTRIBUTION. THAT STRENGTH ABOVE WHICH 95% OF THE POPULATION UES IS CALCULATED.
j l FOR A NORMAL DISTRIBUTION. THIS IS:
hbb) -
r_ .
I 4rea = 0.05 i
U
. ; a dl x,s..pi. me.n i-
/4 3 population mean Renble4frenph 4 95% >Cidence
=f - l ca+ c-a
- ~ , ,a , . ,
. 1 l
l ZERO STRENGTHS IN SOME CASES. A LARGE PROPORTION' OF THE TEST DATA GIVES
, ZERO ANCHORS IN SIGNIFICANT EXPOSED LENGTHS.
IN THESE CASES. NO STRENGTH FROM ANCHORS IS RELIED UPON.
a a
d 9
SIDE BOUNDARY TO STEEL CONCLUSION: 1 ANCHOR EVERY 48' (33% OF SPECIFIED)
SAMPLE: 17 TESTS X = 0.72, s = 0.23.
( = 0.82 1 ZERO DATA POINT (33*)
UTTLE WEIGHT SHOULD BE PLACED ON
- SHORTER
- DATA.
.10
e 4 SIDE ANCHORAGES TO STRUCTURAL STEEL b1 TEST WALL EXPOSED NUMBER OF ANCHOPAGES LENGTH NUMBER NUMBER LENGTH l
ANCHORAGES UNIT LENGTH UNIT ANCHORAGT l
SS-1 64.13 37.0 2 0.05405 18.5 !
SS-2 66.6 33.0 2 0.06061 16.5 SS-3 66.6 33.0 2 ,0.06061 16.5 ,
SS-4 67.1 45.0 2 0.04444 22.5 SS-5 184.4 32.0 1 0.03125 32.0
.SS-6 188.2 42.0 2 0.04762 21.0 SS-7 188.2 37.0 2 0.05405 18.5 SS-8 18'8.3 43.0 2 0.04651 21.5 SS 188.10 38.0 2 0.05263 19.0 SS-10 191.26 33.0 1 0.03030 33.0 SS-11 191.35 54.0 3 0.05556 18.0 SS-12 191.35 65.0 3 0.04615 21.7 SS-13 191.55 69.0 3 0.04348 23.0 SS-14 194.20 32.0 1 0.03125 32.0 SS-15 195.9 55.5 3 0.05405 -
18.5 SS-16 196.6 59.0 3 0.05085 19.7 SS-17 212.0 33.0 0 0.00000 00 Numtier of Tests N
I
]
N N 3
x x l
2' N \
l X N TN N l
\ N \\ \
1-s TN NN N D W W W 9 F Ng N \ \
l 0 1. 04 C'C, 06 10 l Rollo of Spectand Anchorages E
I i
SIDE BOUNDARY TO STEEL i
i 4
USING DATA GREATER THAN 40' (8 TESTSi j
11:%> 1 ANCHCR EVERY 26' (62% OF SPECIFIED)
BUT THIS CONCLUSION IS POSSIBLY VIOLATED BY 1 DATA POINT WITH NO ANCHORS IN 33*.
Q STRENGTH IS LESS THAN 82% -
m l
l l
i 11-l: ..
( _. -
l SIDE BOUNDARY TO STEEL IF WE REPLACE THE ZERO POINT WITH 1 ANCHOR IN 84" (THE CONCt.USION FROM ALL THE DATA):
9 i ANCHOR evERY 4a-S THfS IS SUBSTANTIATED BY 8 TESTS WITH LENGTHS > 40" WHERE AT LEAST ONE ANCHOR WAS FOUND IN EACH CASE.
S ALSO THERE ARE 6 TESTS IN THE RANGE 30" TO 36". ONE OF WHICH OfVES NO ANCHORS.
BASED ON THE 48" SPACING. THIS EVENT HAS A PROSABluTY OF 31 %
AND TWO OUT OF SDC WITH NO ANCHOfB WOULD BE EXPECTED 32%
OF THE TIME.
1 THUS. THE ZERO POINT IS QUITE IN KEEPING WITH THE 48* SPACING.
ON THIS BASIS. WE MPERT 1 ANCHOR EVERif 48*
1 12.
. i
4 l
l 4
i l
i SIDE BOUNDARY TO CONCRETE 4
i CONCLUSION: 1 ANCHOR EVERY 80" GO% OF SPECIFIED)
~
i
- . SAMPLE: 20 TESTS
- i. .
X = 0.57 l s = 0.29 D4 = 0.7 3 ZERO DATA POINTS C50*. 32". 36*)
i
- LITTLE WElGHT SHOULD BE PLACED ON " SHORTER" DATA i
i t
i.
13
SIDE ANCHORAGES TO CONCRETE TEST WALL. EXPOSED NUMBER OF ANCHORAGES LENGTH NUMBER NUMBER LENGTH ANCHORAGES UNIT LENGTH UNIT ANCHORAGE SC-1 45.2 33.0 1 0.03030 33.0 SC-2 62.2 38.0 2 0.05263
~
19.0 SC-3 63.5 ' 30.0 0 0.00000 00 SC-4 64.4 36.0 0 0.V0000 M SC-5 65.21 33.0 1 0.03030 33.0 S.C-6 66.0 33.0 1 0.03030 33.0 SC-7 66.2 48.0 2 0.04167 24.0 SC-8 - 66.5 40.0 2 0.05000 20.0 SC-9 . 67.1 43.0 2 0.04651 21.5 SC-10 68.10 52.0 2 0.03846 26.0 SC-11 68.10 36.0 2 0.05556 18.0 SC-12 184.9 32.0 2 0.06250 16.0 SC-13 188.9 36.0 1 0.02778 36.0 SC-14 194.20 32.0 0 0.00000 00 i SC-15 194.22 54.0 -
3 0.05556 . 18.0 l
SG-16 194.25 32.0 1 0.03125 32.0 SC-17 195.14 48.0 2 0.04167 24.0 SC-18 198.0 31.0 1 0.03226 31.0 SC-19 198.3 36.0 2 0.05556 18.0 ,
SC-20 209.0 38.0 1 0.02632 38.0 Number of Tests l
l f< 3
\
5 -
\ N z .
N \
8 y qN T SN s s\
- 's s\ s sNT s
N y 3
5Z y
04 NN )5 Y N 04 NT s N\
i
- 04' Q
\
i.o Rodo of Spec 0Aed Anchereges
l l
SIDE BOUNDARY TO CONCRETE USIND DATA GREATER THAN 40" (6 TESTS)
% 1 ANCHOR EVERY 32" (50%' OF SPECIFIED)
. HOWEVER. WE HAVE 30*, 32" AND 36* LENGTHS WITH NO ANCHORS GMNG POSSIBLE VIOLATION OF THE ABOVE ASSUMPTION.
% STRENGTH IS LESS THAN 50% .
e l
l t 14
SIDE BOUNDARY TO CONCRETE IF WE REPLACE THE ZERO POINTS WITH AN EQUAL NUMBER OF DATA ,
POINTS ADDING 32* AND 1 ANCHOR (THE CONCLUSION FROM THE 40* DATA)
TO THE ACTUAL RECORDED DATA. AND ANALYZE LENGTH / ANCHORAGE DATA:
Q 1 ANCHOR EVERY 75*
G THERE ARE 11 TESTS IN THE RANGE OF 30* TO 36*, THREE OF WHICH HAVE NO ANCHORS.
) SASED ON A 75* SPACING. TH48 EVENT HAS A PROSABIOTY OF 7%
OTHER PROSA81UTIES ARE:
4 ZERO TESTS: 15%
5 ZERO TESTS: 22%
6 ZERO TESTS: 23%
7 ZERO TESTS: 17%
THUS. THE TNfEE ZERO POINTS AfE AT THE LOW END OF WHAT WOULD BE EXPECTED FROM A 75" SPACING.
ON THIB _"^** WE ANS8'T 1 ANCHOR n= 80*
(AN EVEN MULTIPLE OF THE SLOCK DEIGHT) 15
- - 6 ---+,.m -
--w-w- -
=w.w ~ - -
'9
4 TOP BOUNDARY TO CONCRETE CONCLUSION: 1 ANCHOR EVERY 80* (20% OF SPECIFIED)
, SAMPLE: 10 TESTS X = 0.71 s = 0.29 = .09 NO 22RO DATA POINTS
^
WEIGHT SHOULD BE PLACED EQUALLY ON ALL CIATA POINTS DUE TO LOW CORRELATION BETWEEN NUMBER OF ANCHORS AND EXPOSED LENGTH.
Y 16'
i TOP ANCHORAGES TO CONCRETE l
i 1
l TEST WALL EXPOSED NUMBER OF ANCHORAGES LENGTE NUMBER NUMBER LENGTH ANCHORAGES UNIT LENGTH UNIT ANCHOMGE TC-1 45.2 66.0 2 ' 0.03030 33.0 TC-2 62.2 48.0 2 O.04167 24.0 TC-3 64.12 38.0 2 0.05263 19.0 TC-4 66.6 30.0 2 0.06667 15.0 TC-5 66.11 48.0 1 0.02083 48.0 TC-6 184.7 48.0 3 0.06250 16.0 TC 184,8 49.0 3 0.06122 16.3 TC 188.10 49.0 3 0.06122 16.3 TC-9 209.0 44.0 1 0.02273 44.0 TC-10 212.1 48.0 1 0.02083 48.0 Number of Tests <
JL 3-w uniworvii - em
. 2- g g N -
N 1-
, _ _ _ _ _ _ h y __E___ _
g ___
% _ _ . _ A N
! ds > ~
N N N\ s l 61 o*+ 06 C>6 le MM ." _Z$C- ;_
l i
. j
a x --
9 0
- 9 TOP BOUNDARY TO CONCRETE A UNIFORM DISTFUBUTION IS ASSUMED LOWER AND UPPER UMITS ARE ESTABUSHED BY FITTING MEAN AND STANDARD DEVIATION OF THE SAMPLE TO THOSE OF THE ASSUMED DISTRBUTION.
THIS CONFIRNS AN UPPER UMIT OF 1.0. LOWER UMIT IS ESTABUSHED CONSERVATIVELY AT 0.15.
USING THIS DISTRSUTION. WE ACCEfrT A STFENGTH BASED ON 1 ANCHOR EVERY 80*
THIS IS SOUNDLY CONFIRMED BY ALL DATA POINTS.
l 17
, _ , , - . ., ~
n --
4 -
i.
I i .
I i
l l.
l TOP BOUNDARY TO STFFt 1
i I
i 8 OF 23 TESTS HAVE 2ERO ANCHORS. AND 4 OF THESE 6 l
j ARE FROM TESTS WITH CONSIDERABLE LENGTHS.
9 i
- REtv ou no ANCHORS FoR Tu.S sou.nwir rvre.
l .
d 4
b l
l l
E
- l. 18
TOP ANCHORAGES TQ STRUCTURAL STEEL TEST WALL EXPOSED NUMBER OF ANCHORAGES NUMBER NUMBER LENGTH ANCHORAGES UNIT LENGTH TS-1 64.4 38.0 2 0.05263 TS-2 64.4 36.0 2 0.05556 TS-3 65.16 43.0 1 0.02326 TS-4 65.21 60.0 1 0.01667 TS-5 66.0 42.0 1 0.02381 TS-6 66.6 13.0 1 0.07692 TS-7 66.6 51.0 2 0.03922 TS-8 66.18 50.0 0 0.00000 TS-9 67.1 58.0 1 0.01724 TS-10 .
68.10 48.0 0 0.00000 TS-11 184.2 38.0 0 0.00000
~
"TS-l'2 184.4 48.0 1 0.02083 TS-13 188.2 48.0 0 0.00000 TS-14 188.9 47.0 0 0.00000 TS-15 191.26 73.0 1 0.01370 TS-16 191.35 81.0 1 0.01235 TS-17 191.55 48.0 1 0.02083 TS-18 194.23 49.0 1 0.02041 TS-19 195.14 48.0 2 0.04167 TS-20 195.22 68.0 1 0.01471 TS-21 196.6 57.5 2 0.03478 TS-22 196.7 10.0 0 0.00000 TS-23 196.7 60.5 2 0.03306 .
Number of Tests 6
j ~ (e a m
\
5-4-
N N
3 N
! N %
t- N N
N s N N -
l g
N sh I'
N N '
g N s Nw 5 l N s N s N Ns N> < N _
e y w w V l 04 0'4 04 o.S 8.o i .2, M of Spectned Anchorages
. l
. j
. I 1
l l
TOP BOUNDARY TO O-DECK r
! 14 OF 18 TESTS FOUND ZERO ANCHORS.
I RELY ON NO ANCHORS FOR THIS BOUNDARY TYPE.
w l
l l
l i
19 l .
l ' '
. )
2
, TOP ANCHORAGES TO O-DECK l
l TEST WALL EXPOSED _ NUMBER OF ANCHORAGES !
NUMBER NUMBER' LENGTH ANCHORAGES UNIT LENGTH l l
TQ-1 63.5 53.0 0 0.00000 TQ-2 64.6 51.0 1 0.01961 TQ-3 66.2 72.5 0 0.00000 TQ-4 66.5 28.0 ' 0 0.00000 l TQ-5 ' 66.5 16.0 O 0.00000 TQ-6 66.7 48.0 2 0.04167 TQ-7 66.12 47.0 0 0.00000
. TQ-8 111.7 40.0 2 0.05000 TQ-9 188.3 56.0 3 0.05357
- TQ-10 191.49 64.0 0 0.0000G
. TQ-11 194.21 59.0 0 0.00000 TQ-12 194.21 47.0 0 0.00000 TQ-13 194.22 49.0 0 0.00000 TQ-14 194.25 32.0 0 0.00000 TQ-15 195.18 41.0 0 0.00000 TQ-16 195.23 6.0 0 0.00000 TQ-17 195.23 35.0 0 0.00000 TQ-18 195.23 12.0 0 0.00000 -
I b
Number of Tests O
- g. .
N is. -N N
10 -
\
N S-N 6 N N
4 N N
- 2. , N
\
F 5 5 W W
\ W w 02 04 06 o4 l.o
% of spectnee Anchorages
niinuwiun u ( sup os g a u-- 1>
9 4 so.a u oacas ,
NUMBERS OF WALLS QUALIFIED USING ,
STATISTICAL APPLICATION OF TESTS (TOTAL = 50 WALLS) e e
6 F
b ~
?u 5 5
n I
$4 l i..
I II . .
,2 i'
II II 1.I I II b l.l1.11.1..l..l.l.l.l.l.l..I1..l.
b lO l'5 IO 25 3'O 35 4'O 4'5 5'O 55 6'O 6'5 7O 75 80 8'5 9O 95' 10 0
% OF ACTUAL / ALLOWABLE LOADS
.(SUM OF WALLS IN A 5% INCREMENT) *
. HALL # LOCATION H H T TYP ACT ALL 1. TYP ACT ALL 1. TYP ACT ALL 1. -
45.2 INTAKE 21'-6" 14' 14' 8" C 53 187 28 C 52 . 93 56 C 52 93 56 '
62.3 REACT 23'-0" 8' 6' 30" F 0 0 0 C 52 185 28 F 0 0 0 62.5 REACT 23'-0" 28' 9' 12" F 0 0 0 C 47 . 93 51 ML 47 1 :,0 29 -
63.0 REACT 23'-0" 8' 18' 12" F 0 0 0 C 48 . 93 52 F 0 0 0 63.1 REACT 23'-0" 26' 40' 30" Q2 5 480 1 C 1 185 1 C 20 185 11 63.3 REACT 23'-0" 8' 5' 18" F 0 0 0 C '84 185 45 F 0 0 0 63.5 REACT 23'-0" 26' 29' 30"&24" Q2 117 360 33 .HL 155, 320 48 C 161 185 87 63.7 REACT 23'-0" 26' 7' 24" F 0 0 0 C 154 185 83 ML 154 320 48 63.8 REACT 23'-0" 8' 22' 12" C 38 90 42 C 0 93 0 F 0 0 0 63.9 REACT 23'-0" 8' 5' 12" C 32 90 36 C 0~ 93 0 F 0 0 0 65.10 REACT. 51'-0" 8' 8' 8" C 76 90 84 ML 50 100 50 F 0 0 0 65,2 REACT 51'-0" 20' 23' 45" Q2 477 720 66 ML 188 480 ,39 C 188 278 68 .
65.20 REACT 51'-0" 10' 13' 8"' C 40 90 44 ML 55 100 55 ML 55 100 55 65.4 REACT 51'-0" 20' 29' 48" Q2 230 720 32 ML 200 480 42 C 50 278 17 ,
65.6 REACT 51'-0" 10' 5' 12" Q1 0 0 0 C 84 93 89 F 0 0 0 65.7 REACT 51'-0" 21' 11' 42" Q1 0 0 0 ML 89 278 32 C 86.8 278 31 66.22 REACT 74 -3" 8' 5' 12" F 0 0 0 ML 54 160 34 C 54 93 58 66.24 REACT 74'-3" 8' 5' 12" F 0 0' 0 ML 54 160 34 C 54 93 58 66.5 REACT 74'-3" 15' 22' 8" Q2 '48 120 40 C 31 93 33 ML 31 100 31 66.7 REACT 74'-3" 15' 33' 8" Q2 27 120' 23 MT 0 0 0 C 2.4 93 1 68.10 REACT 74'-3" 14' 9' 12" H 0 0 0 C 16 93 17 C 30 93 32 68.2 REACT 74'-3" 11' 8' 30" F 0 0 0 C 96 185 52 C 96 185 52 68.3 REACT 74'-3" 11' 10' 36" F 0 0 0 C 97 185 52 C_ -97 185 52 68.4 REACT 74'-3" II' 10' 36" Q2 146 480 30 C 131 185 71 C 131 185 71 68.8 REACT 74'-3" 8' 18' 18"&24" F 0 0 0 ML 86 260 33 C 51 185 28 77.0 REACT 13'-9" 7' 10' 18" 02 46 240 19 HL 46 60 77 C 72 185 39 111.2 REACT 61'-4" 11' 11' 30" H 0 0 0 C 83 185 45 F 0 0 Q 111.5 REACT 61'-0" 8- 8' 30" Q2 148 480 3.1 C 178 185 96 ML 226 380 59 116.6 REACT 42'-3" 7' 8' 12" Q1 0 0 0 C 63 127 50 C 40.8 127 32 185.10 AUX (-)l7'-6" 18' 9' 8" C 24.6 90 27 ML 24 100 24 ML 24.3 100 24 185.3 AUX 3'-0" 13' 13' 24" Q1 0 0 0 C 31 185 17 F 0 0 0 185.4 AUX 3'-0" 13' 18' 24" Q1 0 0 0 C 71 185 31 C 70 185 38 185.6 AUX 3'-0" 13' 6' 24" F 0 0 0 ML 36 320 11 C 36 185 19 185.7 AUX (-)17'-6" 18' 6' 8" F 0 0 0 HL 38 100 38 C 38 93 41 185.9 AUX (-)17'-6" 18' 6' 8" F 0 0 0 ML 34 100 34 C 34 93 37 188.1 AUX 23'-0" 13' 6' 8" Q2 64 120 53 MT 0 0 0 C 33 93 35 188.10. AUX 23'-0" 11' 41' 8"&l2" C 52 90 58 ML 25 100 25 HL 78 127 61 188.3 AUX 23'-0" 14' 18' 8" C 39 90 43 ML 51 100 51 C 51 93 '54 188.6 AUX 23'-0" 14' 4' 24" F 0 0 0 C 38 185 21 C 38 185 '21 188.7 AUX 23'-0" 13' 7' 24" F 0 0 0 C 60 185 32 C 60 185 32 191.33 RADHST(-)1'-0" 8' 8' 8" C 11 90 12 HL 13 100 13 ML 13 100 13 Allt.e lw.d Sm. (%>me lo lln's l Q A lI"I'-**c~' 'I l')
TA8LE 4-1 (Sheet 1 of 2)
, TOP SIDE #1 SIDE #2 -
, '.NALL'# LOCATION H W T TYP ACT .ALL % TYP ACT ALL % TYP ACT ALL %
191.55 RADWST(-)1'-0" 21' 17' 8" H 27 37 73 H 22 - . 1;27 17 C 13 93 14 -
194.25 RADHST 23'-O' .11 ' 7' 8" 02 10 120 8 MT 0 0 0 C 11 93 12 195.19 RADNST 3" -0" 11' 19' 8" Q2 19 120 16 MT 0 O O C 62 93 67 196.7 RADHST 5 '-0" 16' 9' 8" N 113 142 80 MT 0 0 0 C 32 93 34 198.1 ' GEN 23'-0" 11' 13' 8" C 34 90 38 ML . 29 100 29 C 29 93 31 198.2 GEN 23'-0" 11' 13' 8" C 58 90 64 C 0 93 0 F 0- 0' 0 198.4 GEN 23'-0" 9' 5' 18" N 0 0 0 C '81 185 44 F 0 0 0 209.0 TURBINE 23'-0" 13' 31' 8" 02 65 120 54 C 35 93 38 MT 0 0 0 210.3 TUR8INE 37'-0" 2' 11' 12" C 9 90 10 ML 0 160 0 F 0 0 0 BOUNDARY TYPES: C- Reinforced Concrete MT Masonry T (non-interlocking)
ML Masonry L (interlocking) -
F Free H Structural Steel -
Q1 Q deck with flutes perpendicular to wall Q2 : Q deck with flutes parallel to wall ,
+
TABLE 4-1 (Sheet 2 of 2)
19 ATTACHMENT 9 b (*) ' N 85 ' b #f 4/hef. ,4 ;.} '- -
. o.
9 SAFETY RELATED'BLOCKOUTS WALL.. WALL LOC WALL LOC SFTY 9 NUMBER (1) (2) RTNG 116.1 REACT 36-3 SR1 4 116.2 REACT 36-3 SR1 116.3 REACT 36-3 SR1
- 116.5 REACT 36-3 SR1 [ ' ;
y 977.1 RADWASTE 51-0 SR1 .
979.11 AUX 23-0 SR1 979.12 AUX 23-0 SR1 y 979.3 TURBINE 30-0 SR1 980.9 TURBINE 37-0 SR1 982.1 REACT 6 SR1
) 982.2 REACT 6 SR1 9,82.3 REACT 6 SR1 982.4 REACT 6 SR1
) 984.1 REACT 6 SR1 984.2 REACT 6 SR1 984.3" REACT 6 SR1
) 984.4 REACT 6 SR1 984.5 REACT 6 SR1
- 984.8 REACT 2-9 SR1 '
) 984.9 REACT 51-0 SR1 1
s . ~s - -
) .
go
== %ew w 3
)
i e
i) d
- 5- ,,
L
. 9
MAenP?stCNw b ** Y *O# #
OESIGN CRITERIA FOR MASONRY BLOCK 0VTS 4
I. Background During the construction phase of the Pilgrim Station, numerous openings in reinforced concrete walls were made to allow for installation of piping and conduit runs later on in the construction phase. At the later phase of construction, when piping and conduit had been installed, the ope.nings in concrete walls were closed in many cases using masonry units. In general, the location of such "blockouts" are shown on the mechanical drawings. It is the position of the Boston Edison Company that masonry blockouts are not masonry walls, and thus do not fall under the scope of NRC I&E Bulletin 80-11, " Masonry Wall Design." It is, l
however, prudent to examine masonry blockouts with consideration of
! potential impact to safety systems. This document details the requirements and acceptance criteria for structural evaluation of
! masonry blockouts.
II. Scope This criteria applies to the use of masonry units wholly surrounded by reinforced concrete. There shall be no size limitation to the use of masonry in this classification as " blockout" except that ' floor to ceiling masonry installations shall be considered walls and shall be
' governed by the criteria deveioped for BEco's response 'to Bulletin 80-11.
5 l III. Desian Assumption The following assumptions shall apply to analysis of masonry blockouts.
i
- 1. There shall be no credit taken for, mechanical agchorage either
} horizontal or vertical between the masonry blockouts and surrounding
. . concrete.
2.
Blockouts do not contribute to the overall structural behavior of the surrounding concrete walls. Drawing C-121, Section A, shows additional wall reinforcing at blockout locations.
! 3. Anchorage is provided by morta'r bond between the masonry and the
!. surrounding concrete. The ultimate shear stress for the mortar.
joint is taken as 83 psi. For conservatism this ultimatc=value-l shall be reduced by a factor of 2 for SSE, P80C, and tornado loads; and by a factor of 4 for 08E loads.
1
- 4. Blockouts shall be assumed to respond.to out of plane. loads as a
-monolithic whole. The. predominant failure mechanism will be a l
shearing of the mortar bond, followed by the sliding of the masonry' blockout out of the concrete in which it is confined.
i l
!. 1
- 5. This criteria is applicable to masonry blockouts that remain uncracked when:
a) assumed simply supported at' top and bottom; and, l Ib) analyzed as a one-way span vertically.
- 6. Actual shear stress is assumed to be uniformly distributed across the mortared surface area (excluding hollow cells). ,
IV. Procedure All masonry blockouts shall be examined in accordance with NEDWI 289, Rev. O, "Walkdown of New Blockwall' Scope," to determine relative potential to impact safety systems. ~0nly those blockouts which are determined to have,the potential to damage a safety related system or
!. . component shall be further analyzed for structural ability to withstand
, loads (safety related blockouts). All safety related blockouts shall be analyzed for the loading combinations set forth in the Design Criteria for Masonry Walls DC-1, Rev.1, September 14, 1981. Acceptance criteria for masonry blockouts shall be based on the assumptions listed in Item III above.
i Modifications, if any, shall be designed to the loading combinaticus and the acceptance criteria for masonry walls and design criteria DC-1.
V. Quality Assurance Reauirements ,
Modifications to' masonry blockouts shall be examined on a case by. case '
basis. Where the blockout is used directly to support a Class I system or component, or where the blockout forms a part of secondary-l containment, the modifications shall be designated Class I and subject to the quality control requirements of quality category "Q." Where the masonry blockout forms an integral part of a designated. three hour. fire
' barrier, modifications shall be des.igna'ted "FPQ." All other masonry blockouts which have an indirect impact on safety systems, i.e., the creation of.a II/I situation, shall be modified to the procurement ~and installation requirements of Class II (quality category "C"). - This is consistent with PNPS FSAR Chapter 12.2 which provides that " Class II-designated structures and/or equipment shall not degrade the integrity of any structures and/or equipment designated Class I."
l i : .-
'W.
L
' , _ . _ ,_ _ , .a . , - . _ .
0 8
.5 7
O
. '7 S
.'6 S .b T G U ,
)
O T K .
. 5 N iC 5 ) F i
nO)
ALS sM p F uBT . 0 5 ( R uY U C uR O SI N vN K C S uO O .
'4 5 E%
S L .
- R5 ivA B i
T vM 0 SA i
0 .
a N 2 I 4 RN I v A usa SL 5 ES HL
.'3 SL
,mRT EA vTO L A
.S T O A W
'3 UF
.R(
.A C O T
.E 5 AM H .
U S '2 S
(
. O
'2
.' 5 l .
1 I l !
1' i
f
.' O l
. 5 l IgEgEsl- II Il 3 8 7 E 5 4 3 2 !
5
=
.b 1 1 1 1 1 1 1
' 9 8 7 6 4 3 2 zw3sz to L ItW z. 3a_h o :mE z
^
a i
O ATTACHMENT 10 ACTI3N ITEMS AND MEETING MINUTES (JULY 18, 1984)
- 1. Based on discussions at the meeting and information provided in Attachment 5, the staff accepted the criterion used by the licensee in determining shear loads for the bottom boundary of walls.
- 2. The NRC staff accepted the block-out criteria (Attachment 9) proposed by the licensee, provided following conditions are met:
- a. The licensee to survey all block-outs (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 block-out 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 direction as the PB0C load.
- 3. The licensee will provide representative calculations to show differences between prior Cygna analysis and subsequent refined analysis for walls qualified without reliance on the statistically determined line loads.
- 4. The licensee provided considerable information on the statistical analysis at the meeting. The NRC staff will inform the licensee regarding the acceptance of statistical concept and need for any further actions pending discussions with the NRC management.
- 5. The licensee is still reviewing alternate qualification scheme for walls 209.13 and 209.14 and will discuss with the staff at a later date.
- 6. The licensee no longer expects to pursue delaying any modification until the 1986 refueling ercage (item 7 of action items of June 7,1984 meetings).
- 7. The licensee provided calculations for walls 64.4, 63.4 and 188.10 at the meeting. The NRC staff will review these calculations and advise the licensee of any outstanding issues resulting from this review.
- , - - - -