ML18058B866

From kanterella
Jump to navigation Jump to search
Resolution of Anchor Bolt Design Issues.
ML18058B866
Person / Time
Site: Palisades Entergy icon.png
Issue date: 05/13/1993
From: Jill Daly
SARGENT & LUNDY, INC.
To:
Shared Package
ML18058B865 List:
References
SL-4841, SL-4841-R, SL-4841-R00, NUDOCS 9306080345
Download: ML18058B866 (64)


Text


(}-=

-"~

-~

-~

--~--

-~-

--=-==

-=-*=

?:~:

~~-~==*=~-- --~

~:::;-:"=~===~

='%::::  :::N-~'.~cl""'e.-:'a*-r g '

= ===;\==-== ;;.-R:::e~l:a*t~ed-

?C..-a°::'fe:::t*"11":_,'-'

~~" ', __ ~->

__ _ --=- -- *-

-:...::-=.-::.:::- =--

--~: ;;:;;.,,.::;:::.,::=;::::::::.:=R *. *.t:~*.:::.o!i>q?::.09-~- ---- -----*- ----*

- -- - - - - - -- --- ~t0j9J; _.~,.,0. _~~1.:1~"'-"' - --- - c-:

~=;;:~=::=:=g;; ~ ~ ~Efpcfrt::*s~ ~i41=~~::;~.:::~~-;_ .. ===

E-E~Hevi&i.O:l':J~O'. .

=t!~~1.:3.1 §~3.

~"""""-""*"'~"""""'""" ~--

UM'-'\WI ~{""' - - ~"'

<G3"!W~-.....,--

$<~'"""' -"'~"'=.,,,,,,..,.

,,~ """"""'""=

M6134.001 05-93 9306080345 930601 PDR ADOCK 05000255 p PDR

I.

CONSUMERS POWER COMPANY PALISADES NUCLEAR PLANT RESOLUTION OF ANCHOR BOLT DESIGN ISSUES REPORT SL-4841 REV.0 NUCLEAR SAFETY RELATED PROJECT NO. 8863-0'J SARGENT & LUNDY

r*

i SARGENT 8 LUNDY SL-4841 Rev. 0 05-13-93 CONTENTS Section 1

INTRODUCTION 1

Scope 1

EVALUATION PROCESS 2

Initial Screening Process Detailed Evaluation 3 Calculation of Drillco/Hilti Interaction 4

SUMMARY

AND CONCLUSION 5 6-7 REFERENCES SL-4841/051393

RESOLUTION OF ANCHOR BOLT 1 SARGENT 8 LUNDY SL-4841 DESIGN ISSUES Rev. 0 05-13-93 INTRODUCTION The objective of this report is to summarize the evaluation of anchor capacities for pipe supports when the design involves the interaction of concrete tension zones for Drillco undercut anchors and Hilti Kwik Bolt anchors.

Various transmittals have been issued over the last 2 years (see References 1 through 6). These transmittals describe the NRC concerns and the evaluations performed over the last 2 years.

The procedure discussed below to resolve NRC concerns was developed at a meeting between the NRC, CPCo and Sargent & Lundy held in NRC offices on March 23, 1993 (Reference 7). This procedure provides an evaluation of concrete anchorage capacity based on ACI 318-95 (draft) guidance.

This evaluation addresses a total of 11 pipe support calculations which represent all cases involving the interaction of Drillco anchors and Hilti Kwik Bolts. These pipe supports were installed during the Palisades Plant Steam Generator Replacement Project (SGRP).

EVALUATION PROCESS The evaluation consists of a two-step process involving an initial screening as the first step followed by a detailed evaluation of two governing pipe supports.

SL-4841/051393

SARGENT 8 LUNDY 2 SL-4841 Rev. 0 05-13-93 Initial Screening Process The first step in the review was to perform an initial screening of the eleven pipe support calculations.

The purpose of the screening was to select one or two calculations that envelop the eleven pipe support designs.

It was determined that the critical supports should be selected based on design margins as given in the original design basis support calculations. In most cases, the original design basis calculations present results in the form of interaction ratios which are determined by taking the design load divided by the allowable load. To minimize the need for support modifications due to load increases, loads used in the original pipe support design had considerable margin over the support loads determined by piping analysis. For this reason, the Interaction Ratios (IR) reported in Table 1 would be considerably reduced if the design basis calculations were updated to reflect the final piping analysis loads .

As shown in Table 1, three supports have interactions greater than 0.6.

TABLE 1 SCREENING

SUMMARY

DBB5-H7 The unistrut anchor shear cone does not overlap with 1/2" diameter maxi-bolt shear cone.

DBB6-H6 Bolt Interaction Ratio (IR) = 0.445. This is low comoared to other cases.

DBB6-H7 Bolt IR=0.324, low compared to other cases.

DBB6-H9 Bolt IR=0.386, low comoared to other cases.

DBB7-H7 Bolt IR=0.731, oerform detailed evaluation.

DBB7-H9 Bolt IR=0.646, oerform detailed evaluation.

DBB8-H4 Bolt IR=0.667, minor interaction with stair handrail and 1/4" conduit strap anchor.

DBB8-H14 Bolt IR=0.567. Maximum loads in ftilti bolts and Drillco bolts do not occur simultaneously.

EB20-H2 Applied forces and moments in these three pipe supports are very small.* Bolt IR for EB20-H5 EB20-H7 = 0.104.

EB20-H7 SL4841/051393

SARGENT 8 LUNDY 3 SL-4841

  • Rev. 0 05-13-93 Of the three supports, DBB8-H4 was not considered to be a controlling case because the Drillco bolts only overlap with Hilti anchors for a stair railing and a 1/4" diameter anchor for a conduit strap. The remaining two supports, DBB7-H7 and DBB7-H9 were selected for detailed evaluation.

Detailed Evaluation Based on discussion held with the NRC in their offices on March 13, 1993 (Ref. 7), it was determined that a detailed evaluation of these two anchor bolt configurations should be performed using the October 1992 version proposed for incorporation as Chapter 22 of ACI 318-95.

Details of the procedure are given in the attached sample calculation which documents the detailed evaluation of pipe support DBB7-H9. The October 1992 version of ACI 318-95, Chapter 22, is included as Attachment A.

The following considerations have been incorporated into the calculation:

1. Pipe support design loads have been scaled to reflect the support loads as determined in the piping analysis (Ref. 8).
2. Concrete is assumed cracked, therefore, Vtacrv is set equal to 1.0 instead of 1.4.
3. Minimum specified concrete compressive strength is used. No credit is taken for the higher in-place concrete strength.
4. When checking the tension/shear interaction for the Hilti bolts, the Hilti allowable loads are determined by taking Hilti Test data and dividing by a factor of safety of 4. These allowables are compared with allowables based on the October 1992 proposed ACI 318-95 criteria, and the smaller values are used to calculate the interaction ratio.

SL-4841/051393

SARGENT 8 LUNDY 4 SL-4841

  • Rev. 0 05-13-93
5. For the check of concrete pullout due to the interaction of Drillco and Hilti anchors, concrete pullout capacity is determined based on ACI 318-95 and the combined interaction is determined with the assumption that both anchor bolt groups experience their maximum applied tension loads simultaneously.

Calculation of Drillco/Hilti Interaction One design issue not addressed in the proposed ACI 318-95 criteria is the method used to evaluate the concrete capacity when two overlapping anchor bolt groups have different effective embedment depths. To avoid further concerns on how cone strength should be distributed between two adjacent bolts with significantly different embedment depths, the interaction equation from proposed ACI 318-95 was utilized as follows:

(...!!E...) . + (..!!E...) s; 1. 0 cl>Nn Drillco cl>Nn Hil ti where Nu is the applied tension load and Nn is the nominal concrete tensile strength per ACI 318-95, Section 22.4. The <P factors are 0.65 and 0.55 for the Drillco and Hilti anchors, respectively. This approach is more conservative than other approaches, such as splitting up the overlapping area between the adjacent anchors or adding the two loads and then comparing with the total pullout capacity. Using the above interaction equation is practical for two anchor groups with substantial cone overlap such as shown in the attached calculation. However, it would not be practical as a general approach since the resulting penalty to the anchor bolt groups is far too severe when the overlap between two adjacent anchor bolt groups is relatively small.

SL-48411051393

SARGENT 8 LUNDY 5 SL-4841

  • Rev. 0 05-13-93 Results of the evaluation are as follows:

Support No. Original Design Basis Interaction Based on Interaction ACI 318-95 DBB7-H7 0.731 0.65 DBB7-H9 0.646 0.48 Based on the above results, it is concluded that the Palisades SGRP pipe support anchorages are adequate and meet the requirements of the concrete capacity method as proposed in the October 1992 version of Chapter 22 of ACI 318-95.

SUMMARY

AND CONCLUSION As a result of NRC concerns over the evaluation of cone overlap between Drillco Maxibolt and Hilti Kwik Bolts during an inspection of the Steam Generator Replacement Project (SGRP), eleven cases have bene identified and screened to select two controlling cases for detailed evaluation.

In a meeting held with the NRC on March 23, 1993 (Ref. 7), it was determined that the October 1992 proposed criteria for Chapter 22 of ACI 318-95 would provide a conservative basis for determining the concrete pullout capacity of the Drillco and Hilti anchors. Results of the detailed evaluation of the two controlling pipe supports indicate a maximum interaction ratio of 0.65 for support DBB7-H7.

Based on this evaluation, it is concluded that the SGRP Drillco/Hilti anchorages are acceptable. The proposed ACI 318-95 criteria have been met using a conservative approach employing the minimum specified concrete compression strength and the assumption of cracked concrete.

SL-4841/051393

SARGENT B LUNDY 6 SL-4841

  • REFERENCES Rev. 0 05-13-93
1. NRC Inspection Report 90025 Items OlM, OlN, 05/24/91 Highlighted NRC inspector concern over Steam Generator Replacement Project (SGRP) evaluation of cone overlap associated with Drillco Maxibolts and Hilti Bolts.
2. CPCo/Bechtel/NRC meeting in White Flint, 04/23/91 Bechtel committed to revisit its analysis methodology associated with SGRP analysis methods.
3. Letter from B Holian (NRC) to DC Kansai (BPC), Evaluation of Palisades Anchor Bolt Design, 06/13/91 NRC formally requested additional information supporting the validity of the design/analysis method employed for the anchor bolts.
4. Letter from DC Kansai (BPC) to B Holian (NRC), Evaluation of Palisades Anchor Bolt Design, 10/21/91 Additional information is* provided to NRC to support SGRP design.
5. Letter from A. Masciantonio (NRC) to GB Slade (CPCo), Evaluation of Palisades Anchor Bolt Design, 06/15/92 NRC provides its conclusions of review of lQ/21/91 submittal. NRC takes specific exception to ACI 349 Appendix B. NRC provides three-step evaluation criteria for CPCo action.

SL-48411051393

7 SARGENT 8 LUNDY SL-4841 Rev. 0 05-13-93

6. NRC/CPCo/BPC/S&L Telecon to Interpret NRC 06/15/92 Letter.

Some clarification was obtained with regard to the three part NRC criteria for anchor evaluation.

7. Memorandum from A.H. Hsia (NRC) to CPCo, "Summary of Meeting Held on March 23, 1993 with Consumers Power Company on Anchor Bolts Analysis for Palisades Nuclear Generating Plant," April 5, 1993.
8. "SGRP Anchor Bolt Loads for ACI-318-95 Assessment," Engineering Analysis No. EA-A-NL-93-016, Rev. 0, dated 5/7/93.

SL-48411051393

CALC NO: PAL-93.2 SARGENT & LUNDY ENGINEERS REV: 0 PAGE: 1 CLIENT: CONSUMERS POWER CO. PREPARED BY:ar:: ~~~..,,.V;Y.DATE:S-/£-9s PROJECT NAME: PALISADES PLANT REVIEWED BY; ~ . I/

DATE: !3 3 I Cf:;

PROJECT NUMBER: 8863-09 APPROVED BY'/'ff(i DATE, .,, /)} lf5 SAFETY RELATED OBJECTIVE THIS CALCULATION IS PERFORMED TO DOCUMENT THE EVALUATION OF ANCHOR BOLT DESIGN FOR SUPPORT NO M101-6050-H9 (ALSO KNOWN AS DBB7-H9) FOR STEAM GENERATOR E50B BLOWDOWN. THE PURPOSE IS TO EVALUATE THE ANCHORAGE USING PROPOSED CRITERIA FOR ACI 318-95. SEE ATTACHMENT A.

REFERENCES/DESIGN INPUT

1. CPCO PALISADES CALCULATION EFWA-PD-H40, REV.5
2. HILTI TECHNICAL GUIDE H-428, DATED 9/88
3. CPCO PALISADES CALCULATION EA-SP-033383-DBB7-H9, REV.0 ASSUMPTIONS ASSUMPTIONS ARE NOTED IN THE TEXT OF THE ANALYSIS.

ATTACHMENTS A. ACI 318-95, 10/92 VERSION OF PROPOSED CHAPTER 22, FASTENING TO CONCRETE B. "SGRP ANCHOR BOLT LOADS FOR ACI 318-95 ASSESSMENT", CPCo ENGINEERING ANALYSIS NO. EA-A-NL-93-016, REV. 0.

C. PIPE SUPPORT DWG.M101-6050-H9, REV.3, (3 SHEETS)

D. FIELD CHANGE REQUEST FCR 116 (2 SHEETS)

ANALYSIS THE ANALYSIS INCORPORATES THE FOLLOWING REQUIREMENTS:

1. CONCRETE IS ASSUMED CRACKED.
2. MINIMUM SPECIFIED CONCRETE COMPRESSIVE STRENGTH OF 4000 PSI IS USED.
3. BOTH PIPE HANGERS ARE ASSUMED TO EXPERIENCE MAXIMUM LOAD AT THE SAME TIME
4. A FACTOR OF SAFETY OF 4 IS APPLIED TO HILTI TEST RESULTS TO OBTAIN ALLOWABLE LOADS.

A. CHECK DRILLCO BOLTS Drillco concrete cone capacity 2

$ := 0.65 fc' := 4000* psi A pn := 871.31*in See attached figure 11' ecn := 1.0 h ef := 8*in A pno := 9 *h el 2

ll'acrn := 1.0 k := 17 A pno = 576*in II' sn := 0.7 + 0.3* 8.5 II' sn =0.913 12

._ A pn ( ) 1.5 ~c' . -1.5

$N n .- $*--*ljl ecn*ll' sn*ll' acrn*k* h ef * -:m *lb A pno PSI

$N n = 21.828*kips

C.ALC NO: PAL-93.2 SARGENT & LUNDY ENGINEERS REV: 0 PAGE: 2 CLIENT: CONSUMER POWER CO. PREPARED

  • PROJECT NAME: PALISADES PLANT PROJECT NOMBER: 8863-09 SAFETY RELATED REVIEWED APPROVED

'CJ

~--y-

"O. <I\

~.\.

/ C,ON~TEi eD&tii-11

___'B'.°=' "'_:_,.~_G.~' .j- U!.' ---(

+e"r

~ ~-

T ..

~*

" qD .-J,;

I

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

. /I

  • ot* * ..,.
  • I. . - .

~*.,

.......--~*+,,,.s +1toA.t ~ [~

1

=-'2. 1

......----r\.-s -- ' \ *

- °' -. .,;,s._-*. ___--* -

~

\l.

A 3:5

~ _ _ _ .$_ _ _

_l.5'."8:1.2n

~

..U).

I ~-

~t

  • II "J -

I ~!

~

I

  • * -. r, I 1

~-

\tll

~

--~~~~~-1o-----.~

.. I'

. \l

_:i_ * - -

__:t 0 - - "

1~

A pn is Drillco concrete area (neglect Hilti concrete area)

=(X DZST)x(Y DZST)- Shaded area due to concrete edge A pn := (8.5 + 6.5 + 12)-(12 + 9 + i + 12)*in 2

- [ 3.5*(12 + i) }in 2

A pn =871.313*in2

CALC NO: PAL-93.2 SARGENT & LUNDY ENGINEERS REV: 0 PAGE: 3 CLIENT: CONSUMERS POWER CO. PREPARED ATE': 6--/z-tj 3 PROJECT NAME: PALISADES PLANT PROJECT NUMBER: 8863-09 ATE: 5/(!J/ I/_j DATE: 5/J )/1)

SAFETY RELATED Force per bolt

~N n

~N n1 := - ~N n1 = 5.457*kips 4

Determine shear allowable for 0.5 in. diameter Drillco bolt As shown on sheet 7 of Ref. 3 the baseplate computer model includes the eccentricity of the load, therefore ljl ecv := 1 *0 ljl sv := 1.0 ljl acrv := 1 *0 S := 9.875* in d 0

= 0.5-in I:= 8-d 0 I= 4*in C 1 := 8.S*in 2

Apv := 1.5*C r(3*C 1 + S) A pv = 451.031 *in A pvo := 4.5*C 12 A pvo =325.125*in 2 15

$V := ~---*ljl n A Apv ecv *ljl sv

  • ljl acrv
  • 6* ( -

d I ).2.~o

- * ~c' in

- * (C psi in 1

- ) * *lb pvo o

~V n = 9.088*kips Shear per bolt

$V n

~Vn1 : = - ~V n 1 = 4.544*kips 2

Allowable bolt loads based on ACI 318-95, Section 22.8 Conservatively assume threads in the shear plane

~ t := 0.8 for tension F y := 110-ksi

~ v := 0.65 for shear Fut:= 125-ksi 2

A st:= 0.142*in 2

A sv := 0.126*in

CALC NO: PAL-93.2 SARGENT & LUNDY ENGINEERS REV: 0 PAGE: 4 CLIENT: CONSUMERS POWER CO. PREPARED BY:cJ..<7.~~ATE: S--/~-t/3 PROJECT NAME: PALISADES PLANT REVIEWED BY: J!J~ll,k~,*~ DATE: S//3/9.:J PROJECT NUMBER: 8863-09 APPROVED BY' 1-:m7/J7 DATE' t;/1"5,/13 SAFETY RELATED

~Ny:= ~ rA sf (.8*F ut) .8*F ut < Fy

~Ny= 11.36*kips ........... Does not control Shear allowable per ACI 318-95

~VY:=~ v*O.S*A sv* (.B*F ut)

~v Y =4.095*kips ........ Controls Check Drillco Bolt Interaction Based on loads given in Attachment B, Sheet 12, Bolt # 4 has the controlling bolt loads as follows Nu := 784-lb vu:= 239-lb Shear v

__u_ = 0.292 < 1.0 OK

.2*Wy Tension Nu

- - = 0.144 < 1.0 OK

~N n1

CALC NO: PAL-93.2 SARGENT & LUNDY ENGINEERS REV: 0 CLIENT: CONSUMERS POWER CO. :ATE: S-/2-f 3 PROJECT NAME: PALISADES PLANT PROJECT NUMBER: 8863-09 REVIEWED BY:

APPROVED BY: J"Jlt TE: S(/J/Cj..J D:ATE : 7j /l 3/ 13 SAFETY RELATED B. CHECK HILTI BOLTS Hilti concrete cone capacity

$ := 0.55 2 2 fc' := 4000* psi A pn := 14.75 -in See figure (page 2) ljl ecn := 1.0 h ef := 2.75-in A pno := 9*h el 2

ljl acrn := 1.0 k := 17 A pno = 68.063*in ljlsn := 1.0 F A pn ( ) 1.5 *~c' . -1.5 *lb n := $*--*ljl A ecn *ljl sn *ljl acrn *k* h et -*m psi pno F n = 8.62*kips Force per bolt

  • - $T n

$T n1 4 $T n1 = 2.155*kips Check allowable based on test results (Ref. 2)

Hilti bolt diameter=S/8 in., EMB length=2.75 in., Spaced 6.5 in. O.C With a factor of safty of 4;

$Ty:= 6~00*1b F y = 1.65*kips Shear

$Sy:= 11:62*1b $Sy = 2.891 *kips Check bolt tension and shear 2 2 A st:= 0.226*in A sv := 0.202*in Tn := $ i*A srO.B*F ut Tn = 18.0B*kips > Test Result Sn := $ v*0.5*A sv*O.B*F ut Sn= 6.565*kips > Test Result Use allowables based on test results

CALC NO: PAL-93.2 SARGENT & LUNDY ENGINEERS REV: 0 PAGE: 6 CLIENT: CONSUMERS POWER CO. PREPARED 6--,/ '2 f .3 PROJECT NAME: PALISADES PLANT PROJECT NUMBER: 8863-09 REVIEWED BY:

APPROVED BY:

5fg!~.J 5 13/qy SAFETY RELATED Check Hilti Bolt Interaction Based on loads given in Attachment B, Sheet 12:

Tu := 730-lb Su := 564-lb Su = 0.976 < 1.0 OK

.2*$S y Tu = 0.442 < 1.0 OK Fy C. COMBINED EFFECT OF DRILLCO & HILTI ANCHORS Check total tension on the concrete due to combined effect of Drillco and Hilti anchors by combining the tension interaction ratios.

Nu = 0.784*kips Tu = 0.73*kips

$N n 1 = 5.457*kips F n1 = 2.155*kips Nu Tu IC comb:= - - + - - IC comb = 0.48 < 1 . o OK

$N n1 $T n1 The allowable loads are calculated per ACI 318-95, Section

22.4 CONCLUSION

PIPE SUPPORTS M101-6050-H9 AND EFWA-PD-H40 MEET THE CRITERIA PROPOSED FOR CHAPTER,22 OF.ACI 318-95.

Cale. No. PAL-93.2 Rev. 0, Attachment A

. -- - **- - . *-- ------------- CODE Page 1 of28 9.3.2.5 - Fastening to Concrete .

(a) Fasteners governed by ductile tensile failure of a metallic element where calculations indicate an initial concrete failure is precluded ...**a.so.

(b) Fasteners governed by a brittle steel failure and fastener applications where calculations indicate an initial concrete breakout or slow pullout failure governs i) Shear Loads .....* 0.65 ii) Tension loads Classification of fasteners as determined from ASTM ZXXX or ASTM ZfYY Class 1 ...... 0.65 Class 2 ...... 0.55 Class 3 ...... 0.45

Cale. No. PAL-93.2 CODE Rev. o, Attachment A

-* Page 2 of28 Chapter 22 - Fastening to Concrete 22.0

  • Notation

'\No = Projected area of one fastener, for tensile strength caJculation, if assumed not

  • limited by actuaJ edge distance or spacing, (Sect.22.4.3), in2.

~N = Actual projected area of a fastener or group of fasteners, for tensile strength calculation (Sect. 22.4.3), in2 *

~Vo = Projected area of one fastener, for shear strength calculation, if assumed not limited by actual corner influences or member thickness (Sect.22.5.3), in2 *

~v = Actual projected area of a fastener or group of fasteners, for shear strength calcul~n (Sect: 22.5.3) in2 .

c = Edge distance from center of a fastener to the edge of concrete, inches.

C1 = Edge distance from the center of a fastener to the edge of concrete in one direction. Where shear is present c 1 is in the direction of the shear force, inches.

C1crN = Critical edge distance fer tensile load case, (1.She1 for headed studs, headed anchor bolts, expansion anchors and undercut anchors), inches.

C1s = Smallest actual edge distance, inches.

C2 = Edge distance from center of a fastener to the edge of concrete in the direction orthogonal to c 1* Where shear is present, ~is in the direction perpendicular to shear force, inches.

C2crV = Critical edge distance for shear force (1.Sc 1 for headed studs, headed anchor bolts, and undercut anchors), inches.

C25 = Smallest actual edge distance perpendicular to the direction of shear force, inches.

do = Outside diameter of fastener or shaft diameter of headed stud, inches.

du = Diameter of head of headed stud, inches.

e = eccentricity of a load on a group of fasteners, distance from the centroid of the'

Cale. No. PAL-93.2

- --*-CODE Rev. O, Attachment A Page 3 of28 fasteners to the point of load application, < s/2, inches.

f' c = Specified compressive strength of concrete, psi.

fy = Minimum specified yield strength of fastener steel, psi.

tut = Minimum specified ultimate tensile strength of fastener steel, psi.

h = Thickness of member in which a fastener is anchored, inches.

h 81 = effective embedment depth measured from the free surface to the bearing surface of the anchor head, inches.

k = an effectiveness factor to be taken as:

20 for cast in place headed studs, 17 for insert anchors .

t = Activated load bearing length of fastener < 8d 0 , inches.

= het for fasteners with a constant overall stiffness, such as headed studs, undercut anchors, and torque controlled and expansion anchors where there is no distance sleeve or the expansion sleeve also has the function of the distance sleeve.

= 2d 0 for torque contro!led expansion anchors with a distance sleeve separated from the expansion sleeve.

Nb = Basic concrete breakout tensile strength of a single fastener in cracked concrete (Sect. 22.4.2), lb.

N" = Nominal tensile strength of a single fastener or group of fasteners, considering effects of edge distance, spacing, number of fasteners, absence of concrete cracking, and other factors (Sect. 22.4.1 ), lb.

NP = Basic pullout tensile strength of a single headed stud in cracked concrete (Sect. 22.9.4), lb.

  • NP" = Nominal pullout tensile strength of a single fastener, considering the effects of absence of cracking, (Sect. 22.9.3.), lb.

NY = Nominal material yield tensile strength of a single fastener (Sect 22.8.2), lb.

s = Fastener center to center spacing, inches.

scrN = Critical spacing for tensile force calculations (3hef for headed studs, headed anchor bolts, and undercut anchors), inches.

Cale. No. PAL-93.2 Rev. 0, Attachment A

- ----.CODE Page 4 of28 ScrV = Critical spacing for shear force calculations (3c1 for headed studs, headed anchor bolts, and undercut anchors) , inches .

vb = Basic concrete breakout she~r strength of a single fastener in cracked concrete (Sect. 22.5.2), lb.

vn = Nominal concrete breakout shear strength of a single fastener or group of fasteners, considering effects of edge distance, spacing, member thickness, number of fasteners, absence of concrete cracking, and other factors (Sect.

22.5.1), lb.

Vy = Nominal material yield shear strength of a single fastener (Sect 22.8.3), lb.

1P'ec:N = Modification factor, for tensile strength, to account for fastener groups which are loaded eccentrically (Sect. 22.4.4).

11'ecv = Modification factor, for shear strength, to account for fastener groups which are loaded eccentrically (Sect. 22.5.4).

11'sN = Modification factor, for tensile strength, to account for edge distances smaller than c 1crN* (Sect. 22.4.5).

1P'sv = Modification factor, for shear strength, to account for edge distance smaller than c 2 crv* (Sect. 22.5.5).

1P'acrN = Modification factor, for tensile strength, to account for the absence or control of cracking, (Sect 22.4.6).

1.l'~crV = Modification factor, for shear strength, to account fer the absence or control of cracking, (Sect 22.5.6).

22.1 - Scope 22.1.1 - This chapter provides design requirements for fasteners used to transmit loads from attachments into concrete structures by means of tension, shear or a combination of tension and shear. The nominal strength of fasteners is governed by the lower of the yielding of the fastener, a concrete breakout type failure, or a slow fastener pullout failure in cracked concrete. Other failure modes, such as pullout failures in uncracked concrete

  • and immediate pullout failures in cracked concrete are precluded by prequalification tests of Section 22. 1.2 which shall demonstrate the suitability of the fastener for use in structural concrete.

Cale. No. PAL-93.2 Rev. 0, Attachment A

  • -**-- CODE---------- - -**- Page 5 of28 22.1.2 - Independent third party prequalification evaluation testing of *aJ1 anchors must be performed to ensure suitability for use in structural concrete. Anchors to be installed in potentially cracked concrete must meet the requiremeats of ASTM ZXXX Standard Specification for Proper Functioning and Suitability of Anchors for Use in Cracked Concrete. In the special case where a fastener is to be located in a region of a structural member wl:lere no cracking at service loads due to other causes is expected, anchors**

meeting the requirements of ASTM ZiYY, Standard Specification for Proper Functioning and Suitability of Anchors for Use in Non-Cracked Concrete. are allowed.

22.1.3 - This chapter applies to both cast in situ type anchors, such as headed studs or headed anchor bolts, and fasteners post-installed in hardened concrete, such as expansion anchors and undercut anchors. Adhesive anchors and direct fasteners such as powder or pneumatic actuated nails or bolts are beyond the current scope of this chapter.

22.2 - Genera~ Requirements 22.2.1 - Fasteners shall be designed for all static load combinations as outlined in Section 9.2. Load applications which are predominantly fatigue, impact or seismic are beyond the current scope of this chapter.

22.2.2 - The strength reduction factors to be used in design of fasteners, as given in Section 9.3.2.5, are:

(a) Fasteners governed by ductile tensile failure of a metallic element where calculations indicate an initial concrete failure is precluded .... q, = 0.80.

(b) Fasteners governed by a brittle steel failure and fastener applications where calculations indicate an initial concrete breakout or slow pullout failure governs i) Shear Loads ... q, = 0.65 ii) Tension loads Classification of fasteners as determined from ASTM ZXXX or ASTM "Z:fYY Class 1 .... rp = 0.65 Class 2 .... rp = 0.55 Class 3 .... rp = 0.45 22.3

  • Strength of Fasteners 22.3. 1 - Strength design of fasteners shall be* based on the computation of the steel tensile and shear strength of ~he fastener and the attachmer.t, the concrete breakout tensile and shear strengths as well as the pullout strength of fasteners in cracked

Cale. No. PAL-93.2 Rev. O, Attachment A CODE Page 6 of28 concrete. The minimum of these strengths is to be taken as the nominaJ strength of the fastener for each load condition. Detailed requirements for concrete capacity design tor concrete breakout strengths are given in Section 22.3.2 through 22.3.5, fastener steel design requirements are given in Section 22.8 and pullout.design requirements are given in Section 22.9. In addition to the normaJ design procedures of ACI 318-95 for other actions, the design of fasteners for tension must recognize that the safety of fasteners is-_

more sensitive to appropriate installation and that some fastening devices are less subject to installation errors and tolerances. Thus, for the design of fasteners:

where ¢ for tensile loading varies with the sensitivity of the fasteners' tensile strength to installation error and tolerances as reflected in ASTM ZXXX and ZfYY.

22.3.2 - The ba?ic desigfl ccnc:-ete breakout tensile strength, Nb* and the basic design concrete breakout shear streng~h. Vb, tor any fastener shall be based on design models which result~ in predictions cf strength in substantial agreement with results of comprehensive tests and which reflect the indications of fracture mechanics theory to account for size effects.

22.3.3 - The nominal design ccncrete breakout tensile strength, N", and the nominal design concrete breakout shear strength, V",for any fastener or group of fasteners shall be based on the basic individual fastener breakout strength, with modifications made for the number of anchors, the effects of close spacing of fasteners, proximity to edges, depth of the structural member, eccentric loadings of fastener groups, absence of cracking and other considerations which effect fastener strength.

22.3.4 - Interaction of tensile and shear loads shall be considered in design using a conservative interaction expression which results in predictions of strength in substantial agreement with results of comprehensive tests.

22.3.5 - The concrete breakct.:t strength requirements of sections 22.3.2 through 22.3.4 are satisfied by the concrete capacity design procedure of sections 22.4 through 22.6.

22.4 - Concrete Capacity Design Tensile Strength of Fasteners 22.4.1 - Nominal concrete breakout tensile strength of a fastener or group of fasteners:

  • (£q.22-1)

Cale. No. PAL-93.2 Rev. 0, Attachment A

.. CODE .... __ Page 7 of28 22.4.2 Basic concrete breakout tensile strength of a single fastener in cracked concrete:

Nb--k.fi!V'c h<<1.5 (Eq.22-2) where k =20 for cast in place headed studs.

k =17 for insert anchors.

22.4.3 - The effects of multiple fasteners, spacing of fasteners, and edge distance on nominal concrete breakout tensile strength shall be included by applying the modification factor '\N /'\No . '\N is the actual projected area of the failure surface for the fastener or group of fasteners which can be approximated as the base of the rectilinear geometrical figure which results from projecting the failure surface outward at a slope of 35 degrees from the fastener, or in the case of a group of fasteners, from the axis lines between adjacen~fasteners. '\No is the maximum possible projected area of a single fastener remote from edges:

  • ApNo = (2*c,aN)2

=(2*1.5h~J 2 = 9h:, (Eq.22-3) 22.4.4 - Modification factor for eccentrically loaded fastener groups:

1 iv l!ICN = s; 1 (Eq.22-4)

(1 + 2e) saN The eccentricity e shall not exceed s/2.

In the special case where eccentric loading exists about two axes, the modification factor, 11'ecN' shall be computed for each axis individually and the product of these factors used as 11'ecN in Eq. 22-1.

22.4.S - Modification factor for the disturbance of stress distribution caused by the proximity of a fastener to an edge:

c 13 = smallest actual edge distance.

Cale. No. PAL-93.2 Rev. 0, Attachment A COOE Page 8 of28 tVsN=1 if C, 5 ~C,aN=1.5h 61 (Eq.22-58) 1VsN=0.7 +0.3...5..!__ if 'isS:'iaN=1.5h61 (Eq.22-Sb) c,crN t 22.4.6 - In the special case where a fastener is located in a region of a structuraJ member where no cracking at service load levels due to other causes is expected, anchors meeting either ASTM ZXXX or ASTM Z'fYY are permitted and the following modification factor is allowed:

,,.actH = 1.4 Torque controlled expansion anchors which have not met the requirements for use in cracked concrete according to ASTM ZXXX and drop-in anchors (anchors not able to re-expand under load) .shall only be used in such uncracked regions.

22.4.7 - Where cracking due to other causes is expected at service load levels, and where such cracking may exceed the level controlled by the adequate distribution of flexural reinforcement in accordance with Section 10.6.4, special confining reinforcement shaH be used with any fastener.

22.5 - Concrete Capacity Design Shear Strength of Fasteners 22.5.1 - Nominal concrete breakout shear strength of a fastener or group of fasteners:

(Eq.22-6) 22.5.2 - Basic concrete breakout shear strength of a single fastener in cracked concrete:

Vb~6*( ~ t\ra;,*~*C,1 "5 (Eq.22-7) 0 22.5.3 - The effects of multiple fasteners, spacing of fasteners, edge distance and depth of the structural member on nominal concrete breakout shear strength shall be included by applying the reduction factor Apv / Apvo* Apv is the actual projected area of the failure surface for a fastener or group of fasteners which can be approximated as the base of the rectilinear geometrical figure which results from projecting the failure surface outward

Cale. No. PAL-93.2 Rev. 0, Attachment A

.. CODE -- Page 9 of28 at a slope of 35 degrees from me point where the fastener crosses the surface of the concrete, or in the case of a gr\:up of fasteners, from the axis lines between adjacent anchors. ~vo is the maximum possible projected area.for a single fastener in a deep member and remote from edges in the direction perpendicular to the shear force :

Apvo=~a'/2~av

=(1.Sc,)-(3c,) = 4.Sc,2 (Eq.22-8) 22.5.4 - Modification factor fer eccentrically loaded fastener groups:

(Eq.22-9)

The eccentricity e shall not exceed s/2.

In the special case where eccentric loading exists about two axes, the modification factor, 11'ecv shall be computed fer each axis individually and the product of these factors used as 11'ecv in Eq. 22-6.

22.5.5 - Modification factor fer the disturbance of stress distribution caused by the proximity of a fastener to an eC:Ge :

~ = smaller edge distance perpendicular to direction of shear,

'V sv=1 if Czs2:~oif 1.Sc, (Eq.22-10a)

Wsv= 0. 7 +0.3 Czs if ~s~~av=1.5c, (Eq.22.-1 Ob)

Czav 22.5.6 - In the special case where a fastener is located in a region of a structural member where no cracking at service load levels due to other causes is expected, anchors meeting either ASTM ZXXX or ASTM zyyy are permitted and the following modification factor is allowed:

1PacrV = 1.4 In the special case where a fastener is located in a region of a structural member where cracking due to other causes is expected at service load levels and where special reinforcement has been provided which can control edge shear concrete breakouts, the

Cale. No. PAL-93.2 Rev. o, Attachment A CODE Page 10 of28 following modification factors are allowed:

V acrV = 1.2 - for fasteners in cracked concrete wwi straight edge reinforcement having a bar diameter of ~ 1/2 inch.

V acrV = 1.4 - for fasteners in cracked concrete with straight edge reinforcement and stirrups spaced at not more than 4 inches.

Torque controlled expansion anc.~ors and drop-in anchors are permitted in cracked concrete under pure shear loadings.

22.6 - Interaction of Shear and Normal Forces Fasteners or groups of fasteners which are subjected to both shear and axiaJ loads shall be designed to satisfy the following requirements:

22.6.1 -

22.6.2-If Vu ::5 0.2¢Vn* then Nu ::S ¢Nn .

If Nu~S 0.2¢Nn* then Vu ::5 ¢Vn*

  • 22.6.3 - if Vu > 0.2¢ V n and Nu > 0.2 ¢ Nn* then (Eqn.22-11) 22.7 - Minimum Spacings and Edge Distances This section provides minimum spacings and edge distances required to prevent splitting failures. Lesser values obtained from tests performed under ASTM ZXXX and "Z:fYY are acceptable if approved by the engineer of record.
22. 7 .1 - Minimum fastener center to center spacirg shall be 1he1.

22.7.2 - Minimum edge distances shall be :

  • 1het - for undercut and headed fasteners, 2h., - for torque controlled and expansion anchors.

22.7.3 - Minimum depth for member, h, in which a fastener is anchored - 2h.,.

22.8 - Design Requirements for Fastener Steel 22.8.1 - The shear and tensile strength of a fastener, Ny and VY' must be evaluated based on the properties of the fastener material and the physical dimensions of the fastener. Test results are also acceptable to establish values of Ny and Vy.

Cale. No. PAL-93.2

  • -- ________CODE_ ___ . Rev. 0, Attachment A Page 12 of28 2 2 4 u -d0 ) 9 f.'c Np =l!(d (Eq.22-13) in a region of a structural member where no cracking at service load levels due to other

=

causes is expected, in which case t ~ 1.4 is allowed.

Cale. No. PAL-93.2 Rev. 0, Attachment A

___C_Q_Q~- --*------------ _______ . Page 11 of28 22.8.2 - The tensile strength cf a fastener may be calculated as :

For ductile fastener material with a well defined yield point .*.*

For fastener material witlicut a well defined yield point *.* tt>N., = <PAa (0.8 fuJ.

In no case shall fy or O.Sfut be taken greater than 120,000psi.

22.8.3 - The shear strength of a fastener may be calculated as:

For ductile fastener mater:aJ with a well defined yield point.. ...

For fastener material withcut a well defined..yield point.. ..

22.9 Design Requirements for Fastener Pullout 22.9.1- The pullout tensile strength of post-installed fasteners in cracked concrete cannot be calculated. The values of NP must be derived from the* results of tests performed according to ASTM ZXXX, using the test mean minus two standard deviations as the nominal pullout tensile strength.

22.9.2 - The pullout tensile strength of cast in place anchors and headed studs in cracked concrete can be evaluated using 22.9.3, which is based on the fastener's physical dimensions and the concrete properties. The values of NP may also be derived from the results :if tests performed according to ASTM ZXXX, using the test mean minus

. two standard deviations as the nominal pullout tensile strength. The cracked concrete values can be adjusted for uncracked concrete applications using 22.9.5.

22.9.3 - Nominal pullout tensile strength of a single headed stud:

(£q.22-12) 22.9.4 - Basic pullout tensile strength of a single headed stud in cracked concrete:

22.9.5 - 'llacrN shall be taken as 1 except for the special case where a fastener is located

Cale. No. PAL-93.2 Rev. o, Attachment A Page 13 of28

~lo_w Ch_?_rt!9r F~;;~e!"ler Design Fastener PrequaJification SactJcn 22. i.2 , Precluce Pull-out Failure In Uncracked Concrete Demonstrate Suitability In Cracked Concrete Uaing ASTM Tests Sec:tlcn 22.9 Oetennlne Basic Fastener Sedlal'l 22.3.1 ta Strength by Equat!M 22.3.3 ....

or Test Results General Requirements of Methods Allowed NP for Calculating Tensile and Shear Strength

~ -c;;;;d~"SddY------------,

Oetannine Modification Factor a....ReQunnwa

'P- Absence of Cracking

.J, Sedlcr1 22.4 I Calculate Nominal '

I-I II Tensile Pullout S1iength N ... N, 'I' ...

Determine Basic Fastener Strength by Equailon or Test Results j

I I Determine Basic Fastener

\ Shnglb by Equadon

or Test Results I

I I Nb I vb I I y y I Detennine Fastener Steel Tensile Strength IDetermine Modification Factors \  ! Determine Modification Factors 1 II~ ~

I I

~

!I Proleded Alea i Projeaed Alea ~

y

\ J 'P.:t. Eccentricity

'Pec:Y Eccantrtcity Determine Fastener \

Steel Shear Strength \ 'PIH Edge Proximity '¥..., Edge Proximity

\

'I'cH Absence of Cracking \ 'P-v Absence of Cracking *

\

I Calculate Nominal I caJculale Nominal I Tensile B~eakout Strength i Shear Breakout Strength

~

i.-Ouig-.-nl.aeda-....1 1~ "'Nb ~'Pa.'l'v.'P mN !  : Vn

  • VD ~ 'Pec:V~V 'PmY
  • --- \

\ I

.f~--------*--_-_-_---

.,J 1 Faclared La.di i ~ FICl!cn from I

Sec:aOn e.21 J

I

\ Determine Smallest

..__ _ _ __.I Nominal Tensile Strength l.I  ; Determine Sm&llest

! Nominal Shear S1rengttl y

I Tension / / """'""-.Shw and~ermai Shur

, SINctum .Analysia I Onty / , , Only

. '4,4 y

...l Section 22.6 I Interaction 1 _..

c~*auon

,..........V_<-~-v-1 u -

"' n I I O.Cermin. R9qutrlld I Slt9nQ11'11 I ~4 Nu & Vu I Section 22.2.2 I 0.CennWie Strength

' Reduction Factors

__., let Shear 111<1 Tension ' S-ic~:on ~ 7*

  • (llllllCl'I ~ w*I vary wun:
dua of last-)

Cale. No. PAL-93.2 Rev. 0, Attachment A

... _COMMENTARY Page 14 of28 Chapter 22 - Fastening to Concrete 22.1 - Scope The 1995 ACI Building Code has included specific provisions for m

fastening to concrete for the first time. This reflects the increased demand from code users for comprehensive coverage of this important application, the considerable research and design development stimulated by ACI Committees 349 and 355, and increased cooperation with CEB.

22.1.1 - Under tension loadings fasteners can experience 4 different types of failures:

1. Pullout -Failure by sliding out of the fastening device or parts of it from the concrete and/or failure by pulling the cone through the sleeve without the breaking out of a fairly substantial portion of the surrounding concrete.
3. Steel - There are two types of steel failures which can occur. The first and more preferable is a ductile failure by yielding of the fastening device or the system fastened to the concrete before any breakout of the concrete occur~ Such ductile failures utilize a ¢=a.so according to Section 22.2.2(a).

However, many fasteners are produced from brittle material and the initiaJ steel failure can be brittle. In such cases design must be based on a lower cp factor reflecting a more brittle concrete breakout, pullout, or splitting failure.

4. Concrete Breakout - Failure by breakout of a cone of concrete from the structural concrete member before failure of the fastener or the attachment.
5. Splitting of Concrete- Failure by splitting of the structural member before failure of the fastener or the attachment.

Pullout failures are a func!ion of the fastener construction, the anchor mechanism, the condition of the drilled hole, the size of the anchor head (headed studs) and the deformability of the concrete. For post-installed fasteners, these types of failures cannot be predicted theoretically and must be precluded through comprehensive prequalification tests. A method is presented in this chapter to calculate the pullout strength of cast-in-place headed studs in cracked or uncracked concrete.

Steel failure loads can be predicted when the* properties of the fastener are known and when concrete breakout and splitting are precluded.

Concrete breakout failures are the primary focus of this chapter. A highly transparent, user-friendly method of failure load calculation is presented. Other calculation methods or test procedures, are acceptable if proven to offer comparable accuracy and conservatism.

Splitting failures are currently addressed only indirectly with the inclusion of

  • requirements for minimum concrete dimensions which prevent this type of failure.

22.1.2 - For a fastener to be used in conjunction with the requirements of this chapter, the results of the ASTM tests must indicate that, at the tested embedment

Cale. No. PAL-93.2 Rev. 0, Attachment A COMMENTARY Page 15 of28 depth, no sudden pullout failure occurs in uncracked concrete or cracked concrete.

22.1.3 - The behavior of adhesive anchors is often sub~tantially different than expansion anchors and undercut anchors. Adhesive anchors are particularly sensitiv~

to installation procedures. Provisions for adhesive anchors are under development but at this time such anchors are outside the scope of this chapter.

Similarly, provisions for powder and pneumatic actuated nails and bolts are under development, but not included in the current scope.

22.2.2 - The factor for ductile failures is indicative of the confidence in the calculation of steel tension failures, and the amount of forewarning prior to a ductile failure. The factor for brittle failures has been chosen based on the factor presented for *a11 other embedments" in Appendix B of the Nuclear Safety Structures Code (ACI 349), and comprehensive probabalistic studies ( , ] which were based on the use of

<P = 0.65 for concrete controlled failures. Ideally this <P factor could vary depending on the amount of reinforcing near the face, the state of stress in the concrete near the face, the life safety requirements of the attachment, and the redundancy of the connection. While such considerations may be reflected in future editions, there is not sufficient data to be more precise, or less conservative at this time.

The strength reduction factor for fasteners under tension loads, however, does reflect that the performance of post-installed fasteners is sensitive to the installation procedures, and that some fastening devices are more subject to installation tolerances and errors than others. The strength reduction factor, <P for tension, reflects this sensitivity.

The ASTM ZXXX and ZiYY installation safety tests determine the <P which is appropriate for a particular fastening device. The three classes of fasteners and their respective ¢ factors are as follows:

<P = 1.0 x0.65 = 0.65 for systems with high installation safety, Class 1

<P = 0.85 x0.85 =0.55 for systems with m:rmal installation safety, Class 2

¢ = o. 70 x0.65 =0.45 for systems with low but still acceptable installation safety, Class 3.

For cast-in-place headed anchors ¢ may be taken as 0.65. For most undercut anchors this factor will also be 0.65, for most torque controlled expansion anchors this factor may be 0.55.

The failure loads of fasteners under shear loads is not as sensitive to installation errors and tolerances, therefore all fasteners shall use <P= 0.65 for shear calculations. /

22.3.2 - The ACI 349 [ ] approach for concrete breakout tensile and shear strength design may meet the requiremer.~s of this section for many a~plications. The approach is not recommended overall because size effects are not clearly considered, the calculations involved in determining projected areas for overlapping cones and near edges are excessively ccrnG!icated, and the standard deviation is excessive for

Cale. No. PAL-93.2 Rev. o, Attachment A COMMENTARY Page 16 of28 some applications [ ] .

The x: method [ ) deve!8ped at the University of Stuttgart is considered to be generally acceptable for both tensile and shear strength calculations.

The method allowed uncer Section 22.3.5, the CCD Method, is an adaptation of the Jc: method and is considered to be accurate, relatively easy and transparent. The method predicts the load bearing capacity of a fastener or group of fasteners by using one basic equation for a single fastener in cracked concrete, and multiplying by factors which account for the number of fasteners, edge distance, spacing, eccentricity and absence of cracking.

The tensile strength calculations are based on the _ic factor method for single fasteners which assumes a breakout cone angle of 35 degrees (see Figure 22.3.2 (a)).

Both the CCD and the x: methods include fracture mechanics theory, which indicates that in the case of brittle concrete failure the failure load increases at a rate less than the increase in the available surtace and that the nominal stress at failure (peak load divided by failure area) decreases with increasing member size. The nominal stress at failure in tension decreases in proportion to 1/ Jh 81 , so the failure load increases with 1.S

  • h~ . . * .

Similarty ,_the shear strength calculations are also based on the Jc: factor method which assumes a breakout cone angle of 35 degrees (see *Figure 22.3.2(b)), and considers fracture mechanics theory. The nominal stress at failure in shear decreases in proportion to 1 / Jc 1 , so the failure load increases with c, 1.s

  • Test procedures may also be used to determine the basic single fastener shear breakout strength, but the test results must be evaluated on a statistical basis so that the basic capacity is not taken greater than the mean value less 2 standard deviations and the number of tests is suffic:ent for statistical validity.

22.4.2 - The basic fastener capacity equation is derived assuming a concrete failure cone with an angle of 35 degrees, and considering fracture mechanics concepts, as follows:

? 1 k h-0.S 3 ef where k 1 Jfc' = Concrete tensile capacity, IS he,2 = Activated load bearing area, k 3 hef -o.s = Size effect.

where k 1 , k2 , and k 3 are calibration factors.

If k =k 1

  • k 2
  • k3 , then:

15 Nb = k Vfji 1

c h Bf*

Cale. No. PAL-93.2 Rev. 0, Attachment A COMMEN-TARY Page 17 of28 For metal anchors and headed studs, the mean values of k and the standard deviations were determined from a large database of test results [ ] . The k value in the design equation is the mean value for specimens tested ill cracked concrete less two standard deviations (one-sided 95% tolerance limit with 95% confidence) [ ].

For fasteners influenced by three or more edges, it is possible that the concrete breakout capacity of the fasteners, which is calculated using the CCD .Method, decreases for a constant edge distance and an increasing embedment depth, h., .

This always yields safe results, but it does not reflect actuaJ behavior. These special cases have been studied for the JC-Method [ ] .

The anomaly can be remedied in two ways. The firSt is to use a fracture

  • mechanics approach for the load reduction. The primary problems with this approach are the very complicated formulas and the loss of transparency in calculating ultimate loads. The second approach is more simple and user-friendly. An effective value for the embedment depth het
  • which predicts the highest capacity for the fastener arrangement may be used. The value may be calculated as:

he1* = C1 /1.5.

22.4.3 - Figure 22.4.3(a) shows ApNo* ~No is the maximum projected area for a single fastener. Figure 22.4.3(b) shows examples of the projected areas for various single fastener and multiple fastener arrangements. Since ~N is the total projected area for a group of fasteners, and ApNo is the area for a single fastener, there is no need to include the number of fasteners in the equation.

22.4.4 - Figure 24.4.4 shows dimension e.

22.4.5 - If fasteners are located so close to an edge that there is not enough space for a complete breakout cone to develop, the load bearing capacity of the fastener is reduced. If the smallest side cover distance is greater than c 1crN (1.5 he,), a complete

=

cone can form and there is no reduction (VsN 1). If the side cover is less than C 1crr:(1.5 he,) the factor, V sN , is required to adjust for the disturbance of the radial symmetric stress distribution.

22.4.6 - Figure 22.4.S(a) shows the design equation for post-installed anchors in uncracked concrete compared to test results. Figure 22.4.S(b) shows the design equation for headed studs in unc:-acked concrete compared to test results.

22.5.2 - As with tensile loadings, the concrete breakout shear capacity does not increase with the failure surface, which is proportional to c 12

  • Instead the capacity increases proportionally to c 1 1*5 , due to the size effect. The failure load is also influenced by the fastener stiffness and the fastener diameter.

The constant 6 in the shear strength equation was determined similarly to the k value in the tensile strength ec;ua::on. It is also the mean val~e of the constant less two standard deviations (one-sided 95% tolerance limit with 95% confidence) times a 0.7 factor to account for cracking.

Cale. No. PAL-93.2 Rev. 0, Attachment A

_-- - COMMEN-TARY Page 19 of28 c,c:rH =1.Shet 1.Sh.,,

1.5h"'

Section Through Failure Cone

1. )(

Plan View Ar;ld- 2c,_..,x 2c,_...

.. 3h.,x 3h., =Sh~

Figure 22.4.3(a) CaJculation of A pN0

Cale. No. PAL-93.2 Rev. 0, Attachment A COMMENTARY Page 18 of 28 For fasteners influenced by three or more edges, it is possible that the concrete breakout shear capacity of the fasteners, which is calculated using the CCD Method, decreases for a constant edge distance and an increasing distance to the edge of concrete, c 1

  • This always yields safe results, but it does not reflect actual behavior.

These special cases have been studied for the 1e-Method [ ] .

The anomaly can be remedied in two ways. The first is to use a fracture mechanics approach for the load reduction. The primary problems with this approach are the very complicated formulas and the loss of transparency in calculating ultimate loads. The second approach is more simple and user-friendly. An effective value for the distance to the concrete edge, c 1

  • which predicts the highest capacity for the fastener arrangement may be used. The value may be Calculated as:

C1 * = <; /1.5.

22.5.3 Figure 22.5.3(a) shows ~vo* ~vo is the maximum projected area for a single fastener, which approximates the surface area of the full breakout cone for a fastener which is unaffected by edge distance or spacing or depth of member. Figure 22.5.3(b) shows ~xamples of the projected areas for various single fastener and multiple fastener arrangements. Apv approximates the full surface area of the breakout cone for the particular arrangement of fasteners. Since ~~ is the total projected area for a group of fasteners, and Apvo is the area for a single fastener, there is no need to include the number of fasteners in the equation.

22.5.4 - Figure 22.5.4 shows the dimension e for shear calculations.

22.5.5 - Figure 22.5.5 shows the dimension C2 for the 11'sV calculation.

22.5.6 - Figure 22.5.6 shows the design equation for fasteners in uncracked concrete compared to test results.

22.6 Traditionally, the shear-tension interaction expression has been expressed as:

where x varies from 1 to 2. The current recommendation is a simplification of the expression where x = 2 (see Figure 22.6} with limits chosen to eliminate computation of interaction effects where very small quantities of the second force are present..

Cale. No. PAL-93.2 Rev. O, Attachment A Page 20 of28 ApN =(c 1 +1.5~1)(2 x 1.5he1) if c, < 1.5h91 .

ApN =(c 1 +s+1.5he,)(2x1.5hef) if c 1 < 1.Shet and s < scrN =3h et ApN = ( c 1 + s1 + 1.5h91 )(c 2 + s 2+1.Shat) if c1 and c 2 < 1.Shet and s 1and s 2 < scrN =3h 9 r

  • Figure 22.4.3(b) Projected areas for single fasteners and groups of fasteners

Cale. No. PAL-93.2 COMMENTARY Rev. o, Attachment A Page 21of28 Figure 22.4.4 Definition of Dimension e

Cale- No- PAL-93.2

. COMMENTARY _. Rev. 0, Attachment A Page 22 of28 Equation vs Test for Post-Installed Anchors in uncracked concrete. and not effected by edges or spacing x

Mean CCD Equation for tension on Post-Installed Anchors 200 in uncracked concrete x

x "O

- x x

co 0

.....J Q)

J co L.l..

100 150 200 250 Effective Embedment Depth, h 91 , mm 22.4.6(a) Mean and Design CCD Equations for anchors in uncracked concrete compared to test de.ta, tor post-installed anchors.

Cale. No. PAL-93.2 Rev. 0, Attachment A

--- - ---COMMENTARY - Page 23 of28 Equation vs Test for Heaaed Studs In &n:radced cancr9te. and not effected by edgee or spacing x

900'I Mean CCD Equation---..._

~'

~

for tension on ~ /

/

Headed Studs /

/

in uncracked /

I concrete /

/

/

l

/

600 /

/

/

/

/

/

/

/

/

x ~ /

Ij ~ / x//

~ _,,.,. Design 300 i.

II ~ /~,.--~Equation i' ,,.,,./ for anchors in

' ~ x /'

i x x ~/ uncracked X *<Z -',,,,

so .... " concrete o.....,
;::;__~~~---~~~~--<1--~~~~-'

0 200 400 600 Effective Embedment Depth, hef ,mm 22.4.6(b) Mean and Design CCD Equations compared to test data, for headed studs in uncracked concrete.

Cale. No. PAL-93.2

_____CQMMElfl'.ARY __ Rev. 0, Attachment A Page 24 of28 Equation v~ Shear Test Results fer single anchors in deep uncracked members European Tests 160-1 z I Mean CCD Equation ~

i J20t tor shear in uncracked concrete x

/,,

~

x ,/

I x

/

  • ca I ,, ,,/

/

u. 80 ~

I /

/

x , "'

/

>O< /

x x "'/

x x ,.,,/,..,Design 40 l ~ ,,.£__ Equation .

I x x xx ,,.""' for shear in I. xx:.

x~ 311

,,.,,.-x ;;s/ uncrac ked 1

h x ""....-

concrete i -~~,

o-*~:;_~~~~~~~~~~~~~----.J 0 100 200 300 Edge Distance in Direciton of Shear, c 1

  • mm Figure 22.5.2 Mean and Design CCD Shear Equations for uncracked concrete compared to test data.

Cale. No. PAL-93.2 Rev. o, Attachment A COMMENTARY Page 25 of 28 c

1.5c 1 /Fastener

.. v 1.5c 1 c 2crV =1.Sc 1

.. )r 1.5c 1 1 1.5c 1

~1


Edge of Concrete Plan View 1.5c 1 y

I_Apvo= 2c2aVx c aV

_.1 Side Section 2

= 3c1 x 1.5c 1 = 4.5c~

Front View Figure 22.5.3(a) Calculation of Apvo

Cale. No. PAL-93.2 COMMENTARY Rev. 0, Attachment A


Page 26 of28 Apv= 2 x 1.5c1 x h if h < 1.Sc 1 Apv= (2 x 1.Sct'~ s ) x h if h < 1.5c1 and s < Scrv = 3c 1 I

ApV -I\ Apv= 1.sc, ( 1.5c1+ c )

r- if c 2 <1.Sc 1 2

0 IO I )

\

,' ~

(

y Figure 22.5.3(b) Projected areas tor single fasteners and groups of fasteners

Cale. No. PAL-93.2 COMMENTARY- -- --* ----**- -- Rev. O, Attachment A Page 27 of28 Figure 22.5.4 Definition of Dimension e c1 I

~ < - ~I r--1',, 3s~II 0

I Ii .. . J.: iI c2. 1 =1.5c 1 .; 1

  • '..'..'.
  • I / Fastener

<*~_-;

-t-+-.--::-_:_ ....

  • I..

I

. II I

v C2.2< 1.sc 1L 7_ _:_____

1 1

\

\__ Edge of Concrete Plan View For calculation of '¥5 v use smaller of cz 1 and c 22

  • Figure 22.5.5 Dimension c 2 for Edge Proximity Modification Factor

Cale. No. PAL-93.2 COMMENTARY Rev. 0, Attachment A Page 28 of28

              • . r(.!i**.* Nu

)~ (::L )2= 1 Vu

    • Trilinear
    • .. Interaction T

Approach

.2N u .. -- ---.----- -* -* -.---------------. ---------------

Figure 22.6 Shear and Tensile Load Interaction Equation

Cale. No. PAL-93.2 Rev. 0, Attachment B Page 1of15 T~s

-***- PALISADES NUCLEAR PLANT ENGINEERING ANALYSIS COVER SHEET SGRP Anchor Bolt Loads for ACI-318-95 Assessment EA- A-NL - q S -o I lo Total N11nber of Sheets 12 +3 A~ ... &.J T.S INITIATION AND REVIEW Calculation Status Preliminary Pending Final Superseded D D s: D InUiated I nit Review Metlx>d Technically Reviewed Revr Rev Appd Appd CPCo I Description By Detai 1 Qual By Appd Bv Date Alt Cale Review Test Bv Date 0 Original Issue ABC\ S{'l[,~

~ I rkil ~/7-/~3 R~

-I. OBJECTIVE The objective of this calculation is to develop Drillco and Hilti bolt loads for an evaluation of bolt capacities for pipe supports SP-033383-DBB7-H7 and SP-033383-DBB7-H9. In addition, pipe supports SP-033382-DBB8-H4 and SP-033382-DBB8-H14 are evaluated to conclude that they are not controlling with respect to concrete/bolt interaction analyses. These pipe supports and bolts were previously evaluated during the steam generator replacement project. However, they (bolts) are being reviewed per a different criteria at this time.

It is not the purpose of this calculation to evaluate bolt loads. It is the purpose of the calculation to define bolt loads for a subsequent evaluation of concrete capacity versus bolt capacity per AC! 318-95 (draft) guidance.

II. ANALYSIS INPUT A. EA-SP-033383-DBB7-H7, Steam Generator E-508 Slowdown, Rev 0 B. EA-SP-033383-DBB7-H9, Steam Generator E-508 Slowdown, Rev 0

c. EA-SP-033382-DBB8-H4, Steam Generator E-508 Recirculations, Rev 0 D. EA-SP-033382-D888-Hl4, Steam Generator E-508 Recirculation, Rev O E. Palisades Technical Specification C-173(Q), "Technical Requirements for the Analysis and Design of Safety Related Pipe Supports, Rev 4

Cale. No. PAL-93.2

@>:.:  :; IC PALISADES NUCLEAR PLANT ANALYSIS CONTINUATION SHEET Rev. 0, Attachment B Page 2of15 EA- A-/\J L- 'f.3-o/b Sheet _2_ Rev # _¢____

Reference/

F. Memo from Anthony H Hsia (NRC) to CPCo, "Summary of Meeting Held on March 23, 1993, with Consumers Power Company on Anchor Bolts Analysis for Palisades Nuclear Circulating Plant",

April 05, 1993 G. AIR-A-NL-93-016 H. Memo from RB Jenkins (CPCo) to AC Eberhardt (Sargent & Lundy),

Anchor Bolt Loads for Drillco/Hilti Bolts Interaction, RBJ93*022 III. ASSUMPTIONS The assumptions will be noted in the text of the analyses.

IV ANALYSIS This evaluation is being conducted in order to determine applied piping loads on the bolts. The analyses of Reference II.A, 8, C and D use design loads. These references will be sorted to arrive at the applied loads.

A. Evaluation of EA-SP-033383-0887-H? (Reference II.A).

Sheet 3 of the enclosed Bechtel calculation yields the information in Table 1. For support Ml01-6048-H7, the smallest ratio of design load to applied load is 1.45.

Support Ml01-6074-Hl3 is also supported from the same frame assembly. Such a ratio cannot be directly taken from Reference II.A, or the attached Table 1. However, per page 8.5 of the Bechtel calculation associated with Reference II.A, the ratio can be derived by 1/.698 = 1.433 (the ratio of the design load/actual load). These numbers (1.45 and 1.433) are essentially the same.

These values do not reflect self weight excitation (SWE) of the support. Exactly how Bechtel incorporated these loads into its computer output is not clear. However, the SWE loads will not be a concern here because they are judged to be very low. Clearly, the 1.5 multi-mode factor employed by Bechtel in calculating them is not necessary for a very stiff frame.

The vertical spectra ordinate is also overpredicted because it should reflect the horizontal ground rather than horizontal amplified response spectra.

Cale. No. PAL-93.2 Rev. 0, Attachment B Page 3of15 PAL I SADES NUCLEAR PLANT EA-A-NL-'la-014>

ANALYSIS CONTINUATION SHEET Sheet _3_ Rev # -11¢---

Reference/Corrment Sheet 2 of Attachment C of the Becthel calculation shows that five load cases were employed in the ME035 calculation for the Drillco plate. How load case 1 (SWE) was managed with the other four is not clear. However, an inspection of the magnitude of these SWE baseplate inputs with respect to the piping design load inputs at the baseplate shows that they are very small. Therefore, their influence over this analysis may reasonably be ignored. Sheet 2 of Attachment E of the associated Bechtel calculation shows three load cases for the Hilti bolt plate. Table 2 of this evaluation lists the loads taken directly from the ME035 outputs for those plates.

Table 2 loads have been reviewed in order to determine how to consolidate and simplify them for ultimate analysis with regard to concrete capacity. The following approach has been taken:

The largest load for each bolt (node point) location in tension and shear (two loads actually) have been selected regardless of load combination which generated them.

These loads are from load cases two through five for the Drillco plate and load case one through three for the Hilti plate.

The commensurate SWE (load case 1) load is added to that load determined above for the Drillco plate only. This is a conservative approach.

The resulting loads are then scaled down by 1.433.

The lowest ratio of the two supports.

An example for the Drillco plate is for node point i/n shear--

(2089 + 34) = 1482 lbs 1.433 An example for the Hilti plate is for node 78 in tension --

1704 = 1189 lbs

1. 433 -

The resulting loads are in Table 3 of this evaluation.

8. Evaluation of EA-SP-033383-DB87-H9 (Reference II.8).

Unlike support calculation EA-SP-033383-DBB7-H7, the associated support with this calculation contains but a single Drillco baseplate. However, this baseplate is in close proximity to a Hilti baseplate from another support.

Cale. No. PAL-93.2

<§>:.:

eg=an; I -

PALISADES NUCLEAR PLANT ANALYSIS CONTINUATION SHEET Rev. 0, Attachment B Page 4 of 15 EA- A-l'JL-ql,01'3 Sheet _4_ Rev # ¢.

Sheet 3 of the associated Bechtel calculation shows the loads in Table 4. Sheet 2 of Attachment A of the Bechtel calculation shows the piping input loads to the support. That MElOl output shows that the 372 lb normal piping load at node point 260 consists of 269 lbs of deadweight, 3 lbs due to thermal expansion and 99 lbs of seismic load. The moments and horizontal forces are all due to friction. However, the friction factor should be based upon weight and thermal support loads and not upon seismic loads. (See Reference II.E). Therefore, all the friction loads should be scaled by 27 2 = 0.731 (note numerical rounding). The result is the ma ulation of corrected loads in Table 4. Comparing the corrected loads with the design load yields a minimum bump factor of 4.03 between support analysis (design) loads and applied loads.

This factor does not apply directly to the SWE loads calculated for this support. That calculation employs a peak-of-the-spectra approach with a 1.5 multi-modal factor and with the vertical input scaling the horizontal amplified response spectra. In addition, the 1.5 multi-modal factor is used to scale gravity as well. The resulting loads are extremely conservative -- perhaps by an order of magnitude.

Sheet 2 of Attachment C of Reference II.B shows the three load cases used the ME035 baseplate analysis. The Table 5 loads here reflect the bolt loads directly from that output. Load case 1 is the SWE load case.

Page 9.3 of the Bechtel calculation provides a 730 lb tensile load and a 564 lb shear load for the Hilti bolts from a referred calculation. It will be assumed that all of the Hilti bolts on that plate have those loads without reviewing the calculation. A sketch of that plate is on sheet 8 of the Bechtel calculation. The process of arriving at bolt loads for the bolt pattern versus concrete capacity will be as for the earlier assessment. That is as follows:

For each node point in Table 5, take the maximum tension and shear load from load case 2 or 3.

Tabulate these loads and add to the corresponding SWE load of load case 1.

Then factor the sum by 4.0 in order to get applied support loads. The result is in Table 6.

Cale. No. PAL-93.2 Rev. 0, Attachment B

@:.: -? cc PALISADES NUCLEAR PLANT ANALYSIS CONTINUATION SHEET Page 5of15 EA- A - N L - q?. -o I la Sheet Rev # ~

-,~-

Reference/Comment

c. Evaluation of EA-SP-033382-DBB8-H4 per Reference II.C.
  • Per Reference II.F, this calculation need not be revisited in detail because the calculated Drillco plate interaction ratio is similar to that determined in the Reference II.A and B calculations and because the Hilti plates in the area are very lightly loaded. It is assumed, then, that the other calculations bound. Is is noted, however, that the design load/applied load ratio in this calculation is a minimum of 2.43. Therefore, it is not judged to be limiting.

D. Evaluation of EA-SP-033382-DBB8-H14 per Reference II.D.

Per Reference II.F, this calculation need not be revisited in detail because the calculated Drillco plate interaction ratio is similar to that determined in the Reference II.A and B calculations and because the Drillco bolts and Hilti bolts do not see their maximum loads simultaneously. It is assumed, then, that the referenced calculations bound. It is noted, however, that the design load/applied load ratio in this calculation is a minimum of 2.0. Therefore, it is not limiting.

E. Evaluation process summary.

There is an infinite number of ways to manage the Bechtel design loads so as to determine more realistic, yet conservative, applied loads. The approach here is merely one of the simpler methods.

It has been assumed that the Bechtel analysis work is accurate. It may have been possible to review the calculation for the Hilti plate in proximity to the support in the Reference II.B calculation above. However, a simplifying assumption was made.

The Table 3 and Table 6 bolt loads replace those of Reference II.F and will be transmitted to Sargent & Lundy to satisfy Reference II.G.

V* CONCLUSION This calculation provides reasonable anchor bolt loads for the analysis required in Reference II.F.

Cale. No. PAL-93.2 Rev. 0, Attachment B Page 6 of 15 PALISADES NUCLEAR PLANT EA- A-NL 0lb ANALYSIS CONTINUATION SHEET Sheet Rev # r£

-,~-

Reference/Conment TABLE 1 Load Data for Support SP-033383-DBB-H7 for the Drillco and Hilti Baseplates Support No Ml01-6048-H7 Pipe Loads (Faulted)

Fx = 889 lbs Fy = 1,035 lbs Fz = 593 1bs Mx = 1,483 in-lbs Support No Ml01-6048-H7 Design Loads (Faulted}

Fx = 2,000 lbs Fy = 1,500 lbs Fz = 1,000 lbs Mx = 2,500 in-lbs Support No Ml01-6074-Hl3 Design Loads (Faulted)

Fx = 1,000 lbs Fy = 1, 500 lbs Fz = 450 lbs Mx = 1,125 in-lbs My= 1,800 in-lbs

Cale. No. PAL-93.2 Rev. 0, Attachment B Page 7of15 PALISADES NUCLEAR PLANT

' I EU ANALYSIS CONTINUATION SHEET

  • Sheet _7_ Rev # ¢ a :a 118 TABLE 2 Baseplate Bolt Loads for Support Cale EA-SP-03383-DBB7-H7 DRILLCO PLATE BOLT LOADS LOAD CASE NODE SHEAR {lbs) TENSION (lbs) 1 11 59 214 1 17 34 80 1 61 70 17 1 65 86 86 2 11 274 0 2 17 1482 1079 '-

2 61 1937 5331 2 65 1775 1728 3 11 1574 0 3 17 2089 1205 3 61 1202 5470 3 65 218 1626 4 11 1323 4197 4 17 352 1955 4 61 1472 573 4 65 2136 1516 5 11 278 1503 5 17 1487 0 5 61" 1935 2864 5 65 1774 3847

Cale. No. PAL-93.2 Rev. 0, Attachment B Page 8 of 15 PALISADES NUCLEAR PLANT EA- A-NL o lb

  • -. JI llEU ANALYSIS CONTINUATION SHEET Sheet _8_ Rev # ¢ Referenca/Conmmt TABLE 2 - cont' Hilti Plate Bolt Loads LOAD CASE NODE SHEAR (lbs) TENSION (lbs) 1 12 616 942 1 19 623 704 1 78 702 1615 1 82 701 827 2 12 639 860 2 19 635 770 2 78 686 1704 2 82 676 739 3 12 627 1090 3 19 618 770 3 78 696 1470

' 3 82 698 804

Cale. No. PAL-93.2 Rev. 0, Attachment B Page 9 of 15 PALISADES NUCLEAR PLANT EA- A-NL. *'ll -Ofb ANALYSIS CONTINUATION SHEET Sheet _9_ Rev # --?-</--

TABLE 3 Applied Loads for The EA-SP-033383-DBB7-H7 Baseplates DRILLCO MAXIBOLT LOADS NODE POINT MAX SHEAR (lbs) MAX TENSION (lbs}

11 1139 3078 17 1482 1420 61 1401 3829 65 1550 2745 HILT! BOLT LOADS NODE POINT MAX SHEAR (lbs) MAX TENSION (1bs) 12 446 761 19* 443 537 78* 490 1189 82 489 577 TOTAL LOADS FOR EIGHT BOLT GROUP TENSION 14,136 lbs SHEAR = 7,440 lbs

  • These are the HILT! bolts closest to the DRILLCO bolt plate.

Cale. No. PAL-93.2 Rev. 0, Attachment B Page 10 of 15 PALISADES NUCLEAR PLANT EA- A-NL _q a -otb ANALYSIS CONTINUATION SHEET Sheet 10 Rev # -

Reference/Comment TABLE 4 Load Data for Support EA-SP-033383-DBB7-H9 Applied Piping Loads for Support No Ml01-6050-H9 Fx = 0.3(372) = 112 lbs Fy = 372 lbs Fz = 0.3(372) = 112 lbs Mx = 1.5(112) = 168 lbs Joint 7 Mz = 1.5(112) = 168 lbs Joint 7 Mx = 5.25(112) = 588 lbs Joint 4 Mz = 5.25(112) = 588 lbs Joint 4 Design Loads to Quality Support No Ml01-6050-H9 Fx = 500 1bs Fy = 1,500 lbs Fz = 500 lbs Mx = 1,750 in-lbs Mz = 1,750 in-lbs ected Piping Loads for Support No Ml01-6050-H9 Fx = 0.3(272) = 82 lbs Fy = 372 lbs Fz = 0.3(272) = 82 lbs Mx = 1.5(82) = 123 lbs Joint 7 Mz = 1.5(82) = 123 lbs Joint 7 Mx = 5.25(82) = 431 lbs Joint 4 Mz = 5.25(82) = 431 lbs Joint 4

Cale. No. PAL-93.2 Rev. 0, Attachment B Page 11 of 15

...... PALISADES NUCLEAR PLANT ANALYSIS CONTINUATION SHEET Sheet 11 Rev # ;6.

Reference/Cooment TABLE 5 Baseplate Bolt Loads (Ori 11 co Pl ate) for Support Cale EA-SP-033383-DBB7-H9 LOAD CASE NODE SHEAR (1 bs} TENSION (1 bs}

1 18 133 24 1 34 166 172 1 23 43 30 1 39 110 262 2 18 549 2218 2 34 318 133 2 23 785 3073 2 39 845 497 3 18 592 319 3 34 726 2384 3 23 264 237 3 39 501 2875

Cale. No. PAL-93.2 Rev. 0, Attachment B Page 12 of 15 PALISADES NUCLEAR PLANT EA- A - NL .. 9}. - 0 Ib ANALYSIS CONTINUATION SHEET Sheet 12 Rev # _rj__

Rafarence/Comment TABLE 6 Applied Loads for the EA-SP-033383-DBB7-H9 Baseplates DRILLCO MAXIBOLT LOADS NODE POINT MAX SHEAR {lbs) MAX TENSION (lbs) 18 181 561 34 223 639 23 207 776 39 239 784 HILT! BOLT LOADS NODE POINT MAX SHEAR (lbs) MAX TENSION {lbs)

ALL 4 BOLTS 564 730 TOTAL LOADS FOR EIGHT BOLT GROUP TENSION 5,680 lbs SHEAR 3, 106 1 bs

Proc No 9.11 T£CllllQL BQIQ CHECKLIST Athchment S Rev1s1on s EA - A-NL- '13 -OJ Ca REY. __..fP__ Page 1 of 1 This ch.cklist provides guidance for the revtaw of engineertng analyses.

Answer questions Yes or No, or N/A if they do not apply. OocU111nt all conaents on a 3110 Form. Satisfactory resolution of cmm11nts 1Hd completion of this checklist is noted by the Techntcally R1viewed signatur! an the Initiation and Review record block of Fon1 3619.

(Y v:N, N/A)

1. Have the proper input codes, standards and design .l.

principles been spectfttd?

z. Have the input codes, standards and d1sign principles b1en prop1rly appl ild? __

y_.___ _

3. Are 111 inputs and assumptions valid and th* basis far y 4.

their use docU111ntld?

Is Vendor 1nfanat1an used as input addressed correctly in V

the analysis? _l__ __

5. If the analysts argUMnt departs fl"Oll Vendor Infan11tton/Reca1111ndations, is the departure Just1f1cat1on dacUlllfttld? ~1~
6. An 1ssQ11Ptton1 1ccurat1ly descr1bld and reasonable? 'i
7. Has* th* use of engfnHrtng Judg-t beea doc._tad and Justtftld? *
  • t.J[A
8. An all constants, variables and fa,...lu cornet lftd properly appltld?

y

9. Have any *tnor (tnstgntft,_cant) 1rron . _ tdeettftld? If yes; Id111ttf1 on the 3110 FoN UICI Justify thetr (

tnstgntftcance. (

10. Does analysts involve *ldtng? If Yes; vwifJ the following tnfarwatiOlt ts 1ccurat1l1 repreHfttld on the analysts dr*tng (CJuqMR doc'tmlnt). N/A
  • Type of W.141
  • Sta of V.lct
  • 1i."t1l Bltng JotnlCI ..
      • Md TM.c**** of Material Being Jotnld Laca&.t09 of Veld(*)
  • Aptwapriata Veld S_..loV U. Has the objecttve of the analysts bHn mt?
12. Have ldllt11t1tr1ttve nqut..-ts sudl 11 nUllblrtng ud far11&t beell sattsftld?

Cale. No. PAL-93.2 Rev. 0, Attachment B Page 13of15

Cale. No. P NUCLEAR OPE~NS DEPARTMENT Rev. 0, Attach Oocum~-view Sheet Page 14 of IS Document Number Revis~Number C.1. 3a- EA-A-~L-'l3-0IC::, \./.) Page J of I Item Page andtor Number Section Number Comments Response or Resolution 1- 1

2. y 3, .,

~

Lj. l"eo.'blt ~ Pri\\

5e~

s. £ }"..,~~,.\ 2. 1:.-r~ c..>...3c:..\ <Ar" *, .,:)

.\\,42> ~

"I..,so....t j 5~

f 4l T:; r-e.t t) 4 R~'\.

q T,.\i\es  !~ti- ~ ~7 5\..DJ\J ~ll. 1-.\h 2. - Co Jo i Organization Review Coordinator Date Date' 5+L ~e, s 1~ 9*~ s  ?- 'fj Form 3110 10-91

Cale. No. PAL-93.2 Form 3698 9-89 PALISADES NUCLEAR PLANT Rev. 0, Attachment B ENGINEERING ANALYSIS CHECKLIST Page 15 of 15 Affected Revision Items Affected By This EA Yes No Required Identify* Closeout Other EAs

2. Design Documents Elec E-38 through E-49
3. Design Documents Mech M259, M664, M665 4.0 LICENSING DOCUMENTS 4.1 Final Safety Analysis Report (FSAR) 0 [8 4.2 Technical Specifications 0 ~

4.3 Standing Order 54 0 ISif 5.0 PROCEDURES 5.1 Administrative Procedures 0 ~

5.2 Working Procedures 0 ~

5.3 Tech Spec Surveillance Procedures 0 eg 6.0 OTHER DOCUMENTS

.1 Q-List 0 Iii

.2 Plant Drawings 0 ~

6.3 Equipment Data Base 0 ~

6.4 Spare Parts (Stock/MMS) 0 ~

6.5 Fire Protection Program Report 0 ~

(FPPR) 6.6 Design Basis Documents 0 ~

6.7 Operating Checklists 0 ~

6.8 SPCC/PIPP Oil and Hazardous Material Spill Prevention Plan 0 ~

6.9 EEQ Documents 0 ~

Do any of the fol lowing documents need to be generated as a result of this EA:

Yes No

1. Corrective Action Document? 0 81 Reference
2. Safety Evaluation? 0 ~ Reference
3. EEQ Evaluation Sheet? 0 ~ Reference PRC Review of this EA Required? 0 ~

Date S-'7- 93..

  • Identify Section, No, Drawing, Document, etc.

ITEM' NO. I PART NO. 1REO'D. I NO. DESCRIPTION MATERIAL 1 1 TS 4 X 4 X Ye* X 2 *-s* LC~ ' - ASTM A-500 GR.B 2 2 TS 3 X 2 X 1/4

  • X 0 *- 7V2
  • LG. A ASTM A-500 GR.B 3 1 TS 3 X 3 X 1/4* X 0'-1-0Y2* LG:'Y.:>\ ASTM A-500 GR.B 4 1 PLATE 11/4
  • X 1 1
  • X 16
  • ASTM A-~6 5 4 V2* C2l X 13 1/4"' LG.ORILLCO MAXI-SOLT MIN.EMS. *8"' SPEC. C-97ACO>

6 2 BAR 1/'4

  • X 3Y'2
  • X 3Y'2
  • CROUNDED CORNERS l ASTM A-36 7 1 BAR 11.*x 2v,*x 2v,* <ROUNDED CORNERS> ASTM A-36 8 1 PL ATE 1/ ..
  • X 5
  • X 5 * - ASTM A-36 9 ' 1 TS 4 X 4 X 1/4* X o*-3* LG."}. ASTM A-500 GR.B

/j\

NOTES:

1. FABRICATE PER SPEC. 20557-M-614(Ql. ..'
2. INSTALL PER SPEC. 20557-M-604f Ql.
3. RESTRAINT DIRECTION: FY FOR HANGER LOCATION. SEE
4. NO REBAR IS TO BE CUT DURING THE INSTALLATION DRAWING M101-6041 OF EXPANSION BOLTS. Cale. No. PAL-93.2 ..

ATTACHMENT Rev. 0, Attachment C TO CALC.. NO.~ Page 1 of 3 TOTAL N~ SHEETS 3 -

'\:i"< ., I \'<>o\ C\ I LOCATION PLAN

~*. ~ l-~0-91 AREA 1 - CTMT.

3 I N/R NIR llJA N/R N/R N/R IRE,_V *!ELECT /CS Nv

  • MECH. CIVIL PLANT DESIGN ARCf.l. NUCLEAR ...

ly N~

1 COORDINATION SIGN-OFF  :

DATA PT. 260 z x .. - n.

I 'I\ 1/'so/41 INCOAP. FCA-227. ISSUED 9AS-BUILT* ~ 9-.. *P~c. YH I 1\ 11/10190 INCOAP. FCR-116 I FCN-171. REVISED AS NOTED cw CHA TDT P.IC "CS -~:

/1\ 8121190 ISSUED FOR CONSTRUCTION PEA FC-193 C8 AP CHA PJC C8 ~

~ 6 /a.. 190 ISSUED FOR FABR I CAT I ON CB RP PJC GB

~.

REY. DATE REVISIONS - 9T Cll ..

$.G. £508 8LCM>OWN BECHTEL POWER CORPORATION SYSTEM STRESS ISO *101-SCMt PALISADES NUCLEAR PLANT G.\ITIOSIUC PIPE SUPPORT .CALC. SG88-PD-H9

'20557 DRAWING NO.

M101-6050 SHEET 1 OF 3 DFN:QS1 :[230,7J6050H9 ftEV.

3

.DGN

Cale. No. PAL-93.2 Rev. 0, Attachment C Page 2 of3

  • PLATE T.o.c.

EL.612'-0"

  • -*i' ~ TOP ONLY 1 "GROUT 1" MIN. 1Y2" MAX ©
  • j 3

ELEVATION .~

LOOKING SOUTHWEST T i .

a .

  • ~

.i BECHTEL POWER CORPORATION SYSTEM S.G. E508 BLOWDOVN STRESS ISO M101-60<tl PALISADES NUCLEAR PLANT ~lll£RSIUG PIPE SUPPORT CALC. SGBB-PO-H9

20557 DRAWING NO.

M101-6050 SHEET 2 OF 3 DF'N=QS 1 : [ 230, 7 l 6050H9 REV.

  • OGN

Cale. No. PAL-93.2 Rev. 0, Attachment C Page 3 of3 9

PLATE

  • 9
  • 21/4* 31/4. 31/4* 21/4*

,..,.,"°

~

~

+

PLATE 8 T.S.

  • -~

~s. 0 DRILLED SECTION A-A HOLE (TYP. 4 l

- *-:a

~

i a i BECHTEL POWER CORPORATION SYSTEM S.G. ESOB BLOWOWN STRESS ISO M101-604'1 PALISADES NUCLEAR PLANT GAJTl£RS8lMG PIPE SUPPORT CALC. SGBB-PD-H9 STEAM GENERATOR JOB NO. DRAWING NO. REV.

REPLACEMENT PROJECT 20557 M101-6050 PIPE SUPPORT NUMBER H9 SHEET 3 OF 3 L_A~ = PS3 . DFN=QS1 :[230,7J6050H9 .DGN

JCe NO. 20H7 PAGE 1 OF .2_

TASK NO. /O. I PALISADES STEAM GENERATOR REPLACEMENT PROJECT WP&IANO.

wA. - '"* 1 - m. 00 (,

"' DRAWINGOA SPECIFICATION NO.

FIELD CHANGE REQUEST/FIELD CHANGE NOTICE AEV.

NO.

DWG. OR SPEC. CHANGES DCN FCA FCN ADD. y Q

N

~NO.\\~

"""AllEIU'f' AHIHIATi

/1~,,.11!. 1'1..\~D&,Of-I'(! ri t1 11 l ____ .L,*_

()

I FOi!!_

lot.J~;o

"'101- ,0 so Jos~-7-Q-17,4.(Q)

I 4

x I><

M.AIOll '°" OWIGl To F 4.c11-1r-A.1~ 0o.i.J ST A." C'T1t:1 l..J IXllTINCI CCllCITIOlll S P£t. C-'r 'JA(G), p!ti.A,.. j. I

  • 1'( A., LG \A.J A. W TH-L ~ 1s T A-r.J cl. 6~ jwJ !.£1>.J ~

"Ni:.~h mA-K L- Ao'- r ,......, 4)

.A-t..> f.').15;-11'.I'- A" cH-,, rt ~ lit f..1\i.£.Z> ttlS.1'1T OI= )\.4-U -601..-r -r U-) x i#l. ])1.+11'1£ 'TEA. o~ TH!. E.x1si"1AJ'- ~11o1c.~.

QUCMITIOll OI' CHAHCIE Ul.l l&.N '-':tc. ,,.'i10 ti\) OF .z. ~<:~

~'"**IS F*r.t Jt4. /ol- "oSO - /+ 1 CJl/l..L AJ~

~1..£. r ~~~.~ s p,+(./ 1',J G- A. f:... Q Y 1A.LMLvrs. A.ta.w.Ls T A'P~"~ of LJ f-S -...; ),Ne:~ /_u;:.)r.Tlo /tJ SJ/.r. ...u-J () AJ }.U..f. .l.

Cale. No. PAL-93.2 AiT'ACHMENT Rev. 0, Attachment D Page 1 of2 TO CALC. NO.:

TOTAL NO. OF SHEETS tJ

'O~: Q.\\--~ \0\~o\ ~1:>

c:~K.P*.~ 11/ '/cio

-~ llQlllA'IUM A110 QATI

//(!]/~

  • I -

/D 1~ 1 ~ CJ OWIOIMPllCMD llO I Dlll'QllTIOll

/I I 12- I /o

  • ao ..,

CMANGI AMDllO '-' :row TO 11 INCO#OMTID klerawuzl\MIMOCATI

'b(.,.. D "° QM/WING O.~CllfCATOa n TD llO 1ik~ 6. Br°"'" 'o/,z,./9l)

MMAMS R~,;Jkef c.h~/UI /5 ,:_~/cl/,,Pd. 7),,5 FcR ck,,v . . h.Jhd..,~

I T, ~

~(.ittY be>lf loc~:c-n5. I'}~ l"PViJI0'7 ~ d,.sn;., .. drekJN~d.$ ,, 1°j Vt'/ J/;..,.:;-I.

1 7/1.;c.,~ q ,,,:d uu""7 ?, C~t:"~I~ t""'J~J_J_ /Ai.,S '

  • .) I."'*

PALISADES ftAGE  ;.. °' )..

STEAM GENERATOR REPLACEMENT PROJECT TASK NO. IO* I FIELD CHANGE REQUEST/FIELD CHANGE NOTICE

-SUPPLEMENT- 'ell*

Cale. No. PAL-93.2

'°' -

110.

Rev* 0, Attachment D Page 2 of2

~* .I 4" I "

4 0

,~1

  • 4 I 4

\

/+- i ~I

~-*

I **

II

+--

a~.. -~-

.I I

+- -+~~-------.J 6~ , ~ITC:fl t; #IOWllJ(,

I~* 9 ~ J. I "'~~IC~

-e> a.u. 1.1~

QI= HOLE

Lo r!Ai! i OP Hgt.

sc:;.g&- ~D .. H9

( l)W(4

  • fnlOI- 'OSrJ)

~"SE. PL..A-\C. F4tl E~r~T***fitt;R* .

(~ * , X "/Jfy* -* AA.ti- ~n,

.S. .D 81!> ' - ti ~ 0 (-tf

  • I A~tlW. bT$)

f" £*d£0 ""'")

( ~ ~~';~~I~ MIO I - SJ+T i1ii)