Citation
Anchor Embedment Requirements for Signal/Sign Structures

Material Information

Title:
Anchor Embedment Requirements for Signal/Sign Structures
Creator:
HALCOVAGE, KATHLEEN M. ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Base courses ( jstor )
Cantilevers ( jstor )
Concretes ( jstor )
Diameters ( jstor )
Failure modes ( jstor )
Specimens ( jstor )
Steels ( jstor )
Strain gauges ( jstor )
Torsion ( jstor )
Transportation ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Kathleen M. Halcovage. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
7/12/2007
Resource Identifier:
659898769 ( OCLC )

Downloads

This item has the following downloads:

halcovage_k ( .pdf )

halcovage_k_Page_048.txt

halcovage_k_Page_015.txt

halcovage_k_Page_059.txt

halcovage_k_Page_095.txt

halcovage_k_Page_093.txt

halcovage_k_Page_119.txt

halcovage_k_Page_073.txt

halcovage_k_Page_062.txt

halcovage_k_Page_111.txt

halcovage_k_Page_123.txt

halcovage_k_Page_009.txt

halcovage_k_Page_116.txt

halcovage_k_Page_031.txt

halcovage_k_Page_094.txt

halcovage_k_Page_058.txt

halcovage_k_Page_101.txt

halcovage_k_Page_019.txt

halcovage_k_Page_038.txt

halcovage_k_Page_088.txt

halcovage_k_Page_063.txt

halcovage_k_Page_106.txt

halcovage_k_Page_018.txt

halcovage_k_Page_017.txt

halcovage_k_Page_032.txt

halcovage_k_Page_107.txt

halcovage_k_Page_035.txt

halcovage_k_Page_074.txt

halcovage_k_Page_003.txt

halcovage_k_Page_049.txt

halcovage_k_pdf.txt

halcovage_k_Page_092.txt

halcovage_k_Page_040.txt

halcovage_k_Page_025.txt

halcovage_k_Page_087.txt

halcovage_k_Page_103.txt

halcovage_k_Page_016.txt

halcovage_k_Page_005.txt

halcovage_k_Page_067.txt

halcovage_k_Page_096.txt

halcovage_k_Page_046.txt

halcovage_k_Page_050.txt

halcovage_k_Page_024.txt

halcovage_k_Page_121.txt

halcovage_k_Page_010.txt

halcovage_k_Page_083.txt

halcovage_k_Page_105.txt

halcovage_k_Page_002.txt

halcovage_k_Page_026.txt

halcovage_k_Page_043.txt

halcovage_k_Page_075.txt

halcovage_k_Page_110.txt

halcovage_k_Page_023.txt

halcovage_k_Page_060.txt

halcovage_k_Page_077.txt

halcovage_k_Page_091.txt

halcovage_k_Page_021.txt

halcovage_k_Page_104.txt

halcovage_k_Page_054.txt

halcovage_k_Page_085.txt

halcovage_k_Page_057.txt

halcovage_k_Page_006.txt

halcovage_k_Page_011.txt

halcovage_k_Page_068.txt

halcovage_k_Page_120.txt

halcovage_k_Page_022.txt

halcovage_k_Page_012.txt

halcovage_k_Page_020.txt

halcovage_k_Page_014.txt

halcovage_k_Page_045.txt

halcovage_k_Page_008.txt

halcovage_k_Page_030.txt

halcovage_k_Page_090.txt

halcovage_k_Page_113.txt

halcovage_k_Page_079.txt

halcovage_k_Page_007.txt

halcovage_k_Page_089.txt

halcovage_k_Page_086.txt

halcovage_k_Page_109.txt

halcovage_k_Page_055.txt

halcovage_k_Page_061.txt

halcovage_k_Page_029.txt

halcovage_k_Page_108.txt

halcovage_k_Page_112.txt

halcovage_k_Page_114.txt

halcovage_k_Page_028.txt

halcovage_k_Page_118.txt

halcovage_k_Page_065.txt

halcovage_k_Page_078.txt

halcovage_k_Page_066.txt

halcovage_k_Page_044.txt

halcovage_k_Page_070.txt

halcovage_k_Page_115.txt

halcovage_k_Page_064.txt

halcovage_k_Page_124.txt

halcovage_k_Page_099.txt

halcovage_k_Page_081.txt

halcovage_k_Page_041.txt

halcovage_k_Page_076.txt

halcovage_k_Page_036.txt

halcovage_k_Page_027.txt

halcovage_k_Page_097.txt

halcovage_k_Page_052.txt

halcovage_k_Page_100.txt

halcovage_k_Page_051.txt

halcovage_k_Page_056.txt

halcovage_k_Page_033.txt

halcovage_k_Page_072.txt

halcovage_k_Page_039.txt

halcovage_k_Page_042.txt

halcovage_k_Page_117.txt

halcovage_k_Page_071.txt

halcovage_k_Page_122.txt

halcovage_k_Page_034.txt

halcovage_k_Page_013.txt

halcovage_k_Page_004.txt

halcovage_k_Page_080.txt

halcovage_k_Page_098.txt

halcovage_k_Page_037.txt

halcovage_k_Page_082.txt

halcovage_k_Page_053.txt

halcovage_k_Page_069.txt

halcovage_k_Page_102.txt

halcovage_k_Page_047.txt

halcovage_k_Page_084.txt

halcovage_k_Page_001.txt


Full Text





ANCHOR EMBEDMENT REQUIREMENTS FOR SIGNAL/SIGN STRUCTURES


By

KATHLEEN M. HALCOVAGE













A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ENGINEERING

UNIVERSITY OF FLORIDA

2007

































2007 Kathleen M. Halcovage

































To my family, for your constant love and support throughout my life. To my parents, Barbara
and George, your dedication to providing me with the best education available has been a pivotal
part of my success. To my siblings, Barbara, George, Sarah, and Christopher, you have always
encouraged me and challenged me to be the very best that I can be. To my niece and nephew,
Grace and Aidan, you light up my life and always remind me of what is important in life.









ACKNOWLEDGMENTS

I thank my advisor, Dr. Ronald A. Cook, for his guidance throughout the course of my

research and tenure at the University of Florida, the Florida Department of Transportation

Structures Research Center Staff for their hard work in building my test apparatus and

orchestrating the testing of the specimen, and my supervisory committee for their assistance in

the preparation of this thesis. I also thank my family for their constant support and

encouragement.









TABLE OF CONTENTS

page

A CK N O W LED G M EN T S ................................................................. ........... ............. .....

LIST OF TABLES .............. ......... ...................................................7

LIST OF FIGURES .................................. .. ..... ..... ................. .8

A B S T R A C T ............ ................... ............................................................ 1 1

CHAPTER

1 INTRODUCTION ............... .......................................................... 13

2 B A CK G R O U N D .......................................... ................ ......................... .... 15

2.1 Literature Review ................................... .. ........... .. ............15
2.2 Site Investigation .................................................................... 19
2.3 A applicable C ode Provisions ........................................... ....................................... 19
2.3.1 Cracking and Threshold Torsion............................................................ .......... 20
2.3.2 N ominal Torsional Strength ...........................................................................22
2.3.3 C om bined Shear and T orsion ..................................................................... .. ....23
2.3.4 ACI Concrete Breakout Strength for Anchors ................................................. 24
2.3.5 Alternate Concrete Breakout Strength Provisions.................... ...............26
2.3.6 ACI 318-05 vs. AASHTO LRFD Bridge Design Specifications........................28

3 DEVELOPMENT OF EXPERIMENTAL PROGRAM ................................. ...............33

3.1 D description of T est A pparatu s ............................................. ......... ..............................33
3.2 Shaft D design .............. ................................. ................. ............... 34
3.2.1 Torsion D design .................... .. .............. .... ........... 34
3.2.2 Longitudinal and Transverse Reinforcement ............................... ................ 35
3 .2 .3 F lex u re ....................................................................................3 5
3.3 A nchor D design ....................................................................................................... ..... 36
3.3.1 D iam eter of A nchor B olts ...................................................................................36
3.3.2 Concrete Breakout Strength of Anchor in Shear Parallel to a Free Edge ..............37
3.3.3 Development Length of the Bolts............... .................................. ...................39
3.4 Steel Pipe A apparatus D design ................................................. ............................... 39
3.5 Concrete Block D design ..................................................... ........ .. ............ 41
3.6 C om bined Shear and T orsion ........................................ ............................................42
3 .7 O v erv iew .......................................................................... 4 2









4 IMPLEMENTATION OF TESTING PROGRAM.....................................................51

4.1 M materials ...................................................... 51
4.1.1 Concrete Strength ..................................... ....... .............. ... .......... 51
4.1.2 Bolt Strength....................................... ..... .......... 51
4.1.3 Carbon Fiber Reinforced Polymer Wrap....................................................51
4.2 Instrum entation ..................... .......... ......... ............ ............ 53
4.2.1 Linear Variable Displacement Transducers ................................. ............... 53
4 .2 .2 Strain G ages.................................................................................... ........ .. ... 54

5 T E ST R E SU L T S ................................................................60

5 .1 In itia l T e st .................................................................................................................. 6 0
5.1.1 Behavior of Specimen During Testing ...................................... ...............60
5.1.2 Behavior of Strain Gages During Testing ..........................................................61
5.1.3 Sum m ary of Initial Test Results ................................................................. 62
5.2 CFR P R etrofit Test .................. .................. .. .... .. ................ ............. ........ 62
5.2.1 Behavior of Specimen with CFRP Wrap During Testing ....................................63
5.2.3 Behavior of Strain Gages During Testing ................................... .................64
5.2.4 Sum m ary of Test R esults............ ......................................... ............... 64

6 SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS .............................75

APPENDIX

A TEST APPARATUS DRAWINGS......................................................... ...............77

B DESIGN CALCULATION S ....................................................... ......... ......82

C IN IT IA L T E ST D A T A ................................................................................... ............ 108

D R E TR O FIT TE ST D A TA ................................................................................. ........... 115

L IST O F R E F E R E N C E S ..................................................................................... ..................122

B IO G R A PH IC A L SK E T C H ......................................................................... .. ...................... 124
















6









LIST OF TABLES

Table page

3 -1 F field dim en sion s .................................................................................................... 50

3-2 Sum m ary of design calculations ......................................................... 50









LIST OF FIGURES


Figure page

1-1 Failed cantilever sign structure ........................................................................... ...... 14

2-1 Cantilever sign structure at Exit 79 on Interstate 4 in Orlando.......................................29

2-2 N ew foundation installed at the site ...................................................................... 29

2-3 Failed foundation during post-failure excavation...........................................................30

2-4 Concrete breakout of an anchor caused by shear directed parallel to the edge for a
circular foundation ............... ................. .............. ................ ......... 30

2-5 Concrete breakout failure for an anchor loaded in shear............................................31

2-6 Determination of Avco based on the 35 failure cone ...................................................31

2-7 Shear load oriented (a) perpendicular to the edge and (b) parallel to the edge .................32

3-1 Schem atic of test apparatus........................................................................................... 44

3-2 Front elevation of test apparatus .......................................................................... ....... 44

3-3 P lan view of test apparatus s .................................................................. ...... .................. 45

3-4 Side elevation of test apparatus....................................................................... 45

3-5 Lever arm for the calculation of bolt flexure.................................................................. 46

3-6 Adjusted cover based on a single anchor and 350 failure cone.......................................46

3-7 Development of the projected failure area for the group of anchors around a circular
fou n d action .................................................................................4 7

3-8 Two anchor arrangement displays the minimum spacing such that no overlap of the
failure cones occurs............ ..................................... .... 47

3-9 O overlap of failure cones.............................................................................. ............ 47

3-10 The contribution of the "legs" of the failure cone to AV, along a straight edge
decreases as the number of bolts increases.................................................... ............. 48

3-11 Overlap of failure cones for a circular foundation........... ................................ 48

3-12 A ssem bled test specim en ............................................................ ............... 49

3-13 Shaft with pipe apparatus attached prior to instrumentation being attached ...................49









4-1 Method for the determination of the tension, TCFRP, that must be resisted by the
C FR P w rap ................ .... ........ ............... .............................55

4-2 Instrum entation layout on the base plate ........................................ ....................... 56

4-3 Instrum entation layout on face of shaft ........................................ ......................... 56

4-4 Instrumentation layout on rear of shaft/face of concrete block ......................................57

4-5 Instrumentation layout of pipe at load location ...... ........ ..........................................57

4-6 Location of LVDTs D1V, D4, and D7 on the test specimen .........................................58

4-7 Strain gage layout on base plate............................................... .............................. 58

4-8 Strain gage on base plate of test specim en............................................ .. ................59

5-1 Initial cracks on face of shaft ....................................................................... 66

5-2 Initial cracks on face and side of shaft (alternate view of Figure 5-1) ............................66

5-3 Face of test specimen after testing exhibits cracks between the bolts along with the
characteristic concrete breakout cracks ........................................ ......................... 67

5-4 Crack pattern on face of shaft after testing depicts characteristic concrete breakout
failure cracks ................ ..................................... ............................67

5-5 Applied Torsion vs. Plate Rotation Plot- Initial Test............................................. 68

5-6 Applied Torsion vs. Bolt Strain Plots for each bolt at the appropriate location on the
base plate with Applied Torsion vs. Plate Rotation plot in center (full size plots in
A pp en dix C ) .............................................................................. 69

5-7 Bolt Strain Comparison Plot for Initial Test exhibits the redistribution of the load
coinciding w ith crack form ations............................................................ .....................70

5-8 Shaft with the CFRP wrap applied prior to testing ............................... ...................70

5-9 Applied Torsion vs. Plate Rotation Plot- Retrofit Test................... ........................... 71

5-10 Shaft exhibiting characteristic torsion cracks from face to base of shaft ........................71

5-11 Applied Torsion vs. Bolt Strain plots for the Retrofit Test at the appropriate bolt
location around the base plate with Applied Torsion vs. Plate Rotation plot in center
(full size plots in A appendix D )................................................. .............................. 72

5-12 Bolt bearing on the bottom of the base plate during loading ....................................73









5-13 Bolt Strain Comparison plot for the retrofit test exhibits slope changes at milestone
lo a d s ..............................................................................................7 3

5-14 Face of shaft after test illustrates yielding of bolts, concrete breakout cracks around
the perimeter, and torsion cracks in the center. ..................................... ............... 74

5-15 Torsion cracks along length of the shaft after the test .....................................................74

A-i Dimensioned front elevation drawing of test apparatus ............... ...... ......... 77

A-2 Dim ensioned plan drawing test apparatus ........................................ ...... ............... 78

A-3 Dimensioned side elevation drawing of test apparatus.................................................79

A-4 Dim ensioned pipe apparatus drawing ..................................................... .... ........... 80

A-5 Dim ensioned channel tie-down drawing ........................................ ....................... 81

C-1 Applied Torsion vs. Rotation Plot .............................................................................108

C-2 Bolt Strain Comparison Plot ........................................................................ 108

C-3 Applied Torsion vs. Strain Plots for each bolt location............................109

D A applied Torsion vs. R rotation Plot .................................................................................115

D -2 B olt Strain C om prison Plot ............................................................................ ............ 115

D-3 Applied Torsion vs. Strain Plots for each bolt location............................116









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Engineering

ANCHOR EMBEDMENT REQUIREMENTS FOR
SIGNAL/SIGN STRUCTURES

By

Kathleen M. Halcovage

May 2007

Chair: Ronald A. Cook
Major: Civil Engineering

During the 2004 hurricane season, several anchor embedment failures of the foundations of

cantilever sign structures occurred. The purpose of this research program was to determine the

cause of the failure of those foundations. After a thorough literature review, in conjunction with

site investigation, and testing, it was determined that the failure originated from the shear load on

the anchors directed parallel to the edge of the foundation. The shear load resulted from the

torsion loading on the anchor group that occurred during the hurricane. Investigation of this

failure mode, based on the ACI 318-05 Appendix D provisions for concrete breakout of anchors,

indicated that this is a failure mode not considered in the current design procedures for these

types of foundations. Furthermore, it was determined that it very well describes the type of

failure noted in the field investigation.

A test specimen was designed to preclude other possible failure modes not exhibited in the

field (e.g. steel failure of the anchors, bending of the anchors, and torsional failure of the

foundation). The results of the testing indicated the failure of the foundations was caused by

concrete breakout due to shear on the anchors directed parallel to the free edge of the foundation.

The test specimen failed at the torsion predicted by the ACI 318-05 Appendix D provisions

based on the expected mean strength of the anchors for concrete breakout with shear directed









parallel to the free edge. Additionally, the cracks that formed were the same type as those noted

in the field investigation, and matched the expected pattern for concrete breakout failure.

After failure, additional testing was performed to determine a viable repair/retrofit option.

The repair/retrofit option used a carbon fiber reinforced polymer (CFRP) wrap around the top of

the foundation. The results of this testing indicated that this repair/retrofit technique strengthens

the foundation such that it not only meets its initial capacity for concrete breakout, but, also, can

exceed this capacity. The results of this test led to the development of guidelines for the

evaluation and repair/retrofit of existing foundations.









CHAPTER 1
INTRODUCTION

During the 2004 hurricane season, the failure of foundations of cantilever sign structures

occurred along Florida highways (Figure 1-1). These failures necessitated a review of the

current design and construction procedures for the foundations of cantilever sign structures.

The main objective of this research program was two-fold: to determine the cause of the

failure of the cantilever sign structures; and, to propose a retrofit option for the foundation. In

order to fulfill this objective, a thorough literature review, site investigation of a failed

foundation, and experimental program were conducted. The findings of the literature review and

site investigation were used to develop the experimental program. The findings of the

experimental program were applied in the development of the retrofit guidelines.

Furthermore, this project tested whether or not the ACI 318-05, ACI (2005), Appendix D

provisions for anchorage to concrete are applicable for circular foundations.

































Figure 1-1. Failed cantilever sign structure









CHAPTER 2
BACKGROUND

While there have not been published reports detailing failures of sign structure

foundations, such as those being investigated in this study, information on the behavior of anchor

installations under various load conditions was found. The main subjects of much of the

literature were the effects of fatigue and wind load on overhead sign structures. Additionally,

there have been studies conducted on the failure modes of anchor installations, but these findings

were not based on circular foundations. In later sections, one of these anchorage failure modes

will be introduced for application in this research program.

This chapter presents the findings of the literature review, the conclusions drawn based on

a site investigation of a failed foundation, and applicable design equations for the determination

of the failure mode. The information presented in the chapter served as the base upon which the

experimental program was developed.

2.1 Literature Review

Keshavarzian (2003) explores the wind design requirements and safety factors for utility

poles and antenna monopoles from various specification manuals. It was found that the

procedure outlined in ASCE (1991), ASCE 74- Guidelines for Electrical Transmission Line

Structural Loading, resulted in the smallest factor of safety for the design. AASHTO (2001),

Standard Specificationsfor Structural Supportsfor Highway Signs, Luminaires and Traffic

Signals, was used as a part of the comparison for the design of the antenna monopole. The

design from this specification was compared to that from ASCE (2000), ASCE 7-98-Minimum

Design Loadsfor Buildings and Other Structures; TIA/EIA (1996), Structural Standards for

Steel Antenna Towers andAntenna Supporting Structures; and, ASCE 74. The wind forces at

the base were the same for ASCE 74, AASHTO, and ASCE 7-98. The forces using TIA/EIA









were higher because it requires that a 1.69 gust response factor be applied to the design.

Therefore, the pole designed using TIA/EIA would have between 30 and 40 percent extra

capacity. ASCE 7-98 and AASHTO resulted in the same margin of safety. The paper did not

include findings that were completely relative to this project, but it provided additional sources

for design of structures for comparative purposes.

Keshavarzian and Priebe (2002) compares the design standards specified in ASCE (2000),

ASCE 7-98, and IEEE (1997), NESC- NationalElectrical Safety Code. The NESC does not

require that utility poles measuring less than 60 feet in height be designed for extreme wind

conditions. Short utility poles were designed to satisfy NESC specifications (i.e. without

extreme wind conditions). The poles were then evaluated according to the ASCE 7-98 wind load

requirements. It was found that the poles did not meet the ASCE 7-98 requirements. Therefore,

it was recommended that the exclusion for short utility poles in the NESC be reevaluated. The

paper also mentioned AASHTO (1994), Standard Specifications for Structural Supports for

Highway Signs, Luminaires and Traffic Signals. It outlined that in the AASHTO specification,

support structures exceeding 50 feet and overhead sign structures must be designed for a 50-year

mean recurrence interval, or extreme wind loading condition.

MacGregor and Ghoneim (1995) presents the background information for the formulation

of the thin-walled tube space truss analogy design method for torsion that was first adopted into

ACI (1995), ACI 318-95. The design methodology was adopted because it was simpler to use

than the previous method and was equally accurate. The basis for the derivation of the new

method was based on tests that were conducted in Switzerland. Both solid and hollow beams

were tested during that research. In comparing the data from both tests, it was discovered that

after cracking the concrete in the center had little effect on the torsional strength of the beam.









Therefore, the center of the cross-section could be ignored, and the beam could be idealized as a

hollow tube.

A space truss was formed by longitudinal bars in the corners, the vertical closed stirrups,

and compression diagonals. The compression diagonals were spiraled around the member and

extended between the torsion cracks. The paper also explained the shear stresses created by

torsion on the member.

In addition to the derivation of the equations for torsion and shear, the authors discussed

the limits for when torsion should be considered and the requirements for minimal torsional

reinforcement. The tests, conducted on both reinforced and prestressed concrete beams, showed

that there was acceptable agreement between the predicted strengths, as determined by the

derived equations, and the test results. This agreement was comparable to the design equations

from the ACI Code.

In addition to these papers, other reports reviewed include Lee and Breen (1966), Jirsa et

al. (1984), Hasselwander et al. (1977), and Breen (1964). These four studies focused on

important information regarding anchor bolt installations. Other reports that were examined for

relevance were from the National Cooperative Highway Research Program (NCHRP). These

are: Fouad et al. (1998), NCHRP Report 411; Kaczinski et al. (1998), NCHRP Report 412; and,

Fouad et al. (2003), NCHRP Report 494.

Fouad et al. (2003) details the findings of NCHRP Project 17-10(2). The authors stated

that AASHTO (2001) does not detail design requirements for anchorage to concrete. The ACI

anchor bolt design procedure was also reviewed. Based on their findings, they developed a

simplified design procedure. This procedure was based on the assumptions that the anchor bolts

are hooked or headed, both longitudinal steel and hoop steel are present in the foundation, the









anchor bolts are cast inside of the reinforcement, the reinforcement is uncoated, and, in the case

of hooked bolts, the length of the hook is at least 4.5 times the anchor bolt diameter. If these

assumptions did not apply, then the simplified procedure was invalid. The anchor bolt diameter

was determined based on the tensile force on the bolt, and the required length was based on fully

developing the longitudinal reinforcement between the embedded head of the anchor. The

authors further stated that shear loads were assumed to be negligible, and concrete breakout and

concrete side face blowout were controlled by adequate longitudinal and hoop steel. The design

procedure was developed based on tensile loading, and did not address the shear load on the

anchors directed parallel to the edge resulting from torsion.

Additionally, the authors presented the frequency of use of different foundation types by

the state Departments of Transportation, expressed in percentages of states reporting use.

According to the survey the most common foundation type used for overhead cantilever

structures was reinforced cast-in-place drilled shafts (67-100%) followed by spread footings (34-

66%) and steel screw-in foundations (1-33%). None of the states reported the use of directly

embedded poles or unreinforced cast-in-place drilled shafts.

ASCE (2006), ASCE/SEI 48-05, entitled Design of Steel Transmission Pole Structures was

obtained to gather information on the foundation design for transmission poles structures. The

intent was to determine whether or not the design of such foundations was relevant to the

evaluation of the foundations under examination in this research. In 9.0 of the standard, the

provisions for the structural members and connections used in foundations was presented. Early

in the section, the standard stated that the information in the section was not meant to be a

foundation design guide. The proper design of the foundation must be ensured by the owner

based on geotechnical principles. The section commented on the design of the anchor bolts. The









standard focused on the structural stability of the bolts in the foundation; it looked at bolts in

tension, bolts in shear, bolts in combined tension and shear, and the development length of such

bolts. The standard did not present provisions for failure of the concrete.

2.2 Site Investigation

A site investigation was conducted at the site of one failed overhead cantilever signal/sign

structure located at Exit 79 on Interstate 4 in Orlando (Figure 2-1). Figure 2-2 is the newly

installed foundation at the site. The failed foundation had the same anchor and spacing

specifications as the new foundation. This site visit coincided with the excavation of the failed

anchor embedment. During the course of the excavation the following information was

collected:

* The anchor bolts themselves did not fail. Rather, they were leaning in the foundation,
which was indicative of a torsional load on the foundation. While the integrity of the
anchor bolts held up during the wind loading, the concrete between the bolts and the
surface of the foundation was cracked extensively (Figure 2-3). The concrete was
gravelized between the anchors and the hoop steel. It should be noted that upon the
removal and study of one anchor bolt, it was evident that there was no deformation of the
bolt itself.

* The hoop steel did not start at the top of the foundation. It started approximately 15 in.
(381 mm) into the foundation.

* The concrete was not evenly dispersed around the foundation. The hoop steel was exposed
at approximately three to four feet below grade. On the opposite side of the foundation
there was excess concrete. It was assumed that during the construction of the foundation,
there was soil failure allowing a portion of the side wall to displace the concrete.

2.3 Applicable Code Provisions

The initial failure mode that was focused on in the background review was torsion.

However, based on the results of the site investigation, it was determined that the most likely

cause of failure was concrete breakout of an anchor (Figure 2-4). The equations for torsion are

presented in this section as they were used during the design of the experimental program to

prove that the concrete breakout failure will occur before the torsional failure.









2.3.1 Cracking and Threshold Torsion

Torsion is the force resulting from an applied torque. In a circular section, such as the

foundation under review, the resulting torsion is oriented perpendicular to the radius or tangent

to the edge. ACI (2005), ACI 318-05, details the equation for the cracking torsion of a

nonprestressed member. In R11.6.1, the equation for the cracking torsion, Tcr, is given

(Equation 2-1). The equation was developed by assuming that the concrete will crack at a stress

of 4\/f'c.

(A 2
T 4 = 4 (2-1)

Where
Tcr = cracking torsion (lb.-in.)
f'c = specified compressive strength of the concrete (psi)
Acp = area enclosed by the outside perimeter of the concrete cross-section (in.2)
= 7r2, for a circular section with radius r (in.)
epp = outside perimeter of the concrete cross-section (in.)
= 27r, for a circular section with radius r (in.)

This equation, when applied to a circular section, results in an equivalent value when

compared to the basic equation (Equation 2-2) for torsion noted in Roark and Young (1975).

The equality is a result of taking the shear stress as 4ff'c.


T = (2-2)
2
Where
T = torsional moment (lb.-in.)
r = shear stress, 4/f'c, (psi)
r = radius of concrete cross-section (in.)

ACI 318-05 11.6.1 (a) provides the threshold torsion for a nonprestressed member

(Equation 2-3). This is taken as one-quarter of the cracking torsion. If the factored ultimate

torsional moment, T, exceeds this threshold torsion, then the effect of torsion on the member

must be considered in the design.










T = AC (2-3)

Where
0 = strength reduction factor

AASHTO (2004), AASHTO LRFD Bridge Design Specifications, also presents equations

for cracking torsion (Equation 2-4) and threshold torsion (Equation 2-5). Equation 2-4

corresponds with the AASHTO (2004) equation for cracking torsion with the exception of the

components of the equation related to prestressing. That portion of the equation was omitted

since the foundation was not prestressed. It must be noted that these equations are the same as

the ACI 318-05 equations.

A2
TC= 0.125 -cp (2-4)
Pc
Where
Tcr = torsional cracking moment (kip-in.)
Acp = total area enclosed by outside perimeter of the concrete cross-section (in.2)
pc = the length of the outside perimeter of the concrete section (in.)

AASHTO (2004) also specifies the same provision as ACI 318-05 regarding the threshold

torsion. In 5.8.2.1, it characterizes the threshold torsion as one-quarter of the cracking torsion

multiplied by the reduction factor. Equation 2-5 corresponds with the threshold torsion portion

of AASHTO (2004) equation.

T = 0.250T1r (2-5)

The above referenced equations considered the properties and dimensions of the concrete.

They did not take into consideration the added strength provided by the presence of

reinforcement in the member. For the purposes of this research, it was important to consider the

impact of the reinforcement on the strength of the concrete shaft.









2.3.2 Nominal Torsional Strength

ACI 318-05 11.6.3.5 states that if the ultimate factored design torsion exceeds the

threshold torsion, then the design of the section must be based on the nominal torsional strength.

The nominal torsional strength (Equation 2-6) takes into account the contribution of the

reinforcement in the shaft.


T, = 2AAf cotO (2-6)
s
Where
T, = nominal torsional moment strength (in.-lb.)
Ao = gross area enclosed by shear flow path (in.2)
A, = area of one leg of a closed stirrup resisting torsion with spacing s (in.2)
fyt = specified yield strengthfy of transverse reinforcement (psi)
s = center-to-center spacing of transverse reinforcement (in.)
0 = angle between axis of strut, compression diagonal, or compression field and
the tension chord of the member

The angle, 0, is taken as 45, if the member under consideration is nonprestressed. This

equation, rather than taking into account the properties of the concrete, takes into account the

properties of the reinforcement in the member. These inputs include the area enclosed by the

reinforcement, the area of the reinforcement, the yield strength of the reinforcement, and the

spacing of the reinforcement. For the purpose of this research, the reinforcement under

consideration was the hoop steel.

AASHTO (2004) also outlines provisions for the nominal torsional resistance in 5.8.3.6.2.

Equation 2-7 is the same equation that ACI 318-05 presents. The only difference is in the

presentation of the equations. The variables are represented by different notation.

2AoAfy cot0
Tn =- (2-7)
s
Where
T, = nominal torsional moment (kip-in.)
Ao = area enclosed by the shear flow path, including any area of holes therein (in.2)
At = area of one leg of closed transverse torsion reinforcement (in.2)
0 = angle of crack










As the above referenced equation evidences, the ACI 318-05 and the AASHTO (2004)

provisions for nominal torsional strength are the same. Based on the code provisions, the

nominal torsional strength represents the torsional strength of the cross-section.

2.3.3 Combined Shear and Torsion

Another area that had to be considered in this research was the effect of combined shear

and torsion. Both ACI 318-05 and AASHTO (2004) outline equations for the combined shear

and torsion. Since the foundation had a shear load applied to it, it had to be determined whether

or not the shear load was large enough to necessitate consideration. The ACI 318-05 equation

(Equation 2-8) and the AASHTO (2004) equation (Equation 2-9) are presented hereafter. The

ACI 318-05 equation is located in 11.6.3.1 of ACI 318-05, and the AASHTO (2004) equation is

presented in 5.8.3.6.2 of that specification. The ACI 318-05 equation is presented with V,

substituted on the left-hand side.


V < Vu + 1p7Ah (2-8)
b" d 1 .7A2 h
Where
V, = factored shear force at section (lb.)
bwd = area of section resisting shear, taken as Aoh (in.2)
T, = factored torsional moment at section (in.-lb.)
ph = perimeter of centerline of outermost closed transverse torsional reinforcement (in.)
Aoh = area enclosed by centerline of the outermost closed transverse torsional
reinforcement (in.2)

The AASHTO (2004) equation that is presented (Equation 2-9) is intended for the

calculation of the factored shear force. For the purpose of this project, the right-hand side of the

equation was considered.










V V = V (.92Ao (2-9)

Where
V, = factored shear force (kip)
ph = perimeter of the centerline of the closed transverse reinforcement (in.)
T, = factored torsional moment (kip-in.)

The determination of whether or not shear had to be considered was made based on a

comparison of the magnitudes of the coefficients of these terms. This is investigated further in

Chapter 3.

2.3.4 ACI Concrete Breakout Strength for Anchors

In ACI 318-05 Appendix D, the concrete breakout strength is defined as, "the strength

corresponding to a volume of concrete surrounding the anchor or group of anchors separating

from the member." A concrete breakout failure can result from either an applied tension or an

applied shear. In this report, the concrete breakout strength of an anchor in shear, D.6.2, will be

studied. The breakout strength for one anchor loaded by a shear force directed perpendicular to a

free edge (Figure 2-5) is given in Equation 2-10.


Vb = 7 f(c)5 (2-10)

Where
Vb = basic concrete breakout strength in shear of a single anchor in cracked concrete
(lb.)
Se = load bearing length of anchor for shear (in.)
do = outside diameter of anchor (in.)
ca, = distance from the center of an anchor shaft to the edge of concrete in one
direction; taken in the direction of the applied shear (in.)

The term te is limited to 8do according to D.6.2.2. The equations in ACI 318-05 were

developed based on a 5% fractile and with the strength in uncracked concrete equal to 1.4 times

the strength in cracked concrete. The mean concrete breakout strength in uncracked concrete is

provided in Fuchs et al. (1995) and given in Equation 2-11.










Vb =13 d, J(c1)15 (2-11)


For a group of anchors, Equation 2-12 applies. This equation is the nominal concrete

breakout strength for a group of anchors loaded perpendicular to the edge in shear.

A,
Vcbg V= ,Vec,v Ved,V c Vb (2-12)

Where
Vcbg = nominal concrete breakout strength in shear of a group of anchors (lb.)
Av = projected concrete failure area of a single anchor or group of anchors, for
calculation of strength in shear (in.2)
Avco = projected concrete failure area of a single anchor, for calculation of strength in
shear, if not limited by corner influences, spacing, or member thickness (in.2)
= 4.5(ca1)2, based on a 350 failure cone (Figure 2-6)
fe, v = factor used to modify shear strength of anchors based on eccentricity of applied
loads, ACI 318-05 D.6.2.5
qed, v = factor used to modify shear strength of anchors based on proximity to edges of
concrete member, ACI 318-05 D.6.2.6
qci,v = factor used to modify shear strength of anchors based on presence or absence of
cracks in concrete and presence or absence of supplementary reinforcement,
ACI 318-05 D.6.2.7, accounted for in Equation 2-11

The resultant breakout strength is for a shear load directed perpendicular to the edge of the

concrete. Therefore, an adjustment had to be made to account for the shear load acting parallel

to the edge since this was the type of loading that resulted from the torsion on the anchor group.

In D.5.2. l(c) a multiplication factor of two is prescribed to convert the value to a shear directed

parallel to the edge (Figure 2-7). Fuchs et al. (1995) notes that the multiplier is based on tests,

which indicated that the shear load that can be resisted when applied parallel to the edge is

approximately two times a shear load applied perpendicular to the edge.

In order to convert the breakout strength to a torsion, the dimensions of the test specimen

were considered to calculate what was called the nominal torsional moment based on the

concrete breakout strength, Tn,breakout.









2.3.5 Alternate Concrete Breakout Strength Provisions

In the book Anchorage in Concrete Construction, Eligehausen et al. (2006), the authors

presented a series of equations for the determination of the concrete strength based on a concrete

edge failure. These equations are presented in Chapter 4, 4.1.2.4 of the text. Equation 2-13 is

the average concrete breakout strength of a single anchor loaded in shear. It must be noted that

this equation is for uncracked concrete.

Vc =3.0 -do fc c0 (2-13)
0 e c200 al
Where
oV,c = concrete failure load of a near-edge shear loaded anchor (N)
do = outside diameter of anchor (mm)
te = effective load transfer length (mm)
fcc200oo = specified concrete compressive strength based on cube tests (N/mm2)
S1.i18f'c
ca, = edge distance, measured from the longitudinal axis of the anchor (mm)

a =0.1- e
\a

S = 0.1. do
C al

As was the case for the ACI 318-05 equations, the term te is limited to 8do. Equation 2-14

accounts for the group effect of the anchors loaded concentrically. The authors stated that cases

where more than two anchors are present have not been extensively studied. They did, however,

state that the equation should be applicable as long as there is no slip between the anchor and the

base plate.


AV A V~ (2-14)

Where
Av = projected area of failure surface for the anchorage as defined by the overlap
of individual idealized failure surfaces of adjacent anchors (mm2)
Avco = projected area of the fully developed failure surface for a single anchor
idealized as a half-pyramid with height ca1 and base lengths 1.5ca1 and 3ca1 (mm2)









ACI 318-05 specifies that, in order to convert the failure shear directed perpendicular to

the edge to the shear directed parallel to the edge, a multiplier of two be applied to the resultant

load. The provisions outlined in this text take a more in-depth approach to determining this

multiplier. The method for calculating this multiplier is detailed in 4.1.2.5 of Eligehausen et al.

(2006). The authors stated that, based on previous research, the concrete edge breakout capacity

for loading parallel to an edge is approximately two times the capacity for loading perpendicular

to the edge if the edge distance is constant. The authors further moved to outline equations to

calculate the multiplier based on the angle of loading. The first equation (Equation 2-15) that is

presented in the text is a generalized approach for calculating the multiplier when the angle of

loading is between 550 and 900 of the axis perpendicular to the edge. For loading parallel to the

edge the angle is classified as 900 (Figure 2-7).

1
=aV = .(2-15)
cosa + 0.5sina
Where
Wa,v = factor to account for the angle between the shear load applied and the
direction perpendicular to the free edge of the concrete member
a = angle of the shear load with respect to the perpendicular load

This equation results in a factor of two for loading parallel to the edge. Equation 2-16

provides the concrete breakout strength for shear directed parallel to the edge using qfa,v.

VUyC = Va'V *.Vu,c (2-16)
Where
Vuc,a= concrete failure load for shear directed parallel to an edge based on qa,v (N)

An alternate equation for calculating this factor is also presented in the Eligehausen et al.

(2006) text. This equation is only valid for loading parallel to the edge. This equation is based

on research proposing that the multiplier to calculate the concrete breakout capacity for loading

parallel to the edge based on the capacity for loading perpendicular to the edge is not constant.









Rather, it suggested that it is based on the concrete pressure generated by the anchor. The base

equation for the application of this factor is Equation 2-17.

Suc, parallel = parallel V ,c (2-17)
Where
V ,= concrete failure load in the case of shear parallel to the edge (N)
parallel = factor to account for shear parallel to the edge
Vu,c = concrete failure load in the case of shear perpendicular to the edge (N)

Equation 2-18 is used for the determination of the conversion factor parallel.


=' parallel = 4- k n' cc (2-18)

Where
k4 = 1.0 for fastenings without hole clearance
0.75 for fastenings with hole clearance
n = number of anchors loaded in shear
fcc = specified compressive strength of the concrete (N/mm2)
conversion tof'c as specified for Equation 2-13

The results of Equation 2-13 through Equation 2-18 are presented alongside the ACI 318-

05 equation results in Chapter 3. These are presented for comparative purposes only.

2.3.6 ACI 318-05 vs. AASHTO LRFD Bridge Design Specifications

In Sections 2.3.1 through 2.3.3, both the applicable design equations in ACI 318-05 and

AASHTO (2004) were presented. As was shown, the ACI and AASHTO equations were the

same. Additionally, the provisions for the concrete breakout failure capacity are only provided

in ACI 318-05. AASHTO does not provide design guidelines for this failure. Therefore, the

ACI 318-05 equations were used throughout the course of this research program.






























Figure 2-1. Cantilever sign structure at Exit 79 on Interstate 4 in Orlando


Figure 2-2. New foundation installed at the site






























Figure 2-3. Failed foundation during post-failure excavation


Figure 2-4. Concrete breakout of an anchor caused by shear directed parallel to the edge for a
circular foundation




















Vb












.. .. . . ... .
: : :. : : : : : : :


. . . . . . . . .
. '. . . . . . . . .





...............................
. . . . a






. . . . . . .


Figure 2-5. Concrete breakout failure for an anchor loaded in shear


. -.. . ..
. . . .
.'." ".'.'.'.'.'.'."
.'.'.'.'.'.'.'.'.'.'."
.'.'.'.'.'.'.'.'.'.'."
.'.'.'.'.'.'.'.'.'.'."
.'.'-.'.'.'.'.'.'.'-".
.'.'.'.'.'.'.'.'.'.'."




. .... . . .. .
. ... . . .. .
. .... . . .. .

. . .... -.. . ..
. . .... ... . ..


1.5cal


A vco=. 5cal 2(1.5cal)

=4.5(al)2


Figure 2-6. Determination ofAvco based on the 35 failure cone


..............................
..............................
..............................
..............................
..............................
..............................
..............................
..............................
..............................
..............................
..............................
..............................
..............................
. . . . . .
...........
..............................
.............. ...............
..............................
.............. ...............
............... ..............
.............. ...............
..............................
.............. ...............
. . . . . . . .
..............
0 . . ......

.............








































::::::::::::::::::::::: ... ...Perpendicular

. . .. . .. .. . . A.i












. .4. ..:. .






.I ...


Figure 2-7. Shear load oriented (a) perpendicular to the edge and (b) parallel to the edge


................................
................................
................................
................................
................................
................................
................................
................................
................................
................................
................................
................................
................................


.............
. . . . . . .

.............
...............
...............
...............
................
.............
.. ...............
...............




...............................
. . . . . . . . .
...............................
. . . . . .
..........
. . . . . .
. . . . . . . .
. . . . . . . .


. . . . . . . .
................................
................................
................................
................................
................................
................................
................................
................................









CHAPTER 3
DEVELOPMENT OF EXPERIMENTAL PROGRAM

After a thorough background investigation, it was determined that the most likely cause of

the failure was the concrete breakout of an anchor loaded by a shear force directed parallel to a

free edge. The shear force on the individual anchors was caused by torsion applied to the bolt

group from the sign post. Based on this determination, an experimental program was formulated

to determine if this was in fact the failure mode of the foundation. Therefore, it was of the

utmost importance to design the test apparatus to preclude other failure modes. This chapter

focuses on the development of the experimental program.

3.1 Description of Test Apparatus

The test apparatus was designed such that the field conditions could be closely modeled for

testing at the Florida Department of Transportation (FDOT) Structures Research Center. A

schematic of the test apparatus is shown in Figure 3-1. The test apparatus consisted of:

* A 30" (762 mm) diameter concrete shaft that extended 3'-0" (914 mm) outward from the
concrete block

* Twelve 37" (940 mm), 1.5" (38.1 mm) diameter F1554 Grade 105 anchor bolts embedded
into the concrete around a 20" (508 mm) diameter

* A 16" (406 mm) diameter steel pipe assembly welded to a 24" (610 mm) diameter, 1"
(25.4 mm) thick steel base plate with holes drilled for the anchor bolts to provide the
connection between the bolts and pipe assembly

* A 6'-0" x 10'-0" x 2'-6" (1830 mm x 3050 mm x 762 mm) reinforced concrete block to
provide a fixed support at the base of the shaft

* Two assemblies of C 12x30 steel channels and plates to attach the block to the floor

The base for the design of the various components of the test apparatus was one half of the

size of the failed foundation investigated during the site visit. The dimensions of the field

foundation are presented in Table 3-1. From that point, the elements of the test apparatus were

designed to preclude all failure modes other than the concrete breakout failure of the anchors.









Information pertaining to the design of the components of the apparatus is presented in the

following sections. Figures 3-2 through 3-4 are drawings of the test apparatus. For large scale

dimensioned drawings, reference Appendix A. Complete design calculations are located in

Appendix B.

3.2 Shaft Design

The starting point for the design of the concrete shaft was based on developing a test

specimen approximately one half of the size of the foundation that was investigated during the

site visit. From there, the various components of the shaft were designed the meet the ACI 318-

05 requirements, and to prevent failure before the concrete breakout strength was reached and

exceeded. All of the strengths were calculated using a concrete strength of 5500 psi (37.9 MPa),

which was the strength indicated on the FDOT standard drawings.

3.2.1 Torsion Design

The basic threshold torsional strength of the shaft, 24.6 kip-ft (33.4 kN-m) was calculated

using the ACI 318-05 torsional strength equation (Equation 2-3). This strength, however, did not

take into account the reinforcement in the shaft. Therefore, it was assumed that the threshold

torsion would be exceeded. As a result, the torsional strength of the shaft was based on the

nominal torsional strength.

In order to calculate the torsional strength that the shaft would exhibit during testing, the

ACI nominal torsional strength equation was applied. Before the strength was calculated, the

minimum requirements for the shaft reinforcement were followed as outlined in ACI 318-05

7.10.5.6 and 11.6.5.1. The nominal torsional strength (Equation 2-6) was then calculated for

the specimen. This value, 253 kip-ft (343 kN-m), was compared to the concrete breakout

strength. The spacing of the hoop steel in the shaft was altered until the nominal torsional

strength exceeded the concrete breakout strength. Hence, if the concrete breakout failure was the









correct failure mode, it would occur before the torsional capacity of the shaft was exceeded

during testing.

3.2.2 Longitudinal and Transverse Reinforcement

As was outlined in the previous section, the required amount of hoop steel to meet the ACI

318-05 specifications was determined using guidelines from Chapters 7 and 11 in the code. The

resultant hoop steel layout was twenty-four #4 bars spaced evenly around a 27 in. (686 mm)

diameter circle. The transverse hoops were comprised of #3 bars at 2.5 in. (635 mm) totaling

fourteen #3 bar hoops. The required splice for the #3 bar was 12 in. (305 mm), and the

development length required for the #4 bar into the concrete block was 8 in. (203 mm) with a 6

in. (152 mm) hook. In the test setup, the #4 bars extended 27 in. (686 mm) into the block, which

exceeded the required length. This length was used for simplicity in design and construction of

the test setup. The #4 bars were tied into one of the cages of reinforcement in the concrete block.

3.2.3 Flexure

Due to the eccentric loading of the bolts, the flexural capacity of the shaft had to be

calculated. It had to be determined that the shaft would not fail in flexure under the load applied

during testing. The flexural reinforcement in the shaft was the longitudinal reinforcement, the #4

bars. The first step to determine the capacity was to assume the number of bars that would have

yielded at the time of failure. From that point, the neutral axis of the shaft was located following

the ACI 318-05 concrete stress block methodology presented in Chapter 10 of the code. It was

then checked if the number of bars that had yielded was a good assumption. Once this was

verified, the nominal moment capacity of the shaft was calculated, and, then, compared to the

maximum flexural moment based on the concrete breakout capacity. The flexural capacity of the

shaft, 262 kip-ft (355 kN-m), exceeded the maximum flexural moment on the shaft, 60.6 kip-ft

(95.2 kN-m).









3.3 Anchor Design


3.3.1 Diameter of Anchor Bolts

The starting point for the diameter of the F1554 Grade 105 anchor bolts to be used in the

test apparatus was based on half the diameter of those in the field specimen. The size determined

using that methodology was 1 in. (25.4 mm). Once the concrete breakout strength capacity of

the anchors was determined, the corresponding shear load on each of the bolts was calculated.

The anchor bolt diameter had to be increased to 1.5 in. (38.1 mm) in order to ensure that the bolts

would not experience steel failure in flexure or shear. The maximum flexure on the bolts was

calculated by taking the maximum shear applied to each bolt and calculating the corresponding

maximum flexural moment (Figure 3-5). The lever arm (Equation 3-1) for the calculation of the

capacity was defined in Eligehausen et al. (2006) Section 4.1.2.2 b.

S= e1 +a3 (3-1)
Where:
/ = lever arm for the shear load (in.)
el = distance between the shear load and surface of concrete (in.)
a3 = 0.5-do, without presence of a nut on surface of concrete, Figure 3-5 (in.)
0, with a nut on surface of concrete

The base plate was restrained against rotation, and translation was only possible in the

direction of the applied shear load. The maximum applied moment for each bolt was calculated

based on these support conditions and the lever arm calculation. Full fixity occurred a distance

a3 into the shaft.

Using the section modulus of the bolts, the stress was then calculated and compared to the

yield strength of the bolts, 105 ksi (724 MPa). The shear strength of the bolts was calculated

using the provisions in Appendix D of ACI 318-05. In both cases it was determined that the

bolts had sufficient strength.









3.3.2 Concrete Breakout Strength of Anchor in Shear Parallel to a Free Edge

The breakout strength provisions outlined in ACI 318-05 Appendix D and the breakout

provisions introduced in Eligehausen et al. (2006) were applied to the design of the shaft.

Equation 2-11, from ACI 318-05, was used as the primary equation for the calculation of the

breakout strength. In order to apply the ACI provisions to the circular foundation a section of the

concrete was ignored (Figure 3-6). If the full cover, c, was used in the calculation, the failure

region would have included area outside of the circle. Rather than extending beyond the edge of

the concrete, the 35 degree failure cone (Figure 2-6) was extended to the edge of the shaft as

shown in Figure 3-6. Equation 3-2 was developed to determine the adjusted cover, cal.

r2 +3.25(r2 r 2)r
a = (3-2)
3.25
Where
rb = radius measured from the centerline of the bolt to the center of the foundation (in.)
(Figure 3-6)
r = radius of circular foundation (in.)

As presented in Section 2.3.4, the projected concrete failure area for a single anchor, Avco,

is equivalent to 4.5(ca1)2. Figure 3-7 illustrates the development of the projected concrete failure

area for a group of anchors, Avc, as a function of the number of bolts, n, the radius of the shaft, r,

and the adjusted cover. The resultant concrete breakout strength using the adjusted cover

approach was conservative relative to assuming the full cover.

Equation 3-3 and Equation 3-4 are used to calculate the concrete breakout torsion,

Tn,breakout, and are based on the ACI provisions for shear parallel to the free edge.


For A_ < sin-' 1 5c
A~ )

],breakout = 2 AVC Vb rb (3-3)
AVco










For A > sin 1.5cal (i.e. no overlap of failure cones)

] =,breakout = 2 n Vb .rb (3-4)
Where
A = angle of circular sector for each bolt (deg) (Figure 3-7)
ca, = adjusted cover (in.) (Equation 3-2)
rb = radius measured from the centerline of the bolt to the center of the foundation (in.)
(Figure 3-6)
Avc = projected concrete failure area of a group of anchors (in.2) (Figure 3-7)
A, = projected concrete failure area of a single anchor (in.2) (Figure 3-6)
Vb = concrete breakout strength in shear for a single anchor calculated using Equation 2-
11 with ca, as calculated in Equation 3-2 (lb.)
n = number of bolts

Using Equation 3-3, the ACI concrete breakout torsion for the test specimen was

determined to be 182 kip-ft (247 kN-m), which was less than the nominal torsional capacity.

During the analysis of the design equations, an issue arose regarding the calculation of the

Eligehausen et al. (2006) factor Vparallel. The result of Equation 2-18 was 4.06 compared to the

ACI 318-05 factor and qa,v of 2.0. This prompted an investigation of the application of the

multiplier to the circular foundation in this research program.

The majority of the tests for the determination of Vu,c (Equation 2-14) were for groups of

two bolts. Therefore, it was investigated how the Avc/Avo term is affected by the spacing

between the bolts and the number of bolts. Figure 3-8 shows that for spacing, s, of 3. Oca or

greater there is no overlap of the breakout cones. In those cases the strength is the sum of the

single anchor strengths. Figure 3-9 illustrates the overlap of the breakout cones. The Avc/Avoc

term is used to calculate the breakout strength for the case where the failure cones overlap.

Av/Avco can be normalized by dividing by the number of bolts. An increase in the number

of bolts at the same spacing along a straight edge leads to a reduction in the normalized A V/AVco

term. This reduction is illustrated in Figure 3-10. The contribution of the failure cone

outstanding "legs" at the ends of the group area, Ave, decreases as the number of bolts increases.









For a circular foundation, with s<3. Ocal, there is a constant overlap of the failure cones with no

outstanding "legs" (Figure 3-11). The equivalent number of bolts along a straight edge is taken

as infinity in order to represent a circular foundation (i.e. no outstanding "legs"). Therefore, the

normalized Avc/Avco term for this case was calculated for an infinite number of bolts at the

prescribed spacing for the foundation. To convert these ratios into a multiplier for WVparallel, the

ratio of the normalized A vc/Avco for an infinite number of bolts to the normalized term for two

bolts was calculated. That multiplier, 0.52, was applied to the parallel term resulting in an

adjusted parallel of 2.1. This resulting value agreed with the ACI 318-05 factor and the

Eligehausen et al. (2006) factor qaW,vof 2.0.

The resultant concrete breakout torsions, based on the Eligehausen et al. (2006) concrete

breakout strength (Equation 2-13), were 167 kip-ft (227 kN-m) using parallel of 2.1 in Equation

2-17, and 159 kip-ft (216 kN-m) using Ia,Tv of 2.0 in Equation 2-16. These torsions were

calculated using the same moment arm, rb, used in Equation 3-3 and Equation 3-4. These results

and the results of the other calculations are summarized in Table 3-2.

3.3.3 Development Length of the Bolts

Another key aspect of the shaft design was to ensure that the anchor bolts were fully

developed. In order to meet the code requirements, the splice length between the #4 bars and

anchor bolts was calculated using the development length equations presented in ACI 318-05

Chapter 12. The bolts needed overlap the #4 bars across 26.7 in. (678 mm), and in the test setup

the overlap was 29 in. (737 mm). Therefore, this requirement was met.

3.4 Steel Pipe Apparatus Design

The components of the steel pipe apparatus included the pipe, which was loaded during

testing, and the base plate. The pipe design was based on the interaction between torsion,

flexure, and shear as presented in AISC (2001), LRFD Manual of Steel Construction-LRFD









Specification for Steel Hollow Structural Sections. Each of the individual capacities was

calculated for various pipe diameters and thicknesses. The individual strengths were compared

to the projected failure loads for testing, the concrete breakout failure loads. In addition to

verifying that the capacity of the pipe exceeded those loads, the interaction of the three capacities

was verified. The purpose was to check that the sum of the squares of the ultimate loads divided

by the capacities was less than one. Based on this analysis, it was concluded that an HSS 16.000

x 0.500 pipe would provide sufficient strength.

In order to load the pipe, it needed to have a ninety degree bend in it. This was achieved

by welding two portions of pipe cut on forty-five degree angles to a steel plate. The weld size

for this connection was determined such that the effective throat thickness would equal the

thickness of the pipe, which was 0.50 in. (12.7 mm).

The factors included in the design of the base plate were the diameter of the pipe, required

weld size, bolt hole diameter, and the required distance between the edge of the bolt hole and the

edge of the plate. The required width of the weld between the base plate and the pipe was

calculated such that the applied torsion could be transferred to the plate without failing the weld.

From that point, the bolt hole location diameter had to be checked to ensure that there was

sufficient clearance between the weld and the nuts. It was important that the nuts could be fully

tightened on the base plate. A 0.25 in. (6.35 mm) oversize was specified for the bolt hole

diameter. This oversize was based on the standard oversize used in the field. Beyond that point,

it was ensured that there would be sufficient cover distance between the bolt hole and the edge of

the plate.

The design of the components of the steel pipe apparatus was crucial because these pieces

had to operate efficiently in order to correctly apply load to the bolts. If the apparatus were to









fail during testing, the objective of the research could not be achieved. The weight of the pipe

apparatus was calculated in order to normalize the load during testing. The load applied to the

anchorage would be the load cell reading less the weight of the pipe apparatus.

3.5 Concrete Block Design

The design of the concrete block was based on several key factors to ensure that it served

its purpose as a fixed support at the base of the shaft. The amount of reinforcement required was

based on a strut-and-tie model of the block as outlined in ACI 318-05 Appendix A and, as an

alternate approach, beam theory to check the shear strength and flexural strength of the block.

For the flexural capacity calculations, the ACI 318-05 concrete stress block provisions were

utilized. Based on the results of both approaches, it was determined that 3 #8 bars, each with a

12 in. (305 mm) hook on both ends, spaced across the top and the bottom of the block were

required. Additionally, two cages of #4 bars were placed in the block on the front and back faces

meeting the appropriate cover requirements to serve as supplementary reinforcement. The

purpose of reinforcing the block was to ensure structural stability of the block throughout the

testing process.

Two channel apparatuses were also designed in order to tie the block to the floor of the

laboratory in order to resist overturning. The loads that had to be resisted by each tie-down were

calculated such that the floor capacity of 100 kips (445 kN) per tie-down would not be exceeded.

The channels were designed in accordance with the provisions set forth in AISC (2001). The

welds between the channels and steel plates had to be sufficiently designed such that the

channels would act as a single unit thereby transferring load from the plates through the

channels. Also, the channels were spaced far enough apart to fit 1.5 in. (38.1 mm) bolts between

the channels. A 0.25 in. (6.35 mm) oversize was specified for the spacing of the channels and the

holes in the steel plates. The construction drawings for the channels are located in Appendix A.









In addition to assuring that the concrete block system had sufficient capacity to resist the

applied load, the bearing strength of the concrete had to be calculated. This was done in order to

verify that the concrete would not fail in the region that was in contact with the steel channels.

The bearing strength was found to be sufficient. As a result, it was concluded that the concrete

block system would efficiently serve as a fixed connection, and under the loading conditions it

would not prematurely fail.

3.6 Combined Shear and Torsion

As was presented in Chapter 2, a calculation had to be carried out to ensure that shear need

not be considered in the design. Rather than inputting the values for the ultimate shear and

ultimate torsion into Equation 2-8, the coefficients of these terms were calculated. The base for

doing so was to input the torsion as a function of the shear. For the test specimen, the ultimate

torsion, T,, was taken as the moment arm multiplied by the ultimate shear, Vu. The moment arm

for the load was 9 ft. (2740 mm). As an alternate approach, the actual concrete breakout strength

and the corresponding shear could have been inputted into the equation rather than the generic

variables. The result of the calculation to determine the coefficients was that the coefficient for

the shear term was 1 compared to a coefficient of 88 for the torsion term. This calculation

sufficiently verified that the shear contribution could be ignored in design.

3.7 Overview

The previous sections detailed the design of the various components of the experimental

program. It was of the utmost important to verify that the apparatuses not pertaining to the

foundation failure would not fail during testing (i.e. concrete block system and pipe apparatus).

Furthermore, all other foundation failure modes had to be precluded in the design. This ensured

that if the concrete breakout failure in shear was the failure mode it would be observed during

testing.









Figure 3-12 and Figure 3-13 show the fully assembled test specimen at the Florida

Department of Transportation Structures Research Center.










--CONCRETE BLOCK

SHAFT

BASE PLATE

PIPE ASSEMBLY
hL I


I LOAD LOCATION

Figure 3-1. Schematic of test apparatus


Figure 3-2. Front elevation of test apparatus




































Figure 3-3. Plan view of test apparatus


Figure 3-4. Side elevation of test apparatus























Leveling Nut


Figure 3-5. Lever arm for the calculation of bolt flexure


Figure 3-6. Adjusted cover based on a single anchor and 350 failure cone









chord = 2r sin -

r^sin
\ /,..2


Figure 3-7. Development of the projected failure area for the group of anchors around a circular
foundation

1.5ca, 1.5ca, 1..5c 1.5c

Cal


s=3. 0c1a

Figure 3-8. Two anchor arrangement displays the minimum spacing such that no overlap of the
failure cones occurs


Figure 3-9. Overlap of failure cones


II r H











"Leg"_ "Leg"




A Vc
Leg" "Leg"





Figure 3-10. The contribution of the "legs" of the failure cone to Avc along a straight edge
decreases as the number of bolts increases


Figure 3-11. Overlap of failure cones for a circular foundation
































Figure 3-12. Assembled test specimen


Figure 3-13. Shaft with pipe apparatus attached prior to instrumentation being attached









Table 3-1. Field dimensions
Component Field Dimension
Shaft Diameter 60 in.
Hoop Steel Diameter 46 in
Hoop Steel Size #5
Longitudinal Steel #9
Size
Anchor Bolt Diameter 2 in.
Anchor Embedment 55 in.
Bolt Spacing 36 in.
Diameter
Base Plate Diameter 42 in.
Base Plate Thickness 11/ in.

Table 3-2. Summary of design calculations
Component Design Type
Shaft
Cracking Torsion
Basic Torsion
Threshold Torsion
Nominal Torsion
Anchor
ACI Concrete
Breakout
Eligehausen et al.
Concrete Breakout
Eligehausen et al.
Concrete Breakout
Bolt Flexure
Bolt Shear


Equation Reference


(2-1)
(2-2)
(2-3)
(2-4)

(2-12)

(2-16)

(2-17)


Result


131 kip-ft
131 kip-ft
24.6 kip-ft
253 kip-ft

182 kip-ft

159 kip-ft

167 kip-ft

253 kip-ft
1756 kip-ft









CHAPTER 4
IMPLEMENTATION OF TESTING PROGRAM

In order to proceed with testing the specimen presented in Chapter 3, important

considerations had to be made. The first area under consideration was the concrete strength. It

was important to determine this to calculate the predicted failure mode prior to testing. Also, the

flexural and shear strengths of the bolts were calculated using the specified yield strength. The

other area that was of key importance was the instrumentation. The instrumentation was

required to produce meaningful data during testing. The other section of this chapter is on the

carbon fiber reinforced polymer (CFRP) wrap used in the retrofit test.

4.1 Materials

4.1.1 Concrete Strength

As it was stated in Chapter 3, the initial calculations for the design of the test setup were

carried out on the assumption of a concrete strength of 5500 psi (37.9 MPa). The concrete

breakout strength was recalculated based on the concrete strength at the time of testing. On the

date of the test, the concrete strength was 6230 psi (43 MPa). This strength was calculated based

on the average of three 6 in. (152 mm) x 12 in. (305 mm) cylinder tests.

4.1.2 Bolt Strength

The yield strength of the F1554 Grade 105 anchor bolts was assumed to be 105 ksi (723.95

MPa). This was used to calculate the flexural strength and shear strength of the bolts.

4.1.3 Carbon Fiber Reinforced Polymer Wrap

The first test was considered concluded after significant cracking and when the test

specimen stopped picking up additional load. The loading was ceased before the specimen

completely collapsed. The reason for doing so was to enable a second test to be performed on

the specimen after it was retrofitted with a carbon fiber reinforced polymer (CFRP) wrap. The









second test verified whether the CFRP wrap was an acceptable means to retrofit the failed

foundation.

The amount of CFRP that was applied to the shaft was determined by calculating the

amount of CFRP required to bring the shaft back to its initial concrete breakout strength. The

CFRP wrap that was used for the retrofit was SikaWrap Hex 230C. The properties of the wrap

were obtained and the ultimate tensile strength was used to calculate the required amount that

needed to be applied. The property specifications for the SikaWrap were based on the mean

strength minus 2 standard deviations. ACI (2002), ACI 440.R-02, 3.3.1 specifies that the

nominal strength to be used for design be based on the mean strength less 3 standard deviations.

Therefore, the design strength provided by Sika was adjusted to ensure that the design met the

ACI specifications.

The method for calculating the amount of CFRP required was to convert the torsion to a

shear load per bolt. The shear load, which was directed parallel to the edge, had to be adjusted to

such that it was directed perpendicular to the edge. In order to do this, the ACI multiplier of 2

was divided from the load. That load per bolt directed perpendicular to the edge was converted

to a pressure around the circumference of the shaft. The equivalent tension that had to be

resisted by the CFRP wrap was then calculated, and the amount of CFRP to provide that tensile

strength was determined. Figure 4-1 illustrates this method.

Two layers of the wrap were prescribed to meet the ACI concrete breakout strength based

on assuming that the full 12 in. (305mm) width of the CFRP wrap would not be effective.

Rather, it was assumed that the depth of the concrete breakout failure cone based on the cover,

1.5-cover, was the effective width, 7.5 in. (191 mm). Three layers of the CFRP wrap were

applied to the specimen. The addition of the extra layer exceeded the required strength, so it was









deemed acceptable. Once the wrap was set, the retrofit test was carried out. Calculations for the

design of the CFRP wrap layout are located in Appendix B.

4.2 Instrumentation

4.2.1 Linear Variable Displacement Transducers

Linear Variable Displacement Transducers (LVDTs) were placed at the location of the

load cell, and at various points along the shaft and base plate. A total often LVDTs were

utilized in the project. Figure 4-2 is a schematic of the layout of the LVDTs on the base plate.

Figure 4-3 and Figure 4-4 show the location of the LVDTs on the shaft, and Figure 4-5 shows

the LVDT at the load location. The denotation for each of the LVDTs is also on the drawings.

These identification codes were used to denote the LVDTs during testing. The purpose of the

LVDTs along the shaft and base plate was to allow for the rotation of the base plate to be

measured during the testing. The LVDTs at the front and back of the shaft were to allow for the

rotation to be measured relative to the rotation of the shaft. The intent in the project was such

that the shaft would not rotate; only the base plate would rotate as the bolts were loaded. The

horizontal LVDT on the base plate was intended to indicate if there was any horizontal

movement of the base plate. The rotation of the base plate was calculated using Equation 4-1.

RDi +-Dz} (4-1)
R = tan (4-1)
D gage
Where
R = base plate rotation (rad)
D1v = displacement of LVDT D1V (in.)
Ds = displacement of LVDT D3 (in.)
Dgage = distance between LVDTs D1V and D3 (in.)

Once the test apparatus was assembled, the distance Dgage was measured. This distance

was 26.31 in. (668 mm). Figure 4-6 shows LVDTs D1V and D4 on the test specimen.









4.2.2 Strain Gages

Strain gages were attached to the base plate on the outer surface adjacent to the bolt holes

in order to determine how may bolts were actively transferring load given the 1.75 in. (44.5 mm)

holes for the 1.5 in. (38.1 mm) anchors. In applying the ACI 318-05 equation for concrete

breakout strength of an anchor in shear directed parallel to an edge (Equation 2-12) it was of key

importance to know how many bolts were carrying the load. For instance, if two bolts were

carrying the load, the concrete would fail at a lower load than if all twelve bolts were carrying

the load. In addition to showing the placement of the LVDTs, Figure 4-2 also details the

location of the strain gages on the base plate. Figure 4-7 shows the denotation of the strain gages

relative to the bolt number, and Figure 4-8 shows a strain gage on the base plate of the test

specimen. Note that the bolt numbering starts at one at the top of the plate and increases as you

move clockwise around the base plate.
















Divide by 2



J T


TCFRP c ^FRP \


Figure 4-1. Method for the determination of the tension, TCFRP, that must be resisted by the
CFRP wrap



















Strain


Figure 4-2. Instrumentation layout on the base plate

















D6


Figure 4-3. Instrumentation layout on face of shaft







I


Figure 4-4. Instrumentation layout on rear of shaft/face of concrete block


I D1


Figure 4-5. Instrumentation layout of pipe at load location



























Figure 4-6. Location of LVDTs D1V, D4, and D7 on the test specimen

Bolt 1
/


Figure 4-7. Strain gage layout on base plate






















Figure 4-8. Strain gage on base plate of test specimen







Figure 4-8. Strain gage on base plate of test specimen









CHAPTER 5
TEST RESULTS

Two tests were performed on the test specimen. The initial test was conducted to

determine whether the concrete breakout failure was the failure mode demonstrated in the field.

The verification of this was based on the crack pattern and the failure load recorded. If the

failure torsion was the concrete breakout failure torsion, then the hypothesized failure mode

would be verified. The retrofit test was performed on the same test specimen. This test was

completed to establish whether a CFRP wrap was an acceptable retrofit for the foundation.

5.1 Initial Test

5.1.1 Behavior of Specimen During Testing

The initial test on the foundation was carried out on 31 August 2006 at the Florida

Department of Transportation Structures Research Center. The test specimen was gradually

loaded during the testing. Throughout the test, the formation of cracks on the surface of the

concrete was monitored (Figures 5-1 and 5-2). At 90 kip-ft (122 kN-m), the first cracks began to

form. When 108 kip-ft (146 kN-m) was reached, it was observed that the cracks were not

extending further down the length of the shaft. Those cracks that had formed began to slightly

widen. These cracks, Figure 5-1, were characteristic of those that form during the concrete

breakout failure. At 148 kip-ft (201 kN-m), cracks spanning between the bolts had formed

(Figure 5-3). The foundation continued to be loaded until the specimen stopped taking on more

load. The torsion load peaked at 200 kip-ft (271 kN-m). Loading ceased and was released when

the applied torsion fell to 190 kip-ft (258 kN-m). The predicted concrete breakout capacity of

the shaft at the time of testing was calculated as 193 kip-ft (262 kN-m) (Equation 3-3).

At failure, the foundation displayed the characteristic cracks that one would see in a

concrete breakout failure (Figure 5-4). As intended, the bolts did not yield, and the shaft did not









fail in torsion. Data was reduced to formulate applied torsion versus plate rotation and applied

torsion versus bolt strain plots. The Applied Torsion vs. Plate Rotation plot (Figure 5-5) shows

that the bolts ceased taking on additional load after the noted concrete breakout failure due to the

shear parallel to the edge resulting from the applied torsion. It also exhibits slope changes at the

loads where crack development started or the existing cracks were altered. The first slope

change at 108 kip-ft (146 kN-m) coincided with the widening of the characteristic diagonal

cracks on the front face of the shaft. The second change occurred at 148 kip-ft (201 kN-m)

corresponding with the formation of cracks between the bolts.

5.1.2 Behavior of Strain Gages During Testing

Figure 5-6 displays the Applied Torsion vs. Bolt Strain plots for each bolt relative to its

location on the foundation. Recall that the term bolt strain refers the measurement of the strain

in the base plate at the bolt location. The strain was a result of the bolt carrying load. The first

line on the plots in Figure 5-6 is 50 kip-ft (67.8 kN-m). At this level, all of the bolts were

carrying load with the exception of bolts one, six, and eight. At the next level, 100 kip-ft (136

kN-m) bolt one picked up load, but bolts six and eight remained inactive.

It must be noted that, at 108 kip-ft (138 kN-m), which was the first slope change on the

Applied Torsion vs. Plate Rotation Plot, a redistribution of the loading occurred. This

redistribution is illustrated in Figure 5-7. As the cracks widened, those bolts that were

transferring the majority of the load were able to move more freely, and, therefore, the other

bolts became more active in transferring the load to the foundation. A similar redistribution to a

lesser degree occurred at approximately 148 kip-ft (201 kN-m), which coincided with the first

observation of cracks between the bolts.

As the various plots illustrate, some of the strain gages recorded negative strains, while

others recorded positive strains. This was most likely due to the bearing location of the bolt on









the base plate. Although this occurred, the relative strain readings were considered acceptable.

To further explore this phenomenon, strain gages were placed on the bottom of the base plate in

addition to those on the top for the second test.

5.1.3 Summary of Initial Test Results

The results of this test indicated that the concrete breakout failure was the failure mode

observed in the site investigation. The characteristic cracks and the structural integrity of the

bolts in the failed foundations, as observed during the site investigation, was the first step to

arriving at this failure mode. The percent difference between the failure torsion and the

predicted failure torsion (Equation 3-3) was 3.6%. Therefore, it was concluded that the

foundation failed at the failure torsion for the predicted failure mode. These results indicated

that the design methodology for cantilever sign foundations should include the concrete breakout

failure due to shear directed parallel to an edge resulting from torsional loading. All plots for the

first test are located in Appendix C.

5.2 CFRP Retrofit Test

After the results of the first test were reviewed, the need for a method to strengthen

existing foundations became apparent. Since the concrete breakout failure had not been

considered in the design of the cantilever sign structure foundations, a system had to be put in

place to evaluate whether or not those existing foundations would be susceptible to failure. One

economical method of retrofitting the existing foundations is the use of Carbon Fiber Reinforced

Polymer (CFRP) wraps.

Recall that, at the conclusion of the first test, the bolts had not yielded, and the concrete

was still intact. This enabled a second test on the failed foundation to be carried out. The key

focus of this second test was to determine if the foundation could reach its initial concrete

breakout strength again. The foundation was retrofitted with three layers of 12 in. (305 mm)









wide SikaWrap Hex 230C (Figure 5-8). This amount of CFRP exceeded the amount required to

attain the concrete breakout strength, 193 kip-ft (262 kN-m). The torsional strength of the shaft

with the CFRP wrap was calculated. The resultant strength based on the effective width, Section

4.1.3, of 1.5-cover, or 7.5 in. (191 mm), was 229 kip-ft (310 kN-m). Since that effective depth

was an assumption for design, the strength based on the full width, 12 in. (305 mm), of the wrap,

367 kip-ft (498 kN-m), was also calculated for reference.

5.2.1 Behavior of Specimen with CFRP Wrap During Testing

The second test was conducted on 13 September 2006. For this test, the concrete strength

was not required to be known, since the concrete had already failed. The containment provided

by the CFRP wrap, along with the anchor bolts, was the source of the strength of the foundation.

As the purpose of the second test was to learn how much load the foundation could take, and if

that load met or exceeded the concrete breakout strength, the load was not held for prolonged

periods at regular intervals during the test. Figure 5-9 is the Applied Torsion vs. Plate Rotation

plot for the second test. The foundation was closely monitored for crack formation along the

shaft, propagation of existing cracks, and failure of the CFRP wrap.

The strength of the foundation exceeded the predicted concrete breakout strength of 193

kip-ft (262 kN-m). It was not until the loading reached 257 kip-ft (348 kN-m) that the first pops

of the carbon fibers were heard. At that torsion load, the strength of the CFRP wrap based on the

effective depth, 229 kip-ft (310 kN-m), was exceeded. Therefore, the effective depth of the wrap

was a conservative assumption.

At approximately 288 kip-ft (390 kN-m) more pops were heard. However, the carbon fiber

did not fail. During the course of the test, characteristic torsion cracks began to form along the

shaft (Figure 5-10) and propagated to the base of the shaft. This occurred because the ACI 318-

05 nominal torsional strength (Equation 2-6) of 253 kip-ft (343 kN-m) was exceeded. Although









these cracks had formed, the foundation still had not failed. Another phenomenon that occurred

was the yielding of the bolts. According to the calculations for the yield strength of the bolts, the

bolts yielded at approximately 253 kip-ft (343 kN-m) of applied torsion. The strength was

determined using the same methodology outlined in Section 3.3.1. This was the within the range

in which the yielding was observed (Figure 5-9). The bolts were yielding, but they did not reach

their ultimate strength. The test abruptly concluded when the concrete block shifted out of place,

causing the load cell to be dislodged from its location on the pipe. This occurred at 323 kip-ft

(438 kN-m).

5.2.3 Behavior of Strain Gages During Testing

For the retrofit test, strain gages were placed on the top and bottom of the base plate.

Figure 5-11 shows each of the Applied Torsion vs. Bolt Strain plots at the appropriate bolt

locations. Note that as the loading increased, the bottom strain gages began to behave similarly

for all of the bolts. The strain was increasing at a higher rate. This illustrated that as the bolts

picked up load and began to bend, they were primarily in contact with the bottom of the base

plate (Figure 5-12). The strains recorded by the bottom gages indicate that all of the bolts

became active during the test.

Similar to the behavior of the bolts throughout the initial test, Figure 5-13 illustrates the

changes in the bolt strain data for the top gages corresponding with milestone loads during the

test.

5.2.4 Summary of Test Results

Upon removal of the pipe apparatus, the crack pattern illustrated the concrete breakout

failure, and torsional cracks in the center of the shaft verified that the torsional capacity was

exceeded during testing (Figure 5-14). Figure 5-15 details the characteristic torsion cracks on

the side of the shaft after testing. The test proved that the CFRP wrap was an acceptable method









for retrofitting the foundation. It exceeded the concrete breakout strength. The success of this

retrofit test led to the development of guidelines for the evaluation of existing foundations and

the guidelines for the retrofit of those foundations in need of repair. All plots for the retrofit test

are located in Appendix D.




















Figure 5-1. Initial cracks on face of shaft


f7


/


em &

it
I I -t4
'tt A\4a6


Figure 5-2. Initial cracks on face and side of shaft (alternate view of Figure 5-1)































Figure 5-3. Face of test specimen after testing exhibits cracks between the bolts along with the
characteristic concrete breakout cracks


Figure 5-4. Crack pattern on face of shaft after testing depicts characteristic concrete breakout
failure cracks

















250




200




S150
o


E 100
B.


50




0


Figure 5-5. Applied Torsion vs. Plate Rotation Plot- Initial Test














































68


Peak Applied
Torsion



Crack Formation
Between Bolts









Cracks Begin to
Widen







0 0.5 1 1.5 2 2.5 3

Rotation (deg)










































Figure 5-6. Applied Torsion vs. Bolt Strain Plots for each bolt at the appropriate location on the
base plate with Applied Torsion vs. Plate Rotation plot in center (full size plots in
Appendix C)




















200
Crack Formation
Between Bolts

S150,,- .7.- .. -. -




100
/ Cracks Begin to
Widen

50 '




0
-750 -600 -450 -300 -150 0 150 300 450 600 750
Microstrain



Figure 5-7. Bolt Strain Comparison Plot for Initial Test exhibits the redistribution of the load
coinciding with crack formations


Figure 5-8. Shaft with the CFRP wrap applied prior to testing
















350


300


S250


S200


150


100


50


0.5 1 1.5 2 2.5
Rotation (deg)


Figure 5-9. Applied Torsion vs. Plate Rotation Plot- Retrofit Test


Figure 5-10. Shaft exhibiting characteristic torsion cracks from face to base of shaft


4.5 5


























I 4I I I I I
10 I I






.... A ....7- ...-...A...-..











Figure 5-11. Applied Torsion vs. Bolt Strain plots for the Retrofit Test at the appropriate bolt
location around the base plate with Applied Torsion vs. Plate Rotation plot in center
(full size plots in Appendix D)


I Ell-













Base Plate







Bearing
Location


Figure 5-12. Bolt bearing on the bottom of the base plate during loading


350


300


250


".
S200


a 150


100


50


0


-1100 -825 -550 -275 0 275 550 825 1100
Microstrain


Figure 5-13. Bolt Strain Comparison plot for the retrofit test exhibits slope changes at milestone
loads


Bolts
Yielded






ACI Concrete
Breakout Strength

________^ H ^ ^ /- ____




-^y^--
































Figure 5-14. Face of shaft after test illustrates yielding of bolts, concrete breakout cracks around
the perimeter, and torsion cracks in the center.


Figure 5-15. Torsion cracks along length of the shaft after the test









CHAPTER 6
SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS

The purpose of this research program was to determine the cause of the failure of

foundations of cantilever sign structures during the 2004 hurricane season. After a thorough

literature review, in conjunction with the site investigation, and testing, it was determined that

the foundations failed as a result of an applied torsion which caused a concrete breakout failure

due to shear directed parallel to the edge on the anchors. This anchorage failure is detailed in

ACI 318-05 Appendix D. Previous to this experimental research, this failure mode was not

considered in the design of the cantilever sign foundations. Cantilever sign foundations need to

be designed for shear parallel to the edge on the anchor resulting from torsion.

Test results indicate that the failure of the foundations was caused by concrete breakout

due to shear directed parallel to the edge on the anchors. The test specimen failed at the torsion

predicted by the ACI 318-05 Appendix D design equations. Additionally, the crack pattern

matched the crack pattern exhibited in the field, and both foundations emulated the characteristic

crack pattern of the shear directed parallel to an edge for concrete breakout failure. It is

recommended that future tests be performed on circular foundations to further investigate the

concrete breakout failure for a shear load directed both parallel and perpendicular to an edge.

Additional testing was performed to determine an acceptable retrofit option. It was

determined that applying a CFRP wrap to the foundation strengthens the foundation such that it

not only meets its initial concrete breakout capacity, but, also, exceeds the capacity. The results

of this test led to the development of guidelines for the evaluation and repair of existing

foundations. The guidelines were based on the following:

S Using either the torsional load from the design or, if not available, using the ACI nominal
torsional strength (ACI 318-05 11.6.3.6), determine the torsional capacity of the
foundation.









* Calculate the concrete breakout strength in accordance with ACI Appendix D.

* If the concrete breakout strength is less than the maximum of the nominal torsional
strength and design torsion, then the foundation is susceptible to failure.

* The amount of the SikaWrap 230C required is calculated using the maximum of the
nominal torsional strength and the design torsion. The amount required is given in layers
of the CFRP wrap.

These guidelines were submitted to the Florida Department of Transportation. The

guidelines will be used to evaluate and, if necessary, repair the existing foundations. It is critical

that such foundations be evaluated in order to determine the susceptibility to this type of failure.

The proper use of the findings of this research program will allow for future prevention of the

failures exhibited during the 2004 hurricane season.







APPENDIX A
TEST APPARATUS DRAWINGS


-e

1]n__
1--IFel


2 2[2_I_


' Sr




9.
E ---------



. . .


--- t-- -


S



em
" / -
| tI


v __


,__/ _
= ~ a*~


1


I,


c:

i'


































Qr4














__ is




F I









Ct

C)n



C
g -e


B3







-e
C)
C
Ct3

C)

S~




64



















































I"I
- -- -- .-. - -. --.. ', . . ..



-- -
n. ,.,._ I





4 '0


S- -









a: -e

SC



.-------------<----------------.
,& Sb
S- P
Ii Q)





































igA


4
K


3
u

r eL
~c16;
fff
Pld

9
Bsd
w8
~C~~el
r


-4



















~1










APPENDIX B
DESIGN CALCULATIONS


Design of Test Program
Input
Shaft
d := 30in Diameter of the Shaft

f := 5500psi Concrete Strength

Hoop Steel
BarSize Hoop := 3 Hoop Steel Size

sh:= 2.5in Spacing of Hoop Steel

ft = 60ksi Yield Strength of Hoop Steel

dh:= 27n Hoop Steel Diameter

Bar Size Long:= 4 Longitudinal Steel Size


torsion = 075 ACI 318-05 9.2.3.2

Moment Arm:= 9f1


ORIGIN 1


Bolts
do:= 1-5in Diameter of the Bolt

db:= 20in Center-Center Diameter
of Bolts
boltt= 105k-i

No Bolts:= 12 Number of Bolts

CFRP Properties
tC. := 0.u5]h in Thickness of \\rap

fn Ct-RI := 91-lksi Tensile Strength of
CFRP
w cF I, := 12in Width of CFRP Sheet


For the Hield model, FI554 Grade 55 Anchor Bolls
were used.
tboltfield:= 55ksi










Failure Equations


Torsion
Cracking Torsion


Acp := Pp 2 cp
2- 2
c ..ACcp
Ter: 4 .psl -
Psi Pep)


Basic Torsion


T:- 4- Ipsi
psi

(ds3

Tbasic 2


Threshold Torsion


\Cp
=p l.- I -2n

lcr = 131 nI L kp. Li


Th.I, 11 'I kLp-Ii


Tthreshold = @torsion -psl--
Spsi t Pcp)
Nominal Torsional StrengthPcp


Nominal Torsional Strength


I l.2cshold = 2- 7 k!p-rt


Ao: .(dh 2)2 Area enclosed by hoop steel

At:= I.[(Bar Size Hoop-in + 8) 2]2 Area of hoop steel

S:- 45 ACI318-05-11.6.3.6 (a) Grad: 0.--
180

2-AoAtf ot
T, torsion-: -"t( rad)
sh


A 5712 56 i

A i II In-

rd l 7


In ls,.UU= '5 kip-tt










Combined Shear and Torsion


T,(Vu) := V.Moment Arm-kip


If V := 1kip T(



Coefficient Shear:=






Coefficient Torsion :


Ah-:= Ao Aoh = 572.56 in"

p11:= n-dh Ph = 84.82 in


V) := Vu.Moment_Arm

SVu ,
kip
Aoh
in2


Tu(Vu) Ph-2
kip-in in
2
1.7- _
1 A
"2J


Coefficient Torsion
CofficientShear88.58
Cocfficient Shcar


The coefficient for the torsion term is 88 times that of the shear
term. Therefore. shear need not be considered.










Concrete Breakout Strength


dsdb
cover :- Bolt Cover
2


c db 2 ds 2 Q) Cdb
2 + 3.25 -- -
cal: 3.25


C I 3 s5 in Cover for the calculation of
the anchor ritcmLthi


= 2r sin --
2


hord




3600 r
A


-


TZ


360
A:- --deg
No Bolts


A = 30 deg







n chord -1.5c,


chord_group:= 2---sinI-

chord_group = 7.76 in


Amiingroup:= 2-asin .

Amingroup = 45.24 deg
If A is greater than Anoup then
there is no group effect


AVc:= NoBolts-chord_group-1.5-Cal


C i r = 111


A\'c 53 AVo: 4.5-cal


A -co o 5 u


`J










Effect:= "Group Effect- Analysis is Valid" if A < Amin group

"Analysis Invalid" if A> Am group


Check_Group_Effect = "Group Effect- Analysis is Valid"

ACI 318-05-1).6.2
1c:= 8-do

Vb :=2 5P 15
Vb. 13i .a lbf
-[do) in ,j psi In i I


ycV: 1.4


Vb = 13.5 kip


ACI 318-05- D.6.2.7


1.0 for cracked concrete with no supplcmenutal rcinlforctmcut or reinforcement
smaller than a No. 4 bar
1.2 for cracked concrete with edge reinforcement of a No.4 bar or greater
1.4 for uncracked concrete or with edge reinforcement of a No.4 bar or greater
enclosed within stirrups spaced no more than 4 in. apart


WecV:- 1.0 ACT 318-05-D.6-2-5


edV :- 1.0 ACI 318-05- D.6-2-6


All anchors are loaded in shear in the same direction


Ave
AVc
Vebg : 'cV.4\''edV"b
\VL-cI


Vcbg parallel: 2"Vcbg


Tn breakoutACI:- Vebg_parallel*(db 2)


Vcbg = 109.01 kip


Vcbg_parallel = 218.03 kip


Tn breakout ACI = 81.69kip.fi


CheckGroup











Eligehausen et al. (2006) Concrete Breakout Provision-

fcc200 := 1.18f IXI i the conm\eI)ion factor between
die cylinder test and cube rest


do 0.5

rmm
a := 0.1- -
cal
mm


0.5
0.-
p :- 0.1. a-
cal
mM


p P1.5
Va,, 3 cc200 cal 5
umm mm N mm



Avc
Vuc :- V'uc
AVco


\,*S2.'6 N

\' = 1 -t LIp


uV = 4 -2'1.1 27 N


L '. tI- p


a := 90---
S 180
1
V: cos( av) + 0.5 sin(V)


Vue V:= VucdI aV


Parallel := 4*k4"


[ mdo cc 200
NoBolts fcc200
mmy NN
2
mm
Vuc
N


jc ~ IYI S Lip


k := 0.75


' pa:iIllel


Equation 2-18 was based on tests with two bolts with a straight edge. Therefore, the applicability
of the equation comes into question when there are more than two bolts being loaded. The
following analysis will determine the ratio of V/, for 2 bolts and V, for arrangements of 2 or
more bolts.
Step 1: For 2 bolt arrangements, calculate Vc as a function of spacings, and V'c.

The spacing will be taken as a function of the cover


A, will be the area of the group of anchors for a unit depth
Ao will be the area of a single anchor for a unit depth


'4'ctV -
















cover


cover = 5 in

Sb:= T.db + No_Bolts

Sb = 5.24 in


Ao(c) := 1.5-2-c


Ao(c) -> 3.0-c


This value does not change as a result of changing
the spacing. The generic variable for the cover is c.


L5ca I 1.5c1 I 1.Sca1 I 1.5c1a







s=3. 0ca

For the case where s>=3.0cal, there is no overlap.


3.0-c
1.5-c
s(c) := 1.0-c
0.5-c
0.c
For
n:= 2



Av(c, s,n)

Ao(c).n


Av(c,s,n) := 2-(1.5.c) + s(c)-(n- 1)


n is the number of bolts


A(c,s,n) -> 30-deg(c,function,2)



0.75 The ratios are normalized by
factoring out the number of bolts
0.67 under consideration such that the
0.58 result may be compared to the
0.5 J ratio for 2 or more bolts.











For a circular foundation,n is considered equal to infinity because the bolts are continuous;
there is no end as there is for a straight edge


n:= 1.1020


AV c,s,n)

Ao(c).n


For n= infinity


0.5
0.33
0.17

0 0


Therefore, to calculate the reduction factor for the parallel conversion multiplier, the ratio of the
Ae!/Ago terms for 2 bolts and an infinite number of bolts will be calculated considering the proper
ratio ofs/c.


For the foundation under invesitgationpz is equal to infinity


sb
= 1.05
cover


sb
sratiotest(c):= -c
cover


Av(c, sratio test, n)
Group Multiplier Infinity:= Ao(
Ao(c)-n


Group Multiplier 2 Bolts :


AV(c, sratio test, 2)

Ao(c).2


Group_I Millriphlr_ruliiur 0 35


( rIup 1_Multilier_ iIfi niry
Groi.p Mulltiphei 2 Boll'


Group Multiplier Infnity
Reduction s bolt Group Multiplier 2 Bolts



Vparallelnew:= V parallel'Wreduction s bolt



Vucparallel := parallel new-Vuc


Tn breakoutaV := VucaV(db + 2)


Tn breakout parallel := Vucparallel'(db 2)


G roupl ulitiplhcr 2 Boll' 11








t'redJiiclnj 1oll = IIe'l



k'pl.illel_ne = 2 1



'tipara.llel = 893,' 0 N

'ucpaiaallel 2111" 4


I l'i ca lL III(|.IV


[1 hi eakr'ur i:par:ilIel lb ~ Lip Ii


I -1 kip-ll










Summary of Failure Equations


Tr = 131.06kip.ft

Tbasic = 131.06 kipl f
Tthreshold = 24.57 kip. fi
Tntorsion = 252.95 kipft

Tn breakout ACI= 181.69 kip-ft
Tnbreakout ctV = 159.4kip.ft


According to ACI 318-05 11.6.3.5, assuring the ultimate
torsion exceeds the threshold torsion, the nominal torsional
capacity is taken as the strcngth of the section.




Tn breakout wpurullc = 167.48kip.ft


For the design of the various components of the test specimen, the ACI Concrete Breakout Strength
will be the maximum moment as it is the predicted failure mode. (breakout ACi

Mmax := Tnbreakout ACI


Mmax = 181.69 kip ft


Vmax:= 1ima + Moment Arm Vmax= 20.19kip Failure Load












Tn torsion- 252.95 kip.ft


Shaft Design

Mmax shaft: Vmax-36in


Mmax shaft- 60.56kip.ft


BarSizeHoop = 3 Bar SizeLong 4

sh 2.5 in := 60ksi
t- 60ksi
dh 27in

Required splice for #3 bars


ACI 318-05 12.2.2


& t-rWf e (Bar Size Hoop
RequiredSpliceHoop := -* --*\ Size inp
ec
25. -.psi
Spsi


Required splice between #4 bars and anchor bolts

s:=- 0.8 Use the simplification for the (%+KtI)/db term


.3 bolt field VtyWe*ys' (
Id: No
d 40 rfc cb Ktr term
S-psi Note: The yi
psi determine th
the field for
Development length of #4 bars
ACI 318-05 12.5

0.02 .- &e Bar Size Long ~
ldh:= -- Pm

Fpsi


RequiredSplihe_ loop 1 2 14 in


cb Ktr term:- 2.5 ACI318-05 12.2.3


Id 2 'im
eld Nti cngth in the field was used to
e splice length to replicate the embedment in
the test setup.


s% Ir1 jij


Hook Length= 12Bar Size Long
8


yt:- 1.0 Ye:- 1.0 k :- 1.0


Hook Length t in











Flexural Capacity of Shaft Calculated using to ACI Stress Block
Bar Size Long-m + 8
R := ds 2 R = I u AIng_ steel:= r Size Long-in

S= 60 ksi number of bars:= 24

numberbars_yielded-Along steer fy
Ac :=
0.85-fc


Acircseg(h) := R2acos{R (R h). 2-Rh h A

h:= n,.. \cr-,,egilh.h,0in,15m)


i 0 reel = 2 i "-



Ac 12 '1 ui





i = .3 .3'i


P (f):- 0.85
0.65


0.85


if f < 4000.psi

if f > 8000-psi

0.05f- 4000-psi
1000-p0.05si
1000-psi 7


ACI 10.2.7.3


h
c:= c = 4.2!. W

p c
y:= .002--- = 2Sini
.003


The assumption was correct!


9.2502in-Al,,;_stee1l2 + 12.0237 in .A\lilgsteel-2 + 15in.Along_steel2 ...
+ 17.9763 in-Along_steel2 + 20.7 N I.All,,g _steer2 + 23.1314in-Along_steer2 ...
+25.0189m-Alo,,nI steer2 + 26.1677 n-Aloi_ steer2 + 26.5in-Along steel
bars Alo
17-Alongsteel


Oflexure:= -90


Mn Shaft:= flexure nuIberbars_yielded-Alot, g steerfy dbars )


Check Flexure Shaft : "Sufficient Strength" if Mn Shaft > Mmax shaft

"Insufficient Strength" if Mn Shaft < Mmax shaft

Check_FI excre_Sh.jll = "Siil'litenl SIrelngil"


n Sh.i = 262.2. p f


l ( I ( Ic I


dbars


19.13 in










Anchor Design


Check Bolt Flexure and Shear


Mmax
Vmax anchor :
--No Bolts
2
Vmax anchor- 18.17 kip


fbolt:= Mmax anchor bolt

uta: min(l .9-_bollt125ki)


max anchor: Vmax anchor"



Mmax anchor 24.98 kip.in


bolt= 75.4ksi Ase bk

futa 125 ksi Vbol :


.5-do I 2m
2


do

4


Sbolt 0.33 in3


t:= 1.405in2

Ase bolt'uta


CheckBolt Flexure : "Sufficient Strengll" if fbol < f bolt

"Insufficient Strcllglh" if fbol> y bolt

Check_Bolt_Flexure = "Sufficient Strength"

Check Bolt Shear:- "Sufficient Strenth" if Vbolt Vmaxanchor

"Insufficient Strength" if Vbolt < Vax anchor

Check_Bolt_Shear = "Sufficient Strength"

Calculate the load that will cause the bolts to yield
fy bolt-Sbolt-db-No Bolts
Mbolt yicld 0.5-do + 2in Mboltield = 253.02 kip-ft

Pboltyield:= Mbolt yield : Moment_Arm Pboltyield = 28.11 kip











Pipe Apparatus Design
Based on AISC LRFD Manual of Steel Construction-
LRFD Specificationfor Steel Hollow Structural Sections
Pipe Properties-HSS 16.000 x 0.500
tppe: 0.465 Apipe: 22.7i2 D t : 34.4 Wpipe: 82.8 -
4 3 ,3
pipe : 685 i Spipe: 85.7in rpipe : 5.49in Zpipe : 112m3

Dpipe: 16in Jpipe : 1370in Cpipe : 1713i

Design of Short Section- Applied Shear, Torsion, and Flexure

LShortPipe : 15in Vma 20.19 p Nlm 15I., ki p i'lorsion


MFlexure : Vmax-LShort Pipe


Design Flexural Strength

b:= 0.90 pipe := Dt
kpApipe 49.3

MdShortPipe:= ip pipe


CheckFlexureShortPipe :=



(heck Flexire


Design Shear Sirength
v:= 0.9

F 1.60-E 0.78E'
Fcr := max ----, -5 3
5 3
LShort_Pipe D t 4 D t2
D-
F "pipe

Fcr:= mm(0.6.Fyipei,Fcr)



Vd Short Pipe:= vFcr Apipe + 2


Fy :ipe: 42ksi

Fupipe 58ksi

E:= 29000ksi


Nie leie 25 23 1,p fi


kpJipe:= 0.0714-E + Fyjipe
Spipe 34.4
pipe
> Xpipe, tbFypipe"Zpipe, "Equation Invalid")
Md_nhoPrt_Plpc 352.8 kip h

"Sufficient Sure ,gh" if Md Short Pipe > MFlexure

"Insufficient Strength" if MdShort Pipe <1 le]..me

_Sholrt_Pilie = "S efficient S. ctimlh"


Fr = 575.22 ksi


i = 2 2 ksi



V1 Mhorl Pipe = 257 V2 jp


Check Shear Short Pipe := "Sufficient Strength" if VdShort Pipe > Vmax

"Insufficient Strength" if V ShortPipe
_heckSheaiShorl_Pipe = 'Sull'ilenti Strenglh"











Design Torsional Strength

T:= 0.90 Tdpipe: T-Fcr-Cpipe


Idjupc 3I 1 krp-i


Check Torsion Short Pipe : "Sufficient Srienliiil" if Tdipe > Mmax

"Insufficient Strength" if Tdpipe < Mmax

tlieck Tor'i 'ii Shirt Pipe "Sitticienir St ieliLh"
Check Interaction


SMFlexure Vmax
Interaction Short Pipe MF:exure + V-max
Md Short PipeJ VdShort Pipe


Mmax

Tdpipe


Interaction Short Pipe 0.48

Check Interaction Short Pipe : "Sufficient Sriei"llg' if Interaction Short Pipe < 1
"Insufficient Cernii-tli' if InteractionShort Pipe > 1

'hec'k_h.eia ion__ hIrlin_Pti [ "Sutli.'litIl Suie pii"
Design of Long Section- Applied Shear and Flexure

LLongPipe: 9ft \Va li19 1)p M11 m l1l 6'J klp-
Design Flexural Strength

Md Long Pipe iflpjipe > pipe b-.Fylipe.Zpipe'"Equation Invalid")
M duonieip nIie:up p i f 1 L n 1
M -1 I 1 ,' ~ e -3 2S kip). h


Check Flexure Long Pipe :


"Sufficient sieLI1" if MdLong Pipe > M ax

"Insufficient Strength" if MdLong Pipe < Mmax


( heck Flexire I -iue Pipe "Sllticient SNtlentlih"
Design Shear Strenglth

Vd Long Pipe := vFcrApipe 2


Check Shear Long Pipe :


VI I on,- Pipe 5' -4' kip


"Sufficient Strength" if Vd LongPipe > Vmax

"Insufficient Strength" if VdLongPipe < Vmax


Check Slie: I [ onp Pipe "Silicient Stlie1rli'"










Check Interaction
/2
Mm 1 Vnax
Interaction LongPipe := --- + max
Md Long_Pipe VdLong_Pipe
Interaction Long Pipe = 0.52

Check Interaction Long Pipe : "Sufficient Sircnilh" if Interaction Long Pipe < 1
"Insufficient Sircngili" if Interaction Long_Pipe> 1

Check_Interaction_Long_Pipe = "Sufficient Strength"

Weight of Pipe Apparatus

lbf 24m 2 lbf
Wipe app := W peLShort Pipe + Wpipe-LLong ipe + 490 -n2 -lin + 26in-20in-0.5in-490--

Base Plate \\ eld Plate

Wpipeapp = 1.05 kip

This weight must be subtracted from the applied load to account for overcoming the weight oftlle
pipe before the bolts were loaded. Will be used to normalize the test results










Concrete Block Design
Reinforcement
Two different methods were employed to check the reinforcement in the block:
A) Strut-and-Tie Model
B) Beam Theory


R




K .4'


A) Design by Strut-and-Tie Model ACI 318-05 Appendix A


Mmax = 181.69 kipft


Mmax
d:= 6ft+ 8in R:=--
d


R = 27.25 kip


NODE A


9 := ata -- 6 =
d~ft)
R
T C sin(O)

T:= C.cos(0)


0.64

C = 45.42 kip

T 36.34kip


R




C














Check Reinforcement

NoBarsBlockReinforcement := 3 BlockReinforcement Bar No := 8 fyBlock Reinforcement : 60ksi



(Block Reinforcement Bar No 8N 2
ABlock Reinforcement:" No Bars Block Reinforcement.. Bl-ck-Reinf-rce-ent Bar No22


\BII, R LIIlOrLCCIILt ll U- 2 _r? L




CheckReinforcement_A := "Sufficient iieigtl" if (ABlock Reinforcementt Block Reinforcement) > T


"Insufficient rleilr'h" if ABlockReinforcement.y_BlockReinforcement < T


li'hel. ReinriloiccIlnent \


B) Beam Theory


R






II
SIi

I
II
II
II
Ii
II
Ii
II
Ii
II
II
II
I I
II
I I
I I
I i
I I
I I
I ,II
I I
I I
I I
I I

I I


Vblo M


Y- '
:-I
, I
' I4 3'-o"


4'-9"


Vblock : R \block 2:.25 'ip M : R.(3ft+ 4in)


121


" tlml'l e'llli ,iret'l li"


M "90.&1 Ip.-ft











Check Shear

Check Shear B := "Sufficient Strength" if ABlock Reinforcementf Block_ RCIIfi.[L !It) > Vblock

"Insufficient "IieC1lh" if ABlockReinforcement _BlockReiforcement < Vblock


( ]ick Sihear B = "Su'iiicinl Sircength"


Check Flexure


i b




d

A,
-- ***


0.85s,


b := 30in
h:= 6ft
d:= 5.5ft


Locate the neutral axis, c, such that (C T)

T:= ABlockReinforcement BlockementBlo Reinforment T 141.37 kip

CT(a):= C(a) T a:= root(CT(a),a,0,h)


C(a) : 0.85-fc-b-a

. = I 1ll 11


1P(fc) := 0.85
0.65


0.85


p (fc)= 0.78


a
c := -


if fc <4000-psi

if fc> 8000-psi


- 0.5-fc 4000-psi)
( 1000-psi j


ACI 10.2.7.1


Calculate the nominal moment capacity, M1
Capacity Reduction Factor

Mn Block:= T'( 2


ACI 9.3.2, ACI 10.3.4, ACI 10.9.3


-1 BI U k = "I 'Lp II


CheckFlexure_B := "Sufficient Strength" if Mn Block > M

"Insufficient Strength" if Min Block < M

Check FILcirc B "SliIt'iciclt Sircenrll"


ACI 10.2.7.3


. = I oI

C 10.1 1 a1










Required Hook Length for a #8 Bar

ookNo8 : 12.(Block Reinforcement BarNoin ACI R12.5
-- 8

Summary of Concrete Block Design Reinforcement
Block Reinforcement Bar No = 8
NoBars_BlockReinforcement = 3 3 bars on top and bottom
Check_Reinforcement_A = "Sufficient Strength"
Check_Shear_B = "Sufficient Stiength"
Check_FlexureB = "Sufficient Strength"


Hook No 8 = 12in




Full Text

PAGE 1

1 ANCHOR EMBEDMENT REQUIREMENTS FOR SIGNAL/SIGN STRUCTURES By KATHLEEN M. HALCOVAGE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Kathleen M. Halcovage

PAGE 3

3 To my family, for your constant love and support throughout my lif e. To my parents, Barbara and George, your dedication to providing me with th e best education availa ble has been a pivotal part of my success. To my siblings, Barbara, George, Sarah, and Christ opher, you have always encouraged me and challenged me to be the very best that I can be. To my niece and nephew, Grace and Aidan, you light up my life and always remind me of what is important in life.

PAGE 4

4 ACKNOWLEDGMENTS I thank my advisor, Dr. Ronald A. Cook, fo r his guidance throughout the course of my research and tenure at the University of Flor ida, the Florida Department of Transportation Structures Research Center Staff for their ha rd work in building my test apparatus and orchestrating the testing of the specimen, and my supervisory committee for their assistance in the preparation of this thesis I also thank my family for their constant support and encouragement.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........8 ABSTRACT....................................................................................................................... ............11 CHAPTER 1 INTRODUCTION..................................................................................................................13 2 BACKGROUND....................................................................................................................15 2.1 Literature Review.......................................................................................................... ...15 2.2 Site Investigation......................................................................................................... .....19 2.3 Applicable Code Provisions.............................................................................................19 2.3.1 Cracking and Threshold Torsion............................................................................20 2.3.2 Nominal Torsional Strength...................................................................................22 2.3.3 Combined Shear and Torsion.................................................................................23 2.3.4 ACI Concrete Breakout Strength for Anchors.......................................................24 2.3.5 Alternate Concrete Brea kout Strength Provisions..................................................26 2.3.6 ACI 318-05 vs. AASHTO LRFD Br idge Design Specifications...........................28 3 DEVELOPMENT OF EXPE RIMENTAL PROGRAM........................................................33 3.1 Description of Test Apparatus..........................................................................................33 3.2 Shaft Design............................................................................................................... .......34 3.2.1 Torsion Design.......................................................................................................34 3.2.2 Longitudinal and Transverse Reinforcement.........................................................35 3.2.3 Flexure.................................................................................................................. ..35 3.3 Anchor Design.............................................................................................................. ....36 3.3.1 Diameter of Anchor Bolts......................................................................................36 3.3.2 Concrete Breakout Strength of Anchor in Shear Parallel to a Free Edge..............37 3.3.3 Development Length of the Bolts...........................................................................39 3.4 Steel Pipe Apparatus Design............................................................................................39 3.5 Concrete Block Design.....................................................................................................41 3.6 Combined Shear and Torsion...........................................................................................42 3.7 Overview................................................................................................................... ........42

PAGE 6

6 4 IMPLEMENTATION OF TESTING PROGRAM................................................................51 4.1 Materials.................................................................................................................. .........51 4.1.1 Concrete Strength...................................................................................................51 4.1.2 Bolt Strength...........................................................................................................51 4.1.3 Carbon Fiber Reinforced Polymer Wrap................................................................51 4.2 Instrumentation............................................................................................................ .....53 4.2.1 Linear Variable Disp lacement Transducers...........................................................53 4.2.2 Strain Gages............................................................................................................54 5 TEST RESULTS................................................................................................................... .60 5.1 Initial Test............................................................................................................... ..........60 5.1.1 Behavior of Specimen During Testing...................................................................60 5.1.2 Behavior of Strain Gages During Testing..............................................................61 5.1.3 Summary of Initial Test Results.............................................................................62 5.2 CFRP Retrofit Test......................................................................................................... ..62 5.2.1 Behavior of Specimen with CFRP Wrap During Testing......................................63 5.2.3 Behavior of Strain Gages During Testing..............................................................64 5.2.4 Summary of Test Results........................................................................................64 6 SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS.........................................75 APPENDIX A TEST APPARATUS DRAWINGS........................................................................................77 B DESIGN CALCULATIONS..................................................................................................82 C INITIAL TEST DATA.........................................................................................................108 D RETROFIT TEST DATA.....................................................................................................115 LIST OF REFERENCES.............................................................................................................122 BIOGRAPHICAL SKETCH.......................................................................................................124

PAGE 7

7 LIST OF TABLES Table page 3-1 Field dimensions........................................................................................................... .....50 3-2 Summary of design calculations........................................................................................50

PAGE 8

8 LIST OF FIGURES Figure page 1-1 Failed cantileve r sign structure..........................................................................................14 2-1 Cantilever sign structure at Ex it 79 on Interstate 4 in Orlando..........................................29 2-2 New foundation installed at the site...................................................................................29 2-3 Failed foundation during post-failure excavation..............................................................30 2-4 Concrete breakout of an anchor caused by shear directed paralle l to the edge for a circular foundation............................................................................................................ .30 2-5 Concrete breakout failure for an anchor loaded in shear...................................................31 2-6 Determination of AVco based on the 35 failure cone.........................................................31 2-7 Shear load oriented (a) perpendicular to the edge and (b) parallel to the edge.................32 3-1 Schematic of test apparatus................................................................................................44 3-2 Front elevation of test apparatus........................................................................................44 3-3 Plan view of test apparatus................................................................................................45 3-4 Side elevation of test apparatus..........................................................................................45 3-5 Lever arm for the calculation of bolt flexure.....................................................................46 3-6 Adjusted cover based on a si ngle anchor and 35 failure cone..........................................46 3-7 Development of the projected failure ar ea for the group of anchors around a circular foundation..................................................................................................................... .....47 3-8 Two anchor arrangement displays the mi nimum spacing such that no overlap of the failure cones occurs........................................................................................................... .47 3-9 Overlap of failure cones................................................................................................... ..47 3-10 The contribution of the legs of the failure cone to AVc along a straight edge decreases as the number of bolts increases........................................................................48 3-11 Overlap of failure cone s for a circular foundation.............................................................48 3-12 Assembled test specimen...................................................................................................49 3-13 Shaft with pipe apparatus attached prior to instrumentation being attached.....................49

PAGE 9

9 4-1 Method for the determination of the tension, TCFRP, that must be resisted by the CFRP wrap...................................................................................................................... ...55 4-2 Instrumentation layout on the base plate...........................................................................56 4-3 Instrumentation layout on face of shaft.............................................................................56 4-4 Instrumentation layout on rear of shaft/face of concrete block.........................................57 4-5 Instrumentation layout of pipe at load location.................................................................57 4-6 Location of LVDTs D1V, D4, and D7 on the test specimen.............................................58 4-7 Strain gage layout on base plate.........................................................................................58 4-8 Strain gage on base plate of test specimen.........................................................................59 5-1 Initial cracks on face of shaft............................................................................................ .66 5-2 Initial cracks on face and side of sh aft (alternate view of Figure 5-1)..............................66 5-3 Face of test specimen afte r testing exhibits cracks be tween the bolts along with the characteristic concre te breakout cracks.............................................................................67 5-4 Crack pattern on face of sh aft after testing depicts char acteristic concrete breakout failure cracks................................................................................................................. .....67 5-5 Applied Torsion vs. Plate Rotation PlotInitial Test.........................................................68 5-6 Applied Torsion vs. Bolt Strain Plots fo r each bolt at the appropriate location on the base plate with Applied Torsi on vs. Plate Rotation plot in center (full size plots in Appendix C).................................................................................................................... ...69 5-7 Bolt Strain Comparison Plot for Initial Test exhibits the redistribution of the load coinciding with crack formations.......................................................................................70 5-8 Shaft with the CFRP wrap applied prior to testing............................................................70 5-9 Applied Torsion vs. Plate Rotation PlotRetrofit Test......................................................71 5-10 Shaft exhibiting characte ristic torsion cracks from face to base of shaft..........................71 5-11 Applied Torsion vs. Bolt Strain plots fo r the Retrofit Test at the appropriate bolt location around the base plate with Applied Torsion vs. Pl ate Rotation plot in center (full size plots in Appendix D)...........................................................................................72 5-12 Bolt bearing on the bottom of the base plate during loading.............................................73

PAGE 10

10 5-13 Bolt Strain Comparison plot for the retrofit test exhibits slope changes at milestone loads.......................................................................................................................... .........73 5-14 Face of shaft after test illustrates yielding of bolts, concrete breakout cracks around the perimeter, and torsion cracks in the center..................................................................74 5-15 Torsion cracks along length of the shaft after the test.......................................................74 A-1 Dimensioned front elevation drawing of test apparatus....................................................77 A-2 Dimensioned plan drawing test apparatus.........................................................................78 A-3 Dimensioned side elevation drawing of test apparatus......................................................79 A-4 Dimensioned pipe apparatus drawing................................................................................80 A-5 Dimensioned channel tie-down drawing...........................................................................81 C-1 Applied Torsion vs. Rotation Plot...................................................................................108 C-2 Bolt Strain Comparison Plot............................................................................................108 C-3 Applied Torsion vs. Strain Plots for each bolt location...................................................109 D-1 Applied Torsion vs. Rotation Plot...................................................................................115 D-2 Bolt Strain Comparison Plot............................................................................................115 D-3 Applied Torsion vs. Strain Plots for each bolt location...................................................116

PAGE 11

11 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Engineering ANCHOR EMBEDMENT REQUIREMENTS FOR SIGNAL/SIGN STRUCTURES By Kathleen M. Halcovage May 2007 Chair: Ronald A. Cook Major: Civil Engineering During the 2004 hurricane season, several anchor embedment failures of the foundations of cantilever sign structures occurred. The purpose of this research program was to determine the cause of the failure of those f oundations. After a thorough literature review, in conjunction with site investigation, and testing, it was determined that the failure or iginated from the shear load on the anchors directed parallel to the edge of the foundation. Th e shear load resulted from the torsion loading on the anchor gr oup that occurred during the hurrica ne. Investigation of this failure mode, based on the ACI 318-05 Appendix D pr ovisions for concrete breakout of anchors, indicated that this is a failure mode not consid ered in the current design procedures for these types of foundations. Furthermore, it was determin ed that it very well describes the type of failure noted in the field investigation. A test specimen was designed to preclude other possible failur e modes not exhibited in the field (e.g. steel failure of the anchors, bending of the anchors, and torsional failure of the foundation). The results of the testing indicate d the failure of the foundations was caused by concrete breakout due to shear on the anchors dir ected parallel to the fr ee edge of the foundation. The test specimen failed at the torsion pr edicted by the ACI 318-05 Appendix D provisions based on the expected mean strength of the anch ors for concrete breakout with shear directed

PAGE 12

12 parallel to the free edge. Additi onally, the cracks that formed were the same type as those noted in the field investigation, and matched the e xpected pattern for concrete breakout failure. After failure, additional testing was performed to determine a viable repair/retrofit option. The repair/retrofit option used a carbon fiber reinforced polymer (CFRP) wrap around the top of the foundation. The results of this testing indicated that this repair/retrofit technique strengthens the foundation such that it not only meets its initial capacity for concrete breakout, but, also, can exceed this capacity. The results of this test led to the development of guidelines for the evaluation and repair/retro fit of existing foundations.

PAGE 13

13 CHAPTER 1 INTRODUCTION During the 2004 hurricane season, the failure of foundations of cantilever sign structures occurred along Florida highways (Figure 1-1). These failures necessitated a review of the current design and construction procedures for the foundations of cantilever sign structures. The main objective of this research program was two-fold: to determine the cause of the failure of the cantileve r sign structures; and, to propose a re trofit option for the foundation. In order to fulfill this objective, a thorough liter ature review, site inve stigation of a failed foundation, and experimental program were conducted. The findings of the literatu re review and site investigation were used to develop the experimental program. The findings of the experimental program were applied in th e development of the retrofit guidelines. Furthermore, this project tested whethe r or not the ACI 318-05, ACI (2005), Appendix D provisions for anchorage to concrete are applicable for circular foundations.

PAGE 14

14 Figure 1-1. Failed cant ilever sign structure

PAGE 15

15 CHAPTER 2 BACKGROUND While there have not been published repor ts detailing failures of sign structure foundations, such as those being investigated in this study, information on the behavior of anchor installations under various load conditions was found. The main subjects of much of the literature were the effects of fatigue and wind load on overhead sign structures. Additionally, there have been studies conducted on the failure modes of anchor installati ons, but these findings were not based on circular foundations. In later sections, one of these anchorage failure modes will be introduced for application in this research program. This chapter presents the findings of the lite rature review, the conclusions drawn based on a site investigation of a failed foundation, and ap plicable design equations for the determination of the failure mode. The information presented in the chapter served as the base upon which the experimental program was developed. 2.1 Literature Review Keshavarzian (2003) explores the wind design requirements and safety factors for utility poles and antenna monopoles from various sp ecification manuals. It was found that the procedure outlined in ASCE (1991), ASCE 74Guidelines for Electrical Transmission Line Structural Loading resulted in the smallest factor of safety for the design. AASHTO (2001), Standard Specifications for Stru ctural Supports for Highway Signs, Luminaires and Traffic Signals was used as a part of the comparison fo r the design of the antenna monopole. The design from this specification was compar ed to that from ASCE (2000), ASCE 7-98Minimum Design Loads for Buildings and Other Structures ; TIA/EIA (1996), Structural Standards for Steel Antenna Towers and Antenna Supporting Structures ; and, ASCE 74. The wind forces at the base were the same for ASCE 74, AASHTO, and ASCE 7-98. The forces using TIA/EIA

PAGE 16

16 were higher because it requires that a 1.69 gust response factor be applied to the design. Therefore, the pole designed using TIA/EIA would have be tween 30 and 40 percent extra capacity. ASCE 7-98 and AASHTO resulted in the same margin of safety. The paper did not include findings that were completely relative to this project, but it pr ovided additional sources for design of structures for comparative purposes. Keshavarzian and Priebe (2002) compares th e design standards specified in ASCE (2000), ASCE 7-98, and IEEE (1997), NESCNational Electrical Safety Code The NESC does not require that utility poles measuring less than 60 feet in height be designed for extreme wind conditions. Short utility pol es were designed to satisfy NE SC specifications (i.e. without extreme wind conditions). The poles were then evaluated according to th e ASCE 7-98 wind load requirements. It was found that the poles did not meet the ASCE 7-98 requirements. Therefore, it was recommended that the exclusion for short u tility poles in the NESC be reevaluated. The paper also mentioned AASHTO (1994), Standard Specifications for Structural Supports for Highway Signs, Luminaires and Traffic Signals It outlined that in the AASHTO specification, support structures exceeding 50 feet and overhead sign structures mu st be designed for a 50-year mean recurrence interval, or extreme wind loading condition. MacGregor and Ghoneim (1995) presents the background information for the formulation of the thin-walled tube space truss analogy design method for torsion that was first adopted into ACI (1995), ACI 318-95. The design methodology was adopted because it was simpler to use than the previous method and was equally accura te. The basis for the derivation of the new method was based on tests that were conducted in Switzerland. Both solid and hollow beams were tested during that researc h. In comparing the data from both tests, it was discovered that after cracking the concrete in the center had littl e effect on the torsional strength of the beam.

PAGE 17

17 Therefore, the center of the cros s-section could be ignored, and th e beam could be idealized as a hollow tube. A space truss was formed by longitudinal bars in the corners, the vertical closed stirrups, and compression diagonals. The compression dia gonals were spiraled around the member and extended between the torsion cracks. The paper also explained the shear stresses created by torsion on the member. In addition to the derivation of the equations for torsion and shear, the authors discussed the limits for when torsion should be consider ed and the requirements for minimal torsional reinforcement. The tests, conducted on both reinfo rced and prestressed concrete beams, showed that there was acceptable agreement between th e predicted strengths, as determined by the derived equations, and the test results. This agreement was comparable to the design equations from the ACI Code. In addition to these papers, other reports re viewed include Lee and Breen (1966), Jirsa et al. (1984), Hasselwander et al. (1977), and Br een (1964). These four studies focused on important information regarding an chor bolt installations. Other re ports that were examined for relevance were from the National Cooperative Highway Research Program (NCHRP). These are: Fouad et al. (1998), NCHR P Report 411; Kaczinski et al (1998), NCHRP Report 412; and, Fouad et al. (2003), NCHRP Report 494. Fouad et al. (2003) details the findings of NCHRP Project 17 -10(2). The authors stated that AASHTO (2001) does not detail design requirements for anc horage to concrete. The ACI anchor bolt design procedure was also reviewe d. Based on their finding s, they developed a simplified design procedure. This procedure was based on the assumptions that the anchor bolts are hooked or headed, both longitudinal steel and hoop steel are presen t in the foundation, the

PAGE 18

18 anchor bolts are cast inside of the reinforcement, the reinforc ement is uncoated, and, in the case of hooked bolts, the length of the hook is at least 4.5 times the an chor bolt diameter. If these assumptions did not apply, then the simplified pr ocedure was invalid. Th e anchor bolt diameter was determined based on the tensile force on the bolt, and the required length was based on fully developing the longitudinal reinforcement betw een the embedded head of the anchor. The authors further stated that shea r loads were assumed to be neg ligible, and concrete breakout and concrete side face blowout were controlled by adequate longitudinal a nd hoop steel. The design procedure was developed based on tensile loadin g, and did not address the shear load on the anchors directed parallel to th e edge resulting from torsion. Additionally, the authors presented the freque ncy of use of differ ent foundation types by the state Departments of Transportation, expresse d in percentages of st ates reporting use. According to the survey the most common f oundation type used fo r overhead cantilever structures was reinforced cast-in-place drilled shafts (67-100%) followed by spread footings (3466%) and steel screw-in foundations (1-33%). None of the states reported the use of directly embedded poles or unreinforced cast-in-place drilled shafts. ASCE (2006), ASCE/SEI 48-05, entitled Design of Steel Trans mission Pole Structures was obtained to gather information on the foundation design for transmission poles structures. The intent was to determine whether or not the desi gn of such foundations was relevant to the evaluation of the foundations under examination in this research. In .0 of the standard, the provisions for the structural members and connec tions used in foundations was presented. Early in the section, the standard stated that the in formation in the section was not meant to be a foundation design guide. The pr oper design of the foundation mu st be ensured by the owner based on geotechnical principles. The section co mmented on the design of the anchor bolts. The

PAGE 19

19 standard focused on the structural stability of th e bolts in the foundation; it looked at bolts in tension, bolts in shear, bolts in combined tensi on and shear, and the develo pment length of such bolts. The standard did not present provi sions for failure of the concrete. 2.2 Site Investigation A site investigation was conducted at the site of one failed overhead cantilever signal/sign structure located at Exit 79 on Interstate 4 in Orlando (Figure 2-1). Figure 2-2 is the newly installed foundation at the site. The faile d foundation had the same anchor and spacing specifications as the new foundation. This site visit coincided w ith the excavation of the failed anchor embedment. During the course of the excavation the following information was collected: The anchor bolts themselves did not fail. Rather, they were leaning in the foundation, which was indicative of a torsional load on the foundation. While th e integrity of the anchor bolts held up during the wind loadi ng, the concrete between the bolts and the surface of the foundation was cr acked extensively (Figure 2-3). The concrete was gravelized between the anchors and the hoop st eel. It should be noted that upon the removal and study of one anchor bolt, it was ev ident that there was no deformation of the bolt itself. The hoop steel did not start at the top of the foundation. It started approximately 15 in. (381 mm) into the foundation. The concrete was not evenly dispersed ar ound the foundation. The hoop steel was exposed at approximately three to four feet below grade. On the opposite side of the foundation there was excess concrete. It was assumed that during the constr uction of the foundation, there was soil failure allowing a portion of the side wall to displace the concrete. 2.3 Applicable Code Provisions The initial failure mode that was focused on in the background review was torsion. However, based on the results of the site invest igation, it was determined that the most likely cause of failure was concrete breakout of an anchor (Figure 2-4). The equations for torsion are presented in this section as they were used during the design of the experimental program to prove that the concrete breakout failure wi ll occur before the torsional failure.

PAGE 20

20 2.3.1 Cracking and Threshold Torsion Torsion is the force resulting from an applied torque. In a circular section, such as the foundation under review, the resultin g torsion is oriented perpendicu lar to the radius or tangent to the edge. ACI (2005), ACI 318-05, details the equation for the cracking torsion of a nonprestressed member. In R11.6.1, the equation for the cracking torsion, Tcr, is given (Equation 2-1). The equation was developed by assuming that the concrete will crack at a stress of 4 fc. cp cp c crp A f T24 (2-1) Where Tcr = cracking torsion (lb.-in.) fc = specified compressive streng th of the concrete (psi) Acp = area enclosed by the outside perimete r of the concrete cross-section (in.2) = r2, for a circular section with radius r (in.) pcp = outside perimeter of the c oncrete cross-section (in.) = 2 r for a circular section with radius r (in.) This equation, when applied to a circular se ction, results in an equivalent value when compared to the basic equation (Equation 2-2) for torsion noted in Roark and Young (1975). The equality is a result of taking the shear stress as 4 fc. 23r T (2-2) Where T = torsional moment (lb.-in.) = shear stress, 4 fc, (psi) r = radius of concrete cross-section (in.) ACI 318-05 .6.1(a) provides the threshold torsion for a nonprestressed member (Equation 2-3). This is taken as one-quarter of the cracking torsion. If the factored ultimate torsional moment, Tu, exceeds this threshold torsion, then the effect of torsion on the member must be considered in the design.

PAGE 21

21 cp cp cp A f T2 (2-3) Where = strength reduction factor AASHTO (2004), AASHTO LRFD Bridge Design Specifications also presents equations for cracking torsion (Equation 2-4) and threshold torsion (E quation 2-5). Equation 2-4 corresponds with the AASHTO (2004) equation for cracking torsion with the exception of the components of the equation related to prestressing. That portion of the equation was omitted since the foundation was not prestre ssed. It must be noted that th ese equations are the same as the ACI 318-05 equations. c cp c crp A f T2125 0 (2-4) Where Tcr = torsional cracking moment (kip-in.) Acp = total area enclosed by outside perimete r of the concrete cross-section (in.2) pc = the length of the outside perimete r of the concrete section (in.) AASHTO (2004) also specifies the same provision as ACI 31805 regarding the threshold torsion. In .8.2.1, it characterizes the threshold torsion as one-qua rter of the cracking torsion multiplied by the reduction factor. Equation 25 corresponds with the threshold torsion portion of AASHTO (2004) equation. crT T 25 0 (2-5) The above referenced equations considered the properties and dimensions of the concrete. They did not take into cons ideration the added strength provided by the presence of reinforcement in the member. For the purposes of this research, it was im portant to consider the impact of the reinforcement on the strength of the concrete shaft.

PAGE 22

22 2.3.2 Nominal Torsional Strength ACI 318-05 .6.3.5 states that if the ultimate factored design torsion exceeds the threshold torsion, then the design of the section mu st be based on the nominal torsional strength. The nominal torsional strength (Equation 2-6) takes into account the contribution of the reinforcement in the shaft. cot 2 s f A A Tyt t o n (2-6) Where Tn = nominal torsional moment strength (in.-lb.) Ao = gross area enclosed by shear flow path (in.2) At = area of one leg of a closed stirr up resisting torsion with spacing s (in.2) fyt = specified yield strength fy of transverse reinforcement (psi) s = center-to-center spac ing of transverse reinforcement (in.) = angle between axis of strut, compre ssion diagonal, or compression field and the tension chord of the member The angle, is taken as 45, if the member under consideration is nonprestressed. This equation, rather than taking into account the properties of the concrete, ta kes into account the properties of the reinforcement in the member. These inputs include the area enclosed by the reinforcement, the area of the reinforcement, th e yield strength of the reinforcement, and the spacing of the reinforcement. For the purpos e of this research, the reinforcement under consideration was the hoop steel. AASHTO (2004) also outlines provisions for the nominal torsional resistance in .8.3.6.2. Equation 2-7 is the same equation that ACI 31805 presents. The only difference is in the presentation of the equations. The vari ables are represented by different notation. s f A A Ty t o n cot 2 (2-7) Where Tn = nominal torsional moment (kip-in.) Ao = area enclosed by the shear flow path, including any area of holes therein (in.2) At = area of one leg of closed transverse torsion reinforcement (in.2) = angle of crack

PAGE 23

23 As the above referenced equation evidences, the ACI 318-05 and the AASHTO (2004) provisions for nominal torsional strength are the same. Based on the code provisions, the nominal torsional strength represents the torsional strength of the cross-section. 2.3.3 Combined Shear and Torsion Another area that had to be considered in th is research was the effect of combined shear and torsion. Both ACI 318-05 and AASHTO (2004) outline equations for the combined shear and torsion. Since the foundation ha d a shear load applied to it, it had to be determined whether or not the shear load was large enough to n ecessitate consideration. The ACI 318-05 equation (Equation 2-8) and the AASHTO (2004) equation (Equation 2-9) are presented hereafter. The ACI 318-05 equation is located in .6.3.1 of ACI 318-05, and the AASHTO (2004) equation is presented in .8.3.6.2 of that specification. The ACI 318-05 equation is presented with Vu substituted on the left-hand side. 2 2 27 1 oh h u w u uA p T d b V V (2-8) Where Vu = factored shear force at section (lb.) bwd = area of section resi sting shear, taken as Aoh (in.2) Tu = factored torsional mome nt at section (in.-lb.) ph = perimeter of centerline of outermost closed transverse torsional reinforcement (in.) Aoh = area enclosed by centerline of the outer most closed transverse torsional reinforcement (in.2) The AASHTO (2004) equation that is presen ted (Equation 2-9) is intended for the calculation of the factored shear fo rce. For the purpose of this proj ect, the right-hand side of the equation was considered.

PAGE 24

24 2 22 9 0 o u h u uA T p V V (2-9) Where Vu = factored shear force (kip) ph = perimeter of the centerline of the closed transverse reinforcement (in.) Tu = factored torsiona l moment (kip-in.) The determination of whether or not shear had to be considered was made based on a comparison of the magnitudes of the coefficients of these terms. This is investigated further in Chapter 3. 2.3.4 ACI Concrete Breakout Strength for Anchors In ACI 318-05 Appendix D, the concrete breakou t strength is defined as, the strength corresponding to a volume of concrete surroundi ng the anchor or group of anchors separating from the member. A concrete breakout failure can result from either an applied tension or an applied shear. In this report, the concrete break out strength of an anchor in shear, D.6.2, will be studied. The breakout strength for one anchor loaded by a shear for ce directed perpendicular to a free edge (Figure 2-5) is given in Equation 2-10. 5 1 1 2 07a c o o e bc f d d V (2-10) Where Vb = basic concrete breakout stre ngth in shear of a single anch or in cracked concrete (lb.) e = load bearing length of anchor for shear (in.) do = outside diameter of anchor (in.) ca1 = distance from the center of an anchor sh aft to the edge of concrete in one direction; taken in the di rection of the applied shear (in.) The term e is limited to 8 do according to D.6.2.2. The equations in ACI 318-05 were developed based on a 5% fractile and with the stre ngth in uncracked concrete equal to 1.4 times the strength in cracked concrete. The mean conc rete breakout strength in uncracked concrete is provided in Fuchs et al. (1995) and given in Equation 2-11.

PAGE 25

25 5 1 1 2 013a c o o e bc f d d V (2-11) For a group of anchors, Equation 2-12 applies. This equation is the nominal concrete breakout strength for a group of anchors load ed perpendicular to the edge in shear. b V c V ed V ec Vco Vc cbgV A A V, , (2-12) Where Vcbg = nominal concrete breakout strength in shear of a group of anchors (lb.) AVc = projected concrete failure area of a single anchor or group of anchors, for calculation of strength in shear (in.2) AVco = projected concrete failure area of a singl e anchor, for calculati on of strength in shear, if not limited by corner influences, spacing, or member thickness (in.2) = 4.5(ca1)2, based on a 35 failure cone (Figure 2-6) ec,V = factor used to modify shear strength of anchors based on eccentricity of applied loads, ACI 318-05 D.6.2.5 ed,V = factor used to modify shear strength of anchors based on proximity to edges of concrete member, ACI 318-05 D.6.2.6 c,V = factor used to modify shear strength of anchors based on presence or absence of cracks in concrete and presence or absence of supplementary reinforcement, ACI 318-05 D.6.2.7, accounted for in Equation 2-11 The resultant breakout strength is for a shear load directed perpendicular to the edge of the concrete. Therefore, an adjustment had to be ma de to account for the shear load acting parallel to the edge since this was the type of loading th at resulted from the torsion on the anchor group. In D.5.2.1(c) a multiplication factor of two is pres cribed to convert the va lue to a shear directed parallel to the edge (Figure 2-7). Fuchs et al. (1995) notes that the multiplier is based on tests, which indicated that the shear load that can be resisted when applied parallel to the edge is approximately two times a shear load applied perpendicular to the edge. In order to convert the breakout strength to a torsion, the dimensions of the test specimen were considered to calculate what was calle d the nominal torsional moment based on the concrete breakout strength, Tn,breakout.

PAGE 26

26 2.3.5 Alternate Concrete Breakout Strength Provisions In the book Anchorage in Concrete Construction Eligehausen et al. (2006), the authors presented a series of equations fo r the determination of the concrete strength based on a concrete edge failure. These equations are presented in Ch apter 4, .1.2.4 of the text. Equation 2-13 is the average concrete breakout streng th of a single anchor loaded in shear. It must be noted that this equation is for uncracked concrete. 5 1 1 5 0 200 0 ,0 3a cc e o c uc f d V (2-13) Where V0 u,c = concrete failure load of a n ear-edge shear loaded anchor (N) do = outside diameter of anchor (mm) e = effective load transfer length (mm) fcc200 = specified concrete compressive st rength based on cube tests (N/mm2) 1.18fc ca1 = edge distance, measured from the longitudinal axis of the anchor (mm) = 5 0 11 0 a ec = 5 0 11 0 a oc d As was the case for the ACI 318-05 equations, the term e is limited to 8do. Equation 2-14 accounts for the group effect of the anchors loaded concentrically. The authors stated that cases where more than two anchors are present have not been extensively studied. They did, however, state that the equation should be a pplicable as long as there is no slip between the anchor and the base plate. 0 , c u Vco Vc c uV A A V (2-14) Where AVc = projected area of failure surface for the anchorage as defined by the overlap of individual idealized failu re surfaces of adjacent anchors (mm2) AVco = projected area of the fully develope d failure surface for a single anchor idealized as a half-pyramid with height ca1 and base lengths 1.5ca1 and 3ca1 (mm2)

PAGE 27

27 ACI 318-05 specifies that, in orde r to convert the failure shear directed perpendicular to the edge to the shear directed parallel to the ed ge, a multiplier of two be applied to the resultant load. The provisions outlined in this text take a more in-depth approach to determining this multiplier. The method for calculati ng this multiplier is detailed in .1.2.5 of Eligehausen et al. (2006). The authors stated that, based on previous research, the concrete edge breakout capacity for loading parallel to an edge is approximately two times the capacity for loading perpendicular to the edge if the edge distance is constant. The authors further moved to outline equations to calculate the multiplier based on the angle of load ing. The first equation (Equation 2-15) that is presented in the text is a generalized approach for calculating the multiplier when the angle of loading is between 55 and 90 of the axis perpendicular to the edge. For loading parallel to the edge the angle is classified as 90 (Figure 2-7). sin 5 0 cos 1, V (2-15) Where ,V = factor to account for the angle between the shea r load applied and the direction perpendicular to th e free edge of the concrete member = angle of the shear load with re spect to the perpendicular load This equation results in a f actor of two for loading parall el to the edge. Equation 2-16 provides the concrete breakout strength for shear directed paralle l to the edge using ,V. c u V V ucV V, , (2-16) Where Vuc, V= concrete failure load for shear dire cted parallel to an edge based on ,V (N) An alternate equation for calculating this factor is also presented in the Eligehausen et al. (2006) text. This equation is only valid for loadi ng parallel to the edge. This equation is based on research proposing that the multiplier to calcul ate the concrete breakout capacity for loading parallel to the edge based on the capacity for load ing perpendicular to the edge is not constant.

PAGE 28

28 Rather, it suggested that it is based on the conc rete pressure generated by the anchor. The base equation for the application of this factor is Equation 2-17. c u parallel parallel c uV V, , (2-17) Where Vu,c,parallel = concrete failure load in the case of shear parallel to the edge (N) parallel = factor to account for shear parallel to the edge Vu,c = concrete failure load in the case of shear perpendicula r to the edge (N) Equation 2-18 is used for the determ ination of the conversion factor parallel. 5 0 2 44 c u cc o parallelV f d n k (2-18) Where k4 = 1.0 for fastenings without hole clearance 0.75 for fastenings with hole clearance n = number of anchors loaded in shear fcc = specified compressive streng th of the concrete (N/mm2) conversion to fc as specified for Equation 2-13 The results of Equation 2-13 through Equa tion 2-18 are presented alongside the ACI 31805 equation results in Chapter 3. These are presented for comparative purposes only. 2.3.6 ACI 318-05 vs. AASHTO LRFD Br idge Design Specifications In Sections 2.3.1 through 2.3.3, both the applicable design equa tions in ACI 318-05 and AASHTO (2004) were presented. As was shown, the ACI and AASHTO equations were the same. Additionally, the provisions for the conc rete breakout failure capacity are only provided in ACI 318-05. AASHTO does not provide design guidelines for this failure. Therefore, the ACI 318-05 equations were used throughout the course of this research program.

PAGE 29

29 Figure 2-1. Cantilever sign structure at Exit 79 on Interstate 4 in Orlando Figure 2-2. New foundation installed at the site

PAGE 30

30 Figure 2-3. Failed foundation during post-failure excavation Figure 2-4. Concrete breakout of an anchor caused by shear direct ed parallel to the edge for a circular foundation

PAGE 31

31 Vb ca1 Vb Concrete Edge Figure 2-5. Concrete breakout failure for an anchor loaded in shear Vb ca1 35 1.5ca1 1.5ca1 Vb 35 1.5ca1 AVco 1.5ca1 1.5ca1 1.5ca1 AVco=1.5ca1(1.5ca1) =4.5(ca1)2 Figure 2-6. Determination of AVco based on the 35 failure cone

PAGE 32

32 Vb a Vb Perpendicular Axis 90 b Figure 2-7. Shear load oriented (a) perpendicula r to the edge and (b) parallel to the edge

PAGE 33

33 CHAPTER 3 DEVELOPMENT OF EXPERIMENTAL PROGRAM After a thorough background investigation, it was determined that the most likely cause of the failure was the concrete breakout of an anchor loaded by a shear force directed parallel to a free edge. The shear force on the individual anc hors was caused by torsi on applied to the bolt group from the sign post. Based on this determina tion, an experimental program was formulated to determine if this was in fact the failure m ode of the foundation. Therefore, it was of the utmost importance to design the test apparatus to preclude other failure modes. This chapter focuses on the development of the experimental program. 3.1 Description of Test Apparatus The test apparatus was designed such that the field conditions could be closely modeled for testing at the Florida Departme nt of Transportation (FDOT) Structures Research Center. A schematic of the test apparatus is shown in Figure 3-1. The test apparatus consisted of: A 30 (762 mm) diameter concrete shaft that extended 3-0 (914 mm ) outward from the concrete block Twelve 37 (940 mm), 1.5 (38.1 mm) diam eter F1554 Grade 105 anchor bolts embedded into the concrete around a 20 (508 mm) diameter A 16 (406 mm) diameter steel pipe assembly welded to a 24 (610 mm) diameter, 1 (25.4 mm) thick steel base plate with holes drilled for the anchor bolts to provide the connection between the bolts and pipe assembly A 6-0 x 10-0 x 2-6 (1830 mm x 3050 mm x 762 mm) reinforced concrete block to provide a fixed support at the base of the shaft Two assemblies of C12x30 steel channels and pl ates to attach the block to the floor The base for the design of the various component s of the test apparatu s was one half of the size of the failed foundation invest igated during the site visit. The dimensions of the field foundation are presented in Table 3-1. From that point, the elements of the test apparatus were designed to preclude all failure modes other than the concrete breakout failure of the anchors.

PAGE 34

34 Information pertaining to the design of the comp onents of the apparatus is presented in the following sections. Figures 3-2 through 3-4 are drawings of the test apparatus. For large scale dimensioned drawings, reference Appendix A. Complete design calculations are located in Appendix B. 3.2 Shaft Design The starting point for the design of the conc rete shaft was based on developing a test specimen approximately one half of the size of the foundation that was in vestigated during the site visit. From there, the various components of the shaft we re designed the meet the ACI 31805 requirements, and to prevent failure before the concrete br eakout strength was reached and exceeded. All of the strengths were calculated us ing a concrete strength of 5500 psi (37.9 MPa), which was the strength indicated on the FDOT standard drawings. 3.2.1 Torsion Design The basic threshold torsional st rength of the shaft, 24.6 kipft (33.4 kN-m) was calculated using the ACI 318-05 torsional strength equation (Equation 2-3). This strength, however, did not take into account the rein forcement in the shaft. Therefore, it was assumed that the threshold torsion would be exceeded. As a result, the to rsional strength of the shaft was based on the nominal torsional strength. In order to calculate the tors ional strength that the shaft w ould exhibit during testing, the ACI nominal torsional strength equation was applie d. Before the strength was calculated, the minimum requirements for the shaft reinforcement were followed as outlined in ACI 318-05 .10.5.6 and .6.5.1. The nominal torsional st rength (Equation 2-6) was then calculated for the specimen. This value, 253 kip-ft (343 kN -m), was compared to the concrete breakout strength. The spacing of the hoop steel in the shaft was altered until the nominal torsional strength exceeded the concrete br eakout strength. Hence, if the concrete breakout failure was the

PAGE 35

35 correct failure mode, it would occur before th e torsional capacity of the shaft was exceeded during testing. 3.2.2 Longitudinal and Transverse Reinforcement As was outlined in the previous section, the required amount of hoop steel to meet the ACI 318-05 specifications was determined using guidelin es from Chapters 7 and 11 in the code. The resultant hoop steel layout was twenty-four #4 bars spaced evenly around a 27 in. (686 mm) diameter circle. The transverse hoops were co mprised of #3 bars at 2.5 in. (635 mm) totaling fourteen #3 bar hoops. The re quired splice for the #3 bar was 12 in. (305 mm), and the development length required for the #4 bar into the c oncrete block was 8 in. (203 mm) with a 6 in. (152 mm) hook. In the test setup, the #4 bars extended 27 in. (686 mm) into the block, which exceeded the required length. This length was us ed for simplicity in design and construction of the test setup. The #4 bars were tied into one of the cages of reinforcement in the concrete block. 3.2.3 Flexure Due to the eccentric loading of the bolts, the flexural capacity of the shaft had to be calculated. It had to be determ ined that the shaft would not fail in flexure under the load applied during testing. The flexural reinforcement in th e shaft was the longitudinal reinforcement, the #4 bars. The first step to determine the capacity wa s to assume the number of bars that would have yielded at the time of failure. From that point, the neutral axis of the shaft was located following the ACI 318-05 concrete stress bl ock methodology presented in Chap ter 10 of the code. It was then checked if the number of bars that ha d yielded was a good assumption. Once this was verified, the nominal moment capacity of the shaf t was calculated, and, then, compared to the maximum flexural moment based on the concrete br eakout capacity. The fle xural capacity of the shaft, 262 kip-ft (355 kN-m), exceeded the maximu m flexural moment on the shaft, 60.6 kip-ft (95.2 kN-m).

PAGE 36

36 3.3 Anchor Design 3.3.1 Diameter of Anchor Bolts The starting point for the diameter of the F 1554 Grade 105 anchor bolts to be used in the test apparatus was based on half the diameter of those in the field specimen. The size determined using that methodology was 1 in. (25.4 mm). On ce the concrete breakout strength capacity of the anchors was determined, the corresponding shear load on each of the bolts was calculated. The anchor bolt diameter had to be increased to 1. 5 in. (38.1 mm) in order to ensure that the bolts would not experience steel failure in flexure or shear. Th e maximum flexure on the bolts was calculated by taking the maximum shear applie d to each bolt and calculating the corresponding maximum flexural moment (Figure 3-5). The lever arm (Equation 3-1) for the calculation of the capacity was defined in Eligehausen et al. (2006) Section 4.1.2.2 b. 3 1a e l (3-1) Where: l = lever arm for the shear load (in.) e1 = distance between the shear load and surface of concrete (in.) a3 = 0.5do, without presence of a nut on surface of concrete, Figure 3-5 (in.) 0, with a nut on surface of concrete The base plate was restrained against rotati on, and translation wa s only possible in the direction of the applied shear load. The maxi mum applied moment for each bolt was calculated based on these support conditions and the lever arm calculation. Full fixity occurred a distance a3 into the shaft. Using the section modulus of the bolts, the stre ss was then calculated and compared to the yield strength of the bolts, 105 ksi (724 MPa). The shear strength of the bolts was calculated using the provisions in Appendix D of ACI 318-05. In both cases it was determined that the bolts had sufficient strength.

PAGE 37

37 3.3.2 Concrete Breakout Strength of Anchor in Shear Parallel to a Free Edge The breakout strength provisions outlined in ACI 318-05 Appendix D and the breakout provisions introduced in Eligehause n et al. (2006) were applied to the design of the shaft. Equation 2-11, from ACI 318-05, was used as th e primary equation for the calculation of the breakout strength. In order to apply the ACI prov isions to the circular foundation a section of the concrete was ignored (Figure 3-6). If the full cover, c, was used in the calculation, the failure region would have included area outside of the circ le. Rather than extending beyond the edge of the concrete, the 35 degree failure cone (Figure 2-6) was extended to the edge of the shaft as shown in Figure 3-6. Equation 3-2 was deve loped to determine the adjusted cover, ca1. 25 3 25 32 2 2 1 b b b ar r r r c (3-2) Where rb = radius measured from the centerline of the bolt to the center of the foundation (in.) (Figure 3-6) r = radius of circular foundation (in.) As presented in Section 2.3.4, the projected concrete failure area for a single anchor, AVco, is equivalent to 4.5(ca1)2. Figure 3-7 illustrates the development of the projected concrete failure area for a group of anchors, AVc, as a function of the number of bolts, n the radius of the shaft, r and the adjusted cover. The resultant concrete breakout stre ngth using the adjusted cover approach was conservative relative to assuming the full cover. Equation 3-3 and Equation 3-4 are used to calculate the concrete breakout torsion, Tn,breakout, and are based on the ACI provisions fo r shear parallel to the free edge. For b ar c A1 15 1 sin b b Vco Vc breakout nr V A A T 2, (3-3)

PAGE 38

38 For b ar c A1 15 1 sin (i.e. no overlap of failure cones) b b breakout nr V n T 2, (3-4) Where A = angle of circular sector fo r each bolt (deg) (Figure 3-7) ca1 = adjusted cover (in.) (Equation 3-2) rb = radius measured from the centerline of the bolt to the center of the foundation (in.) (Figure 3-6) AVc = projected concrete failure area of a group of anchors (in.2) (Figure 3-7) AVco = projected concrete failure area of a single anchor (in.2) (Figure 3-6) Vb = concrete breakout strength in shear for a single anchor calculat ed using Equation 211 with ca1 as calculated in Equation 3-2 (lb.) n = number of bolts Using Equation 3-3, the ACI concrete brea kout torsion for the test specimen was determined to be 182 kip-ft (247 kN-m), whic h was less than the nominal torsional capacity. During the analysis of the design equations, an issue arose regarding the calculation of the Eligehausen et al. (2006) factor parallel. The result of Equation 2-18 was 4.06 compared to the ACI 318-05 factor and ,V of 2.0. This prompted an inve stigation of the application of the multiplier to the circular founda tion in this research program. The majority of the tests for the determination of Vu,c (Equation 2-14) were for groups of two bolts. Therefore, it was investigated how the AVc/AVco term is affected by the spacing between the bolts and the numb er of bolts. Figure 3-8 shows that for spacing, s, of 3.0ca1 or greater there is no overlap of the breakout cones. In those cases the strength is the sum of the single anchor strengths. Figur e 3-9 illustrates the overlap of the breakout cones. The AVc/AVco term is used to calculate the breakout strength for the case where the failure cones overlap. AVc/AVco can be normalized by dividing by the number of bolts. An increase in the number of bolts at the same spacing along a straight edge leads to a reduction in the normalized AVc/AVco term. This reduction is illustrated in Figure 3-10. The contribution of the failure cone outstanding legs at the ends of the group area, AVc, decreases as the number of bolts increases.

PAGE 39

39 For a circular foundation, with s<3.0ca1, there is a constant overlap of the failure cones with no outstanding legs (Figure 3-11). The equivalent number of bolts along a straight edge is taken as infinity in order to represen t a circular foundation (i.e. no outsta nding legs). Therefore, the normalized AVc/AVco term for this case was calculated for an infinite number of bolts at the prescribed spacing for the foundation. To convert these ratios into a multiplier for parallel, the ratio of the normalized AVc/AVco for an infinite number of bolts to the normalized term for two bolts was calculated. That multiplier, 0.52, was applied to the parallel term resulting in an adjusted parallel of 2.1. This result ing value agreed with th e ACI 318-05 factor and the Eligehausen et al. (2006) factor ,V of 2.0. The resultant concrete breakout torsions, base d on the Eligehausen et al. (2006) concrete breakout strength (Equation 2-13), were 167 kip-ft (227 kN-m) using parallel of 2.1 in Equation 2-17, and 159 kip-ft (216 kN-m) using ,V of 2.0 in Equation 2-16. These torsions were calculated using the same moment arm, rb, used in Equation 3-3 and Equation 3-4. These results and the results of the other calcula tions are summarized in Table 3-2. 3.3.3 Development Length of the Bolts Another key aspect of the shaft design was to ensure that the anchor bolts were fully developed. In order to meet the code require ments, the splice length between the #4 bars and anchor bolts was calculated us ing the development length eq uations presented in ACI 318-05 Chapter 12. The bolts needed overlap the #4 bars across 26.7 in. (678 mm), and in the test setup the overlap was 29 in. (737 mm). Therefore, this requirement was met. 3.4 Steel Pipe Apparatus Design The components of the steel pipe apparatus included the pipe, which was loaded during testing, and the base plate. The pipe desi gn was based on the inter action between torsion, flexure, and shear as presented in AISC (2001), LRFD Manual of Steel Construction-LRFD

PAGE 40

40 Specification for Steel Hollow Structural Sections. Each of the indi vidual capacities was calculated for various pipe diameters and thickne sses. The individual strengths were compared to the projected failure loads for testing, the concrete breakout failure loads. In addition to verifying that the capacity of the pipe exceeded t hose loads, the interactio n of the three capacities was verified. The purpose was to check that the su m of the squares of the ultimate loads divided by the capacities was less than one. Based on th is analysis, it was conc luded that an HSS 16.000 x 0.500 pipe would provide sufficient strength. In order to load the pipe, it needed to have a ninety degree bend in it. This was achieved by welding two portions of pipe cu t on forty-five degree angles to a steel plate. The weld size for this connection was determined such that the effective throat thickness would equal the thickness of the pipe, which was 0.50 in. (12.7 mm). The factors included in the design of the base pl ate were the diameter of the pipe, required weld size, bolt hole diameter, and the required di stance between the edge of the bolt hole and the edge of the plate. The required width of the weld between the base plate and the pipe was calculated such that the applied to rsion could be transferred to the plate without failing the weld. From that point, the bolt hole location diameter had to be checked to ensure that there was sufficient clearance between the weld and the nuts. It was important that the nuts could be fully tightened on the base plate. A 0.25 in. (6 .35 mm) oversize was specified for the bolt hole diameter. This oversize was based on the standard oversize used in the fiel d. Beyond that point, it was ensured that there would be sufficient cove r distance between the bo lt hole and the edge of the plate. The design of the components of the steel pipe apparatus was crucial because these pieces had to operate efficiently in order to correctly ap ply load to the bolts. If the apparatus were to

PAGE 41

41 fail during testing, the objective of the research could not be achie ved. The weight of the pipe apparatus was calculated in order to normalize the load during testi ng. The load applied to the anchorage would be the load cell reading less the weight of th e pipe apparatus. 3.5 Concrete Block Design The design of the concrete block was based on se veral key factors to ensure that it served its purpose as a fixed support at the base of th e shaft. The amount of reinforcement required was based on a strut-and-tie model of the block as outlined in ACI 318-05 Appendix A and, as an alternate approach, beam theory to check the shear strength and fle xural strength of the block. For the flexural capacity calculations, the ACI 318-05 concrete stress block provisions were utilized. Based on the results of both approaches it was determined that 3 #8 bars, each with a 12 in. (305 mm) hook on both ends, spaced across the top and the bottom of the block were required. Additionally, two cages of #4 bars were placed in the block on the front and back faces meeting the appropriate cover requirements to serve as supplementary reinforcement. The purpose of reinforcing the block was to ensure st ructural stability of the block throughout the testing process. Two channel apparatuses were also designed in order to tie the block to the floor of the laboratory in order to resist overt urning. The loads that had to be resisted by each tie-down were calculated such that the floor capacity of 100 kips (445 kN) per tie-down would not be exceeded. The channels were designed in accordance with the provisions set forth in AISC (2001). The welds between the channels and steel plates ha d to be sufficiently designed such that the channels would act as a singl e unit thereby transferring load from the plates through the channels. Also, the channels were spaced far en ough apart to fit 1.5 in. (38.1 mm) bolts between the channels. A 0.25 in. (6.35 mm) oversize was sp ecified for the spacing of the channels and the holes in the steel plates. The construction drawin gs for the channels are located in Appendix A.

PAGE 42

42 In addition to assuring that th e concrete block system had su fficient capacity to resist the applied load, the bearing strength of the concrete had to be calculated. This was done in order to verify that the concrete would not fail in the region that was in c ontact with the steel channels. The bearing strength was found to be sufficient. As a result, it was conc luded that the concrete block system would efficiently serve as a fixe d connection, and under the loading conditions it would not prematurely fail. 3.6 Combined Shear and Torsion As was presented in Chapter 2, a calculation had to be carried out to ensure that shear need not be considered in the design. Rather than inputting the values for the ultimate shear and ultimate torsion into Equation 2-8, the coefficients of these terms were calculated. The base for doing so was to input the torsion as a function of the shear. For the test specimen, the ultimate torsion, Tu, was taken as the moment arm multiplied by the ultimate shear, Vu. The moment arm for the load was 9 ft. (2740 mm). As an altern ate approach, the actual co ncrete breakout strength and the corresponding shear could have been inputte d into the equation rather than the generic variables. The result of the calculation to dete rmine the coefficients was that the coefficient for the shear term was 1 compared to a coefficien t of 88 for the torsion term. This calculation sufficiently verified that the shear co ntribution could be ignored in design. 3.7 Overview The previous sections detailed the design of the various components of the experimental program. It was of the utmost important to ve rify that the apparatuses not pertaining to the foundation failure would not fail during testing (i.e. concrete block system and pipe apparatus). Furthermore, all other foundation failure modes had to be precluded in the design. This ensured that if the concrete breakout failure in shear was the failure mode it would be observed during testing.

PAGE 43

43 Figure 3-12 and Figure 3-13 show the fully assembled test specimen at the Florida Department of Transportation Structures Research Center.

PAGE 44

44 LOAD LOCATION SHAFT CONCRETE BLOCK PIPE ASSEMBLY BASE PLATE Figure 3-1. Schematic of test apparatus Figure 3-2. Front elevation of test apparatus

PAGE 45

45 Figure 3-3. Plan view of test apparatus Figure 3-4. Side elevat ion of test apparatus

PAGE 46

46 a3 l e1 Leveling Nut Figure 3-5. Lever arm for the calculation of bolt flexure 1.5ca1 1.5ca1 ca1 c A rb r 2 15 4a Vcoc A Figure 3-6. Adjusted cover based on a single anchor and 35 failure cone

PAGE 47

47 r n A 360 2 sin 2 A r chord 15 1a Vcc chord n A Figure 3-7. Development of the projected failure area for the gr oup of anchors around a circular foundation ca11.5ca11.5ca11.5ca11.5ca1s=3.0ca1 ca11.5ca11.5ca11.5ca11.5ca1s=3.0ca1 Figure 3-8. Two anchor arrangement displays th e minimum spacing such that no overlap of the failure cones occurs AVc AVc Figure 3-9. Overlap of failure cones

PAGE 48

48 AVcLeg Leg AVc Leg Leg AVcLeg Leg AVcLeg Leg AVc Leg Leg AVc Leg Leg Figure 3-10. The contribution of the legs of the failure cone to AVc along a straight edge decreases as the number of bolts increases s s Figure 3-11. Overlap of failure cones for a circular foundation

PAGE 49

49 Figure 3-12. Assembled test specimen Figure 3-13. Shaft with pipe apparatus attach ed prior to instrument ation being attached

PAGE 50

50 Table 3-1. Field dimensions Component Field Dimension Shaft Diameter 60 in. Hoop Steel Diameter 46 in Hoop Steel Size #5 Longitudinal Steel Size #9 Anchor Bolt Diameter 2 in. Anchor Embedment 55 in. Bolt Spacing Diameter 36 in. Base Plate Diameter 42 in. Base Plate Thickness 1 in. Table 3-2. Summary of design calculations Component Design Type Equation Reference Result Shaft Cracking Torsion (2-1) 131 kip-ft Basic Torsion (2-2) 131 kip-ft Threshold Torsion (2-3) 24.6 kip-ft Nominal Torsion (2-4) 253 kip-ft Anchor ACI Concrete Breakout (2-12) 182 kip-ft Eligehausen et al. Concrete Breakout (2-16) 159 kip-ft Eligehausen et al. Concrete Breakout (2-17) 167 kip-ft Bolt Flexure 253 kip-ft Bolt Shear 1756 kip-ft

PAGE 51

51 CHAPTER 4 IMPLEMENTATION OF TESTING PROGRAM In order to proceed with testing the sp ecimen presented in Chapter 3, important considerations had to be made. The first area under consideration was the concrete strength. It was important to determine this to calculate the pr edicted failure mode prior to testing. Also, the flexural and shear strengths of th e bolts were calculated using th e specified yield strength. The other area that was of key importance was the instrumentation. The instrumentation was required to produce meaningful data during testing. The other sect ion of this chapter is on the carbon fiber reinforced polymer (CFRP) wrap used in the retrofit test. 4.1 Materials 4.1.1 Concrete Strength As it was stated in Chapter 3, the initial calculations for the design of the test setup were carried out on the assumption of a concrete st rength of 5500 psi (37.9 MPa). The concrete breakout strength was recalculated ba sed on the concrete strength at the time of testing. On the date of the test, the concrete strength was 6230 ps i (43 MPa). This strength was calculated based on the average of three 6 in. (152 mm) x 12 in. (305 mm) cylinder tests. 4.1.2 Bolt Strength The yield strength of the F1554 Grade 105 anch or bolts was assumed to be 105 ksi (723.95 MPa). This was used to calculate the fle xural strength and shear strength of the bolts. 4.1.3 Carbon Fiber Reinforced Polymer Wrap The first test was considered concluded af ter significant cracking and when the test specimen stopped picking up additional load. The loading was ceased before the specimen completely collapsed. The reason for doing so wa s to enable a second test to be performed on the specimen after it was retrofitted with a carb on fiber reinforced polymer (CFRP) wrap. The

PAGE 52

52 second test verified whether th e CFRP wrap was an acceptable means to retrofit the failed foundation. The amount of CFRP that was applied to th e shaft was determined by calculating the amount of CFRP required to bring the shaft back to its initial concrete br eakout strength. The CFRP wrap that was used for the retrofit was SikaWrap Hex 230C. The properties of the wrap were obtained and the ultimate tensile strength wa s used to calculate the required amount that needed to be applied. The property specifica tions for the SikaWrap were based on the mean strength minus 2 standard deviations. ACI (2002), ACI 440.R-02, .3.1 specifies that the nominal strength to be used for design be based on the mean strength less 3 standard deviations. Therefore, the design strength prov ided by Sika was adjusted to ensure that the design met the ACI specifications. The method for calculating the amount of CFRP required was to convert the torsion to a shear load per bolt. The shear load, which was direct ed parallel to the edge, had to be adjusted to such that it was directed perpendicular to the ed ge. In order to do this, the ACI multiplier of 2 was divided from the load. That load per bolt di rected perpendicular to the edge was converted to a pressure around the circumference of the shaf t. The equivalent te nsion that had to be resisted by the CFRP wrap was then calculated, a nd the amount of CFRP to provide that tensile strength was determined. Figure 4-1 illustrates this method. Two layers of the wrap were prescribed to meet the ACI concrete breakout strength based on assuming that the full 12 in. (305mm) width of the CFRP wrap would not be effective. Rather, it was assumed that the depth of the c oncrete breakout failure cone based on the cover, 1.5cover, was the effective width, 7.5 in. (191 mm). Three layers of the CFRP wrap were applied to the specimen. The addition of the extr a layer exceeded the required strength, so it was

PAGE 53

53 deemed acceptable. Once the wrap was set, the retr ofit test was carried out Calculations for the design of the CFRP wrap layout are located in Appendix B. 4.2 Instrumentation 4.2.1 Linear Variable Displacement Transducers Linear Variable Displacement Transducers (L VDTs) were placed at the location of the load cell, and at various points along the shaft and base plate. A total of ten LVDTs were utilized in the project. Figure 4-2 is a schematic of the layout of the LVDTs on the base plate. Figure 4-3 and Figure 4-4 show the location of the LVDTs on the shaft, and Figure 4-5 shows the LVDT at the load location. The denotation for each of the LVDTs is also on the drawings. These identification codes were used to denote the LVDTs during testing. The purpose of the LVDTs along the shaft and base plate was to allo w for the rotation of the base plate to be measured during the testing. The LVDTs at the fr ont and back of the shaft were to allow for the rotation to be measured relative to the rotation of the shaft. The intent in the project was such that the shaft would not rotate; only the base pl ate would rotate as the bolts were loaded. The horizontal LVDT on the base plate was intende d to indicate if there was any horizontal movement of the base plate. The rotation of the base plate was calculated using Equation 4-1. gage VD D D R3 1 1tan (4-1) Where R = base plate rotation (rad) D1V = displacement of LVDT D1V (in.) D3 = displacement of LVDT D3 (in.) Dgage = distance between LVDTs D1V and D3 (in.) Once the test apparatus was assembled, the distance Dgage was measured. This distance was 26.31 in. (668 mm). Figure 4-6 shows LVDTs D1V and D4 on the test specimen.

PAGE 54

54 4.2.2 Strain Gages Strain gages were attached to the base plat e on the outer surface ad jacent to the bolt holes in order to determine how may bolts were activel y transferring load given the 1.75 in. (44.5 mm) holes for the 1.5 in. (38.1 mm) anchors. In applying the ACI 318-05 equation for concrete breakout strength of an anchor in shear directed parallel to an e dge (Equation 2-12) it was of key importance to know how many bolts were carrying the load. For instance if two bolts were carrying the load, the concrete would fail at a lowe r load than if all twelve bolts were carrying the load. In addition to showing the placemen t of the LVDTs, Figure 4-2 also details the location of the strain gages on the base plate. Fi gure 4-7 shows the denotation of the strain gages relative to the bolt number, and Fi gure 4-8 shows a strain gage on the base plate of the test specimen. Note that the bolt numbering starts at one at the top of the pl ate and increases as you move clockwise around the base plate.

PAGE 55

55 TCFRPTCFRP Divide by 2 TCFRPTCFRP TCFRPTCFRP Divide by 2 Figure 4-1. Method for the dete rmination of the tension, TCFRP, that must be resisted by the CFRP wrap

PAGE 56

56 LVDT Strain Ga g e D1V D1H D2 D3 Figure 4-2. Instrumentation layout on the base plate D4 D5 D6 Figure 4-3. Instrumentation layout on face of shaft

PAGE 57

57 D7 D8 D9 Figure 4-4. Instrumentation layout on r ear of shaft/face of concrete block D10 6 Figure 4-5. Instrumentation layout of pipe at load location

PAGE 58

58 Figure 4-6. Location of LVDTs D1V, D4, and D7 on the test specimen S12 S11 S10 S9 S8 S7 S6 S5 S4 S3 S2 S1 Bolt 1 Bolt 2 Figure 4-7. Strain gage layout on base plate

PAGE 59

59 Figure 4-8. Strain gage on ba se plate of test specimen

PAGE 60

60 CHAPTER 5 TEST RESULTS Two tests were performed on the test specimen. The initial test was conducted to determine whether the concrete breakout failure was the failure mode demonstrated in the field. The verification of this was based on the crack pa ttern and the failure load recorded. If the failure torsion was the concrete breakout failure torsion, then the hypot hesized failure mode would be verified. The retrofit test was perfor med on the same test specimen. This test was completed to establish whether a CFRP wrap was an acceptable retrofit for the foundation. 5.1 Initial Test 5.1.1 Behavior of Specimen During Testing The initial test on the fou ndation was carried out on 31 August 2006 at the Florida Department of Transportation Structures Resear ch Center. The test specimen was gradually loaded during the testing. Th roughout the test, the formation of cracks on the surface of the concrete was monitored (Figures 5-1 and 5-2). At 90 kip-ft (122 kN-m ), the first cracks began to form. When 108 kip-ft (146 kN-m) was reached it was observed that the cracks were not extending further down the length of the shaft. T hose cracks that had formed began to slightly widen. These cracks, Figure 5-1, were character istic of those that form during the concrete breakout failure. At 148 kip-ft (201 kN-m), cracks spanning between the bolts had formed (Figure 5-3). The foundation continued to be lo aded until the specimen stopped taking on more load. The torsion load peaked at 200 kip-ft ( 271 kN-m). Loading ceased and was released when the applied torsion fell to 190 kip-ft (258 kN-m). The predicted concrete breakout capacity of the shaft at the time of testing was calculated as 193 kip-ft (262 kN-m) (Equation 3-3). At failure, the foundation displayed the charac teristic cracks that one would see in a concrete breakout failure (Figure 5-4). As intende d, the bolts did not yield, and the shaft did not

PAGE 61

61 fail in torsion. Data was reduced to formulate applied torsion versus plate rotation and applied torsion versus bolt strain plots. The Applied Torsion vs. Plate Rotation plot (Figure 5-5) shows that the bolts ceased taking on addi tional load after the noted concre te breakout failure due to the shear parallel to the edge resulting from the applied torsion. It also exhibi ts slope changes at the loads where crack development started or the ex isting cracks were altered. The first slope change at 108 kip-ft (146 kN-m) coincided with the widening of the characteristic diagonal cracks on the front face of the shaft. The second change occurred at 148 kip-ft (201 kN-m) corresponding with the formati on of cracks between the bolts. 5.1.2 Behavior of Strain Gages During Testing Figure 5-6 displays the Applied Torsion vs. Bo lt Strain plots for each bolt relative to its location on the foundation. Recall that the term bolt strain refe rs the measurement of the strain in the base plate at the bolt locat ion. The strain was a result of the bolt carrying load. The first line on the plots in Figure 5-6 is 50 kip-ft (67.8 kN-m). At this level, all of the bolts were carrying load with the exception of bolts one, six, and eight. At the next level, 100 kip-ft (136 kN-m) bolt one picked up load, but bolts six and eight remained inactive. It must be noted that, at 108 kip-ft (138 kN-m), which wa s the first slope change on the Applied Torsion vs. Plate Rotation Plot, a redi stribution of the loading occurred. This redistribution is illustrated in Figure 5-7. As the cracks widened, those bolts that were transferring the majority of the load were able to move more freely, a nd, therefore, the other bolts became more active in transf erring the load to the foundation. A similar redistribution to a lesser degree occurred at approximately 148 kipft (201 kN-m), which coincided with the first observation of cracks between the bolts. As the various plots illustrate, some of the st rain gages recorded ne gative strains, while others recorded positive strains. This was most likely due to the bearing location of the bolt on

PAGE 62

62 the base plate. Although this oc curred, the relative strain readi ngs were considered acceptable. To further explore this phenomenon, strain gages were placed on the bottom of the base plate in addition to those on the top for the second test. 5.1.3 Summary of Initial Test Results The results of this test indicated that the concrete breakout failure was the failure mode observed in the site investigation. The characte ristic cracks and the structural integrity of the bolts in the failed foundations, as observed during the site investigation, was the first step to arriving at this failure mode. The percent difference between the failure torsion and the predicted failure torsion (Equa tion 3-3) was 3.6%. Therefor e, it was concl uded that the foundation failed at the failure to rsion for the predicted failure mode. These results indicated that the design methodology for cantilever sign f oundations should include the concrete breakout failure due to shear directed parall el to an edge resulting from torsional loading. All plots for the first test are located in Appendix C. 5.2 CFRP Retrofit Test After the results of the first test were re viewed, the need for a method to strengthen existing foundations became apparent. Since the concrete breakout failure had not been considered in the design of the cantilever sign st ructure foundations, a system had to be put in place to evaluate whether or not those existing foundations would be susceptible to failure. One economical method of retrofitting the existing founda tions is the use of Carbon Fiber Reinforced Polymer (CFRP) wraps. Recall that, at the conclusion of the first test the bolts had not yiel ded, and the concrete was still intact. This enabled a second test on the failed foundati on to be carried out. The key focus of this second test was to determine if the foundation could reac h its initial concrete breakout strength again. The foundation was retr ofitted with three laye rs of 12 in. (305 mm)

PAGE 63

63 wide SikaWrap Hex 230C (Figure 5-8). This amount of CFRP exceeded the amount required to attain the concrete breakout stre ngth, 193 kip-ft (262 kN-m). The torsional strength of the shaft with the CFRP wrap was calculated. The result ant strength based on the effective width, Section 4.1.3, of 1.5cover, or 7.5 in. (191 mm), was 229 kip-ft ( 310 kN-m). Since that effective depth was an assumption for design, the strength based on the full width, 12 in. (305 mm), of the wrap, 367 kip-ft (498 kN-m), was also calculated for reference. 5.2.1 Behavior of Specimen with CFRP Wrap During Testing The second test was conducted on 13 September 2006. For this test, the concrete strength was not required to be known, since the concre te had already failed. The containment provided by the CFRP wrap, along with the anchor bolts, wa s the source of the stre ngth of the foundation. As the purpose of the second test was to learn how much load the foundation could take, and if that load met or exceeded the concrete breakout strength, the load was not held for prolonged periods at regular intervals during the test. Figure 5-9 is the Applied Torsion vs. Plate Rotation plot for the second test. The foundation was cl osely monitored for crack formation along the shaft, propagation of existing cracks, and failure of the CFRP wrap. The strength of the foundati on exceeded the predic ted concrete breakout strength of 193 kip-ft (262 kN-m). It was not until the loadi ng reached 257 kip-ft (348 kN -m) that the first pops of the carbon fibers were heard. At that torsion load, the strengt h of the CFRP wrap based on the effective depth, 229 kip-ft (310 kN-m), was exceeded Therefore, the effective depth of the wrap was a conservative assumption. At approximately 288 kip-ft (390 kN-m) more po ps were heard. However, the carbon fiber did not fail. During the course of the test, characterist ic torsion cracks began to form along the shaft (Figure 5-10) and propagated to the base of the shaft. This occurred because the ACI 31805 nominal torsional strength (Equation 2-6) of 253 kip-ft (343 kN-m) was exceeded. Although

PAGE 64

64 these cracks had formed, the foundation still had not failed. Another phenomenon that occurred was the yielding of the bolts. According to the calc ulations for the yield strength of the bolts, the bolts yielded at approximately 253 kip-ft (343 kN-m) of applied torsion. The strength was determined using the same methodology outlined in S ection 3.3.1. This wa s the within the range in which the yielding was observed (Figure 5-9). The bolts were yi elding, but they did not reach their ultimate strength. The test abruptly concluded when the conc rete block shifted out of place, causing the load cell to be dislodged from its lo cation on the pipe. This occurred at 323 kip-ft (438 kN-m). 5.2.3 Behavior of Strain Gages During Testing For the retrofit test, strain gages were placed on the top and bottom of the base plate. Figure 5-11 shows each of the Applied Torsion vs. Bo lt Strain plots at the appropriate bolt locations. Note that as the loading increased, th e bottom strain gages began to behave similarly for all of the bolts. The strain was increasing at a higher rate. This illustrated that as the bolts picked up load and began to bend, they were pr imarily in contact with the bottom of the base plate (Figure 5-12). The strain s recorded by the bottom gages indicate that all of the bolts became active during the test. Similar to the behavior of the bolts throughout the initia l test, Figure 5-13 illustrates the changes in the bolt strain data for the top ga ges corresponding with m ilestone loads during the test. 5.2.4 Summary of Test Results Upon removal of the pipe apparatus, the cr ack pattern illustrated the concrete breakout failure, and torsional cracks in the center of the shaft verified that the torsional capacity was exceeded during testing (Figure 5-14). Figure 5-15 details the characteristic torsion cracks on the side of the shaft after tes ting. The test proved that the CFRP wrap was an acceptable method

PAGE 65

65 for retrofitting the foundation. It exceeded the concrete breakout strength. The success of this retrofit test led to the development of guidelin es for the evaluation of existing foundations and the guidelines for the retrofit of t hose foundations in need of repair. All plots for the retrofit test are located in Appendix D.

PAGE 66

66 Figure 5-1. Initial cr acks on face of shaft Figure 5-2. Initial cracks on face and side of shaft (alternate view of Figure 5-1)

PAGE 67

67 Figure 5-3. Face of test specimen after testing exhibits cracks between the bolts along with the characteristic concre te breakout cracks Figure 5-4. Crack pattern on face of shaft after testing depicts ch aracteristic concrete breakout failure cracks

PAGE 68

68 0 50 100 150 200 250 00.511.522.53Rotation (deg)Applied Torsion (kip-ft) Crack Formation Between Bolts Cracks Begin to Widen Peak Applied Torsion Figure 5-5. Applied Torsion vs. Pl ate Rotation PlotInitial Test

PAGE 69

69 1 12 2 3 7 4 6 5 11 10 9 8 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 0 0.5 1 1.5 2 2.5 3 Rotation (deg) Applied Torsion (kip-ft) Figure 5-6. Applied Torsion vs. Bolt Strain Plot s for each bolt at the appropriate location on the base plate with Applied Torsi on vs. Plate Rotation plot in center (full size plots in Appendix C)

PAGE 70

70 0 50 100 150 200 250 -750-600-450-300-1500150300450600750MicrostrainApplied Torsion (kip-ft) Crack Formation Between Bolts Cracks Begin to Widen Figure 5-7. Bolt Strain Comparison Plot for Initial Test exhibits the redistribution of the load coinciding with crack formations Figure 5-8. Shaft with the CFRP wrap applied prior to testing

PAGE 71

71 0 50 100 150 200 250 300 350 00.511.522.533.544.55Rotation (deg)Applied Torsion (kip-ft) ACI Concrete Breakout Strength Bolts Yielded First "Pops" of CFRP Specimen Shifted Figure 5-9. Applied Torsion vs. Pl ate Rotation PlotRetrofit Test Figure 5-10. Shaft exhibiting characteristic torsion cracks from face to base of shaft

PAGE 72

72 1 12 2 3 7 4 6 5 11 10 9 8 0 50 100 150 200 250 300 350 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Rotation (deg) Applied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) 0 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Figure 5-11. Applied Torsion vs. Bolt Strain plot s for the Retrofit Test at the appropriate bolt location around the base plate with Applied Torsion vs. Pl ate Rotation plot in center (full size plots in Appendix D)

PAGE 73

73 Bearing Location Base Plate Bolt V Figure 5-12. Bolt bearing on the bottom of the base plate during loading 0 50 100 150 200 250 300 350 -1100-825-550-27502755508251100MicrostrainApplied Torsion(kip-ft) ACI Concrete Breakout Strength Bolts Yielded Figure 5-13. Bolt Strain Comparison plot for the retrofit test exhi bits slope changes at milestone loads

PAGE 74

74 Figure 5-14. Face of shaft after te st illustrates yielding of bolts concrete breakout cracks around the perimeter, and torsion cracks in the center. Figure 5-15. Torsion cracks along le ngth of the shaft after the test

PAGE 75

75 CHAPTER 6 SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS The purpose of this research program was to determine the cause of the failure of foundations of cantilever sign structures duri ng the 2004 hurricane season. After a thorough literature review, in conjunction wi th the site investigation, and testing, it was determined that the foundations failed as a result of an applied torsion which caused a concrete breakout failure due to shear directed parallel to the edge on the anchors. This anchorage failure is detailed in ACI 318-05 Appendix D. Previous to this experi mental research, this failure mode was not considered in the design of the cantilever sign foundations. Cantilever sign foundations need to be designed for shear parallel to the edge on the anchor resulting from torsion. Test results indicate that th e failure of the foundations wa s caused by concrete breakout due to shear directed parallel to the edge on the anchors. The test specimen failed at the torsion predicted by the ACI 318-05 Appendix D design equations. Additionally the crack pattern matched the crack pattern exhibited in the field, and both foundations emul ated the characteristic crack pattern of the shea r directed parallel to an edge for concrete breakout failure. It is recommended that future tests be performed on ci rcular foundations to further investigate the concrete breakout failure for a sh ear load directed bot h parallel and perpendi cular to an edge. Additional testing was performed to determine an acceptable retrofit option. It was determined that applying a CFRP wrap to the foundation strengthens th e foundation such that it not only meets its initial concrete breakout capac ity, but, also, exceeds the capacity. The results of this test led to the development of guide lines for the evaluation and repair of existing foundations. The guidelines we re based on the following: Using either the torsional load from the desi gn or, if not available, using the ACI nominal torsional strength (ACI 318-05 .6.3.6), de termine the torsional capacity of the foundation.

PAGE 76

76 Calculate the concrete breakout strength in accordance with ACI Appendix D. If the concrete breakout strength is less than the maximum of the nominal torsional strength and design torsion, then th e foundation is susceptible to failure. The amount of the SikaWrap 230C required is calculated using the maximum of the nominal torsional strength and the design torsi on. The amount required is given in layers of the CFRP wrap. These guidelines were submitted to the Florida Department of Transportation. The guidelines will be used to evaluate and, if necessa ry, repair the existing f oundations. It is critical that such foundations be evaluated in order to determine the susceptibi lity to this type of failure. The proper use of the findings of this research program will allow for future prevention of the failures exhibited during the 2004 hurricane season.

PAGE 77

77 APPENDIX A TEST APPARATUS DRAWINGS Figure A-1. Dimensioned front elev ation drawing of test apparatus Fi g ure A-1. Dimensioned fr ont elevation drawin g of test a pp aratus

PAGE 78

78 Figure A-2. Dimensioned plan drawing test apparatus Fi g ure A-2. Dimensioned p lan drawin g test a pp aratus

PAGE 79

79 Figure A-3. Dimensioned side elev ation drawing of test apparatus Fi g ure A-3. Dimensioned side elevation drawin g of test a pp aratus

PAGE 80

80 Figure A-4. Dimensioned pipe apparatus drawing Fi g ure A-4. Dimensioned p i p e a pp aratus drawin g

PAGE 81

81 Figure A-5. Dimensioned channel tie-down drawing Fi g ure A-5. Dimensioned ch annel tie-down drawin g

PAGE 82

82 APPENDIX B DESIGN CALCULATIONS

PAGE 83

83

PAGE 84

84

PAGE 85

85

PAGE 86

86

PAGE 87

87

PAGE 88

88

PAGE 89

89

PAGE 90

90

PAGE 91

91

PAGE 92

92

PAGE 93

93

PAGE 94

94

PAGE 95

95

PAGE 96

96

PAGE 97

97

PAGE 98

98

PAGE 99

99

PAGE 100

100

PAGE 101

101

PAGE 102

102

PAGE 103

103

PAGE 104

104

PAGE 105

105

PAGE 106

106

PAGE 107

107

PAGE 108

108 APPENDIX C INITIAL TEST DATA 0 50 100 150 200 250 00.511.522.53Rotation (deg)Applied Torsion (kip-ft) Crack Formation Between Bolts Cracks Be g in to Widen Peak Applied Torsion Figure C-1. Applied Tors ion vs. Rotation Plot 0 50 100 150 200 250 -750-600-450-300-1500150300450600750MicrostrainApplied Torsion (kip-ft) Crack Formation Between Bolts Cracks Be g in to Widen Figure C-2. Bolt Stra in Comparison Plot

PAGE 109

109 Bolt 10 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) a Bolt 20 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) b Figure C-3. Applied Torsion vs. St rain Plots for each bolt location

PAGE 110

110 Bolt 30 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) c Bolt 40 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) d Figure C-3. Continued

PAGE 111

111 Bolt 50 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) e Bolt 60 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) f Figure C-3. Continued

PAGE 112

112 Bolt 70 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) g Bolt 80 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) h Figure C-3. Continued

PAGE 113

113 Bolt 90 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) i Bolt 100 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) j Figure C-3. Continued

PAGE 114

114 Bolt 110 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) k Bolt 120 50 100 150 200 250 -500-400-300-200-1000100200300400500MicrostrainApplied Torsion (kip-ft) l Figure C-3. Continued

PAGE 115

115 APPENDIX D RETROFIT TEST DATA 0 50 100 150 200 250 300 350 00.511.522.533.544.55Rotation (deg)Applied Torsion (kip-ft) ACI Concrete Breakout Strength Bolts Yielded First "Pops" of CFRP Specimen Shifted Figure D-1. Applied Tors ion vs. Rotation Plot 0 50 100 150 200 250 300 350 -1100-825-550-27502755508251100MicrostrainApplied Tors ion(kip-ft) ACI Concrete Breakout Strength Bolts Yielded Figure D-2. Bolt Stra in Comparison Plot

PAGE 116

116 Bolt 10 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage a Bolt 20 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage b Figure D-3. Applied Torsion vs. St rain Plots for each bolt location

PAGE 117

117 Bolt 30 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage c Bolt 40 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage d Figure D-3. Continued

PAGE 118

118 Bolt 50 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage e Bolt 60 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage f Figure D-3. Continued

PAGE 119

119 Bolt 70 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage g Bolt 80 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage h Figure D-3. Continued

PAGE 120

120 Bolt 90 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage i Bolt 100 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage j Figure D-3. Continued

PAGE 121

121 Bolt 110 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage k Bolt 120 50 100 150 200 250 300 350 -2000-1500-1000-5000500100015002000MicrostrainApplied Torsion (kip-ft) Top Gage Bottom Gage l Figure D-3. Continued

PAGE 122

122 LIST OF REFERENCES American Association of State Highway a nd Transportation Offici als (AASHTO). (1994). AASHTO standard specifications for structural supports for highway signs, luminaries, and traffic signals, 3rd Ed., Washington, D.C. American Association of State Highway a nd Transportation Offici als (AASHTO). (2001). AASHTO standard specifications for structural supports for highway signs, luminaries, and traffic signals, 4th Ed., Washington, D.C. American Association of State Highway a nd Transportation Offici als (AASHTO). (2004). AASHTO LRFD bridge design specifications, 3rd Ed., Washington, DC. American Concrete Institute (ACI). (1995). Build ing code requirements fo r structural concrete (ACI 318-95) and commentary (ACI 318R-95). ACI 318-95, Farmington Hills, Mich. American Concrete Institute (AC I). (2002). Guide for the design and construction of externally bonded FRP systems for strengthening conc rete structures (ACI 440.2R-02). ACI 440.2R02, Farmington Hills, Mich. American Concrete Institute (ACI). (2005). Build ing code requirements fo r structural concrete (ACI 318-05) and commentary (ACI 318R-05). ACI 318-05, Farmington Hills, Mich. American Institute of Steel Construction (AISC). (2001). Manual of steel construction load and resistance factor design, 3rd Ed., Chicago, Ill. American Society of Civil E ngineers (ASCE). (1991). Guidelin es for electrical transmission line structural loading. ASCE Manuals and Reports on Engineering Practice No. 74, Reston, Va. American Society of Civil Engi neers (ASCE). (2000). Minimum design loads for buildings and other structures (ASCE 7-98). ASCE Standard No. 7-98, Reston, Va. American Society of Civil E ngineers (ASCE). (2006). Desi gn of steel transmission pole structures (ASCE/SEI 48-05). ASCE Manuals and Reports on Engineering Practice No. 72, Reston, Va. Breen, J. E. (1964). Development length for anchor bolts. Center for Transportation Research Report 55-1F, Austin, Texas. Eligehausen, R., Malle, R., Silva, J.F. (2006). Anchorage in Concrete Construction, Ernst & Sohn, Berlin, 108-128. Fouad, F.H., Calvert, E. A., and Nunez, E. (1998). Structural supports for highway signs, luminaires, and traffic signals. National Cooperative Highway Research Program Report 411, Washington, D.C.

PAGE 123

123 Fouad, F.H., Davidson, J.S., Delatte, N., Calver t, E.A., Chen, S., Nunez, E., and Abdalla, R. (2003). Structural supports for highway signs, luminaries, a nd traffic signals. National Cooperative Highway Research Program Report 494, Washington, D.C. Fuchs, W., Eligehausen, R., and Breen, J.E. ( 1995). Concrete capacity design (CCD) approach for fastening to concrete. ACI Struct. J., 92(1), 73-94. Hasselwander, G. B., Jirsa, J. O., Breen, J. E ., and Lo, K. (1977). Strength and behavior of anchor bolts embedded near edges of concrete piers. Center for Transportation Research Report 29-2F, Austin, Texas. Institute of Electrical an d Electronics Engineers (IEEE). (1997) National electrical safety code (NESC). Piscataway, N.J. Jirsa, J. O., Cichy, N. T., Calzadilla, M. R., Smart, W. H., Pavluvcik, M. P., and Breen, J. E. (1984). Strength and behavior of bolt inst allations anchored in concrete piers. Center for Transportation Research Report 305-1F, Austin, Texas. Kaczinski, M. R., Dexter, R. J., and Van Dien J. P. (1998). Fatigue-resistant design of cantilevered signal, sign and light supports. National Cooperative Highway Research Program Report 412, Washington, D.C. Keshavarzian, M. (2003). Extreme wind design of self-supported steel struct ures: critical review of related ASCE publications. Practice Periodical on Structural Design and Construction, 8(2), 102-106. Keshavarzian, M., and Priebe, C. H. (2002) Wind performance of short util ity pole structures. Practice Periodical on Structural Design and Construction, 7(4), 141-146. Lee, D. W., and Breen J. E. (1966). Fact ors affecting anchor bolt development. Center for Transportation Research Report 88-1F, Austin, Texas. MacGregor, J. G., and Ghoneim, M. G. (1995). Design for torsion. ACI Struct. J., 92(2), 211218. Roark, R. J., and Young, W. C. (1975). Formulas for Stress and Strain, 5th Ed., McGraw Hill, New York, 286-323. Telecommunications Industry A ssociation/Electrical Industries Associati on (TIA/EIA). (1996). Structural standards for steel antenna to wer and antenna supporting structures. ANSI/TIA/EIA 222-F, Arlington, Va.

PAGE 124

124 BIOGRAPHICAL SKETCH Kathleen M. Halcovage is the daughter of George F. Halcovage, Jr. and Barbara M. Halcovage. She was born in Pottsville, Pennsylva nia on November 10, 1983. She is the second oldest of five children. Kathleen graduate d from Nativity B.V.M. High School in 2001 where she was the class valedictorian. She received a Presidential Scholarship to continue her education at Villanova University in Villanova Pennsylvania. While attending Villanova University, she spent a semester abroad studying at the University of Sheffield in Sheffield, England. She was selected to the Tau Beta Pi Engineering Honor Society, the Chi Epsilon Civil Engineering Honor Society, the Phi Kappa Phi A ll-Discipline Honor Society, the Delta Epsilon Sigma All-Discipline Catholic Honor Societ y, and Whos Who in American Colleges and Universities. Kathleen graduated Summa Cum Laude from Villanova University in May 2005. She received a Bachelor of Science (B.S.) in civ il engineering degree with a business minor. At graduation, she was honored with the Departme nt of Civil and Environmental Engineering Faculty Award Medallion and the Dean Robert D. Lynch Award, which recognizes the scholastic achievements of an outstanding new graduate of the Villanova College of Engineering. Upon graduating, she entered the University of Florida in Gain esville, FL to continue her studies in structural engineeri ng. During her tenure at Florid a, she worked as a Graduate Assistant on a research project sponsored by the Florida Department of Transportation. She will graduate with a Master of Engin eering (M.E.) degree in May 2007.