Citation
Shear strength at the interface of bonded concrete overlays

Material Information

Title:
Shear strength at the interface of bonded concrete overlays
Creator:
Rosen, Christian Joahn ( auhor )
Place of Publication:
Denver, Colo.
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
1 electronic file (113 pages) : ;

Thesis/Dissertation Information

Degree:
Master's ( Master of science)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Civil Engineering, CU Denver
Degree Disciplines:
Civil engineering

Subjects

Subjects / Keywords:
Concrete -- Testing ( lcsh )
Shear (Mechanics) ( lcsh )
Concrete construction -- Maintenance and repair ( lcsh )
Concrete construction -- Maintenance and repair ( fast )
Concrete -- Testing ( fast )
Shear (Mechanics) ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
This thesis investigated multiple methods for measuring shear strength at the bonded interface between a concrete substrate and a concrete overlay as it is influenced by different substrate surface treatments. The results were collected from an experiment conducted at the University of Colorado Denver where six different test pads, each with a unique surface treatment, were exposed to four types of tests. From the different tests, the shear strength at the bonded interface was measured. A discussion of analysis with respect to the variations in surface treatments and test methods is presented. The mechanisms associated with the different tests are discussed. The results are dependent on the test method used to evaluate the shear strength. The findings of the research illustrate the significance that surface treatment of a substrate has on the shear strength at a bonded overlay.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: Adobe Reader.
Statement of Responsibility:
Christian Johan Rosen.

Record Information

Source Institution:
University of Colorado Denver Collections
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
985116930 ( OCLC )
ocn985116930
Classification:
LD1193.E53 2016m R67 ( lcc )

Downloads

This item has the following downloads:


Full Text
SHEAR STRENGTH AT THE INTERFACE OF BONDED CONCRETE OVERLAYS
by
CHRISTIAN JOHAN ROSEN
B.S., Linkoping University, 2014
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements of the degree of Master of Science Civil Engineering
2016


This thesis for the Master of Science degree by Christian Johan Rosen has been approved for the Civil Engineering Program by
Frederick R. Rutz, Chair Kevin L. Rens Chengyu Li
July 30,2016
11


Rosen, Christian Johan (M.S., Civil Engineering)
Shear Strength at the Interface of Bonded Concrete Overlays Thesis directed by Professor Frederick R. Rutz
ABSTRACT
This thesis investigated multiple methods for measuring shear strength at the bonded interface between a concrete substrate and a concrete overlay as it is influenced by different substrate surface treatments. The results were collected from an experiment conducted at the University of Colorado Denver where six different test pads, each with a unique surface treatment, were exposed to four types of tests. From the different tests, the shear strength at the bonded interface was measured. A discussion of analysis with respect to the variations in surface treatments and test methods is presented. The mechanisms associated with the different tests are discussed. The results are dependent on the test method used to evaluate the shear strength. The findings of the research illustrate the significance that surface treatment of a substrate has on the shear strength at a bonded overlay.
The form and content of this abstract are approved. I recommend its publication.
Approved: Frederick R. Rutz
iii


ACKNOWLEDGEMENTS
There are numerous people that in one way or another have helped me make this research successful. First of all I would like to thank Dr. Frederick Rutz who has served as advisor of this research. His enthusiasm and dedication have been really motivating and he has been very supportive in every stage of the research, from planning and preparation to testing and writing the thesis. I would like to thank Tom Thuis, Peter Sillstrop and Jac Corless at the University of Colorado Denver Electronic Calibration and Repair Lab who trained me on the MTS machines, helped me with the pouring of the concrete pads and for manufacturing additional steel pucks for the pull-off test. I would also like to express my gratitude to Bud Warner and Zack Ballard at CTL|Thompson who have been generous with input on the testing process, lent me testing equipment, and for letting me use their facility and equipment for the slant shear test. And last my fellow graduate student Andy Pultorak who spent several freezing winter days helping me with the not so glamorous work of coring and saw cutting the test pads as well as conducting the jacking test.
IV


TABLE OF CONTENTS
Chapter
1 Overview...................................................................1
1.1 Introduction........................................................1
1.2 Goal................................................................2
1.3 Outline.............................................................2
2 Background.................................................................4
2.1 Introduction........................................................4
2.2 Bond................................................................4
2.3 Shear Strength......................................................7
2.3.1 Shear Strength Test Methods........................................8
2.3.1.1 Slant Shear Test.................................................8
2.3.1.2 Guillotine Test..................................................9
2.3.1.3 Jacking Test....................................................10
2.4 Tensile Strength...................................................10
2.4.1 Pull-Off Test.....................................................11
3 Experiment................................................................13
3.1 Construction of Test Pads.........................................13
3.2 Surface Preparation................................................16
4 Testing...................................................................19
v


4.1 Compression Test..................................................19
4.2 Pull-Off Test.....................................................21
4.3 Slant Shear Test..................................................22
4.4 Guillotine Test...................................................25
4.5 Jacking Test......................................................27
5 Experiment Results.......................................................29
5.1 Compression Test..................................................29
5.2 Pull-Off Test.....................................................31
5.3 Slant Shear Test..................................................35
5.4 Guillotine Test...................................................37
5.5 Jacking Test......................................................41
6 Discussion...............................................................45
6.1 Conducting Experiments............................................45
6.2 Experiment Results................................................47
6.2.1 Slant Shear Test..................................................48
6.2.2 Guillotine Test...................................................49
6.2.3 Jacking Test......................................................51
6.2.4 Pull-Off Test.....................................................53
6.2.5 Summary...........................................................55
7 Conclusions..............................................................60
vi


References..................................................................61
Appendix
A. Compression Test...................................................63
B. Pull-Off Test......................................................68
C. Guillotine Test Results............................................75
D. Slant Shear Test Results...........................................88
E. Jacking Test Results...............................................95
vii


LIST OF TABLES
Table
5.1: Surface treatments.............................................................29
5.2: Compression test results.......................................................29
5.3: Interpolated compressive strength at the time of testing.......................30
5.4a: Pull-off test results (SI units)...............................................31
5.4b: Pull-off test results (Imperial units).........................................32
5.5a: Pull-off test results Average tensile stress (SI units)......................33
5.5b: Pull-off test results Average tensile stress (Imperial units)................33
5.6: Pull-off test results Influence from vibration...............................33
5.7a: Pull-off test results Statistical results (SI units).........................34
5.7b: Pull-off test results Statistical results (Imperial units)...................34
5.8a: Slant shear test results (SI units)...........................................35
5.8b: Slant shear test results (Imperial units)......................................35
5.9a: Slant shear results Average shear stress (SI units)..........................36
5.9b: Slant shear results Average shear stress (Imperial units)....................36
5.10: Slant shear results Influence from vibration.................................36
5.11a: Guillotine test results (SI units)............................................37
5.11b: Guillotine test results (Imperial units)......................................38
5.12a: Guillotine test results Average shear stress (SI units).....................39
5.12b: Guillotine test results Average shear stress (Imperial units)...............39
5.13: Guillotine test results Influence from vibration............................39
5.14a: Guillotine test results Statistical results (SI units).....................40
viii


5.14b: Guillotine test results Statistical results (Imperial units)...............40
5.15a: Jacking test block data (SI units)...........................................41
5.15b: Jacking test block data (Imperial units).....................................42
5.16a: Jacking test results - Average shear stress (SI units).......................43
5.16b: Jacking test results - Average shear stress (Imperial units).................43
5.17: Jacking test results Influence from vibration..............................43
5.18a: Jacking test results - Statistical results (SI units)........................44
5.18b: Jacking test results - Statistical results (Imperial units)..................44
6.1: Slant shear test Ranking by capacity.........................................49
6.2: Guillotine test Ranking by capacity..........................................51
6.3: Jacking test Ranking by capacity.............................................53
6.4: Pull-off test Ranking by capacity............................................55
6.5: Average influence from vibration..............................................57
6.6: Summarized ranking by capacity................................................58
6.7: Guillotine test and pull-off test Correlation...............................59
IX


LIST OF FIGURES
Figure
2.1. Diagram of overlay and substrate.........................................6
2.2. Overlay without compaction...............................................7
2.3. Overlay with compaction..................................................8
2.4. Diagram of slant shear test..............................................9
2.5. Diagram of guillotine shear test........................................10
2.6. Diagram of jacking test.................................................10
2.7. Diagram of pull-off test................................................11
2.8. Pull-off test failure modes.............................................12
3.1. Experiment design: Plan.................................................14
3.2. Experiment design: Elevation............................................14
3.3. Sub strate Formwork.....................................................15
3.4. Vibrator used for compaction............................................16
3.5. Broom finish............................................................16
3.6. Rake finish.............................................................17
3.7. Bush hammer finish......................................................17
3.8. Bush hammer used for surface finish.....................................18
4.1. MTS 810 load frame......................................................20
4.2. Compression test set up.................................................20
4.3. 007 James bond tester...................................................21
4.4. Core drilling for slant shear test......................................22
4.5. Core sample with slanted interface......................................23
x


4.6. Slant shear test setup.................................................23
4.7. Close-up on slant shear sample..........................................24
4.8. Vertical core drilling for guillotine test.............................25
4.9. Shear jig design........................................................26
4.10. Guillotine test setup..................................................26
4.11. Blocks on pads 5 and 6 prepared for jacking test......................27
4.12. Jack set up for testing...............................................28
6.1. Compressive strength development........................................46
6.2a. Slant shear test (SI units)............................................48
6.2b. Slant shear test (Imperial units)......................................48
6.3a. Guillotine test Comparison (SI units)................................50
6.3b. Guillotine test Comparison (Imperial units)..........................50
6.4a. Jacking test Comparison (SI units)...................................52
6.4b. Jacking test Comparison (Imperial units).............................52
6.5a. Pull-off test Comparison (SI units)..................................54
6.5b. Pull-off test Comparison (Imperial units)............................54
6.6a. Summary of test results (SI units)....................................56
6.6b. Summary of test results (Imperial units)..............................56
xi


1
Overview
1.1 Introduction
Concrete has been use by mankind in structures for thousands of years. Because of its durability and ability to shape as desired, it is used in all types of construction. However, concrete does not last forever and to extend the lifetime of a damaged concrete structure, deteriorated concrete can be removed and replaced with a patch. Depending on how severe the damage is, different types of repairs can be installed. If a concrete slab is only damaged in the top portion, there is no need to remove concrete all the way to the bottom. Unsound concrete is removed and a patch can be used to fill the void. A goal of patching is to create a repair that resembles the original structure as much as possible where the bonded interface between the new and old concrete behaves monolithically. The bond strength components are shear and tensile strength and even though tension tests on repaired concrete structures are most common, shear strength is of greater interest because the interface is only subject to very small zones of pure tension, while shear better describes the overall bond strength (Silfwerbrand 2003). The reason why tension tests are more common is simply because they are easier to perform. A review of the literature reveals that there are no accepted tests for concrete shear strength because there is no general agreement in the structural engineering community for a shear strength test. The problem with shear tests is that it is difficult to separate direct shear from bending moment, and thus it is hard to compare results to each other.
1


1.2 Goal
The goal of this research is to examine multiple test methods for shear strength. Toward this end, how different surface treatments on a concrete substrate affects the shear strength at the interface of a bonded concrete overlay is evaluated. The tests were conducted on six pads with different surface treatments and where three of them are consolidated. Pull-off, guillotine, slant shear and jacking test were conducted on each pad. The outcome of the tests are compared and analyzed.
1.3 Outline
This thesis consists of seven chapters. The first chapter includes an overview and a description of the goal.
Chapter 2 contains a literature review of current research on shear strength and different methods to test the shear strength at the bonded interface.
Chapter 3 presents the design of the experiment, location and which methods that were used to create the different surface treatments.
Chapter 4 explains all the tests done on the pads, the conditions under which the testing was conducted, and the equations used to calculate shear strength based on the test results.
Chapter 5 presents tables of the data gathered from the four tests.
Chapter 6 compares the results from the four different tests and discusses possible sources of error throughout the experiment.
Chapter 7 contains a summary of the conclusions that can be made from the experiment.
2


Appendices A-E shows the raw data gathered from the tests and pictures of all the
samples.
3


2 Background
2.1 Introduction
When repairing or rehabilitating a concrete structure the unsound concrete is removed and replaced with repair materials; for the case of slabs this is known as an overlay. To make the structure last in a long term perspective, the goal is to create a combined section that resembles the original structure and possesses equivalent strength. A key factor is to have a shear strength at the repair interface that approaches the concrete itself. While bond strength is a measurement of how well the old and the new concrete surfaces are adhered to each other and can be measured by standard tests, shear strength is more elusive and is difficult to measure. It is influenced by several different factors discussed below. There are numerous ways to test the shear strength of a repair, yet none have achieved the status of an accepted industry standard. A repaired concrete structure is made up from an existing substrate and a new overlay with the crucial bonded interface in between.
2.2 Bond
Bond describes how well the overlay and substrate are adhered to each other. The interface between the concrete layers is important because its purpose is to create a bond that resembles the former structure so that the forces from the overlay gets distributed down to the substrate. It is these forces that the bond has to be able to resist. A good bond is when the bond strength is equal to or higher than the tensile strength of either the substrate or the overlay.
The bond can be described with either adhesion or from a mechanical point of view.
In terms of mechanical bonding, the bond mechanism is described by the interlock between
4


the two surfaces and is mainly influenced by the roughness of the substrate and the degree with which the overlay interlocks with the surface of the substrate. Adhesion describes the bonded interface and is usually quantified as the amount of stress or energy required to separate the two layers. It takes into account the mechanical interaction, thermodynamics and chemical bonding. The mechanical adhesion is highly dependent on the surface roughness and the substrate needs to have open cavities for the fresh concrete to anchor to. The surface moisture is very important for a good adhesion because the cement will with capillary absorption anchor to the old concrete. If the substrate surface moisture condition is too low, the fresh cement will not be able to tie to the substrate because it pulls to much water from the overlay, not allowing the overlay to hydrate which leads to a weak bond (Bissonnette et al. 2012). According to tests conducted at the Louisiana State University, the best moisture condition is saturated surface dry with a bond strength almost double compared to an air dry surface (Wan 2010).
When a former bond has been reduced to a level where it is no longer capable of resisting the differential forces between the substrate and overlay, debonding occurs. In a worst case it leads to delamination and slipping. The reasons behind failure can be many. In a long term perspective, a common factor is differential shrinkage and thermal expansion and contraction. If the two layers have different properties and do not act as a homogenous concrete section, the failure will occur at the weak link, which in a repair usually is the bonded interface and that is where the debonding occurs. Reasons behind different properties between them can be that the two concrete layers had different water to cement ratio, the aggregates are of different size, variations in the surface preparation of the substrate and consolidation of the overlay can affect bond or environmental differences at the time of
5


curing. These differences lead to a difference in modulus of elasticity between the substrate and the overlay. Shrinkage in the layers will not match each other. Because of different movement of the pads, over time the tension between them will be so high that the bond strength cannot resist the forces and will cause debonding (Bakhsh 2010).
The forces behind failure can be shear stresses originating from service loads on the concrete or tension or compression stresses caused by thermal influences and shrinkage. When the bond fails it fails in either tension or shear. To test the potential bond and/or shear strength there are many different methods, some that can be done in the field and some that have to be done in a laboratory (Wall and Shrive 1988). These are discussed below.
BONDED INTERFACE
Figure 2.1. Diagram of overlay and substrate According to Dr. Silfwerbrand at the Royal Institute of Technology, in order to achieve a good bond one of the most significant factors is to have a clean substrate surface when the overlay is placed. Dust, grease and other loose particles could significantly reduce the bond strength because it prevents the adhesion that is the key to a good bond. Silfwerbrand also mentions the importance of avoiding micro cracks in order to achieve a good bond. Some surface preparation methods may induce the presence of micro cracks and he asserts that even if a rougher surface is obtained with a certain method, the increasing amount of micro cracks reduces the gain from a rough surface and could lead to an overall weaker bond. Silfwerbrand asserts that the best method to ensure a minimum amount of micro cracks is to use water-jetting instead of mechanical methods such as chipping
6


(Silfwerbrand 1990). However, an experiment conducted by Talbot et al. shows that if a low mass jack hammer is carefully used on the surface, the presence of micro cracks will not be significant (Talbot et al. 1994).
2.3 Shear Strength
When it comes to shear, the interface mainly gets its strength from mechanical interlock and adhesion. Previous studies shows that there are significant differences in the shear strength with different interface textures. Higher macro roughness means a higher bond strength. It means that the most significant factor for good shear strength comes from the surface roughness. To take advantage of the surface roughness and make sure that it reaches its full potential, it is important to make sure that the overlay satisfactory penetrates the substrate surface. This can be done by compacting the overlay which also makes the overall strength of the concrete increase. A visualization of a compacted and non-compacted bond is shown in Figures 2.2 and 2.3 (Silfwerbrand and Paulsson 1998). Another important factor for the interface to resist shear forces is to have a good compressive concrete strength. For a rough surface to actually be able to hold the overlay in place, the concrete has to hold together to avoid slipping.
Figure 2.2. Overlay without compaction
7


BONDED
INTERFACE
Figure 2.3. Overlay with compaction
2.3.1 Shear Strength Test Methods
There are numerous tests to determine the shear resistance in the bonded interface. The ones presented below were used in the conducted experiment.
2.3.1.1 Slant Shear Test
The slant shear test, also known as Arizona slant shear test, is a combination of a compression test and a shear test where the concrete core sample is slanted with an inclined bonded interface. The incline is usually 30 degrees from the longitudinal axis. Because of the geometry of the specimen, the compression force will add to the measured bond strength due to a clamping effect (Ferraro 2008). A compression force is applied at the ends of the sample until failure happens at the bonded plane. If the two concrete layers have the same compressive strength and the shear strength at the interface is equal to or higher than the concrete itself, which is the ultimate goal, the failure mode should be of the same type as for a regular compression test. If the shear strength at the interface is less than the concrete itself, the failure occurs at the interface. The force is then divided by the area of the bonded plane which gives the shear strength. To get the normal and shear stresses of the sample the resultants of the axial compression force is calculated and divided by the bonded area. This test is hard to perform in the field because of the difficulties to acquire the slanted samples.
8


In a laboratory the specimens can be created using a mold with an insertion that creates samples with an inclined plane (Kriegh 1976).
BONDED INTERFACE
Figure 2.4. Diagram of slant shear test
2.3.1.2 Guillotine Test
The guillotine test is a method where a core sample with the bonded interface perpendicular to the longitudinal axis is placed in a guillotine shear jig. The core is resting in a half circle with the same diameter as the sample cut into the steel jig. Another steel piece with the same diameter fitting inside the bottom half of the jig is put on top of the core so that the bonded interface is located between the top and the bottom half of the jig. This exposes the core to two adjacent, equally sized and opposite forces. As seen in Figure 2.2 the core is subjected to a third force that gives support to the core. When a compression force is applied on top of the guillotine device, if positioned correctly, the core will shear off right at the interface. Because of the thickness of the steel used for the jig and the gap between the inner and outer jig, in practice the forces cannot be perfectly adjacent and a small eccentricity will occur which introduces a minor bending moment. This moment however is not considered to influence the failure significantly. The guillotine test is mainly a laboratory test because in the field with thicker overlays, it is difficult to extract undamaged cores (Delatte et al. 2000).
9


BONDED
INTERFACE
Figure 2.5. Diagram of guillotine shear test
2.3.1.3 Jacking Test
The jacking test is conducted by applying horizontal forces parallel to the interface. If conducted in the field, a jack can be used to apply pressure on the sample until a failure happens. A potential problem with this type of test is that a bending moment is introduced because of the eccentricity of the applied force. This bending moment is heavily depending on the jacks ram diameter. A larger ram means that the center of the applied force will have a larger eccentricity. Compared to the bending moment that occurs for the guillotine test, this moment is much more significant to the debonding of the specimen.
BONDED
Figure 2.6. Diagram of jacking test
2.4 Tensile Strength
For a bonded concrete interface to fail in tension there has to be tensile forces. If the surface is relatively rough, the overlay needs to be lifted by an upwards force in order to move. In general, forces like that are not very common. Still the tensile test is the most used
10


to determine the bond strength of a repaired structure simply because it is relatively easy to perform on site.
2.4.1 Pull-Off Test
The pull-off test is a direct tension test and the most common test method to evaluate bond strength. Even though shear strength is the most significant factor for bonding, the pull-off test is simple to use in the field, making it the most used bond test method. A steel puck is attached to the top of a core that is drilled at least 25.4 mm (1 in.) or half the core diameter into the substrate. The steel puck is attached to a testing apparatus that gradually increases a centric upward force normal to the specimen to the point of failure where the core is loose.
Figure 2.7. Diagram of pull-off test
It is of high importance that the load is applied perpendicular and without eccentricity to the core to avoid that a bending moment develops (Delatte et al. 2000). Depending on where the failure happens, at the overlay, substrate, or at the bonded interface, it gives a first implication of the strength of the bond. The failure could also happen at the epoxy used to attach the steel puck to the core or a combination of the failure mode (ASTM C1583 2015).
11


t
t
1
i
VVVVVV KVVVVVV _______
WWWWWV| -------
Figure 2.8. Pull-off test failure modes
a: Failure in substrate, b: Failure at bonded interface, c: Failure in overlay, d: Epoxy failure.
A satisfactory test would be if the break happens perfectly at the bonded interface, because that would indicate the true strength of the bond between the substrate and the overlay. If the break happens in the substrate it is a sign that the bond strength is higher than the substrate itself, which is the ultimate goal for a concrete repair (Bonaldo et al. 2005).
Because of the big differences in the mechanisms behind what makes a good bond between shear and tension, where the mechanical adhesion for shear mainly relies on the overall surface roughness while the tensile mechanical adhesion mainly get its strength from the vertical anchorage in pores and voids, it is hard to find a universal correlation ratio (Bissonnette et al. 2012). It could be situations where the bonded interface has a good shear resistance and a poor tensile resistance or the other way around, meaning that a correlation factor based on the first situation would not be applicable for the second case. However, Delatte et al. found during shear and tensile testing of concrete overlays that the shear stress is more than twice than the tension stress. Their results showed a shear stress over tension stress ratio that varied from 1.42 to 3.39, which gave an average of 2.04 and a standard deviation of 0.33 (Delatte et al. 2000).
12


3 Experiment
The experiment was conducted to evaluate the bond strength between a concrete overlay and a substrate for different types of substrate surface treatments. Three different surface treatments on six concrete test pads were prepared. Half of the overlays were compacted by vibration to evaluate the influence of compaction at each surface treatment, thus six different combinations of surface treatments and compaction were made available for testing. On each pad four different tests were conducted to evaluate the shear and tensile resistance for the different treatments. Cylinders were obtained for concrete compression tests. To acquire samples for the testing, on each pad three full depth core samples were gathered, three cores were drilled 25.4 mm (1 in.) into the substrate, two slanted full depth core samples were drilled and after all coring had been completed, the overlay was cut into rectangular blocks. Because pull-off tensile tests are the most common practice in the field, the results from that test was compared to the Guillotine test and a ratio factor is introduced.
3.1 Construction of Test Pads
The experiment was conducted outdoors at the University of Colorado Denver Civil Engineering Lab. The substrate forms rested on the underlying concrete floor. The pads were covered in a plastic sheet for curing and to serve as weather protection.
Six concrete pads named Pad 1 to Pad 6 with the substrate dimensions 1422 mm x 610 mm (56 in. x 24 in.) and overlay dimensions 991 mm x 559 mm (39 in. x 22 in.) as seen in figures 3.1 and 3.2 were cast. The size of the pads was mainly influenced by the length required to drill for the two slanted cores inclined at 30 degrees.
13


OVERLAY SUBSTRATE INSULATION BOARDS EXISTING CONCRETE FLOOR
Figure 3.1. Experiment design: Plan
Figure 3.2. Experiment design: Elevation Due to the dimensions of the standard lumber that the forms were made of the thickness of the substrate and overlay were 82.6 mm (3 Vi in.) and 88.9 mm (3'A in.) respectively. The substrate was made longer than the overlay to fit a steel angle acting as support to the jack for the jacking test. Three 19.1 mm (3/i in.) diameter bolts were cast into the substrate to anchor the steel angle. 9.53 mm (No. 3) hairpin reinforcing bars were installed around each bolt to resist tension when applying a horizontal force on the steel angle. The hairpins were located approximately 25.4 mm (1 in.) down from the surface. The substrates were cast on a 102 mm (4 in.) thick layer of insulation boards that served as protection to prevent damage to the underlying concrete floor when coring.
14


Figure 3.3. Substrate Formwork
Forms for six substrate pads with insulation boards and 9.525 mm (#3) rebar hairpins can
be seen.
The concrete used for both the substrate and overlay was made using Quikrete 5000 with a labeled 28 day compressive strength of 34500 kPa (5000 psi). The Quikrete was mixed with water in a drum mixer. The substrates were cast on two different days, with pads 5 and 6 were cast four days after the first four. To avoid big differences in shrinkage between the overlay and the substrate, the overlays were cast only 14 days after the first substrates were cast. Before the overlay was cast, the substrate surface was cleaned with high pressure water and compressed air. The overlays were compacted using an electric three horsepower vibrator with a 34.9 mm (l3/s in.) square head.
15



Figure 3.4. Vibrator used for compaction Model: Wyco Sure Speed WSD1
3.2 Surface Preparation
The six substrate surfaces were prepared with three different treatments and for each type of surface treatment, one overlay was compacted by vibration and one was not, giving six unique combinations affecting the bond. The six combinations are listed in Table 5.1.
The broom finish was performed on pad 1 and 2 with a stiff poly fibered push broom shortly after the concrete had been placed.
Figure 3.5. Broom finish Made with stiff bristle broom
16


The rake finish was performed on pad 3 and 4 short after the concrete was placed. The rake used was a garden rake with a tooth spacing of 25.4 mm (1 in.) and the rake marks were approximately 19.1 mm (3A in.) deep.
Figure 3.6. Rake finish
Grooves were approximately 19.1 mm (3A in.) deep and 25.4 mm (1 in.) apart The bush hammered finish was performed on pad 5 and 6 after eight days of curing using an electric jackhammer with a bush hammer attachment. The surface is shown in Figure 3.7 and the bush hammer attachment is shown in Figure 3.8.
Figure 3.7. Bush hammer finish All laitance was removed and the surfaced roughened as shown
17


Figure 3.8. Bush hammer used for surface finish
18


4 Testing
For all tests except the jacking test, a core bit with 76.2 mm (3 in.) diameter and 406 mm (16 in.) of travel length was used to acquire the samples. The real diameter of all specimens using the core bit is 66.7 mm (25/s in.) which was used in all calculations. The guillotine and compression tests were conducted in the University of Colorado Denver Civil Engineering Lab using MTS machines. The jacking and pull-off tests were conducted directly on the pads and the slant shear test was conducted at the CTL|Thompson testing facility in Denver, CO.
4.1 Compression Test
For each substrate surface preparation, three samples of the concrete mix were collected and for the overlay concrete mix a total of three samples were collected. The samples were collected in 203 mm (8 in.) tall Forney cylinder molds with a diameter of 102 mm (4 in.). The first testing took place when the concrete had cured 49 days for substrate 1 to 4, 45 days for substrate 5 and 6 and 35 days for the overlays. At this time two cylinders from each pad were tested. The second day of testing took place 21 days after the first test, when the concrete had cured 70 days for substrate 1 to 4, 66 days for substrate 5 and 6 and 56 days for the overlays. The testing was conducted in a MTS 810 load frame, seen in Figure 4.1, with a maximum load capacity of 980 kN (220 kips). The cylinders were equipped with neoprene pad caps on each end which can be seen in Figure 4.2 and the tests were run under displacement control at a rate of 1.25 mm/minute (0.05 in/minute).
19


Figure 4.1. MTS 810 load frame
Figure 4.2. Compression test set up
r
c
F
A
(4.1)
Where:
f'c = Compressive Strength F = Applied Force A = Cross-section Area
20


4.2 Pull-Off Test
The pull-off test consisted of 18 specimens with three tests for each pad. The test specimens were prepared by core drilling through the overlay and into the substrate by approximately 25.4 mm (1 in.) to make sure that failure could occur in all three stages, overlay, substrate and at the bonded interface. The surface of the specimens was cleaned with acid to remove all laitance and for cleaning and thoroughly rinsed with water before 76.2 mm (3 in.) diameter steel pucks, twelve 20.6 mm (13/ie in.) and seven 12.7 mm (V2 in.) thick, were attached with epoxy. The epoxy was left to cure for three days before the pull-off test was conducted. A 007 James Bond Tester from James Instruments was attached to the steel pucks via a threaded rod and the three bolts on the mount seen in Figure 4.3 were used to level the tester to ensure that the force was applied perpendicular to the specimens.
Figure 4.3. 007 James bond tester
The force required to detach the specimens from the pads was recorded and divided by the area of the specimens to acquire the tensile stress. All specimens failed at or approximately 3.18 mm (Vs in.) from the bonded interface.
21


(4.2)
F
Where:
a = Tensile Stress F = Applied Tensile Force A = Surface Area of Specimen
4.3 Slant Shear Test
The slant shear test was conducted on a total of 12 cores, two from each pad. To acquire the core samples inclined at 30 degrees, a special stand with an angle of 15 degrees was constructed because the core drill mount could only be inclined 45 degrees, as seen in Figure 4.4.
Figure 4.4. Core drilling for slant shear test
The core samples were cut off at each end about 25.4 mm (1 in.) out from the bonded interface to make sure that the applied force was distributed equally over the core surface. Figure 4.5 shows a diagram of the slant core sample.
22


BONDED INTERFACE
Figure 4.5. Core sample with slanted interface
The cores were placed in the testing machine with neoprene pad caps on each end and a compression force was applied. Figure 4.6 shows the testing machine and Figure 4.7 a close-up on the core sample.
Figure 4.6. Slant shear test setup
23


Figure 4.7. Close-up on slant shear sample The applied force was measured and the shear (r) and normal (a) stresses were calculated using the equations presented below.
r
a =
F* cos 30 A
F sin 30 A
(4.3)
(4.4)
Where:
r = Shear Stress a = Normal Stress
F = Applied Axial Compression Force A = Bonded Area
All samples failed in shear at the bonded interface.
24


4.4 Guillotine Test
The guillotine test was conducted on 18 specimens, three from each pad. The core drill was leveled with the pads to ensure that the samples were acquired perpendicular to the surface of the pads. The samples were approximately 171 mm (63A in.) long, the full depth of overlay and substrate.
Figure 4.8. Vertical core drilling for guillotine test The test was conducted in a MTS load frame, model 312 with a load capacity of 90 kN (20 kips). The samples were placed in the guillotine device shown in Figure 4.9 with the bonded interface between the inner and outer jig.
25


Figure 4.9. Shear jig design
A compression force was applied on top of the shear jig and was run under displacement control at the rate of 0.5 mm/minute (0.02 in/minute). The force required to shear off the sample was monitored. Because of the design of the shear jig, with the core sample supported by the jig at the rear end and at the bonded plane, the applied force had to be divided by 2 to obtain the shear failure force at the bond plane, as shown in Equation 4.5. The test setup can be seen in Figure 4.10.
Where:
r = Shear Stress F = Applied Force
A = Cross-section Area at Bonded Interface
(4.5)
26


4.5 Jacking Test
For the jacking test, the pads were saw cut into blocks using a concrete saw as seen in Figure 4.11. For each pad, three solid blocks plus a various amount of blocks with holes from previous coring were prepared. The solid blocks were approximately 102 mm x 203 mm (4 in. x 8 in.) and the cut was made through the entire overlay and approximately 6.35 mm (Vi in.) into the substrate.
Figure 4.11. Blocks on pads 5 and 6 prepared for jacking test The steel angle was mounted with the bolts and the jack, a Simplex RC306C with a 300 kN (30 ton) capacity and 152 mm (6 in.) of travel length, put in place. Because of the length of the jack, for all of the solid blocks the steel angle had to be rotated 180 degrees, meaning that the bolts supported the rear end of the angle inducing a minor upwards bending of the angle as seen in Figure 4.12. Due to the diameter of the jack and the variable surface treatments, the ram head encountered the blocks at different heights, resulting in variations in eccentricity and bending moments between the blocks.
27


Figure 4.12. Jack set up for testing
Once the blocks were detached from the pads, the exact areas were measured and adjusted due to discontinuities from coring and rebar chairs.
t = - (4.6)
Where:
r = Shear Stress F = Applided Jacking Force A = Bonded Area
28


5 Experiment Results
The results from the four different tests are presented in this chapter. In Table 5.1 the different surface treatments are summarized.
Table 5.1: Surface treatments
Pad Surface Treatment Vibrated
1 Broom Yes
2 Broom No
3 Rake Yes
4 Rake No
5 Bush Hammer Yes
6 Bush Hammer No
5.1 Compression Test
The results from testing the three specimens from each surface treatment category and the overlays are shown in Table 5.2.
________________________Table 5.2: Compression test results___________________
Sample Age Compression Force Compressive Strength
(Days) kN (lb) kPa (psi)
Substrate 1,2 49 323 (72 700) 39 900 (5 780)
Substrate 1,2 49 333 (74 900) 41 100 (5 960)
Substrate 3,4 49 310 (69 800) 38 300 (5 550)
Substrate 3,4 49 291 (65 500) 35 900 (5 210)
Substrate 5,6 45 179 (40 300) 22 100 (3 200)
Substrate 5,6 45 167 (37 600) 20 600 (2 990)
Overlay 1-6 35 242 (54 500) 29 900 (4 340)
Overlay 1-6 35 262 (58 900) 32 300 (4 680)
Substrate 1,2 70 354 (79 500) 43 600 (6 330)
Substrate 3,4 70 340 (76 400) 41 900 (6 080)
Substrate 5,6 66 177 (39 900) 21 900 (3 170)
Overlay 1-6 56 334 (75 100) 41 200 (5 970)
29


The increase in compressive strength with time was accounted for. In Table 5.3 are the compressive strengths for each pad calculated based on the age of the concrete at the time of each test. A quadratic relationship based on the test data in Table 5.2 was used to approximate the strength at different dates, as illustrated in Figure 6.1.
_______Table 5.3: Interpolated compressive strength at the time of testing________
Test
Poured In
Compressive Strength kPa (psi)
Slant Shear Test Substrate 1,2 43 600 (6 330)
Slant Shear Test Substrate 3,4 41 900 (6 080)
Slant Shear Test Substrate 5, 6 21 900 (3 170)
Slant Shear Test Overlay 1-6 41 200 (5 970)
Guillotine Test Substrate 1,2 43 600 (6 320)
Guillotine Test Substrate 3,4 41 200 (5 980)
Guillotine Test Substrate 5, 6 21 900 (3 170)
Guillotine Test Overlay 1-6 39 000 (5 650)
Jacking Test Substrate 1,2 43 600 (6 330)
Jacking Test Substrate 3,4 41 900 (6 080)
Jacking Test Substrate 5, 6 21 900 (3 170)
Jacking Test Overlay 1-6 41 200 (5 970)
Pull-Off Test Substrate 1,2 41 200 (5 970)
Pull-Off Test Substrate 3,4 37 900 (5 490)
Pull-Off Test Substrate 5, 6 21 600 (3 140)
Pull-Off Test Overlay 1-6 32 300 (4 690)
Table 5.3 shows that the overlays for each pad all have the same measured compressive strength due to the fact that a total of three cylinders from one batch of overlay mix were collected.
30


5.2 Pull-Off Test
The results from the three specimens from each pad is presented in Table 5.4.
Table 5.4a: Pull-off test results (SI units)
fc Substrate f c Overlay (kPa) Tensile Stress (o) (kPa) a a
Sample
(kPa) f'c J min f'c \J min
IB 41 200 32 300 2480 0.08 13.80
ID 41 200 32 300 1780 0.06 9.90
IF 41 200 32 300 1560 0.05 8.68
2A 41 200 32 300 1240 0.04 6.90
2C2 41 200 32 300 1150 0.04 6.40
2E 41 200 32 300 1340 0.04 7.46
3B 37 900 32 300 1800 0.06 10.02
3D 37 900 32 300 2290 0.07 12.74
3F 37 900 32 300 1780 0.06 9.90
4A 37 900 32 300 990 0.03 5.51
4C 37 900 32 300 1310 0.04 7.29
4E 37 900 32 300 1560 0.05 8.68
5B 21 600 32 300 890 0.04 6.06
5D 21 600 32 300 1470 0.07 10.00
5F 21 600 32 300 760 0.04 5.17
6A 21 600 32 300 1150 0.05 7.82
6C 21 600 32 300 990 0.05 6.74
6E 21 600 32 300 600 0.03 4.08
31


Table 5.4b: Pull-off test results (Imperial units)
fc Substrate (psi) f c Overlay (psi) Tensile Stress (o) (psi) a a
Sample f'c J min If'c jJ min
IB 5 970 4 690 360 0.08 5.26
ID 5 970 4 690 260 0.06 3.78
IF 5 970 4 690 230 0.05 3.31
2A 5 970 4 690 180 0.04 2.63
2C2 5 970 4 690 170 0.04 2.43
2E 5 970 4 690 190 0.04 2.83
3B 5 490 4 690 260 0.06 3.80
3D 5 490 4 690 330 0.07 4.86
3F 5 490 4 690 260 0.06 3.78
4A 5 490 4 690 140 0.03 2.09
4C 5 490 4 690 190 0.04 2.77
4E 5 490 4 690 230 0.05 3.31
5B 3 140 4 690 130 0.04 2.30
5D 3 140 4 690 210 0.07 3.77
5F 3 140 4 690 110 0.03 1.97
6A 3 140 4 690 170 0.05 2.95
6C 3 140 4 690 140 0.05 2.54
6E 3 140 4 690 90 0.03 1.54
Table 5.5 shows that vibrated broom and vibrated rake surfaces, pad 1 and 3, have almost identical capacity while the vibrated bush hammered surface has a capacity of slightly higher than half of the others. From Table 5.6 the results show that for broom and rake surfaces, the vibration increased the capacity more than 50% while the bush hammered surfaces only increased 14% after vibration.
32


Table 5.5a: Pull-off test results Average tensile stress (SI units)
Pad Avg. Tensile Stress (a) (kPa) a
If'c J min
1 1 940 10.79
2 1 240 6.92
3 1 960 10.89
4 1 280 7.16
5 1 040 7.08
6 910 6.21
Table 5.5b: Pull-off test results Average tensile stress (Imperial units)
Pad Avg. Tensile Stress (a) (psi) a
Ifc J min
1 282 4.11
2 180 2.63
3 284 4.15
4 186 2.72
5 151 2.68
6 132 2.35
Table 5.6: Pull-off test results Influence from vibration
Surface Treatment Vibrated Pad Capacity Non Vibrated Pad Capacity
Broom (Pad 1 & 2) 156%
Rake (Pad 3 & 4) 152%
Bush Hammer (Pad 5 & 6) 114%
33


Table 5.7 presents the statistical results for the test.
Table 5.7a: Pull-off test results Statistical results (SI units)
Pad J ^ C min) Standard deviation Coefficient of Variation
1 10.79 2.67 24.8%
2 6.92 0.53 7.6%
3 10.89 1.61 14.8%
4 7.16 1.59 22.2%
5 7.08 2.57 36.3%
6 6.21 1.92 31.0%
Table 5.7b: Pull-off test results Statistical results (Imperial units)
Pad / ^ C min) Standard deviation Coefficient of Variation
1 4.11 1.02 24.8%
2 2.63 0.20 7.7%
3 4.15 0.62 14.8%
4 2.72 0.61 22.4%
5 2.68 0.96 35.9%
6 2.35 0.73 30.9%
34


5.3 Slant Shear Test
The result from each slant shear specimen is presented in Table 5.8.
Table 5.8a: Slant shear test results (SI units)
Normal stress and shear stress calculated from Eg, 4,3 and 4,4
f c Substrate (kPa) f c Overlay (kPa) Normal Stress (a) (kPa) Shear Stress (t) (kPa) T
Sample lf'C J min
1G 43 600 41 200 9 100 15 700 77.3
IQ 43 600 41 200 8 400 14 600 71.9
2G 43 600 41 200 6 100 10 500 51.7
2Q 43 600 41 200 5 700 9 800 48.3
3G 41 900 41 200 8 500 14 800 72.9
3Q 41 900 41 200 8 500 14 800 72.9
4G 41 900 41 200 6 300 10 900 53.7
4Q 41 900 41 200 6 000 10 300 50.7
5G 21 900 41 200 6 500 11 200 75.7
5Q 21 900 41 200 5 800 10 000 67.6
6G 21 900 41 200 4 300 7 400 50.0
6Q 21 900 41 200 3 900 6 700 45.3
Table 5.8b: Slant shear test results (Imperial units) Normal stress and shear stress calculated from Eq. 4.3 and 4.4
f c Substrate (psi) f c Overlay (psi) Normal Stress (a) (psi) Shear Stress (t) (psi) T
Sample lf'C J min
1G 6 330 5 970 1 320 2 280 29.5
IQ 6 330 5 970 1 220 2 120 27.4
2G 6 330 5 970 880 1 520 19.7
2Q 6 330 5 970 820 1 420 18.4
3G 6 080 5 970 1 240 2 150 27.8
3Q 6 080 5 970 1 240 2 150 27.8
4G 6 080 5 970 910 1 580 20.4
4Q 6 080 5 970 870 1 500 19.4
5G 3 170 5 970 940 1 620 28.8
5Q 3 170 5 970 840 1 450 25.8
6G 3 170 5 970 630 1 080 19.2
6Q 3 170 5 970 560 970 17.2
35


Table 5.9 shows that pad 1 and 3, vibrated rake and vibrated broom surface, had the highest capacity. The results in Table 5.10 shows that the influence from vibration is significant for all different surface treatments.
Table 5.9a: Slant shear results Average shear stress (SI units)
Pad Avg. Shear Stress (t) (kPa) T
If'c J min
1 15 200 74.6
2 10 100 50.0
3 14 800 72.9
4 10 600 52.2
5 10 600 71.6
6 7 100 47.6
Table 5.9b: Slant shear results Average shear stress (Imperial units)
Pad Avg. Shear Stress (t) (psi) T
If'c J min
1 2 200 28.5
2 1 470 19.0
3 2 150 27.8
4 1 540 19.9
5 1 540 27.4
6 1 030 18.3
Table 5.10: Slant shear results Influence from vibration
Surface Treatment Vibrated Pad Capacity Non Vibrated Pad Capacity
Broom (Pad 1 & 2) 150%
Rake (Pad 3 & 4) 140%
Bush Hammer (Pad 5 & 6) 150%
36


5.4 Guillotine Test
In Table 5.11 the results from three specimens from each pad is presented.
Table 5.11a: Guillotine test results (SI units)
fc Substrate (kPa) f c Overlay (kPa) Shear Stress (t) (kPa) T
Sample Ifc J min
1A 43 600 39 000 2 640 13.37
1C 43 600 39 000 3 260 16.51
IE 43 600 39 000 4 340 21.98
2B 43 600 39 000 1 730 8.76
2D 43 600 39 000 1 780 9.01
2F 43 600 39 000 1 210 6.13
3A 41 200 39 000 3 710 18.79
3C 41 200 39 000 2 500 12.66
3E 41 200 39 000 3 240 16.41
4B 41 200 39 000 3 150 15.95
4D2 41 200 39 000 2 750 13.93
4F 41 200 39 000 2 360 11.95
5A 21 900 39 000 1 860 12.57
5C 21 900 39 000 2 570 17.37
5E 21 900 39 000 1 650 11.15
6B3 21 900 39 000 2 430 16.42
6D 21 900 39 000 2 090 14.12
6F 21 900 39 000 1 100 7.43
37


Table 5.11b: Guillotine test results (Imperial units)
fc Substrate (psi) f c Overlay (psi) Shear Stress (t) (psi) T
Sample If'c J mm
1A 6320 5650 383 5.10
1C 6320 5650 473 6.30
IE 6320 5650 630 8.38
2B 6320 5650 251 3.35
2D 6320 5650 258 3.43
2F 6320 5650 175 2.33
3A 5980 5650 538 7.16
3C 5980 5650 362 4.82
3E 5980 5650 470 6.25
4B 5980 5650 457 6.08
4D2 5980 5650 399 5.31
4F 5980 5650 343 4.56
5A 3170 5650 270 4.80
5C 3170 5650 373 6.62
5E 3170 5650 239 4.25
6B3 3170 5650 352 6.26
6D 3170 5650 302 5.37
6F 3170 5650 160 2.84
Table 5.12 shows that pad 3, vibrated rake surface, has the highest capacity followed by vibrated broom and vibrated bush hammer surface. Table 5.13 indicates that the vibration for the broom surface is of high significance while for the rake and bush hammer surfaces it did not increase the capacity with more than 14 and 8 percent respectively.
38


Table 5.12a: Guillotine test results Average shear stress (SI units)
Pad Avg. Shear Stress (t) (kPa) T
If'c J min
1 3 420 17.3
2 1 570 8.0
3 3 150 16.0
4 2 760 13.9
5 2 030 13.7
6 1 870 12.7
Table 5.12b: Guillotine test results Average shear stress (Imperial units)
Pad Avg. Shear Stress (t) (psi) T
If'c J min
1 496 6.60
2 228 3.03
3 457 6.08
4 400 5.32
5 294 5.22
6 272 4.82
Table 5.13: Guillotine test results Influence from vibration
Surface Treatment Vibrated Pad Capacity Non-Vibrated Pad Capacity
Broom (Pad 1 & 2) 217%
Rake (Pad 3 & 4) 114%
Bush Hammer (Pad 5 & 6) 108%
39


Table 5.14 presents the statistical results for the test.
Table 5.14a: Guillotine test results Statistical results (SI units)
Pad /r j ( ^ Standard Deviation C oeffi ci ent of Vari ati on
C min)
1 17.3 4.36 25.2%
2 8.0 1.60 20.1%
3 16.0 3.09 19.4%
4 13.9 2.00 14.3%
5 13.7 3.26 23.8%
6 12.7 4.67 36.9%
Table 5.14b: Guillotine test results Statistical results (Imperial units)
Pad M | < \ Standard Deviation C oeffi ci ent of V ari ati on
C min)
1 6.60 1.66 25.2%
2 3.03 0.61 20.3%
3 6.08 1.18 19.4%
4 5.32 0.76 14.3%
5 5.22 1.24 23.8%
6 4.82 1.77 36.8%
40


5.5 Jacking Test
The results from the three solid blocks from each pad is presented in table 5.15.
Table 5.15a: Jacking test block data (SI units)
f c Substrate (kPa) f c Overlay (kPa) Shear Stress (t) (kPa) T
Sample lf'C J mm
1H 43 600 41 200 570 2.81
11 43 600 41 200 810 3.99
1J 43 600 41 200 600 2.96
2H 43 600 41 200 720 3.55
21 43 600 41 200 620 3.05
2J 43 600 41 200 820 4.04
3H 41 900 41 200 970 4.78
31 41 900 41 200 960 4.73
3J 41 900 41 200 1 320 6.50
4H 41 900 41 200 570 2.81
41 41 900 41 200 1 220 6.01
4J 41 900 41 200 600 2.96
5N 21 900 41 200 1 180 7.97
50 21 900 41 200 1 850 12.50
5P 21 900 41 200 1 940 13.11
6H 21 900 41 200 1 270 8.58
61 21 900 41 200 820 5.54
6J 21 900 41 200 850 5.74
41


Table 5.15b: Jacking test block data (Imperial units)
f c Substrate (psi) f c Overlay (psi) Shear Stress (t) (psi) T
Sample Ifc J min
1H 6 330 5 970 83 1.07
11 6 330 5 970 117 1.52
1J 6 330 5 970 87 1.12
2H 6 330 5 970 105 1.36
21 6 330 5 970 90 1.17
2J 6 330 5 970 119 1.54
3H 6 080 5 970 141 1.83
31 6 080 5 970 139 1.80
3J 6 080 5 970 191 2.48
4H 6 080 5 970 83 1.08
41 6 080 5 970 178 2.30
4J 6 080 5 970 87 1.13
5N 3 170 5 970 171 3.04
50 3 170 5 970 268 4.76
5P 3 170 5 970 282 5.01
6H 3 170 5 970 184 3.27
61 3 170 5 970 119 2.11
6J 3 170 5 970 123 2.18
Table 5.16 shows that the vibrated bush hammered treatment has the highest capacity followed by vibrated rake surface and vibrated broom surface. Table 5.17 indicates that vibration is of high significance for the bush hammer treatment and moderately impacts the raked surface while the broom surface experienced a decrease in capacity after vibration, which is the only test where the vibrated pad had a lower capacity than the non-vibrated pad with the same surface treatment.
42


Table 5.16a: Jacking test results Average shear stress (SI units)
r
Pad Shear Stress (t) (kPa) If'c J min
1 660 3.25
2 720 3.55
3 1 080 5.34
4 800 3.92
5 1 660 11.19
6 980 6.62
Table 5.16b: Jacking test results Average shear stress (Imperial units)
Shear Stress (t) T
Pad If'c J min
(psi)
1 96 1.24
2 105 1.35
3 157 2.04
4 116 1.50
5 240 4.27
6 142 2.52
Table 5.17: Jacking test results Influence from vibration
Surface Treatment Vibrated Pad Capacity Non Vibrated Pad Capacity
Broom (Pad 1 & 2) 91%
Rake (Pad 3 & 4) 135%
Bush Hammer (Pad 5 & 6) 169%
43


Table 5.18 presents the statistical results for the test.
Table 5.18a: Jacking test results Statistical results (SI units)
Pad M | ( a \ Standard Deviation Coefficient of Variation
C min)
1 3.25 0.64 19.8%
2 3.55 0.49 13.9%
3 5.34 1.01 18.9%
4 3.92 1.81 46.1%
5 11.19 2.81 25.1%
6 6.62 1.70 25.7%
Table 5.18b: Jacking test results Statistical results (Imperial units)
Pad M j Standard Deviation Coefficient of Variation
C min)
1 1.24 0.25 19.9%
2 1.35 0.18 13.6%
3 2.04 0.38 18.8%
4 1.50 0.69 45.9%
5 All 1.07 25.1%
6 2.52 0.65 25.7%
44


6
Discussion
6.1 Conducting Experiments
Because the compressive strength was only tested on two different occasions, an approximate relationship was assumed for calculation of compression strength for tests conducted on days between the two test dates. The time-strength relationship is in reality a curve that approaches an asymptote once the concrete reaches its maximum strength. At the time of the second test the specimens were so old that it can be assumed that the concrete had reached its maximum strength, which was used for the jacking and slant shear tests that were tested at times after the last compression test. The other two tests uses a quadratic relationship based on the two known strengths and zero strength at the time of casting. The first test was conducted relatively late, when the concrete had cured between 35 and 49 days, which is a time when the concrete time-strength curve should have started to approach a horizontal asymptote and it should make the margin of error relatively small and therefore acceptable. To calculate the concrete compressive strength at the time for each test, a relationship for each pad and test was created based on the three known compressive strengths. The relationship can be seen in figure 6.1, which shows the development of compressive strength for substrate 1 and 2.
45


Developement of Compressive Strength
S
£
m cd > tn
M '
(D

a
£
o
U
50 000
Concrete Age (Days)
8000 6000 4000 2000 0
Figure 6.1. Compressive strength development The Slant Shear test resulted in higher capacity than the other three tests due to the clamping effect. Because of the specimens geometry with the inclined interface and the angle of load application, the axial compression load is both shearing off the specimen but also adds to the bonding by a compression force, a clamping effect. This effect is depending on the coefficient of friction, which directly relates to the shear bonding. A good bond will result in a higher clamping effect, and with that a higher capacity. Because the specimens all have the same geometry with the same inclination, it can be assumed that the clamping effect will develop equally in the different tests and therefore is an expression of the bond.
Because of the different surface treatments, the eccentricity for the jack in the jacking test differed for the different pads. The broom and bush hammer finished pads should, due to the relatively flat surface have similar eccentricity. The rake surface was very uneven which made the eccentricity differ between the blocks, meaning that the bending moment most likely was higher for the raked pads and that it could have affected the results.
46


6.2 Experiment Results
Scatter plots for the average results per pad for each type of test are shown in Figures
6.1 to 6.4 and in Figure 6.5 are the plots combined and the total average is shown. In order to compare the different pads to each other and see which pads that performed best on each test, the figures are accompanied by a table, Tables 6.1 to 6.4. In those tables the pads are ranked against each other using equation 6.1.
Where:
r = Shear Stress f'c = Compressive Strength
Equation 6.1 gives each pad a ratio based on the pad that performed best for each test. The nominator is the specific pads value and the denominator is the value from the best performing pad, meaning that the pad with the best result will get the number 100 % in Tables 6.1 to 6.4 and the remaining pads will show the capacity in relation to the best performing pad.
47


6.2.1
Slant Shear Test
Average (rNfc) (SI units) ooooooooo _
1 1 1 1 1

c ] c ]
c ]




2 3 4 5 6 Pad Compacted by Vibration Without Compaction
Figure 6.2a. Slant shear test (SI units)
The influence of compaction by vibration can be seen
3 X 25 5 1 20 Oh E 'X 15 4 ^ 10 0 :
l 1 1 1 1
r l
C ] L J c ]



[2 3 4 5 6 Pad Compacted by Vibration Without Compaction
Figure 6.2b. Slant shear test (Imperial units)
The influence of compaction by vibration can be seen
48


Table 6.1: Slant shear test Ranking by capacity
Ratio based on highest compressive strength of either substrate or overlay.
Rank by Capacity Pad (X/7v),
( A/TgL*
1 1 100%
2 3 97%
3 5 96%
4 4 70%
5 2 67%
6 6 64%
Table 6.1 shows that the results from the vibrated pads, 1, 3 and 5, are very close to each other and the difference is only 4% between pad 1 and pad 5 and 3% between pad 1 and pad 3. The non-vibrated pads shows results with a capacity of 64% to 70% of the top ranked pad 1. Because of the relatively small difference comparing the three compacted pads to each other and comparing the three non-compacted pads to each other, a conclusion can be reached that for the slant shear test it is mainly the vibration that influenced on the result, and only minor influence due to the different surface treatments. It also shows that the clamping effect discussed earlier results in a very high shear capacity which in combination with the small differences between the pads means that the clamping effect is quite significant and evens out the differences between the pads surface treatments.
6.2.2 Guillotine Test
Figure 6.3 shows that the non-vibrated broom finish has an unusually low shear strength, significantly lower than the other pads in the Guillotine test.
49


20 18 < K 16 1 14 X12 10 8
A k
% r /
\ > i O

/ s
\


L. 0 ]
12 3 4 5 6 Pad Compacted by Vibration O Witliout Compaction
Figure 6.3a. Guillotine test Comparison (SI units)
7
Average (t/yfc) (Imperial units) O'tOGJ.pi.OiO',- 4
< > <
< >
< >



[2 3 4 5 6 Pad Compacted by Vibration O Without Compaction
Figure 6.3b. Guillotine test Comparison (Imperial units)
50


Table 6.2: Guillotine test Ranking by capacity
Ratio based on highest compressive strength of either substrate or overlay
Rank by Capacity Pad (T/V7v), ( a/^L
1 1 100%
2 3 93%
3 4 81%
4 5 79%
5 6 73%
6 2 46%
Table 6.2 shows that the two pads with the highest capacity, pad 1 and pad 3, are the compacted broom and raked surfaces. The third best capacity has pad 4, non-compacted raked surface which has a capacity slightly higher than the compacted bush hammer pad meaning that the bush hammer surfaces overall did not perform well on the Guillotine Test.
6.2.3 Jacking Test
Figure 6.4 shows that the compacted bush-hammered surface has an unusually high strength and that the non-vibrated bush-hammered surface has a strength higher than all other pads, even the compacted ones.
51


12 10 r/3 2 f 8 go f6 A k

L \
A k > V
5-H A i 2 0 ] A L Z 1 L A

12 3 4 5 6 Pad Compacted by Vibration A Without Compaction
Figure 6.4a. Jacking test Comparison (SI units)
4.5 | 4
A i
Average (W/c) (Imperial units) O H- to u

A

/
k ' L


1 2 3 4 5 6 Pad A Compacted by Vibration A Without Compaction
Figure 6.4b. Jacking test Comparison (Imperial units)
52


Table 6.3: Jacking test Ranking by capacity
Ratio based on highest compressive strength of either substrate or overlay
Rank by Capacity Pad (T/V7v).
( a/^L
1 5 100%
2 6 59%
3 3 48%
4 4 35%
5 2 32%
6 1 29%
Table 6.3 shows that the top ranked pad 5, vibrated bush hammer surface has a capacity much higher than the second ranked pad 6, non-vibrated bush hammer surface that has a capacity of 59% compared to pad 5. Going down the list we can see that the ranking is sorted surface treatment by surface treatment, where bush hammer surface is ranked first, raked surface second and broom surface last. The broom surface has a non-vibrated capacity higher than the vibrated broom surface. The jacking test is the only test in which a non-vibrated surface treatment shows a higher capacity than its vibrated counterpart.
6.2.4 Pull-Off Test
Figure 6.5 shows that the compacted broom and compacted rake surfaces had a shear strength significantly higher than the other pads.
53


Average (Wfc) (SI units) -> Os oo o to
>

c ) c 5 < o


0 ]
[2 3 4 5 6 Pad Compacted by Vibration o Without Compaction
Figure 6.5a. Pull-off test Comparison (SI units)
4.5 ^ 4.0 1 § 3.5 1 3.0 Oh 2.5 <2.0 < ^ 1.5 > 4


c ) c >
c )




1 2 3 4 5 6 Pad Compacted by Vibration o Without Compaction
Figure 6.5b. Pull-off test Comparison (Imperial units)
54


Table 6.4: Pull-off test Ranking by capacity
Ratio based on highest compressive strength of either substrate or overlay
Rank by Capacity Pad (a/V7v),
( A/TcL*
1 3 100%
2 1 99%
3 4 66%
4 5 65%
5 2 63%
6 6 57%
From Table 6.4 it is shown that pad 3 has the highest capacity but that pad 1 has a capacity 99% of pad 3, which shows that the two pads surfaces have a very similar tensile strength. After the two top ranked pads pad 4, 5 and 2 with capacities 66%, 65% and 63% of pad 3 respectively is on the list with a gap down to pad 6 with only 57% of pad 3.
6.2.5 Summary
The different test results illustrate different ways of describing the same bonded interface. Figure 6.6 clearly shows that no matter which test is used, the same pattern for the pads appears.
55


80
2 70 E
50
40
30
20
10
0

r
L J


: o i { O
i A L l a £ a
1 2 3 4 5 i 6
Pad
Slant Shear Test Guillotine Test
A Jacking Test Pull-Off Test
Figure 6.6a. Summary of test results (SI units)
Solid symbols represent overlays consolidated by vibration while hollow symbols represent
unconsolidated overlays.
1 1
r
c ] L ]

4 o /S
A L . i ) A O 4 i A A [ 0 A
1 2 3 4 5 6
Pad
Slant Shear Test Guillotine Test
AJackingTest Pull-OfFTest
Figure 6.6b. Summary of test results (Imperial units)
Solid symbols represent overlays consolidated by vibration while hollow symbols represent
unconsolidated overlays.
56


Table 6.5: Average influence from vibration
Surface Treatment Vibrated Pad Capacity Non Vibrated Pad Capacity
Broom (Pad 1 & 2) 154%
Rake (Pad 3 & 4) 135%
Bush Hammer (Pad 5 & 6) 135%
Looking at the results in Figure 6.6, it is shown that the influence from compaction overall is of high significance. For all surface treatments in all tests, except the broom finish in the jacking test, the compacted pad for each surface treatment has a significantly higher capacity than the non-compacted pad. The broom finish in the jacking test has a non-compacted capacity about 10% higher than its compacted counterpart. The compaction significance is made clear in Table 6.5, where the average influence from compaction for the three surface treatments is shown. The broom surface shows the highest increase in capacity with a 54% increase while the rake and bush hammer surfaces both have an average capacity increase of 35%.
57


To see how well the different surface treatments performed, Table 6.6 summarizes all
pads and how they ranked on each test.
Table 6.6: Summarized ranking by capacity
Rank by Capacity Slant Shear Test Guillotine Test Jacking Test Pull-Off Test Total
1 Pad 1 Pad 1 Pad 5 Pad 3 Pad 3
2 Pad 3 Pad 3 Pad 6 Pad 1 Pad 1
3 Pad 5 Pad 4 Pad 3 Pad 4 Pad 5
4 Pad 4 Pad 5 Pad 4 Pad 5 Pad 4
5 Pad 2 Pad 6 Pad 2 Pad 2 Pad 6
6 Pad 6 Pad 2 Pad 1 Pad 6 Pad 2
It is clear that Pad 1, compacted broom finish, and Pad 3, compacted rake finish, appears in the top for most of the tests and after adding it all together, the pad that overall performed best was Pad 3. The table also illustrates the increase in strength from compaction. When looking at the total average ranking, the three compacted pads performed better overall than the non-compacted pads.
58


When there is need for a test in the field, the most commonly used method is the pull-off test. Compared to the other tests conducted in this experiment, the pull-off test is the only one that can be performed relatively easily in the field on an existing structure. The issues with the pull-off test is that it gives the tensile strength of the bonded interface, while in most cases, it is the shear strength that is of importance. To be able to use the data from the pull-off test and get an approximation of the shear strength, a correlation factor between the Guillotine test and the pull-off test can be used. In Table 6.7 the shear stress and tensile stress from each pad have been compared and a ratio has been calculated. It shows a span between 1.27 and 2.15 for the six pads with the calculated average at 1.80 and a standard deviation of 0.33. This number can be compared with Delattes studies, discussed in chapter 2.4.1, where the average calculated ratio was 2.04 (Delatte 2000).
Table 6.7: Guillotine test and pull-off test Correlation Avg. shear stress is from the guillotine test, avg. tensile stress is from the pull-off test.
_____________Ratio = (Avg, Shear Stress) / (Avg, Tensile Stress)___________
Pad Avg. Shear Stress (t) kPa (psi) Avg. Tensile Stress (a) kPa (psi) Ratio
1 3 420 (496) 1 940 (282) 1.76
2 1 570 (228) 1 240 (180) 1.27
3 3 150 (457) 1 960 (284) 1.61
4 2 760 (400) 1 280 (186) 2.15
5 2 030 (294) 1 040 (151) 1.95
6 1 870 (272) 910 (132) 2.06
Average 1.80
59


7
Conclusions
The experiment clearly shows that different surface treatments have big impact on the shear strength at the bonded interface. Within the same test the differences between the pad with best and worst result ranges from a factor of 1.6 to 3.4. Compacting the pads makes a difference for the shear strength of the pad with an average performance increase of 38 percent after compacting. For all tests combined the pad with the overall highest shear strength is pad number 3, raked substrate surface with a compacted overlay.
The experiment shows that the measured shear strength differs significantly depending on which test method that was used. The two shear tests, expected to show similar shear strength, the guillotine and jacking test, showed a guillotine shear strength for the pads ranging from 1.2 to 4.4 times the measured shear from the jacking test with an average factor of 2.7.
The average data from the tests shows that from the pull-off test the tensile strength is %2^f'c for SI units and 3.1 ^f'c for imperial units, the shear strength from the slant shear test is 61.5 ^ f'c for SI units and 23.5 ^ f'c imperial units, the shear strength from the guillotine test is 13.6 f'c for SI units and 52jjrc for imperial units and the shear strength from the jacking test is 5.1^J f'c for SI units and 22^j f'c for imperial units. The big slant shear factor is attributed to the clamping force developed from the compression force applied to the specimen due to its geometry.
The results shows that it is important to know which tests that are influenced by which factors and to be aware of that there is no universal factor to convert results from one test to another. To get the most reliable results when comparing shear strength at the interface from different projects, the same type of test should be conducted.
60


REFERENCES
1. ASTM C1583 (2015), Standard test methodfor tensile strength of concrete surfaces and the bond strength or tensile strength of concrete repair and overlay materials by direct tension (pull'-off Method), American Society of Testing Materials, West Conshohocken, PA.
2. Bakhsh, K. N. (2010), Evaluation of bond strength between overlay and substrate in concrete repairs, Master Thesis, School of Architecture and the Built Environment, Royal Institute of Technology, Stockholm, Sweden
3. Bissonnette, et al. (2012), Best practices for preparing concrete surface overlays, Report No. MERL 12-17, U.S. Bureau of Reclamation, Dept, of the Interior, Technical Service Center, Denver, CO.
4. Bissonnette, B. et al. (2011), Bonded Cement-Based Material Overlays for the Repair, the Linking or the Strengthening of Slabs or Pavements, Springer, Netherlands.
5. Bonaldo, E., Barros, J.A.O and Lourenco, P.B. (2005). Bond Characterization between Concrete Substrate and Repairing SFRC Using Pull-Off Testing, International Journal of Adhesion and Adhesives, Elsevier, United kingdom, Dec 2005, 463-474.
6. Delatte, N.J. et al. (2000), Bond strength development with maturity of high-early-strength bonded concrete overlays, ACIMaterials Journal', American Concrete Institute, Farmington Hills, MI, Mar-Apr 2000, 201-207.
7. Delatte, N.J. et al. (2000), Laboratory and field testing of concrete bond development for expedited bonded concrete overlays, ACI Materials Journal', American Concrete Institute, Farmington Hills, MI, June 2000, 272 280.
8. Ferraro, C. (2008) Investigation of concrete repair, State of Florida Department of Transportation Structural Materials Laboratory, Gainesville,
FL
9. Kriegh, J.D. (1976), Arizona slant shear test: a method to determine epoxy bond strength, ACI Journal, American Concrete Institute, Farmington Hills, MI, July 1976, 372-373.
10. Silfwerbrand, J. (1990), Improving concrete bond in repaired bridge decks, Concrete International', American Concrete Institute, Farmington Hills, MI, Sept 1990, 61-66.
11. Silfwerbrand, J. and Paulsson, J. (1998), Better bonding of bridge deck overlays, Concrete International', American Concrete Institute, Farmington Hills, MI, Oct 1998, 56-61.
12. Silfwerbrand, J. (2003) Shear bond strength in repaired concrete structures, Materials and Structures, Springer, Netherlands, July 2003, 419-424.
13. Talbot,, C. et al. (1994), Influence of surface preparation on long-term bonding of structures, ACI Materials Journal, American Concrete Institute, Farmington Hills, MI, Nov-Dec 1994, 560-566.
61


14. Vaysburd and McDonald, 1999. Technical Report REMR-CS-61, An Evaluation of equipment and procedures for tensile bond testing of concrete repairs, US Army Corps of Engineers,Washington, D.C.
15. Wall, J.S. and Shrive, N.G. (1988), Factors affecting bond between new and old concrete, ACIMaterials Journal, American Concrete Institute, Farmington Hills, MI, Mar-Apr 1988, 117-125.
16. Wan, Z. (2010) Interfacial shear bond strength between old and new concrete, Master Thesis, Department of Civil and Enviromental Engineering, Louisiana State University, Baton Rouge, LA
62


APPENDIX
Introduction
The following appendices present the raw data gathered from the tests and is the origin to all results and conclusions.
A. Compression Test
Appendix A shows the data for the compressive strength tests. It is presented by which pads the sample was gathered from and shows the axial compressive strength required to break the test cylinders and the compressive strength calculated using the cylinder size of 102 mm (4 in.).
63


Substrate 1 & 2
Cylinder Sample Size 101.6 mm (4 in.)
80000 7nnm a 250
/UUUU M 6.0000 a 200
UUUUU w soooo a Sample Date 10/05/15
n 40000 a 150 £ Test Date 11/23/15
r-CJ 4- \J\J M 72 mm a X i m Force 323 kN (72676 lbs)
20000 I y so ioooo s o 0 Compressive Strength 39841 kPa (5783 lbs)
80000 70000 J 6.0000 a 250
900
uuuuu a <0000 / zuu Sample Date 10/05/15
rs 40000 / 150 Test Date 11/23/15
r-cj a 72 0000 a Force 333 kN (74865 lbs)
oaa m a Compressive Strength 41074 kPa (5958 lbs)
ZUUUU M i mm AL > 50
o 0
90000 om 250
70000 A1 200 Sample Date 10/05/15 150 ^ Test Date 12/14/15 Force 354 kN 100



Compressive Strength 43664 kPa
50
o 0
(79531 lbs) (6329 lbs)
64


Substrate 3 & 4
Cylinder Sample Size 101.6 mm (4 in.)
80000
250
200
150
100
50
0
70000
250
200
150
100
50
0
90000 80000 70000 60000 M 50000 £ 40000 30000 20000 10000 0
250
200
150
100
50
0
Sample Date 10/05/15 Test Date 11/23/15 Force 310kN
Compressive Strength 38237 kPa
Sample Date 10/05/15 Test Date 11/23/15 Force 291 kN
Compressive Strength 35893 kPa
Sample Date 10/05/15 Test Date 12/14/15 Force 340 kN
Compressive Strength 41937 kPa
(69788 lbs) (5554 lbs)
(65446 lbs) (5208 lbs)
(76420 lbs) (6081 lbs)
65


Substrate 5 & 6
Cylinder Sample Size 101.6 mm (4 in.)
45000 40000 35000 30000 M 25000 ^ 20000 15000 10000 5000 0
250
200
150
100
50
0
40000
250
200
150
100
50
0
45000 40000 35000 30000 M 25000 £ 20000 15000 10000 5000 0
250
200
150
100
50
0
Sample Date 10/09/15 Test Date 11/23/15 Force 179 kN
Compressive Strength 22079 kPa
Sample Date 10/09/15 Test Date 11/23/15 Force 167 kN
Compressive Strength 20599 kPa
Sample Date 10/09/15 Test Date 12/14/15 Force 177 kN
Compressive Strength 21832 kPa
(40271 lbs) (3205 lbs)
(37550 lbs) (2988 lbs)
(39883 lbs) (3174 lbs)
66


Overlay 1-6
Cylinder Sample Size 101.6 mm (4 in.)
70000
250
200
150
100
50
0
80000 70000 60000 50000 J3 40000 30000 20000 10000 0
250
200
150
100
50
0
Sample Date 10/19/15 Test Date 11/23/15 Force 242 kN
Compressive Strength 29850 kPa
Sample Date 10/19/15 Test Date 11/23/15 Force 262 kN
Compressive Strength 32316 kPa
Sample Date 10/19/15 Test Date 12/14/15 Force 334 kN
Compressive Strength 41197 kPa
(54489 lbs) (4336 lbs)
(58856 lbs) (4684 lbs)
(75070 lbs) (5974 lbs)
67


B. Pull-Off Test
The data for the pull-off test is presented below. It is presented pad by pad and shows the required pull-off force to detach the sample from the substrate and the force converted to stress using the actual core diameter of 66.7 mm (25A in.).
68


Surface Treatment
Vibrated?
Actual Core Diameter
PAD 1
Broom
Yes
66.7 lmn (2 5/8 in.)
Sample IB
Pull-Off Force 8.67 kN (1950 lbs)
Tensile Stress 2484 kPa (360 psi)
Sample ID
Pull-Off Force 6.23 kN (1400 lbs)
Tensile Stress 1784 kPa (259 psi)
Sample IF
Pull-Off Force 5.45 kN (1225 lbs)
Tensile Stress 1561 kPa (226 psi)
69


Surface Treatment
Vibrated?
Actual Core Diameter
PAD 2
Sample 2A Pull-Off Force Tensile Stress
Broom
No
66.7 lmn (2 5/8 in)
Sample 2C2
4.34 kN (975 lbs) Pull-Off Force 4.00 kN (900 lbs)
1242 kPa (180 psi) Tensile Stress 1147kPa (166 psi)
Sample 2E
Pull-Off Force 4.67 kN (1050 lbs)
Tensile Stress 1338 kPa (194 psi)
70


PAD 3
Surface Treatment Rake
Vibrated? Yes
Actual Core Diameter 66.7 lmn (2 5/8 in)
Sample 3B
Pull-Off Force 6.27 kN (1410 lbs)
Tensile Stress 1796 kPa (261 psi)
Sample 3D
Pull-Off Force 8.01 kN (1800 lbs)
Tensile Stress 2293 kPa (333 psi)

Sample 3F
Pull-Off Force 5.23 kN (1400 lbs)
Tensile Stress 1784 kPa (259 psi)
71


Surface Treatment
Vibrated?
Actual Core Diameter
Sample 4A Pull-Off Force Tensile Stress
PAD 4
Rake
No
66.7 lmn (2 5/8 in)
3.45 kN (775 lbs) 987 kPa (143 psi)
Sample 4C
Pull-Off Force 4.56 kN (1025 lbs)
Sample 4E
Pull-Off Force 5.45 kN (1225 lbs)
Tensile Stress 1561 kPa (226 psi)
72


PAD 5
Surface Treatment
Vibrated?
Actual Core Diameter
Bush Hammer Yes
66.7 lmn (2 5/8 in.)
Sample 5B
Pull-Off Force 3.11 kN (700 lbs)
Tensile Stress 892 kPa (129 psi)
Sample 5D
Pull-Off Force 5.12 kN (1150 lbs)
Tensile Stress 1465 kPa (212 psi)
Sample 5F
Pull-Off Force 2.67 kN (600 lbs)
Tensile Stress 764 kPa (111 psi)
73


PAD 6
Surface Treatment
Vibrated?
Actual Core Diameter
Bush Hammer No
66.7 lmn (2 5/8 in.)
Sample 6 A
Pull-Off Force 4 kN (900 lbs)
Sample 6C
Pull-Off Force 3.45 kN (775 lbs)
Sample 6E
Pull-Off Force 2.09 kN (470 lbs)
Tensile Stress 599 kPa (87 psi)
74


C. Guillotine Test Results
Below is the data acquired during the Guillotine test presented. The results are presented pad by pad and shows the applied axial compressive force that was required to shear off the samples. Each sample has a picture showing the failure and each sample is accompanied by a graph showing the development of the axial compressive force prior to failure.
75


PAD 1
Surface Treatment
Vibrated?
Actual Core Diameter
Broom
Yes
66.7 lmn (2 5/8 in.)
Sample 1A
Shear Force 9.23 kN (2074 lbs)
Sample 1C
Shear Force 11.4 kN (2562 lbs)
Sample IE
Shear Force 15.18 kN (3412 lbs)
Average shear force 11.94 kN (2683 lbs)
76


SAMPLE IE
4000
77


Surface Treatment
Vibrated?
Actual Core Diameter
PAD 2
Broom
No
66.7 lmn (2 5/8 in.)
Sample 2B
Shear Force 6.05 kN (1361 lbs)
Sample 2D
Shear Force 6.21 kN (1395 lbs)
Sample 2F
Shear Force
4.21 kN (946 lbs)
Average shear force (lbs)
5.49 kN
(1234 lbs)
78


SAMPLE 2B
1600
SAMPLE 2D
1600 ------------------------------- 7
SAMPLE 2F
1000
79


Surface Treatment
Vibrated?
Actual Core Diameter
PAD 3
Rake
Yes
66.7 lmn (2 5/8 in.)
Sample 3A
Shear Force 12.96 kN (2913 lbs)
Sample 3C
Shear Force 8.72 kN (1960 lbs)
Sample 3E
Shear Force 11.31 kN (2543 lbs)
Average shear force (lbs)
11.00 kN
(2472 lbs)
80


SAMPLE 3E
3000 7
81


Surface Treatment
Vibrated?
Actual Core Diameter
PAD 4
Rake
No
66.7 mm (2 5/8 in.)
Sample 4B Sample 4D2
Shear Force 11.00 kN (2474 lbs) Shear Force 9.61 kN (2160 lbs)
Sample 4F
Shear Force 8.26 kN (1856 lbs)
Average shear force (lbs)
8.94 kN
(2163 lbs)
82


SAMPLE 4B
3000
12
10
8
83


PAD 5
Surface Treatment
Vibrated?
Actual Core Diameter
Bush Hammer
Yes
66.7 lmn (2 5/8 in.)
Sample 5A
Shear Force 6.51 kN (1463 lbs)
Sample 5C
Shear Force 8.97 kN (2017 lbs)
Sample 5E
Shear Force
5.76 kN (1294 lbs)
Average shear force (lbs)
7.08 kN
(1591 lbs)
84


SAMPLE 5A
SAMPLE 5C
SAMPLE 5E
85


PAD 6
Surface Treatment
Vibrated?
Actual Core Diameter
Bush Hammer No
66.7 lmn (2 5/8 in.)
Sample 6B3
Shear Force 8.48 kN (1907 lbs)
Sample 6D
Shear Force 7.28 kN (1637 lbs)
Sample 6F
Shear Force 3.85 kN (865 lbs)
Average shear force (lbs)
6.54 kN
(1470 lbs)
86


87


D. Slant Shear Test Results
The data collected from the slant shear test is presented below. The test included two core samples from each pad. The data presented are the applied compressive force that resulted in failure and the calculated normal and shear forces with a picture showing the failure for each sample.
88


PAD 1
Surface Treatment Vibrated?
Actual Core Diameter Bonded Area (A)
Broom
Yes
66.7 mm (2 5/8 in.) 6980 mm2 (10.82 in2)
Sample 1G
Compression Force (F) 126.8 kN
Shear Force (V) 109.8 kN
Normal Force (N) 63.4 kN
Shear Stress (V/A) 15729 kPa
Sample IQ
Compression Force (F) 117.9 kN
Shear Force (V) 102.1 kN
Normal Force (N) 58.9 kN
Shear Stress (V/A) 14625 kPa
(28500 lbs) (24682 lbs) (14250 lbs) (2281 psi)
(26500 lbs) (22950 lbs) (13250 lbs) (2121 psi)
89


Full Text

PAGE 1

SHEAR STRENGTH AT THE INTERFACE OF BONDED CONCRETE OVERLAYS by CHRISTIAN JOHAN ROSEN B.S., Link ping University, 2014 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements of the degree of Master of Science Civil Engineering 2016

PAGE 2

ii This thesis for the Master of Science degree by Christian Johan Rosen has been approved for the Civil Engineering Program by Frederick R. Rutz, Chair Kevin L. Rens Chengyu Li July 30, 2016

PAGE 3

iii Rosen, Christian Johan (M.S., Civil Engineering) Shear Strength at the Interface of Bonded Concrete Overlays Thesis directed by Professor Frederick R. Rutz ABSTRACT This thesis investigated multiple methods for measuring shear strength at the bonded interface between a concrete substrate and a concrete overlay as it is influenced by different substrate surface treatments The results were collected from an experiment conducted at the University of Colorado Denver where six different test pads, each with a unique surface treatment, were exposed to four types of test s From the different tests, the shear strength at the bonded interface was measured A discussion of analysis with respect to the variations in surface treatment s and test method s is presented The mechanisms associated with the different tests are discussed. The results are dependent on the test method used to evaluate the shear strength. The find ings of the research illustrate the significance that surface treatment of a substrate has on the shear strength at a bonded overlay The form and content of this abstract are approved. I recommend its publication. Approved : Frederick R. Rutz

PAGE 4

iv ACKNOWLEDGEMENTS There are numerous people that in one way or another have helped me make this research successful First of all I would like to thank Dr. Frederick Rutz who has served as a dvisor of this research. His enthusiasm and dedication have been really motivating and he ha s been very supporti ve in every stage of the research, from planning and preparation to testing and writing the thesis I would like to thank Tom Thuis Peter Sillstrop and Jac Corless at the University of Colorado Denver Electronic Calibration and Repair Lab who trained me on the MTS machines helped me with the pouring of the concrete pads and for manufacturing additional steel pucks for the pull off test. I would also like to express my gratitude to Bud Warner and Zack B allard at CTL|Thompson who have been generous with input on the testing process, lent me testing equipment and for letting me use their facility and equipment for the slant shear test. And last my fellow graduate student Andy Pultorak who spent several fr eezing winter days helping me with the not so glamorous work of coring and saw cutting the test pad s as well as conducting the jacking test

PAGE 5

v TABLE OF CONTENTS Chapter 1 Overview ................................ ................................ ................................ ................ 1 1.1 Introduction ................................ ................................ ................................ 1 1.2 Goal ................................ ................................ ................................ ............. 2 1.3 Outline ................................ ................................ ................................ ......... 2 2 Background ................................ ................................ ................................ ............. 4 2.1 Introduction ................................ ................................ ................................ 4 2.2 Bond ................................ ................................ ................................ ............. 4 2.3 Shear Strength ................................ ................................ .............................. 7 2.3.1 Shear Strength Test Methods ................................ ................................ .... 8 2.3.1.1 Slant Shear Test ................................ ................................ ...................... 8 2.3.1.2 Guillotine Test ................................ ................................ ........................ 9 2.3.1.3 Jack ing Test ................................ ................................ .......................... 10 2.4 Tensile Strength ................................ ................................ ......................... 10 2.4.1 Pull Off Test ................................ ................................ ............................ 11 3 Experi ment ................................ ................................ ................................ ............ 13 3.1 Construction of Test Pads ................................ ................................ .......... 13 3.2 Surface Preparation ................................ ................................ .................... 16 4 Te sting ................................ ................................ ................................ .................. 19

PAGE 6

vi 4.1 Compression Test ................................ ................................ ...................... 19 4.2 Pull Off Test ................................ ................................ .............................. 21 4.3 Slant Shear Test ................................ ................................ ......................... 22 4.4 Guil lotine Test ................................ ................................ ........................... 25 4.5 Jacking Test ................................ ................................ ............................... 27 5 Experiment Results ................................ ................................ ............................... 29 5.1 Compression Test ................................ ................................ ...................... 29 5.2 Pull Off Test ................................ ................................ .............................. 31 5.3 Slant Shear Test ................................ ................................ ......................... 35 5.4 Guillotine Test ................................ ................................ ........................... 37 5.5 Jacking Test ................................ ................................ ............................... 41 6 Discussion ................................ ................................ ................................ ............. 45 6.1 Conducting Experiments ................................ ................................ ........... 45 6.2 Experiment Results ................................ ................................ .................... 47 6.2.1 Slant Shear Test ................................ ................................ ....................... 48 6.2.2 Guillotine Test ................................ ................................ ......................... 49 6.2.3 Jacking Test ................................ ................................ ............................. 51 6.2.4 Pull Off Test ................................ ................................ ............................ 53 6.2.5 Summary ................................ ................................ ................................ 55 7 Conclusions ................................ ................................ ................................ .......... 60

PAGE 7

vii References ................................ ................................ ................................ ................... 61 Appendix A. Compression Test ................................ ................................ ...................... 63 B. Pull Off Test ................................ ................................ .............................. 68 C. Guillotine Test Results ................................ ................................ .............. 75 D. Slant Shear Test Results ................................ ................................ ............ 88 E. Jacking Test Results ................................ ................................ .................. 95

PAGE 8

viii LIST OF TABLES Table 5.1: Surface treatments ................................ ................................ ................................ 29 5.2: Compression test results ................................ ................................ ...................... 29 5.3: Interpolated compressive strength at the time of testing ................................ ..... 30 5.4a: Pull off test results (SI units) ................................ ................................ ............. 31 5.4b: Pull off test results (Imperial units) ................................ ................................ ... 32 5.5a: Pull off test results Average tensile stress (SI units) ................................ ....... 33 5.5b: Pull off test results Average tensile stress (Imperial units) ............................. 33 5.6: Pull off test results Influence from vibration ................................ .................... 33 5.7a: Pull off test results Statistical results (SI units) ................................ .............. 34 5.7b: Pull off test results Statistical results (Imperial units) ................................ .... 34 5.8a: Slant shear test results (SI units) ................................ ................................ ........ 35 5.8b: Slant shear test results (Imperial units) ................................ .............................. 35 5.9a: Slant shear results Average shear stress (SI units) ................................ .......... 36 5.9b: Slant shear results Average shear stress (Imperial units) ................................ 36 5.10: Slant shear results Influence from vibration ................................ .................... 36 5.11a: Guillotine test results (SI units) ................................ ................................ ....... 37 5.11b: Guillotine test results (Imperial units) ................................ ............................. 38 5.12a: Guillotine test results Average shear stress (SI units) ................................ ... 3 9 5.12b: Guillotine test results Average shear stress (Imperial units) ......................... 39 5.13: Guillotine test results Influence from vibration ................................ ............... 39 5.14a: Guillotine test resul ts Statistical results (SI units) ................................ ........ 40

PAGE 9

ix 5.14b: Guillotine test results Statistical results (Imperial units) .............................. 40 5.15a: Jacking test block data (SI units) ................................ ................................ ..... 41 5.15b: Jacking test block data (Imperial units) ................................ ........................... 42 5.16a: Jacking test results Average shear stress (SI units) ................................ ....... 43 5.16b: Jacking test results Average shear stress (Imperial units) ............................ 43 5.17: Jacking test results Influence from vibration ................................ .................. 43 5.18a: Jacking test results Statistical results (SI units) ................................ ............ 44 5.18b: Jacking test results Statistical results (Imperial units) ................................ .. 44 6.1: Slant shear test Rank ing by capacity ................................ ................................ 49 6.2: Guillotine test Ranking by capacity ................................ ................................ .. 51 6.3: Jacking test Ranking by capacity ................................ ................................ ...... 53 6.4: Pull off test Ranking by capacity ................................ ................................ ...... 55 6.5: Average influence from vibration ................................ ................................ ........ 57 6.6: Summarized ranking by capacity ................................ ................................ ......... 58 6.7: Guillotine test and pull off test Correlation ................................ ...................... 59

PAGE 10

x LIST OF FIGURES Figure 2.1. Diagram of overlay and substrate ................................ ................................ .......... 6 2.2. Overlay without compaction ................................ ................................ .................. 7 2.3. Overlay with compaction ................................ ................................ ....................... 8 2.4. Diagram of slant shear test ................................ ................................ ..................... 9 2.5. Diagram of guillotine shear test ................................ ................................ ........... 10 2.6. Diagram of jacking test ................................ ................................ ........................ 10 2.7. Diagram of pull off test ................................ ................................ ....................... 11 2.8. Pull off test failure modes ................................ ................................ .................... 12 3.1. Experiment design: Plan ................................ ................................ ...................... 14 3.2. Experiment design: Elevation ................................ ................................ .............. 14 3.3. Substrate Formwork ................................ ................................ ............................. 15 3.4. Vibrator used for compaction ................................ ................................ .............. 16 3.5. Broom finish ................................ ................................ ................................ ........ 16 3.6. Rake finish ................................ ................................ ................................ ........... 17 3.7. Bush hammer finish ................................ ................................ ............................. 17 3.8. Bush hammer used for surface finish ................................ ................................ ... 18 4.1. MTS 810 load frame ................................ ................................ ............................ 20 4.2. Compression test set up ................................ ................................ ....................... 20 4.3. 007 James bond tester ................................ ................................ .......................... 21 4.4. Core drilling for slant shear test ................................ ................................ ........... 22 4.5. Core sample with slanted interface ................................ ................................ ...... 23

PAGE 11

xi 4.6. Slant shear test setup ................................ ................................ ............................ 23 4.7. Close up on slant shear sample ................................ ................................ ............ 24 4.8. Vertical core drilling for guillotine test ................................ ............................... 25 4.9. Shear jig design ................................ ................................ ................................ .... 26 4.10. Guillotine test setup ................................ ................................ ........................... 26 4.11. Blocks on pads 5 and 6 prepared for jacking test ................................ .............. 27 4.12. Jack set up for testing ................................ ................................ ......................... 28 6.1. Compressive strength development ................................ ................................ ..... 46 6.2a. Slant shear test (SI units) ................................ ................................ .................... 48 6.2b. Slant shear test (Imperial units) ................................ ................................ ......... 48 6.3a. Guillotine test Comparison (SI units) ................................ ............................. 50 6.3b. Guillotine test Comparison (Imperial units) ................................ ................... 50 6.4a. Jacking test Comparison (SI units) ................................ ................................ 52 6.4b. Jacking test Comparison (Imperial units) ................................ ....................... 52 6.5a. Pull off test Comparison (SI units) ................................ ................................ 54 6.5b. Pull off test Comparison (Imperial units) ................................ ....................... 54 6.6a. Summary of test results (SI units) ................................ ................................ ...... 56 6.6b. Summary of test results (Imperial units) ................................ ............................ 56

PAGE 12

1 1 Overview 1.1 Introduction Concrete has been use by mankind in structures for thousands of years. Because of its durability and ability to shape as desired, it is used in all types of construction. However, concrete does not last forever and to extend the lifetime of a damaged concr ete structure, deteriorated concrete can be removed and replaced with a patch. Depending on how severe the damage is, different types of repairs can be installed. If a concrete slab is only damaged in the top portion, there is no need to remove concrete al l the way to the bottom. Unsound concrete is removed and a patch can be used to fill the void. A goal of patching is to create a repair that resembles the original structure as much as possible where the bonded interface between the new and old concrete be haves monolithically. The bond strength components are shear and tensile strength and even though tension tests on repaired concrete structures are most common, shear strength is of greater interest because the interface is only subject to very small zones of pure tension, while shear better describes the overall bond strength (Silfwerbrand 2003). The reason why tension tests are more common is simply because they are easier to perform. A review of the literature reveals that there are no accepted tests for concrete shear strength because there is no general agreement in the structural engineering community for a shear strength test. The problem with shear tests is that it is difficult to separate direct shear from bending moment, and thus it is hard to comp are results to each other.

PAGE 13

2 1.2 Goal The goal of this research is to examine multiple test methods for shear strength. Toward this end, how different surface treatments on a concrete substrate affects the shear strength at the interface of a bonded concrete overlay is evaluated. The tests were conducted on six pads with different surface treatments and where three of them are consolidated. Pull off, guillotine, slant shear and jacking test were conducted on each pad. The outcome of the tests are compared and analyzed. 1.3 Outline This thesis consists of seven chapters. The first chapter include s an overview and a description of the goal. Chapter 2 contains a literature review of current research on shear strength and different methods to test the shear strength at the bonded interface. Chapter 3 presents the design of the experiment, location and which methods that were used to create the different surface treatments. Chapter 4 explains all the tests done on the pads, the conditions under which the testing was co nducted and the equations used to calculate shear strength based on the test results. Chapter 5 presents tables of the data gathered from the four tests Chapter 6 compares the results from the four different tests and discusses possible sources of error throughout the experiment. Chapter 7 contains a summary of the conclusions that can be made from the experiment.

PAGE 14

3 Appendi ces A E shows the raw data gathered from the tests and pictures of all the samples.

PAGE 15

4 2 Background 2.1 Introduction When repairing or rehabil itating a concrete structure the unsound concrete is removed and replaced with repair materials ; for the case of slabs this is known as an overlay. To make the structure last in a long term perspective, the goal is to create a com bined section that resembl es the original structure and possesses equivalent strength A key factor is to have a shear strength at the repair interface that approaches the concrete itself. While bond strength is a measurement of how well the old and the new concrete surfaces are ad hered to each other and can be measured by standard tests, shear strength is more elusive and is difficult to measure It is influenced by several different factors discussed below. T here are numerous ways to test the shear strength of a repair yet none have achieved the status of an accepted industry standard. A repaired concrete struct ure is made up from a n existing substrate and a new overlay with the crucial bonded interface in between 2.2 Bond Bond describes how well the overlay and substrate are adher ed to each other. The interface between the concrete layers is important because its purpose is to create a bond that resemble s the former structure so that the forces from the overlay get s distributed down to the substrate. It is these forces that the bon d has to be able to resist. A good bond is when the bond strength is equal t o or higher than the tensile strength of either the substrate or the overlay The bond can be described with either adhesion or from a mechanical point of view. In terms of mechan ical bonding, the bond mechanism is described by the interlock between

PAGE 16

5 the two surfaces and is mainly influenced by the roughness of the substrate and the degree with which the overlay interlocks with the surface of the substrate. Adhesion describes the bonded interface and is usually quantified as the amount of stress or energy required to separate the two layers. It takes into account the mechanical interaction, thermodynamics and chemical bonding. The mechanical adhesion is highl y dependent on the surface roughness and the substrate needs to have open cavities for the fresh concrete to anchor to. The surface moisture is very important for a good adhesion because the cement will with capillary absorption anchor to the old concrete. If the substrate surface moisture condition is too low, the fresh cement will not be able to tie to the substrate because it pulls to much water from the overlay, not allowing the overlay to hydrate which leads to a weak bond (Bissonnette et al. 2012). Ac cording to tests conducted at the Louisiana State University, the best moisture condition is saturated surface dry with a bond strength almost double compared to an air dry surface (Wan 2010). When a former bond has been reduced to a level where it is no longer capable of resisting the differential forces between the substrate and overlay, debonding occurs. In a worst case it leads to delamination and slipping. The reasons behind failure can be many. In a long term perspective, a common factor is differential shrinkage and thermal expansion and contraction. If the two layers have different properties and do not act as a homogenous concrete section, the failure will occur at the weak link, which i n a repair usually is the bonded interface and that is where the debonding occurs. Reasons behind different properties between them can be that the two concrete layers had different water to cement ratio, the aggregates are of different size, variations in the surface preparation of the substrate and consolidation of the overlay can affect bond or environmental differences at the time of

PAGE 17

6 curing. These differences lead to a difference in modulus of elasticity between the substrate and the overlay. Shrinkage in the layers will not match each other. Because of different movement of the pads, over time the tension between them will be so high that the bond strength cannot resist the forces and will cause debonding (Bakhsh 2010). The forces behind failure can be shear stresses originating from service loads on the concrete or tension or compression stresses caused by thermal influences and shrinkage. When the bond fails it fails in either tension or shear. To test the potential bond and/or shear strength there ar e many different methods, some that can be done in the field and some that have to be done in a laboratory (Wall and Shrive 1988). These are discussed below. Figure 2 1 Diagram of o verlay and s ubstrate Acc ording to Dr. Silfwerbrand at the Royal Institute of Technology, i n order to achieve a good bond one of the most significant factors is to have a clean substrate surface when the overlay is placed Dust, grease and other loose particles could significantly reduce the bond strength because it prevents the adhesion that is the key to a good bond Silfwerbrand also mentions the importance of avoiding micro cracks in order to achieve a good bond. Some surface preparation methods may induce the presence o f micro cracks and he asserts that even if a rougher surface is obtained with a certain method, the increasing amount of micro cracks reduces the gain from a rough surface and could lead to an overall weaker bond. Silfwerbrand asserts that t he best method to ensu re a minimum amount of micro cracks is to use water jetting instead of mechanical methods such as chi pping

PAGE 18

7 (Silfwerbrand 1990). However, an experiment conducted by Talbot et al. shows that if a low mass jack hammer is carefully used on the surface, the presence of micro cracks will not be significant ( Talbot et al. 1994). 2.3 Shear Stre ngth When it comes to shear, the interface mainly gets its strength from mechanical interlock and adhesion Previous studies shows that there are significant differences in the shear strength with different interface textures. Higher macro roughness means a higher bond strength. It means that the most si gnificant factor for good shear strength comes from the surface roughness. To take advantage of the surface roughness and make sure that it reaches its full potential, it is important to make sure that the overlay satisfactory penetrates the substrate surf ace. This can be done by compacting the overlay which also makes the overall strength of the concrete increase. A visualization of a compacted and non compacted bond is shown in Figures 2.2 and 2.3 (Silfwerbrand and Paulsson 1998). Another important factor for the interface to resist shear forces is to have a good compressive concrete strength. For a rough surface to actually be able to hold the overlay in place, the concrete has to hold together to avoid slipping. Figure 2 2 Overlay without c ompaction

PAGE 19

8 Figure 2 3 Overlay with c ompaction 2.3.1 Shear Strength Test Methods There are numerous tests to determine the shear resistance in the bond ed interface The ones presented below w ere used in the conducted experiment. 2.3.1.1 Slant Shear Test T he slant shear test also known as Arizona slant shear test, is a combination of a compression test and a shear test where the concrete core sample is slanted with an incl ined bonded interface The incline is usually 30 degrees from the longitudinal axis. Because of the geometry of the specimen, the compression force will add to the measured bond strength due to a clamping effect ( Ferraro 2008). A compression force is appli ed at the ends of the sample until failure happens at the bonded plane. If the two concrete layers have the same compressive strength and the shear strength at the interface is equal to or higher than the concrete itself, which is the ultimate goal, the fa ilure mode should be of the same type as for a regular compression test. If the shear strength at the interface is less than the concrete itself, the failure occurs at the interface. The force is then divided by the area of the bonded plane which gives the shear strength. To get the n ormal and s hear stresses of the sample the resultants of the axial compression force is calculated and divided by the bonded area. This test is hard to perform in the field because of the difficulties to acquire the slanted samples.

PAGE 20

9 In a laboratory the specimens can be created using a mold with an insertion that creates samples with an inclined plane (Kriegh 1976) Figure 2 4 Diagram of s lant s hear t est 2.3.1.2 Guillotine Test The guillotine test is a method where a core sample with the bonded interface perpendicular to the longitudinal axis is placed in a gu illoti ne shear jig The core is resting in a half circle with the same diame ter as the sample cut into the steel jig. Another steel piece with the same diameter fitting inside the bottom half of the jig is put on top of the core so that the bond ed i nterface is located between the top and the bottom half of the jig. This exposes the core to two adjacent, equally sized and opposite forces. As seen i n F igure 2 .2 the core is subjected to a third force that gives support to the core. When a compression force is applied on top of the gu illotine device, if positioned correctly the core will shear off right at the interface Because of the thickness of the steel used for the jig and the gap between the inner and outer jig, in practice the forces cannot be perfectly adjacent and a small e c centricity will occur which introduces a minor bending moment. This moment however is not considered to influence the failure significantly The guillotine test is mainly a laboratory test because in the field with thicker overlays, it is difficult to extr act undamaged cores (Delatte et al. 2000)

PAGE 21

10 Figure 2 5 Diagram of g uillotine s hear t est 2.3.1.3 Jacking Test The jacking test is conducted by applying horizontal forces parallel to the interface. If conducted in the field, a jack can be used to apply pressure on the sample until a failure happens. A potential problem with this type of test is that a bending moment is introduced because of the eccentricity of the applied force. This bending moment is heavily depending on the jack s ram diameter A larger ram means that the center of the applied force will have a larger eccentricity. Compared to the bending moment that occurs for the guillotine test, this moment is much more significant to the debonding of the specimen. Figure 2 6 Diagram of j acking t est 2.4 Tensile St rength For a bonded concrete interface to fail in tension there has to be tensile f orces. If the surface is relatively rough, the overlay needs to be lifted by an upwards force in order to move. In general, forces like that are not very common. Still the tensile test is the most used

PAGE 22

11 to determine the bond strength of a repaired structure simply because it is relatively easy to perform on site. 2.4.1 Pull Off Test The pull off test is a direct tension test and the most common test method to evaluate bond strength. Even though shear strength is the most significant factor for bonding, the pull off test is simple to use in the field making it the most used bond test method. A steel puck is attached to the top of a core that is drilled at least 25 .4 mm ( 1 in ) or half the core diameter into the substrate. The steel puck is atta ched to a testing apparatus that gradually increases a centric upward force normal to the specimen to the point of failure where the core is loose Figure 2 7 Diagram of p ull o ff t est It is of high import ance that the load is applied perpendicular and without eccentricity to the core to avoid that a bending moment develops (Delatte et al. 2000) Depending on where the failure happen s at the overlay, substrate or at the bond ed interface, it gives a first implication of the strength of the bond. The failure could also happen at the epoxy used to attach the steel pu ck to the core or a combination of the failure mode (ASTM C1583 2015).

PAGE 23

12 Figure 2 8 Pull off test failure modes a: Failure in substrate. b: Failure at bonded interface. c: Failure in overlay. d: Epoxy failure A satisfactory test would be if the break happens perfectly at the bonded interface, because that would indicate the true strength of the bond between the substrate and the overlay. If the break happe ns in the substrate it is a sign that the bond strength is higher than the substrate itse lf, which is the ultimate goal for a concrete repair (Bonaldo et al. 2005) Because of the big differences in the mechanisms behind what makes a good bond between shear and tension, where the mechanical adhesion for shear mainly relies on the overall surface roughness while the tensile mechanical adhesion mainly get its strength from the vertical anchorage in pores and voids, it is hard to find a universal correlation ratio (Bissonnette et al. 2012). It could be situations where the bonded interface has a good shear resistance and a poor tensile resistance or the other way around, meaning that a correlation factor based on the first situation would not be applicable for the second case. However, Delatte et al. found during s hear and tensile testing of concrete overlays that the shear stress is more than twice than the tension stress. Their results showed a shear stress over tension stress ratio that varied from 1.42 to 3.39, which gave an average of 2.04 and a standard deviat ion of 0.33 (Delatte et al. 2000).

PAGE 24

13 3 Experiment The experiment wa s conducted to evaluate the bond strength between a concrete overlay and a substrate for different types of substrate surface treatments. Three different surface treatments on six concrete test pads were prepared H alf of the overlays were compacted by vibration to evaluate the influence of compaction at each surface treatment, thus six different combinations of su rface treatments and compaction were made available for testing On each pad four different test s were conducted to evaluate the shear and tensile resistance for the different treatments. Cylinders were obtained for concrete compression tests. To acquire samples for the testing, on each pad three full depth core samples were gather ed, three cores were drilled 25 .4 mm (1 in.) into the substrate, two slanted full depth core samples were drilled and after all coring had been completed, the overlay was cut into rectangular blocks. Because pull off tensile tests are the most common pract ice in the field, the results from tha t test was compared to the Guillotine test and a ratio factor is introduced. 3.1 Construction of Test Pads The experiment was conducted outdoors at the University of Colorado Denver Civil Engineering Lab. The substrate forms rested on the underlying concrete floor. T he pads were covered in a plastic sheet for curing and to serve as weather protection. Six concrete pads named Pad 1 to Pad 6 with the substrate dimensions 14 22 mm x 61 0 mm ( 56 in. x 24 in. ) and overlay dime nsions 9 91 mm x 55 9 mm ( 3 9 in. x 22 in. ) as seen in figures 3.1 and 3.2 were cast. The size of the pads was mainly influenced by the length required to drill for the two slanted cores inclined at 30 degrees.

PAGE 25

14 Figure 3 1 Experiment d esign: Pla n Figure 3 2 Experiment d esign: Elevation Due to the dimensions of the standard lumber that the forms were made of t he thickness of the substrate and overlay were 82 6 mm ( 3 in ) and 8 8 9 mm ( 3 in ) respectively. The substrate was made l onger than the overlay to fit a steel angle acting as support to the jack for t he jacking test. Three 1 9 1 mm ( in ) diameter bolts were cast into the substrate to anchor the steel angle 9.53 mm ( N o. 3 ) h airpin re inforcing bars were installed around each bolt to resist tension when applying a horizontal force on the steel angle. The hairpins were located approximately 25 .4 mm ( 1 in ) down from the surface The substrates were cast on a 102 mm ( 4 in ) thick layer of insulation boards that served as protection to prevent damage to the underlying concrete floor when coring.

PAGE 26

15 Figure 3 3 Substrate Formwork Forms for six substrate pads with i nsulation boards and 9.525 mm ( #3 ) be seen. The concrete used for both the substrate and overlay was made using Q uikrete 5000 with a labeled 28 day compressive strength of 34500 kPa ( 5000 psi ) The Q uikrete was mix ed with water in a drum mixer The substrates were cast on two different days with pads 5 and 6 were cast four days after the first four To avoid big differences in shrinkage between the overlay and the substrate, the overlays were cast only 14 days afte r the first substrates were cast. Before the overlay was cast, the substrate surface was cleaned with high pressure water and compressed air. The overlays were compacted using an electric three horsepower vibrator with a 34 .9 mm ( in. ) square head.

PAGE 27

16 Figure 3 4 Vibrator u sed for c ompaction Model: Wyco Sure Speed WSD1 3.2 Surface Preparation The six substrate surfaces were prepared with three different treatments and for each type of surface treatment, one ov erlay was compacted by vibration and one was not, giving six unique combinations affecting the bond The six combinations are listed in T able 5.1 The broom finish was performed on pad 1 and 2 with a stiff poly fibered push broom short ly after the concrete had been p laced Figure 3 5 Broom f inish Made with stiff bristle broom

PAGE 28

17 The rake finish was performed on pad 3 and 4 short after the concrete was p lac ed. The rake used was a garden rake with a tooth spacing of 25 .4 mm ( 1 in ) and the rake marks were approximately 19 .1 mm ( in ) deep. Figure 3 6 Rake f inish Grooves were ap proximately 19 .1 mm ( in. ) deep and 25 .4 mm ( 1 in. ) apart The bush hammered finish was performed o n pad 5 and 6 after eight days of curing using an electric jackhammer with a bush hammer attachment The surface is shown in Figure 3.7 and the bush hammer attachment is shown in Figure 3.8. Figure 3 7 Bush h ammer f inish All laitance was removed and the surfaced roughened as shown

PAGE 29

18 Figure 3 8 Bush h ammer u sed for s urface f inish

PAGE 30

19 4 Testing For all tests except the jacking test, a core bit with 7 6.2 mm ( 3 in. ) diameter and 40 6 mm ( 16 in ) of travel length was used to acquire the samples. The real diameter of all specimens using the core bit is 66 7 mm ( 2 in ) which was used in all calculatio ns. The g uillotine and compression tests were conducted in the University of Colorado Denver Civil Engineering Lab using MTS machines. The jacking and pull off tests were conducted directly on the pa ds and the slant shear test w as conducted at the CTL|Thom pson testing facility in Denver, CO. 4.1 Compression Test For each substrate surface preparation, three samples of the concrete mix were collected and for the overlay concrete mix a total of three samples were collected. The samples were collected in 203 mm ( 8 in ) tall Forney cylinder molds with a diameter of 10 2 mm ( 4 in ) The first testing took place when the concrete had cured 49 days for substrate 1 to 4, 45 days for substrate 5 and 6 and 35 days for the overlays. At this time two cylinders from each pad were tested. The second day of testing took place 21 days after the first test, when the concrete had cured 70 days for substrate 1 to 4, 66 days for substrate 5 and 6 and 56 days for the overlays. The testing was conducted in a MTS 810 load frame seen in Figure 4.1, with a maximum load capacity of 980 kN ( 220 kips ) The cylinders were equipped with neoprene pad caps on each end which can be seen in Figure 4.2 and the tests were run under displacement control at a rate of 1.25 mm/minute ( 0.05 in/min ute)

PAGE 31

20 Figure 4 1 MTS 810 l oad f rame Figure 4 2 Compression t est s et u p ( 4.1 ) Where:

PAGE 32

21 4.2 Pull Off Test The pull off test consisted of 18 specimens with three tests for each pad. The test specimens were prepared by core drilling through the overlay and into the substrate by approximately 25 .4 mm ( 1 in. ) to make sure that failure c ould occur in all three stages, overlay, substrate and at the bonded interface. The surface of the specimens was cleaned with acid to remove all laitance and for cleaning and thoroughly rinsed with water before 7 6.2 mm ( 3 in. ) diamet er steel pucks, twelve 2 0.6 mm ( 13 / 16 in. ) and seven 1 2.7 mm ( in. ) thick, were attached with epoxy. The epoxy was left to cure for three days before the pull off test was conducted. A 007 James Bond Tester from James Instruments was attached to the steel pucks via a threaded rod and the three bolt s on the mount seen in Figure 4.3 were used to level the tester to ensure that the force was applied perpendicular to the specimens. Figure 4 3 007 James bond t ester The force required to detach the specimens from the pads was recorded and divided by the area of the specimens to acquire the tensile stress. All specimens failed at or approximately 3 .18 mm ( in. ) from the bonded interface.

PAGE 33

22 ( 4.2 ) Where: 4.3 Slant Shear Test The slant shear test was conducted on a total of 12 cores, two from each pad. T o acquire the core samples inclined at 30 degrees, a special stand with an angle of 15 degrees was const ructed because th e core drill mount could only be inclined 45 degrees, as seen in Fig ure 4.4 Figure 4 4 Core d rilling for s lant s hear t est The core samples were cut off at each end about 25 .4 mm ( 1 in. ) out from the bonded interface to make sure that the applied force was distributed equally over the core surface. Figure 4.5 shows a diagram of the slant core sample.

PAGE 34

23 Figure 4 5 Core sample with slanted in terface The cores were placed in the testing machine with neoprene pad caps on each end and a compression force was applied. Figure 4.6 shows the testing machine and Figure 4.7 a close up on the core sample. Figure 4 6 Slant s hear t est setup

PAGE 35

24 Figure 4 7 Close u p on s lant s hear s ample The applied force was measured a nd the shear ( and normal ( ) stresses were calculate d using the equations presented below ( 4.3 ) ( 4.4 ) Where : All samples failed in shear at the bonded interface

PAGE 36

25 4.4 Guillotine Test The guillotine test was conducted on 18 specimens, three from each pad. The core drill was leveled with the pads to ensure that the samples were acquired perpendicular to the surface of the pads. The samples were approximately 17 1 mm ( 6 in. ) l ong, the ful l depth of overlay and substrate Figure 4 8 Vertical c ore d rilling for g uillotine t est The test was conducted in a MTS load frame model 312 with a load capacity of 90 kN ( 20 kips ) The samples were placed in the guillotine d evice shown in Figure 4.9 with the bonded interface between the inner and outer jig.

PAGE 37

26 Figure 4 9 Shear j ig d esign A compression force was applied on top of the shear jig and was run under displacement control at the rate of 0.5 mm/minute ( 0.02 in/minute ) T he force required to shear off the sample was monitored Because of the design of the shear jig, with the core sample supported by the jig at the rear end and at the bonded plane, the applied force had to be divided by 2 to obtain the shear failure force at the bond plane as shown in E quation 4.5 The test setup can be seen in Figure 4.10. Figure 4 10 Guillotine t est setup ( 4.5 ) Where:

PAGE 38

27 4.5 Jacking Test For the jacking test, t he pads were saw cut into blocks using a concrete saw as seen in Figure 4.11 For each pad, three solid blocks plus a various amount of blocks with holes from previous coring were prepared. The solid blocks were approximately 10 2 mm x 20 3 mm ( 4 in. x 8 in. ) and the cut was made through the entire overlay and approximately 6.35 mm ( in ) into the substrate. Figure 4 11 Blocks on pad s 5 and 6 p repared for j acking t est The steel angle was mounted with the bol ts and the jack a Simplex RC306C with a 300 kN ( 30 ton ) capacity and 152 mm ( 6 in ) of travel length, put in place Because of the length of the jack, for all of the solid blocks the steel angle had to be rotated 180 degrees, meaning that the bolts suppor ted the rear end of the angle inducing a minor upwards bending of the angle as seen in Figure 4.12. Due to the diameter of the jack and the variable surface treatments, the ram head encountered the blocks at different heights, resulting in variations in ec centricity and bending moments between the blocks.

PAGE 39

28 Figure 4 12 Jack s et u p for t esting Once the blocks were detached from the pads the exact area s were measured and adjusted due to discontinuities from coring and rebar chairs. ( 4.6 ) Where:

PAGE 40

29 5 Experiment Results The results from the four different tests are presented in this chapter. In Table 5.1 the different surface treatments are summarized. Table 5 1 : Surface treatments Pad Surface Treatment Vibrated 1 Broom Yes 2 Broom No 3 Rake Yes 4 Rake No 5 Bush Hammer Yes 6 Bush Hammer No 5.1 Compression Test The results from testing the three specimens from each surface treatment catego ry and the overlay s are shown in T able 5.2. Table 5 2 : Compression t est r esults Poured In Sample Age (Days) Compression Force kN (lb) Compressive Strength kPa (psi) Substrate 1,2 49 323 (72 700) 39 900 (5 780) Substrate 1,2 49 333 (74 900) 41 100 (5 960) Substrate 3,4 49 310 (69 800) 38 300 (5 550) Substrate 3,4 49 291 (65 500) 35 900 (5 210) Substrate 5,6 45 179 (40 300) 22 100 (3 200) Substrate 5,6 45 167 (37 600) 20 600 (2 990) Overlay 1 6 35 242 (54 500) 29 900 (4 340) Overlay 1 6 35 262 (58 900) 32 300 (4 680) Substrate 1,2 70 354 (79 500) 43 600 (6 330) Substrate 3,4 70 340 (76 400) 41 900 (6 080) Substrate 5,6 66 177 (39 900) 21 900 (3 170) Overlay 1 6 56 334 (75 100) 41 200 (5 970)

PAGE 41

30 The increase in compressive strength with time was accounted for. In Table 5.3 are the compressive strengths for each pad calculated based on the age of the concrete at the time of each test. A quadratic relationship based on the test data in Table 5.2 was used to approximate the strength at different dates, as illustrated in Figure 6.1. Table 5 3 : Interpolated c ompressive s trength at the t ime of t esting Test Poured In Compressive Strength kPa (psi) Slant Shear Test Substrate 1,2 43 600 (6 330) Slant Shear Test Substrate 3,4 41 900 (6 080) Slant Shear Test Substrate 5, 6 21 900 (3 170) Slant Shear Test Overlay 1 6 41 200 (5 970) Guillotine Test Substrate 1,2 43 600 (6 320) Guillotine Test Substrate 3,4 41 200 (5 980) Guillotine Test Substrate 5, 6 21 900 (3 170) Guillotine Test Overlay 1 6 39 000 (5 650) Jacking Test Substrate 1,2 43 600 (6 330) Jacking Test Substrate 3,4 41 900 (6 080) Jacking Test Substrate 5, 6 21 900 (3 170) Jacking Test Overlay 1 6 41 200 (5 970) Pull Off Test Substrate 1,2 41 200 (5 970) Pull Off Test Substrate 3,4 37 900 (5 490) Pull Off Test Substrate 5, 6 21 600 (3 140) Pull Off Test Overlay 1 6 32 300 (4 690) Table 5.3 shows that t he overlays for e ach pad all have the same measured compressive strength due to the fact that a total of three cylinder s from one batch of overlay mix were collected

PAGE 42

31 5.2 Pull Off Test The results from the three specimens from each pad is presented in Table 5.4. Table 5 4 a: Pull o ff t est r esults (SI units) Sample f'c Substrate (kPa) f'c Overlay (kPa) (kPa) 1B 41 200 32 300 2480 0.08 13.80 1D 41 200 32 300 1780 0.06 9.90 1F 41 200 32 300 1560 0.05 8.68 2A 41 200 32 300 1240 0.04 6.90 2C2 41 200 32 300 1150 0.04 6.40 2E 41 200 32 300 1340 0.04 7.46 3B 37 900 32 300 1800 0.06 10.02 3D 37 900 32 300 2290 0.07 12.74 3F 37 900 32 300 1780 0.06 9.90 4A 37 900 32 300 990 0.03 5.51 4C 37 900 32 300 1310 0.04 7.29 4E 37 900 32 300 1560 0.05 8.68 5B 21 600 32 300 890 0.04 6.06 5D 21 600 32 300 1470 0.07 10.00 5F 21 600 32 300 760 0.04 5.17 6A 21 600 32 300 1150 0.05 7.82 6C 21 600 32 300 990 0.05 6.74 6E 21 600 32 300 600 0.03 4.08

PAGE 43

32 Table 5 4 b : Pull o ff t est r esults (Imperial units) Sample f'c Substrate (psi) f'c Overlay (psi) (psi) 1B 5 970 4 690 360 0.08 5.26 1D 5 970 4 690 260 0.06 3.78 1F 5 970 4 690 230 0.05 3.31 2A 5 970 4 690 180 0.04 2.63 2C2 5 970 4 690 170 0.04 2.43 2E 5 970 4 690 190 0.04 2.83 3B 5 490 4 690 260 0.06 3.80 3D 5 490 4 690 330 0.07 4.86 3F 5 490 4 690 260 0.06 3.78 4A 5 490 4 690 140 0.03 2.09 4C 5 490 4 690 190 0.04 2.77 4E 5 490 4 690 230 0.05 3.31 5B 3 140 4 690 130 0.04 2.30 5D 3 140 4 690 210 0.07 3.77 5F 3 140 4 690 110 0.03 1.97 6A 3 140 4 690 170 0.05 2.95 6C 3 140 4 690 140 0.05 2.54 6E 3 140 4 690 90 0.03 1.54 Table 5.5 shows that vibrated broom and vibrated rake surface s pad 1 and 3, have almost identical capacity while the vibrated bush hammered surface has a capacity of slightly higher than half of the others. From Table 5.6 the results show that for broom and rake surfaces, the vibration increased the capacity more than 50% while the bu sh hammered surfaces only increased 14% after vibration.

PAGE 44

33 Table 5 5 a : Pull o ff t est r esults Average t ensile s tress (SI units) Pad (kPa) 1 1 940 10.79 2 1 240 6.92 3 1 960 10.89 4 1 280 7.16 5 1 040 7.08 6 910 6.21 Table 5 5 b : Pull o ff t est r esults Average t ensile s tress (Imperial units) Pad (psi) 1 282 4.11 2 180 2.63 3 284 4.15 4 186 2.72 5 151 2.68 6 132 2.35 Table 5 6 : Pull o ff t est r esults Influence from v ibration Surface Treatment Vibrated Pad Capacity Non Vibrated Pad Capacity Broom (Pad 1 & 2) 156% Rake (Pad 3 & 4) 152% Bush Hammer (Pad 5 & 6) 114%

PAGE 45

34 Table 5.7 presents the statistical results for the test. T a ble 5 7 a : Pull off test results Statistical results (SI units) Pad Standard deviation Coefficient of Variation 1 10.79 2.67 24.8% 2 6.92 0.53 7.6% 3 10.89 1.61 14.8% 4 7.16 1.59 22.2% 5 7.08 2.57 36.3% 6 6.21 1.92 31.0% T a ble 5 7 b : Pull off test results Statistical results (Imperial units) Pad Standard deviation Coefficient of Variation 1 4.11 1.02 24.8% 2 2.63 0.20 7.7% 3 4.15 0.62 14.8% 4 2.72 0.61 22.4% 5 2.68 0.96 35.9% 6 2.35 0.73 30.9%

PAGE 46

35 5.3 Slant Shear Test The result from each slant shear specimen is presented in Table 5. 8 Table 5 8 a : Slant s hear t est r esults (SI units) Normal s tress and s hear s tress c alculated from Eq. 4.3 and 4.4 Sample f'c Substrate (kPa) f'c Overlay (kPa) (kPa) (kPa) 1G 43 600 41 200 9 100 15 700 77.3 1Q 43 600 41 200 8 400 14 600 71.9 2G 43 600 41 200 6 100 10 500 51.7 2Q 43 600 41 200 5 700 9 800 48.3 3G 41 900 41 200 8 500 14 800 72.9 3Q 41 900 41 200 8 500 14 800 72.9 4G 41 900 41 200 6 300 10 900 53.7 4Q 41 900 41 200 6 000 10 300 50.7 5G 21 900 41 200 6 500 11 200 75.7 5Q 21 900 41 200 5 800 10 000 67.6 6G 21 900 41 200 4 300 7 400 50.0 6Q 21 900 41 200 3 900 6 700 45.3 Table 5 8 b: Slant s hear t est r esults (Imperial units) Normal s tress and s hear s tress c alculated from Eq. 4.3 and 4.4 Sample f'c Substrate (psi) f'c Overlay (psi) (psi) (psi) 1G 6 330 5 970 1 320 2 280 29.5 1Q 6 330 5 970 1 220 2 120 27.4 2G 6 330 5 970 880 1 520 19.7 2Q 6 330 5 970 820 1 420 18.4 3G 6 080 5 970 1 240 2 150 27.8 3Q 6 080 5 970 1 240 2 150 27.8 4G 6 080 5 970 910 1 580 20.4 4Q 6 080 5 970 870 1 500 19.4 5G 3 170 5 970 940 1 620 28.8 5Q 3 170 5 970 840 1 450 25.8 6G 3 170 5 970 630 1 080 19.2 6Q 3 170 5 970 560 970 17.2

PAGE 47

36 Table 5.9 shows that pad 1 and 3, vibrated rake and vibrated broom surface, had the highest capacity. The results in Table 5. 10 shows that the influence from vibration is significant for all different surface treatments. Table 5 9 a : Slant s hear r esults Average s hear s tress (SI units) Pad Avg. Shear Stress ( ) (kPa) 1 15 200 74.6 2 10 100 50.0 3 14 800 72.9 4 10 600 52.2 5 10 600 71.6 6 7 100 47.6 Table 5 9 b : Slant s hear r esults Average s hear s tress (Imperial units) Pad Avg. Shear Stress ( ) (psi) 1 2 200 28.5 2 1 470 19.0 3 2 150 27.8 4 1 540 19.9 5 1 540 27.4 6 1 030 18.3 Table 5 10 : Slant s hear r esults Influence from v ibration Surface Treatment Vibrated Pad Capacity Non Vibrated Pad Capacity Broom (Pad 1 & 2) 150% Rake (Pad 3 & 4) 140% Bush Hammer (Pad 5 & 6) 150%

PAGE 48

37 5.4 Guillotine Test In Table 5.1 1 the results from three specimens from each pad is presented. Table 5 11 a : Guillotine t est r esults (SI units) Sample f'c Substrate (kPa) f'c Overlay (kPa) (kPa) 1A 43 600 39 000 2 640 13.37 1C 43 600 39 000 3 260 16.51 1E 43 600 39 000 4 3 40 21 98 2B 43 600 39 000 1 730 8.76 2D 43 600 39 000 1 780 9.01 2F 43 600 39 000 1 210 6.13 3A 41 200 39 000 3 710 18.79 3C 41 200 39 000 2 500 12.66 3E 41 200 39 000 3 240 16.41 4B 41 200 39 000 3 150 15.95 4D2 41 200 39 000 2 750 13.93 4F 41 200 39 000 2 360 11.95 5A 21 900 39 000 1 860 12.57 5C 21 900 39 000 2 570 17.37 5E 21 900 39 000 1 650 11.15 6B3 21 900 39 000 2 430 16.42 6D 21 900 39 000 2 090 14.12 6F 21 900 39 000 1 100 7.43

PAGE 49

38 Table 5 11 b: Guillotine test results (Imperial units) Sample f'c Substrate (psi) f'c Overlay (psi) (psi) 1A 6320 5650 383 5.10 1C 6320 5650 473 6.30 1E 6320 5650 630 8 38 2B 6320 5650 251 3.35 2D 6320 5650 258 3.43 2F 6320 5650 175 2.33 3A 5980 5650 538 7.16 3C 5980 5650 362 4.82 3E 5980 5650 470 6.25 4B 5980 5650 457 6.08 4D2 5980 5650 399 5.31 4F 5980 5650 343 4.56 5A 3170 5650 270 4.80 5C 3170 5650 373 6.62 5E 3170 5650 239 4.25 6B3 3170 5650 352 6.26 6D 3170 5650 302 5.37 6F 3170 5650 160 2.84 Table 5.1 2 shows that pad 3, vibrated rake surface, has the highest capacity followed by vibrated broom and vibrate d bush hammer surface. Table 5.1 3 indicates that the vibration for the broom surface is of high significance while for the rake and bush hammer surfaces it did not increase the capacity with more than 14 and 8 percent respectively.

PAGE 50

39 Table 5 12 a : Guillotine t est r esults Average s hear s tress (SI units) Pad (kPa) 1 3 420 1 7 3 2 1 570 8.0 3 3 150 16.0 4 2 760 13.9 5 2 030 13.7 6 1 870 12.7 Table 5 12 b : Guillotine t est r esults Average s hear s tress (Imperial units) Pad (psi) 1 496 6 6 0 2 228 3.03 3 457 6.08 4 400 5.32 5 294 5.22 6 272 4.82 Table 5 13 : Guilloti n e t est r esults Influence from v ibration Surface Treatment Vibrated Pad Capacity Non Vibrated Pad Capacity Broom (Pad 1 & 2) 2 1 7 % Rake (Pad 3 & 4) 114% Bush Hammer (Pad 5 & 6) 108%

PAGE 51

40 Table 5.14 presents the statistical results for the test. Table 5 14 a : Guillotine test results Statistical results (SI units) Pad Standard Deviation Coefficient of Variation 1 1 7 3 4 36 25 2 % 2 8.0 1.60 20.1% 3 16.0 3.09 19.4% 4 13.9 2.00 14.3% 5 13.7 3.26 23.8% 6 12.7 4.67 36.9% Table 5 14 b : Guillotine test results Statistical results (Imperial units) Pad Standard Deviation Coefficient of Variation 1 6 6 0 1 .6 6 25 2 % 2 3.03 0.61 20.3% 3 6.08 1.18 19.4% 4 5.32 0.76 14.3% 5 5.22 1.24 23.8% 6 4.82 1.77 36.8%

PAGE 52

41 5.5 Jacking Test The results from the three solid blocks from each pad is presented in table 5.15. Table 5 15 a : Jacking t est b lock d ata (SI units) Sample f'c Substrate (kPa) f'c Overlay (kPa) (kPa) 1H 43 600 41 200 570 2.81 1I 43 600 41 200 810 3.99 1J 43 600 41 200 600 2.96 2H 43 600 41 200 720 3.55 2I 43 600 41 200 620 3.05 2J 43 600 41 200 820 4.04 3H 41 900 41 200 970 4.78 3I 41 900 41 200 960 4.73 3J 41 900 41 200 1 320 6.50 4H 41 900 41 200 570 2.81 4I 41 900 41 200 1 220 6.01 4J 41 900 41 200 600 2.96 5N 21 900 41 200 1 180 7.97 5O 21 900 41 200 1 850 12.50 5P 21 900 41 200 1 940 13.11 6H 21 900 41 200 1 270 8.58 6I 21 900 41 200 820 5.54 6J 21 900 41 200 850 5.74

PAGE 53

42 Table 5 15 b : Jacking t est b lock d ata (Imperial units) Sample f'c Substrate (psi) f'c Overlay (psi) (psi) 1H 6 330 5 970 83 1.07 1I 6 330 5 970 117 1.52 1J 6 330 5 970 87 1.12 2H 6 330 5 970 105 1.36 2I 6 330 5 970 90 1.17 2J 6 330 5 970 119 1.54 3H 6 080 5 970 141 1.83 3I 6 080 5 970 139 1.80 3J 6 080 5 970 191 2.48 4H 6 080 5 970 83 1.08 4I 6 080 5 970 178 2.30 4J 6 080 5 970 87 1.13 5N 3 170 5 970 171 3.04 5O 3 170 5 970 268 4.76 5P 3 170 5 970 282 5.01 6H 3 170 5 970 184 3.27 6I 3 170 5 970 119 2.11 6J 3 170 5 970 123 2.18 Table 5.1 6 shows that the vibrated bush hammered treatment has the highest capacity followed by vibrated rake surface and vibrated broom surface. Table 5.1 7 indicates that vibration is of high significance for the bush hammer treatment and moderately impacts the raked surface while the broom surface experienced a decrease in capacity after vibration, which is the only test where the vibrated pad had a lower c apacity than the non vibrated pad wi th the same surface treatment.

PAGE 54

43 Table 5 16 a : Jacking t est r esults Average s hear s tress (SI units) Pad (kPa) 1 660 3.25 2 720 3.55 3 1 080 5.34 4 800 3.92 5 1 660 11.19 6 980 6.62 Table 5 16 b : Jacking t est r esults Average s hear s tress (Imperial units) Pad (psi) 1 96 1.24 2 105 1.35 3 157 2.04 4 116 1.50 5 240 4.27 6 142 2.52 Table 5 17 : Jacking t est r esults Influence f rom v ibration Surface Treatment Vibrated Pad Capacity Non Vibrated Pad Capacity Broom (Pad 1 & 2) 91% Rake (Pad 3 & 4) 135% Bush Hammer (Pad 5 & 6) 169%

PAGE 55

44 Table 5.18 presents the statistical results for the test. Table 5 18 a : Jacking test results Statistical results (SI units) Pad Standard Deviation Coefficient of Variation 1 3.25 0.64 19.8% 2 3.55 0.49 13.9% 3 5.34 1.01 18.9% 4 3.92 1.81 46.1% 5 11.19 2.81 25.1% 6 6.62 1.70 25.7% Table 5 18 b : Jacking test results Statistical results (Imperial units) Pad Standard Deviation Coefficient of Variation 1 1.24 0.25 19.9% 2 1.35 0.18 13.6% 3 2.04 0.38 18.8% 4 1.50 0.69 45.9% 5 4.27 1.07 25.1% 6 2.52 0.65 25.7%

PAGE 56

45 6 Discussion 6.1 Conducting Experiments Because the compressive strength w as only tested on two different occasions an approximate relationship was assumed for calculation of compression strength for tests conducted on days between the two test dates. The time strength relationship is in reality a curve th at approaches an asymptote once the concrete reaches its maximum strength. At the time of the second test the specimens were so old that it can be assumed that the concrete had reached its maximum strength, which was used for the jacking and slant shear te sts that were tested at times after the last compression test. The other two tests uses a quadratic relationship based on the two known strengths and zero strength at the time of casting. The first test was conducted relatively late, when the concrete had cured between 35 and 49 days, which is a time when the concrete time strength curve should have started to approach a horizontal asymptote and it should make the margin of error relatively small and therefore acceptable. To calculate the concrete compressi ve strength at the time for each test, a relationship for each pad and test was created based on the three known compressive strengths. The relationship can be seen in figure 6.1, which shows the development of compressive strength for substrate 1 and 2.

PAGE 57

46 Figure 6 1 Compressive s trength d evelopment The Slant Shear test resulted in higher capacity than the other three tests due to the with the inclined interface and the angle of load application the axial compression load is both shearing off the specimen but also a dds to the bonding by a compression force a clamping effect. This effect is depending on the coefficient of friction, wh ich directly relates to the shear bonding. A good bond will result in a higher clamping effect, and with that a higher capacity. Because the specimens all have the same geometry with the same inclination, it can be assumed that the clamping effect will dev elop equally in the different tests and therefore is an expression of the bond. Because of the different surface treatments, the eccentricity for the jack in the jacking test differed for the different pads. The broom and bush hammer finished pads should, due to the relatively flat surface have similar eccentricity. The rake surface was very uneven which made the eccentricity differ between the blocks, meaning that the bending moment most likely was higher for the raked pads and that it could have affected the results.

PAGE 58

47 6.2 Experiment Results Scatter plots for the average results per pad for each type of test are shown in F igures 6.1 to 6.4 and in F igure 6.5 are the plots combined and the total average is shown. In order to compare the different pads to each ot her and see which pads that performed best on each test, t he figures are accompanied by a table, T ables 6.1 to 6.4. In those tables the pads are ranked against each other using equati on 6.1 ( 6.1 ) Where: Equation 6.1 gives each pad a ratio based on the pad that performed best for each test. performing pad, meaning that the pad with the best result will get the number 100 % in Tables 6.1 to 6.4 and the remaining pads will show the capacity in relation to the best performing pad.

PAGE 59

48 6.2.1 Slant Shear Test Figure 6 2 a Slant s hear t est (SI units) The influence of compaction by vibration can be seen Figure 6 2 b. Slant s hear t est (Imperial units) The influence of compaction by vibration can be seen

PAGE 60

49 Table 6 1 : Slant s hear t est Ranking by c apacity Ratio based on highest compressive strength of either substrate or overlay Rank by Capacity Pad 1 1 100% 2 3 97% 3 5 96% 4 4 70% 5 2 67% 6 6 64% Table 6.1 shows that the results from the vibrated pads, 1, 3 and 5, are very close to each other and the difference is only 4% between pad 1 and pad 5 and 3% between pad 1 and pad 3. The non vibrated pads shows results with a capacity of 64% to 70% of the top ranked pad 1. Because of the relatively small difference comparing the three compacted pads to each other and comparing the three non compacted pads to each other, a conclusion can be reached that for the slant shear test it is mainly the vibration th at influenced on the result, and only minor influence due to the different surface treatments. It also shows that the clamping effect discussed earlier results in a very high shear capacity which in combination with the small differences between the pads m eans that the clamping effect is quite significant and evens out the differences between the pads surface treatments. 6.2.2 Guillotine Test Figure 6.3 shows that the non vibrated broom finish has an unusually low shear strength, significantly lower than the othe r pads in the Guillotine test.

PAGE 61

50 Figure 6 3 a. Guillotine t est Comparison (SI units) Figure 6 3 b. Guillotine t est Comparison (Imperial units)

PAGE 62

51 Table 6 2 : Guillotine t est Ranking by c apacity Ratio based on highest compressive strength of either substrate or overlay Rank by Capacity Pad 1 1 100% 2 3 93% 3 4 81% 4 5 79% 5 6 73% 6 2 46% Table 6.2 shows that the two pads with the highest capacity, pad 1 and pad 3, are the compacted broom and rake d surfaces The third best capacity has pad 4, non compacted rake d surface which has a capacity slightly higher than the compacted bush hammer pad meaning that the bush hammer surfaces overall did not perform well on the Guillotine Test. 6.2.3 Jacking Test Figure 6.4 shows that the compacted bush hammered surface has an unusua lly high strength and that the non vibrated bush hammered surface has a strength higher than all other pads, even the compacted ones.

PAGE 63

52 Figure 6 4 a. Jacking t est Comparison (SI units) Figure 6 4 b. Jacking t est Comparison (Imperial units)

PAGE 64

53 Table 6 3 : Jacking t est Ranking by c apacity Ratio based on highest compressive strength of either substrate or overlay Rank by Capacity Pad 1 5 100% 2 6 59% 3 3 48% 4 4 35% 5 2 32% 6 1 29% Table 6.3 shows that the top ranked pad 5 vibrated bush hammer surface has a capacity much higher than the second ranked pad 6 non vibrated bush hammer surface that has a capacity of 59% compared to pad 5. Going down the list we can see that the ranking is sorted surface treatment by surface treatment, where bush hammer surface is ranked first, rake d surface second and broom surfa ce last. The broom surface has a non vibrated capacity higher than the vibrated b room surface The jacking test is the only test in which a non vibrated surface treatment shows a higher capacity than its vibrated counterpart. 6.2.4 Pull Off Test Figure 6.5 shows that the compacted broom and compacted rake surfaces had a shear strength significantly higher than the other pads.

PAGE 65

54 Figure 6 5 a. Pull o ff t est Comparison (SI units) Figure 6 5 b. Pull o ff t est Comparison (Imperial units)

PAGE 66

55 Table 6 4 : Pull o ff t est Ranking by c apacity Ratio based on highest compressive strength of either substrate or overlay Rank by Capacity Pad 1 3 100% 2 1 99% 3 4 66% 4 5 65% 5 2 63% 6 6 57% From T able 6.4 it is shown that pad 3 has the highest capacity but that pad 1 has a capacity 99% of pad 3, which shows that the two pads surfaces have a very similar tensile strength. After the two top ranked pads pad 4, 5 and 2 with capacities 66%, 65% and 63% of pad 3 respectively is on the list with a gap down to pad 6 with only 57% of pad 3. 6.2.5 Summary The different test results illustrate different wa ys of describing the same bonded interface. Figure 6.6 clearly show s that no matter which test is used, the same pattern for the pads appears.

PAGE 67

56 Figure 6 6 a. Summary of t est r e s ults (SI units) Solid symbol s represent overlays consolidated by vibration while hollow symbols represent unconsolidated overlays. Figure 6 6 b. Summary of t est r esults (Imperial units) Solid symbols represent overlays consolidated by vibration while hollow symbols represent unconsolidated overlays.

PAGE 68

57 Table 6 5 : Average i nfluence from v ibration Surface Treatment Vibrated Pad Capacity Non Vibrated Pad Capacity Broom (Pad 1 & 2) 154 % Rake (Pad 3 & 4) 135% Bush Hammer (Pad 5 & 6) 135% Looking at the results in F igure 6. 6 it is shown that the influence from compaction overall is of high significance. For all surface treatments in all tests, except the broom finish in the jacking test, the compacted pad for each surface treatment has a significantly higher capacity than the non compacted pad The broom finish in the jacking test has a non compacted capacity about 10% higher than its compacted counterpart. The compac tion significance is made clear in T able 6. 5 where the average influence from compaction for the three surface treatments is shown. The broom surface shows the highest increase in capacit y with a 54 % increase while the rake and bush hammer surfaces both h ave an average capacity increase of 35%.

PAGE 69

58 To see how well the different surface treatments performed, Table 6.6 summarizes all pads and how they ranked on each test. Table 6 6 : Summarized r anking by c apacity It is clear that Pad 1, compacted broom finish, and Pad 3, compacted rake finish, appears in the top for most of the tests and after adding it all together, the pad that overall performed best was Pad 3. The table also illu strates the increase in strength from compaction. When looking at the total average ranking, the three compacted pads performed better overall than the non compacted pads. Rank by Capacity Slant Shear Test Guillotine Test Jacking Test Pull Off Test Total 1 Pad 1 Pad 1 Pad 5 Pad 3 Pad 3 2 Pad 3 Pad 3 Pad 6 Pad 1 Pad 1 3 Pad 5 Pad 4 Pad 3 Pad 4 Pad 5 4 Pad 4 Pad 5 Pad 4 Pad 5 Pad 4 5 Pad 2 Pad 6 Pad 2 Pad 2 Pad 6 6 Pad 6 Pad 2 Pad 1 Pad 6 Pad 2

PAGE 70

59 When there is need for a test in the field, the most commonly used method is the pull off test. Compared to the other tests conducted in this experiment, the pull off test is the only one that can be performed relatively easily in the field on an existing structure. The issues with the pull off test is that it gives the tensile strength of the bonded interface, while in most cases, it is the shear strength that is of importance. To be able to use the data from the pull off test and get an approximation of the shear strength, a correlation factor between the Guillotine test and th e pull off test can be used. In Table 6.7 the shear stress and tensile stress from each pad have been compared and a ratio has been calculated. It shows a span between 1.27 and 2.15 for the six pads wi th the calculated average at 1.80 and a standard deviat ion of 0.3 3 This number can be compared with Delatte discussed in chapter 2.4.1 where the average calculated ratio was 2.04 (Delatte 2000) Table 6 7 : Guillotine t est and p ull o ff t est Correlation Avg. shear s tress is from the g uillotine test, a vg. t ensile s tress is from the p ull o ff t est. Ratio = (Avg. Shear Stress) / (Avg. Tensile Stress) Pad Avg. Shear Stress ( ) kPa (psi) kPa (psi) Ratio 1 3 420 (4 96 ) 1 940 (282) 1. 7 6 2 1 570 (228) 1 240 (180) 1.27 3 3 150 (457) 1 960 (284) 1.61 4 2 760 (400) 1 280 (186) 2.15 5 2 030 (294) 1 040 (151) 1.95 6 1 870 (272) 910 (132) 2.06 Average 1.80

PAGE 71

60 7 Conclusions The experiment clearly shows that different surface treatments have big impact on the shear strength at the bonded interface. Within the same test the differences between the pad with best and worst result ranges from a factor of 1.6 to 3.4. Compacting the pads makes a difference for the shear strength of the pad with an average performance increase of 38 percent after compacting. For all tests combined the pad with the overall highest shear strength is pad number 3, raked substrate surface with a compacted overlay. The experiment shows that the measured shear strength differs significantly depending on which test method that was used. The two shear tests, expected to show similar shear strength, the guillotine and jacking test, showed a guillotine shear str ength for the pads ranging from 1.2 to 4.4 times the measured shear from the jacking test with an average factor of 2.7. The average data from the tests shows that from the pull off test the tensile strength is 8.2 for SI units and 3.1 for imperial units the shear strength from the slant shear test is 61.5 for SI units and 23.5 imperial units the shear strength from the guillotine test is 13. 6 for SI units and 5. 2 for imperial units and the shear strength from the jacking test is 5.7 for SI units and 2.2 for imperial units The big slant shear factor is attributed to the clamping force developed from the compression force applied to the specimen due to its geometry. The results shows that it is important to know which tests that are influenced by which factors and to be aware of that there is no universal factor to convert results from one test to another. To get the most reliable results when comparing shear strength at the interface from different projec ts, the same type of test should be conducted.

PAGE 72

61 REFERENCES 1. ASTM C1583 (2015), Standard test method for tensile strength of concrete surfaces and the bond strength or tensile strength of concrete repair and overlay materials by direct tension (pull off Metho d) American Society of Testing Materials, West Conshohocken, PA. 2. Bakhsh, K. N. (2010), Evaluation of bond strength between overlay and substrate in concrete repairs Master Thesis, School of Architecture and the Built Environment, Royal Institute of Techn ology, Stockholm, Sweden 3. Bissonnette, et al. (2012), Best practices for preparing concrete surface overlays, Report No. MERL 12 17 U.S. Bureau of Reclamation, Dept. of the Interior, Technical Service Center, Denver, CO. 4. Bissonnette B. et al. (2011), Bonded Cement Based Material Overlays for Springer Netherlands. 5. Bonaldo, E., Barros, J.A.O and Lourenco, P.B. (2005). Bond Characterization between Concrete Substrate and Repairing SFRC Using Pull Off Testing International Journal of Adhesion and Adhesives Elsevier, United kingdom, Dec 2005 463 474. 6. Delatte, N.J. et al. early ACI Materials Journal American Concrete Institute, Farmington Hills, MI, Mar Apr 2000, 201 207. 7. Delatte, N.J. et al. ACI Materials Journal American Concrete Ins titute, Farmington Hills, MI, June 2000, 272 280. 8. Ferraro C. (2008) I Department of Transportation Structural Materials Laboratory, Gainesville, FL 9. hod to determine epoxy ACI Journal American Concrete Institute, Farmington Hills, MI, July 1976, 372 373. 10. Silfwerbrand, J. (1990), Concrete International American Concrete Institute, F armington Hills, MI, Sept 1990 61 66. 11. Concrete International American Concrete Institute, Farmington Hills, MI, Oct 1998, 56 61 12. Silfwerbrand Materials and Structures Springer, Netherlands, July 2003, 419 424 13. term American Concrete Institute, Farmington Hills, MI, Nov Dec 1994, 560 566.

PAGE 73

62 14. Vaysburd and McDonald, 1999. Technical Report REMR CS 61, An Evaluation of equipment and procedures for tensile bond testing of concrete repairs US Army Corps of Engineers,Washing ton, D.C. 15. ACI Materials Journal American Concrete Institute, Farmington Hills, MI, Mar Apr 1988, 117 125. 16. Interfacial shear bond strength between ol d and new concrete Louisiana State University, Baton Rouge, LA

PAGE 74

63 APPENDIX Introduction The following appendices present the raw data gathered from the tests and is the origin to all resu lt s and conclusions A. Compression Test Appendix A shows the data for the compressive strength tests It is presented by which pads the sample was gathered from and shows the axial compressive strength required to break the test cylinders and the compressive strength calculated using the cylinder size of 10 2 mm ( 4 in )

PAGE 75

64

PAGE 76

65

PAGE 77

66

PAGE 78

67

PAGE 79

68 B. Pull Off Test The data for the pull off test is presented below It is presented pad by pad and shows the required pull off force to detach the sample from the substrate and the force converted to stress using the actual core diameter o f 6 6 .7 mm ( 2 )

PAGE 80

69

PAGE 81

70

PAGE 82

71

PAGE 83

72

PAGE 84

73

PAGE 85

74

PAGE 86

75 C. Guillotine Test Results Below is the data acquired during the Guillotine test presented. The results are presented pad by pad and shows the applied axial compressive force that was required to shear off the samples. Each sample ha s a picture showing the failure and each sample is accompanied by a graph showing the development of the axial compressive force prior to failure.

PAGE 87

76

PAGE 88

77

PAGE 89

78

PAGE 90

79

PAGE 91

80

PAGE 92

81

PAGE 93

82

PAGE 94

83

PAGE 95

84

PAGE 96

85

PAGE 97

86

PAGE 98

87

PAGE 99

88 D. Slant Shear Test Results The data collected from the slant shear test is presented below. The test included two core samples from each pad. The data presented are the applied compressive force that resulted in failure and the calculated normal and shear forces with a picture showing the failure for each sample.

PAGE 100

89

PAGE 101

90

PAGE 102

91

PAGE 103

92

PAGE 104

93

PAGE 105

94

PAGE 106

95 E. Jacking Test Results Below are the unanalyzed data from the jacking test The first page shows the location of each block on the pads followed by the dimensions and failure force for the solid blocks from each pad with a picture of the failure face.

PAGE 107

96

PAGE 108

97

PAGE 109

98

PAGE 110

99

PAGE 111

100

PAGE 112

101

PAGE 113

102