University of Tennessee, Knoxville TRACE: Tennessee Research and Creative Exchange
Masters Theses Graduate School
12-2006
Rebound Ace Tennis Court Surface: The Effect of Temperature on the Coefficient ofriction F
Denise Helen Bauer University of Tennessee - Knoxville
Follow this and additional works at: https://trace.tennessee.edu/utk_gradthes
Part of the Engineering Commons
Recommended Citation Bauer, Denise Helen, "Rebound Ace Tennis Court Surface: The Effect of Temperature on the Coefficient of Friction. " Master's Thesis, University of Tennessee, 2006. https://trace.tennessee.edu/utk_gradthes/1501
This Thesis is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Masters Theses by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [email protected]. To the Graduate Council:
I am submitting herewith a thesis written by Denise Helen Bauer entitled "Rebound Ace Tennis Court Surface: The Effect of Temperature on the Coefficient ofriction. F " I have examined the final electronic copy of this thesis for form and content and recommend that it be accepted in partial fulfillment of the equirr ements for the degree of Master of Science, with a major in Industrial Engineering.
John Hungerford, Major Professor
We have read this thesis and recommend its acceptance:
Dongjoon Kong, Tyler Kress, Songning Zhang
Accepted for the Council: Carolyn R. Hodges
Vice Provost and Dean of the Graduate School
(Original signatures are on file with official studentecor r ds.) To the Graduate Council:
I am submitting herewith a thesis written by Denise Helen Bauer entitled “Rebound Ace Tennis Court Surface: The Effect of Temperature on the Coefficient of Friction.” I have examined the final electronic copy of this thesis for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Master of Science, with a major in Industrial Engineering.
John Hungerford Major Professor
We have read this thesis and recommend its acceptance:
Dongjoon Kong
Tyler Kress
Songning Zhang
Acceptance for the Council:
Linda Painter
Interim Dean of Graduate Studies
(Original signatures are on file with official student records.)
Rebound Ace Tennis Court Surface: The Effect of Temperature on the Coefficient of Friction
A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville
Denise Helen Bauer December 2006
Acknowledgments
I am extremely grateful to my thesis committee, John Hungerford, Dongjoon
Kong, Tyler Kress, and Songning Zhang, for their time and commitment. I would like to thank Dr. Tyler Kress for his guidance and support throughout both the thesis process and masters program, especially in the initial stages of formulating my research focus. I would also like to thank Dr. John Hungerford for his guidance, support, patience, and professionalism he brought to the thesis process and my experience as a masters student at the University of Tennessee. In spite of personal health issues, he still found the time to fulfill his role as my thesis advisor. For that, I am eternally grateful.
I would like to give special acknowledgment to Franz Fasold of Rebound Ace™,
Rapid Transit Sports, and the University of Tennessee women’s tennis coaching staff for providing the material necessary for the research. I also want to thank the Industrial
Engineering Department and the Engage Program for the use of experimental equipment and support throughout my masters program.
I want to pay special tribute to my life partner, Tim, who showed never-ending patience and encouragement. He has always stood by me, and this thesis would not be possible without his support and tireless efforts.
Finally, I would especially like to thank my mom and dad for instilling the importance of faith, family, and education. Their continuous encouragement and belief in me has guided me throughout my life, and I would not be where I am today without them. Thank you.
ii Abstract
The study was conducted to determine if temperature affects the safety of
Rebound Ace™ surface. The objective of this study is to determine how the COF changes as the Rebound Ace™ tennis court surface temperature increases. The lack of literature on this subject leads to the question of the surface’s safety at higher temperatures. A sample of Rebound Ace™ tennis court surface was tested at eight temperatures related to the climate of Melbourne in January with four different shoes to determine how the COF changes with an increase in temperature.
Temperature and shoe brand were both found to be significant factors in the COF.
However, the COF did not show a steady increase as the surface temperature of the
Rebound Ace™ tennis court increased; the COF instead varied over the range of
temperatures. The COF actually decreased over the higher temperatures (115 to 155
degrees F) and was lowest at the highest temperature of 155 degrees F. The Nike Air
Zoom Thrive had the lowest COF across all temperatures.
The results of this study indicate that the Rebound Ace™ tennis court surface may
not be a safety issue at higher temperatures. This finding indicates that the surface
properties may change with the temperature. One such change could be that the surface
liquefies at higher temperatures. The results of the temperature’s effect on the COF in
this study could lead one to believe that the surface is actually safer at higher
temperatures. However, there were limitations in the study with respect to sample size,
normal load amount, measurement of rotational friction, and measurements in the field that prevent this conclusion from be generalized for real life playing situations. Further iii research is needed to examine the true frictional behavior of the shoe-surface interface and introduce real playing forces.
iv Table of Contents
Chapter 1. Introduction ...... 1 Study Limitations and Delimitations ...... 3 Limitations ...... 3 Delimitations...... 4 Objectives and Hypotheses...... 5
Chapter 2. Background...... 6 Rebound Ace™ Surface...... 6 Coefficient of Friction...... 8 Literature Review...... 10 Friction in Sports...... 11 Friction in Tennis...... 16
Chapter 3. Methods...... 19 Equipment, Instrumentation, and Procedure...... 19 Shoes...... 19 Test Surface...... 19 Heating Apparatus...... 25 Static Coefficient of Friction Measurement...... 28 Experimental Design...... 28
Chapter 4. Results ...... 33 Three-way Repeated Measures ANOVA Model ...... 33 Temperature Comparisons...... 42 Shoe Comparisons ...... 47
Chapter 5. Discussion...... 52 Effect of Temperature on the COF ...... 53 Effect of Shoe on the COF...... 55 Temperature Comparisons...... 56 Shoe Comparisons ...... 58
Chapter 6. Conclusions and Recommendations ...... 59 Recommendations for Future Research...... 60
List of References...... 62
Appendix...... 69
Vita ...... 77
v List of Tables
Table 3.1. Shoe Characteristics...... 20
Table 3.2. Combinations of Surface Temperature and Pull Direction for One Shoe ...... 31
Table 4.1. ANOVA for COF...... 34
Table 4.2. COF Values for Temperature ...... 35
Table 4.3. COF Values for Shoe...... 37
Table 4.4. COF Values for Adidas Barricade...... 38
Table 4.5. COF Values for New Balance 650 ...... 39
Table 4.6. COF Values for Nike Air Zoom Thrive...... 40
Table 4.7. COF Values for Wilson DST...... 41
Table 4.8. Tukey’s Test for Adidas Barricade...... 43
Table 4.9. Tukey’s Test for New Balance 650 ...... 44
Table 4.10. Tukey’s Test for Nike Air Zoom Thrive ...... 45
Table 4.11. Tukey’s Test for Wilson DST...... 46
Table 4.12. Tukey’s Test for 72 degrees F ...... 47
Table 4.13. Tukey’s Test for 85 degrees F ...... 48
Table 4.14. Tukey’s Test for 100 degrees F ...... 48
Table 4.15. Tukey’s Test for 110 degrees F ...... 49
Table 4.16. Tukey’s Test for 115 degrees F ...... 49
Table 4.17. Tukey’s Test for 130 degrees F ...... 50
Table 4.18. Tukey’s Test for 140 degrees F ...... 50
Table 4.19. Tukey’s Test for 155 degrees F ...... 51
Table A.1. Minimum Sample Size Needed ...... 70 vi List of Figures
Figure 2.1. Composition of Rebound Ace™ – Grand Slam Surface...... 7
Figure 2.2. Forces Present When Sliding an Object Across a Horizontal Surface...... 9
Figure 3.1. Side View of Adidas Barricade ...... 21
Figure 3.2. Sole View of Adidas Barricade ...... 21
Figure 3.3. Side View of New Balance 650...... 22
Figure 3.4. Sole View of New Balance 650...... 22
Figure 3.5. Side View of Nike Air Zoom Thrive...... 23
Figure 3.6. Sole View of Nike Air Zoom Thrive...... 23
Figure 3.7. Side View of Wilson DST...... 24
Figure 3.8. Sole View of Wilson DST...... 24
Figure 3.9. Rebound Ace™ Tennis Court Test Surface ...... 26
Figure 3.10. Philips BR40 Infrared Heat Lamps ...... 26
Figure 3.11. Acu-Rite 00890 Wired Digital Indoor-Outdoor Thermometer ...... 27
Figure 3.12. Test Surface Set-up...... 27
Figure 3.13. Force Ten Horizontal Pull Force Gauge...... 29
Figure 3.14. Force Ten Horizontal Pull Force Gauge with 222 ± 0.9 N Cell Attached ....29
Figure 3.15. Shoe and Weight Set-up ...... 30
Figure 4.1. Residuals Versus the Order of the Data: Response is COF...... 34
Figure 4.2. Mean COF for the Eight Temperature Conditions...... 35
Figure 4.3. Mean COF for the Four Shoe Brands...... 37
Figure 4.4. Mean COF for Adidas Barricade...... 38
Figure 4.5. Mean COF for New Balance 650 ...... 39 vii Figure 4.6. Mean COF for Nike Air Zoom Thrive ...... 40
Figure 4.7. Mean COF for Wilson DST ...... 41
Figure A.1. Normal Probability Plot of COF...... 70
Figure A.2. Normal Probability Plot of Adidas Barricade COF...... 71
Figure A.3. Normal Probability Plot of New Balance 650 COF ...... 71
Figure A.4. Normal Probability Plot of Nike Air Zoom Thrive COF ...... 72
Figure A.5. Normal Probability Plot of Wilson DST COF...... 72
Figure A.6. Normal Probability Plot of 72 degrees F COF ...... 73
Figure A.7. Normal Probability Plot of 85 degrees F COF ...... 73
Figure A.8. Normal Probability Plot of 100 degrees F COF ...... 74
Figure A.9. Normal Probability Plot of 110 degrees F COF ...... 74
Figure A.10. Normal Probability Plot of 115 degrees F COF ...... 75
Figure A.11. Normal Probability Plot of 130 degrees F COF ...... 75
Figure A.12. Normal Probability Plot of 140 degrees F COF ...... 76
Figure A.13. Normal Probability Plot of 155 degrees F COF ...... 76
viii Chapter 1
INTRODUCTION
With the increasing popularity of artificial surfaces in the sports arena, there is
also an increasing scrutiny over the safety of these surfaces. Football turf has been the
focus of the natural versus artificial debate, but artificial surfaces are also used in many
other sports, such as tennis. Tennis is normally thought to be played on grass, clay, or
concrete hard courts; however, one surface is growing in popularity – Rebound Ace™.
Rebound Ace™ is the surface of choice for Melbourne Park where the Australian Open is
held. In recent years, the Rebound Ace™ surface has faced some criticism, mainly from the media, about the injuries that have occurred while playing on the surface. Their claim
is that the court surface temperature becomes too hot in the summer January heat causing
the court to become sticky and result in ankle and knee ligament injuries (Cross, 2004).
Although Australian Open officials and distributors of the product deny any
problems with the surface, tennis officials, doctors, and even players are still worried
about the injuries sustained on the court. In 2004, the Women’s Tennis Association
conducted an informal survey to discover whether the Rebound Ace™ surface was
thought to be contributing to the high amount of injuries among women players (Pearce,
2004). The findings indicated that it was hard to find a pattern that would lead to the
conclusion that the surface was indeed a factor in the injuries. Andreas Bisaz, the
chiropractor for Australia’s Davis Cup team, is a critic of the surface despite these
findings and believes the courts at Melbourne Park should be changed. He believes it is
the worse of all grand slam courts in that it anchors the players’ feet to the court and 1 increases the risk of ankle injuries (Schlink, 2006). Players Lleyton Hewitt and James
Blake are also opponents of the Rebound Ace™ courts. Hewitt not only criticizes the surface for its alleged stickiness but also for its slow speed (Sheppard, 2006). Blake agrees with Bisaz and believes the surface contributes to foot and ankle injuries (AFP,
2006).
These complaints from a wide range of individuals would make one think that the surface is indeed dangerous. However, there has been no evidence to support their claims.
The only indication that there may be a problem is the large number of players suffering injuries during play at the Australian Open. The mechanisms of the injuries lead many to call the safety of the surface into question. Players are appearing to catch their feet on the surface while either trying to pivot or slide across the court (Bowers, 2002). This is similar to ankle and knee injuries that have been linked to the shoe-surface interface in other sports (Ekstrand & Nigg, 1989; Lambson, Barnhill, & Higgins, 1996; Nigg &
Segesser, 1992; Orchard et al., 1999; Scranton et al., 1997; Valiant, 1990). Still others relate the high frequency of injuries to the early timing of the tournament and believe the court is safe.
Whether the Rebound Ace™ surface is a safety hazard or not is still up for debate.
Australian Open officials stand by the surface and are not planning to remove the surface any time soon. Critics voice their opinions year after year even without hard evidence.
There is no debate, however, about the weather conditions in Melbourne, Australia in
January. The maximum temperature averages approximately 79 degrees F (26 degrees C)
(Victoria Climate Center, 2006), but during the Australian Open, temperatures as high as
130 degrees F (54.4 degrees C) have been recorded (How hot is hot?, 1998; Stadiums and
2 the heat island phenomenon, 2002). With the court temperature averaging about 10
degrees F (6 degrees C) higher, the temperature becomes a larger factor during the
usually hot Australian Open. If the Rebound Ace™ surface does indeed become sticky as
the temperature rises, there may be claim behind all the complaints.
The purpose of this study is to determine if temperature does affect the safety, or
the probability of injury, of the Rebound Ace™ surface. Lack of literature in this area leaves open the debate on the court’s safety. The static coefficient of friction between the surface and a tennis shoe will be examined to establish whether the court may be more prone to injury at higher temperatures. To determine how the coefficient of friction changes with an increase in temperature, a sample of Rebound Ace™ tennis court surface was tested at eight temperatures related to the climate of Melbourne in January.
Study Limitations and Delimitations
There were limitations and delimitations of the study that must be considered
when analyzing the results and generalizing for real-life playing situations.
Limitations
The first limitation of the study is that Coulomb’s law of friction is used a crude
approximation of the coefficient of friction. This causes problems in that Coulomb’s law
assumes that the coefficient of friction is independent of surface area and weight.
However, when adhesion and stiction become important aspects, this assumption is not valid. The surface area and weight, which were not examined in this study, then become significant factors in the coefficient of friction. Another limitation is the fact that the shoe
3 soles were kept at room temperature during the testing. This would not be the case in actual playing surfaces and could affect the coefficient of friction values since both the court surface and the shoe soles are compressible materials that can change surface properties as they became hotter. A final limitation of the study was that the pulls on the shoe might not have been constant since the force gauge used was manual. The main problem arising from this situation is that the force exerted might not have always been exactly horizontal. This is a concern because slightly exerting an upward force could raise the shoe off the surface while a downward force could increase the adhesion between the shoe and surface.
Delimitations
A delimitation of the study is that only the Rebound Ace™ surface was examined.
There are many varieties of court surfaces that react differently to heat. Therefore, the results can only be applied to the Rebound Ace™ surface for the conditions studied.
These conditions are also delimitations of the study. Only four shoe brands in one shoe size were examined, which limits the results to be generalized for all tennis shoes. Two other conditions are that the normal force used in the study was not similar to a player’s body weight and only the static coefficient of friction was examined. These delimitations limit the results to be generalized for normal playing conditions since every player has a different weight, playing speed, and landing force.
4 Objectives and Hypotheses
The objective of this study is to determine how the COF changes as the Rebound
Ace™ tennis court surface temperature increases. The lack of literature on this subject leads to the question of the surface’s safety at higher temperatures. To guide the study,
the following hypotheses will be examined:
1. There will be a significant difference in the COF with a change in temperature.
H0: µT1 ≠ µT2 ≠ µT3 ≠ µT4 ≠ µT5 ≠ µT6 ≠ µT7 ≠ µT8
H1: µT1 = µT2 = µT3 = µT4 = µT5 = µT6 = µT7 = µT8
2. There will be no significant difference in the COF among the different shoe
brands.
H0: µS1 = µS2 = µS3 = µS4
H1: µS1 ≠ µS2 ≠ µS3 ≠ µS4
3. The COF will increase as the surface temperature of the Rebound Ace™ tennis
court increases.
H0: µT1 < µT2 < µT3 < µT4 < µT5 < µT6 < µT7 < µT8
H1: µT1 ≥ µT2 ≥ µT3 ≥ µT4 ≥ µT5 ≥ µT6 ≥ µT7 ≥ µT8
5 Chapter 2
BACKGROUND
Rebound Ace™ Surface
The Rebound Ace™ surface used on the courts at Melbourne Park (home of the
Australian Open) is a product of Rebound Ace Sports. Rebound Ace Sports is based in
Brisbane, Australia and is a division of AVSyntec Pty Ltd. The company is a leader in
surface technology and has over 30 years of experience in specialty coatings for sport
courts. Their surfaces, including the Rebound Ace™ – Grand Slam surface used at
Melbourne Park, are made to improve performance while at the same time reduce leg,
ligament, and lower back strains and injuries (A world leader, 2005). The company also
claims the surface will not lose resilience from aging or hard wear and benefits all styles
of play (The Grand Slam tennis surface, n.d.). This can be beneficial to customers since
little maintenance will be required to keep the court playable.
Rebound Ace™ surfaces are not limited to tennis courts. The use of the surfaces
is quite diverse as they can be found in schools, clubs, private residences, playgrounds, hotels, and parks throughout the world. Although the surface may be used in many places, the main composition of the surfaces is the same: a layer of cushion of polyurethane rubber on top of the base, fiberglass, layers of special filler coats, and at
least one layer of the playing surface. The composition of the Rebound Ace™ – Grand
Slam can be seen in Figure 2.1 (The Grand Slam tennis surface, n.d.). In most years, the surface at Melbourne Park is coated with two layers of the playing surface.
6
Figure 2.1. Composition of Rebound Ace™ – Grand Slam Surface.
7 Coefficient of Friction
When two surfaces are in contact, a force resists any motion or tendency of
motion between the two surfaces. This force is known as friction and is subdivided into
translational and rotational friction (Nigg & Yeadon, 1987). Translational friction acts
along a straight line while rotational friction occurs when the object is moving about an
axis (Nigg, 1990). This study only examined the translational friction due to limitations
in testing capabilities. In the simplest case, the translational friction can be approximated
by the equation recognized as Coulomb’s law of friction (Fast Car, 2004; Nigg, 1990).
Ff = µ*N
Where
Ff = the friction force between the two surfaces
µ = the coefficient of friction (COF)
N = the force normal to the contact surface
The friction force acts between the two surfaces parallel to the area of contact.
There must be a force (F) causing motion or potential motion of one of the objects in
relation to the other in order for a friction force to be present (Dixon, Batt, & Collop,
1999). The friction force acts in the opposite direction of the force causing the motion
(Figure 2.2). If the friction and normal forces are known, the COF can be found through
the ratio of friction force to normal force. This results in a dimensionless value that depends only on the material of the two surfaces; it is assumed independent of the contact
surface area and normal force (Nigg, 1990). Two smooth surfaces, such as ice and metal,
8
F
Ff
N
Figure 2.2. Forces Present When Sliding an Object Across a Horizontal Surface.
would have a low COF while two rougher surfaces, such as rubber and concrete, would
have a higher COF (Friction, 2006).
There are two types of translational friction, static and kinetic (dynamic). Static
friction occurs when an object is at rest and there is potential motion across a surface.
This occurs when the external force is less than the friction force. The COF is known as
static COF for this situation and is usually denoted as µs. When an object is on the brink
of motion (the external force is equal to the friction force), the maximum static COF can
be found from the maximum friction force that occurs at this point and the normal force
of the object (Coefficient of friction, 2006). Dynamic (kinetic) friction is present when an object is in motion along a surface (the external force is greater than the friction force).
The COF is denoted as µk for the dynamic COF and can be found through the friction force that is required to keep the object moving at a constant velocity (Bell, Baker, &
9 Canaway, 1985). The dynamic COF is used most often with contaminated surfaces and is
usually less than the static COF (Friction, 2006).
The interest for this study is the maximum friction force present while playing
tennis on the Rebound Ace™ court surface. This corresponds to the maximum static
COF, which will be referred to as COF throughout the rest of the report.
Two phenomenon that are of interest in this study are the concepts of adhesion and stiction. Adhesion is based on the idea that even two seemingly smooth surfaces may have a roughness to them (Bhushan, 2003). This roughness is comprised of peaks and valleys on the surface, which cause the actual contact area between the two surfaces to differ from the geometrical surface area. As one object moves over another, the contact between the peaks and valleys changes causing the objects to adhere to each other during sliding (Asperities, n.d.). Stiction occurs when there is a presence of a liquid film that increases the static friction due to the viscosity (resistance of a fluid to deform under shear stress) or meniscus properties (Bhushan, 2003). An important factor in the amount of adhesion or stiction experienced by the surfaces is the temperature. The surfaces can soften at high temperatures causing the material to become more fluid and increase the actual contact area. This occurrence is especially evident when the two surfaces are both polymers. Increases in the fluidity and contact area in turn increase the possibility of adhesion and stiction (Bhushan, 2003).
Literature Review
This section recounts some of the current literature on friction testing in sports as
well as the importance of friction in tennis.
10 Friction in Sports
Research in the interaction between the shoe and surface in sports began with the
development of synthetic football turf surfaces in the sixties (Bell, Baker, & Canaway,
1985). The reason for the sudden interest was due to conflicting injury rate statistics for artificial surfaces versus natural surfaces. Statistics from the company at the forefront of artificial turf development claimed that injury frequencies could be reduced by up to 80 percent on artificial turf (Morehouse & Morrison, 1975). However, a comparison on data collected from high school football players indicated that recorded injuries on artificial turf were 50 percent higher than on natural grass (Bramwell, Requa, & Garrick, 1972).
These two studies were only the beginning of the research on the playing surface’s influence on injuries sustained in sports, such as American football, soccer, rugby, and
Australian football.
Foot fixation is of great interest when examining the difference between artificial and natural surfaces. This occurrence, due to high levels of surface friction, is important in both contact and non-contact ankle and knee injuries (D’Ambrosia, 1985; Torg, 1982).
Lambson, Barnhill, and Higgins (1996) state that anterior cruciate ligament (ACL) injuries caused by contact could be avoided if the foot is loosened from the surface.
Surface friction has been found to be the main risk factor in non-contact injuries, which fall into two categories (impact of frictional forces when contacting the playing surface; muscle and tendon fatigue overload) and can vary from simple mat burns to severe knee dislocations (Heidt et al., 1996). Stanitski, McMaster, and Ferguson (1974) believed that the friction between the football shoe and surface would increase on artificial turf and in turn increase the probability of injury to larger linemen. Nigg and Segesser (1992) later
11 supported this idea in a study of non-contact injuries in soccer. They found that up to
two-thirds of the non-contact injuries were most likely due to excessive friction. These
and other similar studies suggest that the frictional properties of the shoe-surface
interface may possibly have a significant influence on short term and long term injuries to
the athlete (Milburn & Barry, 1998).
In their study on injuries in the National Football League (NFL), Powell and
Schootman (1992) determined that the shoe-surface interface was an important factor in
the risk of knee ligament injuries. They also found that the ACL injuries per team were
significantly higher on AstroTurf than on natural grass when surface type was treated as a single factor. Briner and Ely (1999) assessed the volleyball injuries during the 1995
United States Olympic Festival and discovered that fewer injuries were reported on sand courts, which are softer and possess lower foot-surface friction than hard courts. Bjordal
et al. (1997) found that the rate of ACL injuries in soccer was higher on natural grass
than gravel. The results were attributed to a higher shoe-surface friction level on the grass
surface. Similar results were seen in tennis where injuries were more frequent on grass
than hard courts, which produced more injuries than clay, due to the friction
characteristics of the surfaces (Bastholt, 2000; Nigg & Segesser, 1988). McClements and
Baker (1994) believed the shoe-surface interface is the most important playing attribute
in rugby and other sports and saw the need to examine all aspects of the playing surface,
such as the hardness, contamination, and temperature. These factors may influence the
frictional characteristics and thus the safety of the playing surface.
A review on the relationship between the ground and climatic conditions and
football injuries by Orchard (2002) found that, along with friction, the surface’s hardness
12 is a main factor relating to football injuries. The hardness of a surface can be attributed to
the material composition (concrete, hardwood, etc.) or the moisture content (natural turf
after a rainfall). These factors can also influence the friction of the surface; harder
surfaces will result in greater impact forces and can show larger frictional values. This is
especially evident when comparing a dry natural turf field to wet natural turf. Orchard et
al. (1999) discovered that the risk for ACL injuries in the Australian Football League
(AFL) increased when there was low moisture content in the soil. A review on non-
contact ACL injuries in American football by Scranton et al. (1997) revealed that nearly
all the injuries that occurred on natural grass where in dry conditions. These findings
agreed with the study by Heidt et al. (1996) that found the friction was higher for drier
natural turf conditions.
The presence of water on playing surfaces other than natural turf is categorized as
a contaminant rather than a measure of hardness. Contaminants can also be foreign
particles that are not part of the normal playing surface, such as dust and liter. In the case
of contamination, the surface COF usually decreases as found in the study by Newton et al. (2002) on friction changes with the presence of a liquid contaminant. Their results indicated that the friction of a wrestling mat was reduced by 14 percent when covered in a saline solution. However, results from a study by Stanitski, McMaster, and Ferguson
(1974) contradicted the findings of Newton et al. Their research on three artificial surfaces and natural grass under both wet and dry conditions showed no difference in the wet and dry conditions. This indicates that the playing surface properties may dictate whether contamination is a problem.
13 The temperature of the playing surface may affect both the hardness and the friction level. In a study of AFL fields, Orchard (2001) found that over the winter season the ground slightly softened. However, Milner (1985) discovered that if the temperature drops low enough, the ground becomes frozen and as hard as artificial turf. It was also established by Baker (1991) that the surface friction declines constantly over the winter season for natural turf fields. Torg, Stilwell, and Rogers (1996) found similar results for artificial turf. The results of their study on AstroTurf fields in the NFL showed that the friction increased with warmer temperatures for open stadiums. They believed that an increase in surface temperature could put the athlete’s knee and ankle at a higher risk of injury due to an increase in friction. They also alleged that the shoe worn by the athlete was an equally important factor.
Since friction is dependent on the two contacting surfaces, the shoe is another aspect that should be considered (Milburn & Barry, 1998). Torg and Quedenfeld (1971) studied injuries to high school football players and found that the number of cleats on the shoe was a factor in the risk of injury. They concluded that increasing the number of cleats reduced the translational and rotational friction, thus reducing the risk of injuries.
Lambson, Barnhill, and Higgins (1996) also studied the frictional property difference among cleat designs and found that the pattern of cleats may also be a risk factor. Their results showed that the shoes that contained cleats along the outside edge of the shoe had higher frictional values than the shoes without cleats on the outer edge. Orchard (2002) also concluded in his review of injuries in football that longer cleats increased the friction on natural grass. Outside of football, it has been found that the COF varies among shoes on the same flooring for team handball (Olsen et al., 2003). Newton et al. (2002) also
14 found that wrestling shoes with unworn or slightly worn soles had a higher COF than
wrestling shoe with worn soles on the same mat.
Although there are many factors that may affect the COF between the shoe and
surface in sports, there still should be an optimal value to ensure safety while providing
sufficient traction for performance (Bell, Baker, & Canaway, 1985; Ekstrand & Nigg,
1989; Nigg & Segesser, 1988). Ichii (1987) suggested the optimal COF value for sports to be between 0.5 and 0.7, which was based on objective and suggestive assessments. It
has also been recommended that a COF of 0.5 for tennis would reduce loads to the
athlete’s body (Nigg & Segesser, 1988). However, Barrett (1956) concluded that a COF
of 1.1 or above was required for running. Van Gheluwe, Deporte, and Hebbelink (1983)
found similar results for soccer shoes on artificial turf and established that a COF of 1.2
or higher was enough for all maneuvers while slipping occurred with a COF less than 1.0.
The findings of most other studies on the shoe-surface friction in sports correspond to the suggestions of Barrett and Van Gheluwe, Deporte, and Hebbelink. Garrick and LaVigne
(1972) and Milner (as cited in Bell, Baker, & Canaway, 1985) conducted two separate studies testing various shoes on AtroTurf. Garrick and LaVigne reported COF values of
1.1 to 2.45, and Milner found comparable results of 1.0 to 1.5. Stanitski, McMaster, and
Ferguson (1974) found COF values of 0.92 to 1.54 when testing three different artificial surfaces and a natural grass surface while Canaway (1979) tested five different natural grass species and found a COF range of 1.10 to 1.73. Similar COF values of 0.6 to 1.54 were shown by Newton et al. (2002) for three different wrestling shoes on a wet and dry mat. Although the values from the mentioned studies are comparable, the large range (0.6
15 to 2.45) indicates more research is needed to find the optimal range that enhances
performance while still maintaining safety.
Friction in Tennis
Like most other sports, enough friction must be present in tennis to provide ample
traction and prevent slipping (BSI, 1990) due to the many starts, stops, and cutting motions that require an optimal level of friction for performance (Dixon, Batt, & Collop,
1999; Ekstrand & Nigg, 1989). In addition, friction is an important factor in the bounce and spin of the tennis ball. However, too much friction can restrict a player’s foot movement and result in injury (BSI, 1990). These injuries are usually a result of high friction causing the loadings on the body to exceed the threshold (Dixon, Batt, & Collop,
1999). Sliding motions are also used in tennis to minimize the loading during a turn and require a lower level of friction between the shoe and surface. This causes a dilemma in tennis since relatively high friction is needed for starting, stopping, and cutting and low friction is required for sliding. Thus, the friction between the shoe and surface is a main factor in both the game and injury mechanisms (Dixon, Batt, & Collop, 1999).
Researchers have agreed that shoes and playing surfaces are potential hazards in
tennis and may cause overload injuries to the lower extremities (Van Gheluwe & Depote,
1992). The diversity of surfaces in tennis, from clay to concrete and natural grass to synthetic materials, has required researchers to pay special attention to the difference in
frictional properties. Renström (1995) found that it is more likely for the foot to stick to the ground on a fast court, which has a high level of friction, than a slower court. In addition, Nirshl and Sobel (1994) established that potential sudden unexpected stops, a
16 cause of ligament sprains in the knee, increased on a sticky tennis court surface. With the
use of a wide variety of surfaces and players wearing different shoes according to the
playing surface, it is important to understand the interactions between the player, shoe,
and surface (Renström, 1995).
Nigg and Segesser (1988) discovered in a literature review and retrospective
study on tennis injuries that there was a significant difference in the frequency of pain or
injury for different surfaces. They believed one reason for this difference was the
frictional properties of the various surfaces. Surfaces with a low but minimal COF were
considered safer and to have a lower frequency of injuries than surfaces with a high COF
(Nigg & Segesser, 1988). These results lead Nigg and Segesser to suggest that the main
reason for the differences in the frequency of pain and injury among tennis court surfaces
was the frictional properties of the surfaces. This conclusion was supported by Van
Gheluwe and Deporte (1992) in their research on frictional forces and torques produced during a forehand drive for different surfaces and shoes both in the field and in the laboratory. Their results showed that the friction varied over the different surfaces, but not among shoes. Van Gheluwe and Deporte also concluded that fluid surfaces, such as those covered in clay or sand, had the lowest friction values and were the least likely to cause injuries due to frictional overload. Luethi and Nigg (1985) studied the influence of the shoe construction in tennis and found that the type of outsole did affect the COF.
Their results conflicted with Van Gheluwe and Deporte (1992) by showing that the COF changed depending on the traction characteristics of the outsole.
Due to the findings that surface type is a factor in the frictional properties, it is
important to understand the characteristics of the Rebound Ace™ surface. Rebound
17 Ace™ is not considered a fluid surface, but it is claimed to be a surface for all types of
play (The Grand Slam tennis surface, n.d.). The surface is designed to reduce loads on
the body through the cushioned system, which reduces player injury and fatigue (The
Grand Slam tennis surface, n.d.). However, since the surface is covered in an acrylic
(polymer) topcoat, the properties of the surface seem to change as the temperature changes. The claims are that the surface becomes sticky, thus making the interface between the shoe and court more adhesive. This may make the surface more hazardous in certain environments, but there has been no research conducted on this topic to support the claims. The cushioned system has addressed the loading during running or jumping, but has this in turn caused the friction level to increase when heated and therefore increased the chance of ligament injuries by players’ feet sticking to the surface?
18 Chapter 3
METHODS
Equipment, Instrumentation, and Procedure
All the methods involved in this study were conducted under the same
environmental conditions as the temperature of the room was kept at 72 degrees F. There
was a total of three days in the study and eight different temperatures for each run. A complete run of the trials was conducted in one day. At a given temperature, each shoe was pulled in four directions with two replicates.
Shoes
Four women’s tennis shoes were the subjects for this study; each was a different
major tennis shoe brand. All shoes were a United States size 6 (European size 36.5),
which was chosen because it would fit well on the testing surface while also allowing
ample space for the shoe to slide. The tread for each shoe was a variation of the
Herringbone pattern, which is placed on certain areas of the sole to allow for maximum traction during side-to-side and front-to-back motions (How to buy a tennis shoe, n.d.).
Table 3.1 shows the characteristics of each shoe, and the shoes and soles of the shoes can
be seen in Figures 3.1 through 3.8.
Test Surface
The test surface used in the study was a 45.7 cm by 44.7 cm sample of Rebound
Ace™ court surface (Ace Surfaces North America, Inc., 800-678-9223). The sample 19 Table 3.1 Shoe Characteristics.
Length Mass Subject Brand Model Size (cm) (kg) Tread Pattern
1 Adidas Barricade US 6 25.5 0.9 Herringbone (EU 36.5)
2 New 650 US 6 25.8 0.9 Herringbone Balance (EU 36.5)
3 Nike Air Zoom US 6 25.3 0.8 Herringbone Thrive (EU 36.5)
4 Wilson DST US 6 25.9 0.8 Herringbone (EU 36.5)
Mean US 6 25.625 0.85 (EU 36.5)
20
Figure 3.1. Side View of Adidas Barricade.
Figure 3.2. Sole View of Adidas Barricade. 21
Figure 3.3. Side View of New Balance 650.
Figure 3.4. Sole View of New Balance 650. 22
Figure 3.5. Side View of Nike Air Zoom Thrive.
Figure 3.6. Sole View of Nike Air Zoom Thrive.
23
Figure 3.7. Side View of Wilson DST.
Figure 3.8. Sole View of Wilson DST.
24 consisted of a piece of compressed wood with layers of adhesive, rubber shock pad (0.7
cm), sealant, reinforcement material, and filler on top of the wood. This was covered by
two layers of the Rebound Ace™ acrylic surface. The thickness of the test surface was
0.8 cm (not including the wood) and is shown in Figure 3.9.
Heating Apparatus
Two Philips BR40 (12.7 cm diameter) clear infrared heat lamps with aluminum reflectors were used to heat the test surface (Figure 3.10). One lamp was 125 W while the
other was 250 W to obtain varying amounts of heat to the surface as both lamps were not
always on at the same time. The distance between the lamps and the surface was also
adjusted depending on the desired surface temperature. An Acu-Rite 00890 wired digital
indoor-outdoor thermometer (Figure 3.11) was used to measure the temperature of the
room and surface. The room temperature was measured through the indoor sensor
(located on the top right of the thermometer) while the surface temperature was measured
with the wired outdoor sensor. A piece of tape was used to ensure the outdoor sensor stayed on the testing surface. The surface was constantly measured over the full testing
area throughout the testing to ensure the pull area remained at the desired temperature.
The complete set-up is seen in Figure 3.12. The surface was heated to 85, 100, 110, 115,
130, 140, and 155 degrees F (29.5, 37.8, 43.3, 46.1, 54.4, 60.0, and 65.6 degrees C) for
the study. A test was also performed at room temperature, 72 degrees F (22.2 degrees
C).The temperature range was selected to correspond to the normal average January
temperature in Melbourne (79 degrees F) as well as the maximum court temperature
experienced during the Australian Open (140 degrees F).
25
Figure 3.9. Rebound Ace™ Tennis Court Test Surface.
Figure 3.10. Philips BR40 Infrared Heat Lamps.
26
Figure 3.11. Acu-Rite 00890 Wired Digital Indoor-Outdoor Thermometer.
Figure 3.12. Test Surface Set-up.
27 Static Coefficient of Friction Measurement
A Model FDX Force Ten horizontal pull force gauge (Wagner Instruments, 800-
345-4188) was used to find the static COF between the Rebound Ace™ surface and each
of the shoes (Figure 3.13). The 250 x 0.1 N (50 x 0.02 lbs) cell was attached to the unit
for the testing and can be seen in Figure 3.14. The fundamental idea of ASTM Standard
C 1028 (1996) was used as a basis for the friction testing. The changes made were a lower amount of weight used due to the maximum weight able to stay on the shoe while pulling, the entire shoe was used in the pull instead of a sample of the sole attached to a sled, and an additional pull direction was added in the present study. Before each test, the surface was brushed with a cloth to ensure there were no contaminants and the force
gauge was reset to zero. A shoe was placed on the surface with 66.7 N (15 lbs) of weights
placed on top of the shoe. This set-up can be seen in Figure 3.15. The weights were
wrapped in a cloth to keep the weights from sliding off the shoe while being pulled. To
pull the shoes, the hook of the force gauge unit was attached to the top of the heel pad.
The shoe with weights was then pulled until the shoe started to slide. The maximum
reading on the dynamometer was the maximum friction force, and the normal force was
the total weight of the shoe and additional weights. The COF was then found for each pull.
Experimental Design
The static COF is the dependent variable for this study while the temperature,
shoe brand, and pull direction are the independent variables. There were a total of eight
temperatures, four shoe brands, and four pull directions tested. The study was not fully
28
Figure 3.13. Force Ten Horizontal Pull Force Gauge.
Figure 3.14. Force Ten Horizontal Pull Force Gauge with 222 ± 0.9 N Cell Attached.
29
Figure 3.15. Shoe and Weight Set-up.
randomized in that each shoe was tested at a given temperature before testing at the next
temperature. To simulate how the surface would heat up during the day, the temperatures
were increased throughout the study (from 72 to 155 degrees F) instead of randomized.
The order combination of the shoe and pull direction were completely randomized for each temperature.
Each of the four shoes was pulled in all four directions (toward the front, to the
left, diagonal to the front left, and diagonal to the front right) with two replicates per day.
This resulted in 24 pulls for every shoe at each of the eight temperatures. Over the three
days of testing, there was a total sample size of 768 (8 temperatures*4 shoes*4 pull directions*2 replicates*3 days). Table 3.2 shows the combinations of surface temperature and pull direction for one replicate of one shoe. This sample size was shown to be adequate using the maximum differences between means (Table A.1., Appendix).
30 Table 3.2. Combinations of Surface Temperature and Pull Direction for One Shoe.
Surface Temperature (F) Pull Direction 72 Front 72 Left 72 Diagonal Front Left 72 Diagonal Front Right 85 Front 85 Left 85 Diagonal Front Left 85 Diagonal Front Right 100 Front 100 Left 100 Diagonal Front Left 100 Diagonal Front Right 110 Front 110 Left 110 Diagonal Front Left 110 Diagonal Front Right 115 Front 115 Left 115 Diagonal Front Left 115 Diagonal Front Right 130 Front 130 Left 130 Diagonal Front Left 130 Diagonal Front Right 140 Front 140 Left 140 Diagonal Front Left 140 Diagonal Front Right 155 Front 155 Left 155 Diagonal Front Left 155 Diagonal Front Right
31 A three-way repeated measures ANOVA was performed to find the significance
of the independent variables (temperature, shoe, and pull direction) and the interactions between them. A residual versus order plot was also examined to determine the repeatability of the measurements over the three days. Tukey’s method of testing pairwise mean-comparisons was performed to find any significant differences between shoe brands at each temperature and temperatures for each shoe brand.
32 Chapter 4
RESULTS
All the data was analyzed using Minitab 14.0 statistical software, and a 5% level
of significance was used. A check of the normality assumption indicated that the overall
data was not normally distributed (Figure A.1 in the Appendix). Then, a three-way repeated measures ANOVA model was found for the dependent variable COF with
Temperature, Shoe, and Pull Direction as the independent variables. The interactions between Temperature and Shoe, Temperature and Pull Directions, and Shoe and Pull
Direction were also examined.
Three-way Repeated Measures ANOVA Model
Table 4.1 shows the results of the three-way repeated measures ANOVA for the
study. The repeatability over days was examined by plotting the order of the data, and it
was observed in Figure 4.1 that there were no differences between days due to the
randomness of the plot.
The ANOVA Model revealed a significant difference in COF with a change in temperature (p = 0.000) as the mean COF varied from 1.45 ± 0.008 to 1.68 ± 0.002
(Table 4.2). There was a noticeable decreasing mean COF (1.65 ± 0.003 to 1.45 ± 0.008) from 115 to 155 degrees F and overall lowest mean COF at 155 degrees F (Table 4.2 and
Figure 4.2). The minimum and maximum COF values follow the same trend with the exception of the maximum COF value at 130 degrees F.
33 0.15
0.10
0.05
0.00 Residual -0.05
-0.10
-0.15 1 20015010050 250 550500450400350300 600 750700650 Observation Order
Figure 4.1. Residuals Versus the Order of the Data: Response is COF.
Table 4.1. ANOVA for COF.
Source DF Seq. SS Adj. SS Adj. MS F P
Temperature 7 4.26780 4.26780 0.60969 293.81 0.000 Shoe 3 3.35488 3.35488 1.11829 538.91 0.000 Pull Direction 3 0.00217 0.00217 0.00072 0.35 0.790 Temp*Shoe 21 0.80363 0.80363 0.03827 18.44 0.000 Temp*Pull 21 0.04972 0.04972 0.00237 1.14 0.299 Shoe*Pull 9 0.01400 0.01400 0.00156 0.75 0.664 Error 703 1.45879 1.45879 0.00208 Total 767 9.95100
34 Table 4.2. COF Values for Temperature.
Standard Temperature (F) Low High Mean Deviation
72 1.54 1.83 1.68 0.0016 85 1.48 1.76 1.64 0.0038 100 1.49 1.81 1.67 0.0023 110 1.36 1.69 1.55 0.0017 115 1.46 1.77 1.65 0.0030 130 1.42 1.80 1.62 0.0048 140 1.30 1.76 1.56 0.0146 155 1.25 1.66 1.45 0.0083
1.8
1.75
1.7
1.65
1.6
1.55
1.5 Mean COF Mean
1.45
1.4
1.35
1.3 70 72 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150 155 Temperature (F)
Figure 4.2. Mean COF for the Eight Temperature Conditions.
35 A significant difference in COF was also present for the shoe brand (p = 0.000).
The Nike brand had the lowest mean COF at 1.49 ± 0.004 while the Wilson shoe had the
highest mean COF at 1.67 ± 0.006 (Table 4.3 and Figure 4.3). The Nike brand also had
the lowest minimum and maximum COF values. While the highest COF value of 1.83
was recorded for the Adidas shoe. The pull direction was found not to be significant (p =
0.790).
The interaction between the temperature and shoe proved to be statistically
significant (p = 0.000), while the interactions between temperature and pull direction and
shoe and pull direction were found to be insignificant (p = 0.229 and p = 0.664,
respectively).
The mean of COF for each shoe was examined at each temperature since the
interaction between temperature and shoe was found to be significant (Tables 4.4 through
4.7 and Figures 4.4 through 4.7). For all four shoe brands, a dip in COF occurs at 110
degrees F. The COF then rises at 115 degrees F and follows a decreasing trend from 115
to 155 degrees F. The pattern for the COF over the temperature range is the same for
Adidas Barricade and Nike Air Zoom Thrive. New Balance 650 and Wilson DST share a
COF trend that is slightly different from the other two shoe brands. The COF at the
highest temperature (155 degrees F) is the lowest for all shoe brands.
Although the Nike brand shows the lowest overall COF values, it has one of the greatest variances at all temperatures except 130 and 140 degrees F. On the other hand,
the Wilson brand has the highest overall COF, but one of the smallest variances at all temperatures except 72, 140, and 155 degrees F.
36 Table 4.3. COF Values for Shoe.
Standard Shoe Low High Mean Deviation
Adidas 1.35 1.83 1.60 0.0008 New Balance 1.32 1.81 1.64 0.0054 Nike 1.25 1.67 1.49 0.0044 Wilson 1.39 1.80 1.67 0.0057
1.8
1.75
1.7
1.65
1.6
1.55
1.5 Mean of COF 1.45
1.4
1.35
1.3 Adidas New Balance Nike Wilson Shoe
Figure 4.3. Mean COF for the Four Shoe Brands.
37 Table 4.4. COF Values for Adidas Barricade.
Standard Temperature (F) Low High Mean Deviation
72 1.73 1.83 1.77 0.0004 85 1.58 1.73 1.66 0.0036 100 1.59 1.75 1.67 0.0077 110 1.48 1.69 1.57 0.0028 115 1.60 1.67 1.64 0.0016 130 1.50 1.69 1.59 0.0129 140 1.42 1.59 1.51 0.0134 155 1.35 1.49 1.44 0.0098
1.8
1.75
1.7
1.65
1.6
1.55
1.5 Mean of COF 1.45
1.4
1.35
1.3 70 72 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150 155 Temperature (F)
Figure 4.4. Mean COF for Adidas Barricade.
38 Table 4.5. COF Values for New Balance 650.
Standard Temperature (F) Low High Mean Deviation
72 1.60 1.72 1.66 0.0010 85 1.60 1.71 1.65 0.0048 100 1.62 1.81 1.71 0.0185 110 1.57 1.67 1.63 0.0141 115 1.64 1.76 1.69 0.0061 130 1.59 1.74 1.66 0.0111 140 1.55 1.70 1.64 0.0224 155 1.32 1.59 1.44 0.0019
1.8
1.75
1.7
1.65
1.6
1.55
1.5 Mean of COF Mean 1.45
1.4
1.35
1.3 70 72 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150 155 Temperature (F)
Figure 4.5. Mean COF for New Balance 650.
39 Table 4.6. COF Values for Nike Air Zoom Thrive.
Standard Temperature (F) Low High Mean Deviation
72 1.54 1.64 1.59 0.0082 85 1.48 1.57 1.54 0.0076 100 1.49 1.67 1.58 0.0146 110 1.36 1.49 1.41 0.0179 115 1.46 1.58 1.54 0.0040 130 1.42 1.55 1.49 0.0079 140 1.30 1.51 1.42 0.0039 155 1.25 1.48 1.38 0.0110
1.8
1.75
1.7
1.65
1.6
1.55
1.5 Mean of COF 1.45
1.4
1.35
1.3 70 72 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150 155 Temperature (F)
Figure 4.6. Mean COF for Nike Air Zoom Thrive.
40 Table 4.7. COF Values for Wilson DST.
Standard Temperature (F) Low High Mean Deviation
72 1.65 1.76 1.71 0.0024 85 1.67 1.76 1.71 0.0016 100 1.63 1.80 1.73 0.0101 110 1.49 1.64 1.57 0.0080 115 1.67 1.77 1.74 0.0038 130 1.65 1.80 1.73 0.0106 140 1.51 1.76 1.65 0.0212 155 1.39 1.66 1.53 0.0230
1.8
1.75
1.7
1.65
1.6
1.55
1.5 Mean of COF Mean 1.45
1.4
1.35
1.3 70 72 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150 155 Temperature (F)
Figure 4.7. Mean COF for Wilson DST.
41 Normality tests were conducted on each shoe to examine visually any differences
in the data (Figures A.2 through A.5 in the Appendix). The only shoe to follow a normal
distribution according to the Anderson-Darling test was the Adidas Barricade. As before,
the Adidas showed a trend similar to the Nike shoe while the New Balance was similar to
the Wilson shoe.
Normality tests were also conducted for each temperature to examine the data for
any differences among shoes (Figures A.6 through A.13 in the Appendix). It was shown
that the data for 72 and 155 degrees F were distributed normally according to the
Anderson-Darling test. The COF for temperatures of 100, 130, and 140 degrees F were
borderline normal with p-values of 0.036, 0.015, and 0.012, respectively.
Temperature Comparisons
The results of ANOVA indicated that there was a significant difference in COF over temperature. To determine further which temperatures may be similar or dissimilar,
Tukey’s method of testing pairwise mean-comparisons was used. The COF means for each temperature were compared for each shoe brand and the results can be seen in
Tables 4.8 through 4.11. There was no significant difference in four pairings for Adidas
Barricade, 14 pairings for New Balance 650, three pairings for Nike Air Zoom Thrive, and 11 pairings for Wilson DST at confidence level of 95%. For a confidence level of
99%, there was no significant difference in five pairings for Adidas, 17 pairings for New
Balance, seven pairings for Nike, and 11 pairings for Wilson.
42 Table 4.8. Tukey’s Test for Adidas Barricade.
Absolute Difference Treatment Pair in Means T0.05 = 0.0395 T0.01 = 0.0472
72 F vs. 85 F 0.1142 * ** 72 F vs. 100 F 0.1071 * ** 72 F vs. 110 F 0.2050 * ** 72 F vs. 115 F 0.1362 * ** 72 F vs. 130 F 0.1808 * ** 72 F vs. 140 F 0.2625 * ** 72 F vs. 155 F 0.3325 * ** 85 F vs. 100 F 0.0071 NS NS* 85 F vs. 110 F 0.0908 * ** 85 F vs. 115 F 0.0221 NS NS* 85 F vs. 130 F 0.0667 * ** 85 F vs. 140 F 0.1483 * ** 85 F vs. 155 F 0.2183 * ** 100 F vs. 110 F 0.0979 * ** 100 F vs. 115 F 0.0292 NS NS* 100 F vs. 130 F 0.0738 * ** 100 F vs. 140 F 0.1554 * ** 100 F vs. 155 F 0.2254 * ** 110 F vs. 115 F 0.0687 * ** 110 F vs. 130 F 0.0242 NS NS* 110 F vs. 140 F 0.0575 * ** 110 F vs. 155 F 0.1275 * ** 115 F vs. 130 F 0.0446 * NS* 115 F vs. 140 F 0.1263 * ** 115 F vs. 155 F 0.1962 * ** 130 F vs. 140 F 0.0817 * ** 130 F vs. 155 F 0.1517 * ** 140 F vs. 155 F 0.0700 * ** * - significant difference, NS – no significant difference at α = 0.05 ** - significant difference, NS* – no significant difference at α = 0.01
43 Table 4.9. Tukey’s Test for New Balance 650.
Absolute Difference Treatment Pair in Means T0.05 = 0.0385 T0.01 = 0.0461
72 F vs. 85 F 0.0108 NS NS* 72 F vs. 100 F 0.0417 * NS* 72 F vs. 110 F 0.0287 NS NS* 72 F vs. 115 F 0.0242 NS NS* 72 F vs. 130 F 0.0017 NS NS* 72 F vs. 140 F 0.0192 NS NS* 72 F vs. 155 F 0.2179 * ** 85 F vs. 100 F 0.0525 * ** 85 F vs. 110 F 0.0179 NS NS* 85 F vs. 115 F 0.0350 NS NS* 85 F vs. 130 F 0.0125 NS NS* 85 F vs. 140 F 0.0083 NS NS* 85 F vs. 155 F 0.2071 * ** 100 F vs. 110 F 0.0704 * ** 100 F vs. 115 F 0.0175 NS NS* 100 F vs. 130 F 0.0400 * NS* 100 F vs. 140 F 0.0608 * ** 100 F vs. 155 F 0.2596 * ** 110 F vs. 115 F 0.0529 * ** 110 F vs. 130 F 0.0304 NS NS* 110 F vs. 140 F 0.0096 NS NS* 110 F vs. 155 F 0.1892 * ** 115 F vs. 130 F 0.0225 NS NS* 115 F vs. 140 F 0.0433 * NS* 115 F vs. 155 F 0.2421 * ** 130 F vs. 140 F 0.0208 NS NS* 130 F vs. 155 F 0.2196 * ** 140 F vs. 155 F 0.1987 * ** * - significant difference, NS – no significant difference at α = 0.05 ** - significant difference, NS* – no significant difference at α = 0.01
44 Table 4.10. Tukey’s Test for Nike Air Zoom Thrive.
Absolute Difference Treatment Pair in Means T0.05 = 0.0378 T0.01 = 0.0452
72 F vs. 85 F 0.0533 * ** 72 F vs. 100 F 0.0083 NS NS* 72 F vs. 110 F 0.1758 * ** 72 F vs. 115 F 0.0537 * ** 72 F vs. 130 F 0.0987 * ** 72 F vs. 140 F 0.1700 * ** 72 F vs. 155 F 0.2150 * ** 85 F vs. 100 F 0.0450 * NS* 85 F vs. 110 F 0.1225 * ** 85 F vs. 115 F 0.0004 NS NS* 85 F vs. 130 F 0.0454 * ** 85 F vs. 140 F 0.1167 * ** 85 F vs. 155 F 0.1617 * ** 100 F vs. 110 F 0.1675 * ** 100 F vs. 115 F 0.0454 * ** 100 F vs. 130 F 0.0904 * ** 100 F vs. 140 F 0.1617 * ** 100 F vs. 155 F 0.2067 * ** 110 F vs. 115 F 0.1221 * ** 110 F vs. 130 F 0.0771 * ** 110 F vs. 140 F 0.0058 NS NS* 110 F vs. 155 F 0.0392 * NS* 115 F vs. 130 F 0.0450 * NS* 115 F vs. 140 F 0.1163 * ** 115 F vs. 155 F 0.1612 * ** 130 F vs. 140 F 0.0713 * ** 130 F vs. 155 F 0.1162 * ** 140 F vs. 155 F 0.0450 * NS* * - significant difference, NS – no significant difference at α = 0.05 ** - significant difference, NS* – no significant difference at α = 0.01
45 Table 4.11. Tukey’s Test for Wilson DST.
Absolute Difference Treatment Pair in Means T0.05 = 0.0389 T0.01 = 0.0466
72 F vs. 85 F 0.0000 NS NS* 72 F vs. 100 F 0.0221 NS NS* 72 F vs. 110 F 0.1417 * ** 72 F vs. 115 F 0.0275 NS NS* 72 F vs. 130 F 0.0200 NS NS* 72 F vs. 140 F 0.0567 * ** 72 F vs. 155 F 0.1804 * ** 85 F vs. 100 F 0.0221 NS NS* 85 F vs. 110 F 0.1417 * ** 85 F vs. 115 F 0.0275 NS NS* 85 F vs. 130 F 0.0200 NS NS* 85 F vs. 140 F 0.0567 * ** 85 F vs. 155 F 0.1804 * ** 100 F vs. 110 F 0.1638 * ** 100 F vs. 115 F 0.0054 NS NS* 100 F vs. 130 F 0.0021 NS NS* 100 F vs. 140 F 0.0788 * ** 100 F vs. 155 F 0.2025 * ** 110 F vs. 115 F 0.1692 * ** 110 F vs. 130 F 0.1617 * ** 110 F vs. 140 F 0.0850 * ** 110 F vs. 155 F 0.0388 NS NS* 115 F vs. 130 F 0.0075 NS NS* 115 F vs. 140 F 0.0842 * ** 115 F vs. 155 F 0.2079 * ** 130 F vs. 140 F 0.0767 * ** 130 F vs. 155 F 0.2004 * ** 140 F vs. 155 F 0.1237 * ** * - significant difference, NS – no significant difference at α = 0.05 ** - significant difference, NS* – no significant difference at α = 0.01
46 Shoe Comparisons
Tukey’s method was then used at each temperature to determine any significant difference in the means between shoe pairs since the ANOVA showed a significant difference in the COF. Tables 4.12 though 4.19 show these comparisons. The only non- significant differences at a confidence level of 95% are Adidas and New Balance at 85 and 155 degrees F, Adidas and Wilson at 110 degrees F, and New Balance and Wilson at
100 and 140 degrees F. The only additional non-significant difference at a confidence level of 99% was between Adidas and New Balance at 100 degrees F.
Table 4.12. Tukey’s Test for 72 degrees F.
Absolute Difference in Treatment Pair Means T0.05 = 0.0237 T0.01 = 0.0298
Adidas vs. New Balance 0.1100 * ** Adidas vs. Nike 0.1821 * ** Adidas vs. Wilson 0.0642 * ** New Balance vs. Nike 0.0721 * ** New Balance vs. Wilson 0.0458 * ** Nike vs. Wilson 0.1179 * **
* - significant difference at α = 0.05 ** - significant difference at α = 0.01
47
Table 4.13. Tukey’s Test for 85 degrees F.
Absolute Difference in Treatment Pair Means T0.05 = 0.0244 T0.01 = 0.0307
Adidas vs. New Balance 0.0067 NS NS* Adidas vs. Nike 0.1213 * ** Adidas vs. Wilson 0.0500 * ** New Balance vs. Nike 0.1146 * ** New Balance vs. Wilson 0.0567 * ** Nike vs. Wilson 0.1712 * **
* - significant difference, NS – no significant difference at α = 0.05 ** - significant difference, NS* – no significant difference at α = 0.01
Table 4.14. Tukey’s Test for 100 degrees F.
Absolute Difference in Treatment Pair Means T0.05 = 0.0373 T0.01 = 0.0468
Adidas vs. New Balance 0.0388 * NS* Adidas vs. Nike 0.0833 * ** Adidas vs. Wilson 0.0650 * ** New Balance vs. Nike 0.1221 * ** New Balance vs. Wilson 0.0263 NS NS* Nike vs. Wilson 0.1483 * **
* - significant difference, NS – no significant difference at α = 0.05 ** - significant difference, NS* – no significant difference at α = 0.01
48 Table 4.15. Tukey’s Test for 110 degrees F.
Absolute Difference in Treatment Pair Means T0.05 = 0.0296 T0.01 = 0.0372
Adidas vs. New Balance 0.0662 * ** Adidas vs. Nike 0.1529 * ** Adidas vs. Wilson 0.0008 NS NS* New Balance vs. Nike 0.2192 * ** New Balance vs. Wilson 0.0671 * ** Nike vs. Wilson 0.1521 * **
* - significant difference, NS – no significant difference at α = 0.05 ** - significant difference, NS* – no significant difference at α = 0.01
Table 4.16. Tukey’s Test for 115 degrees F.
Absolute Difference in Treatment Pair Means T0.05 = 0.0231 T0.01 = 0.0290
Adidas vs. New Balance 0.0504 * ** Adidas vs. Nike 0.0996 * ** Adidas vs. Wilson 0.0996 * ** New Balance vs. Nike 0.1500 * ** New Balance vs. Wilson 0.0492 * ** Nike vs. Wilson 0.1992 * **
* - significant difference at α = 0.05 ** - significant difference at α = 0.01
49 Table 4.17. Tukey’s Test for 130 degrees F.
Absolute Difference in Treatment Pair Means T0.05 = 0.0353 T0.01 = 0.0443
Adidas vs. New Balance 0.0725 * ** Adidas vs. Nike 0.1000 * ** Adidas vs. Wilson 0.1367 * ** New Balance vs. Nike 0.1725 * ** New Balance vs. Wilson 0.0642 * ** Nike vs. Wilson 0.2367 * **
* - significant difference at α = 0.05 ** - significant difference at α = 0.01
Table 4.18. Tukey’s Test for 140 degrees F.
Absolute Difference in Treatment Pair Means T0.05 = 0.0431 T0.01 = 0.0541
Adidas vs. New Balance 0.1333 * ** Adidas vs. Nike 0.0896 * ** Adidas vs. Wilson 0.1417 * ** New Balance vs. Nike 0.2229 * ** New Balance vs. Wilson 0.0083 NS NS* Nike vs. Wilson 0.2313 * **
* - significant difference, NS – no significant difference at α = 0.05 ** - significant difference, NS* – no significant difference at α = 0.01
50 Table 4.19. Tukey’s Test for 155 degrees F.
Absolute Difference in Treatment Pair Means T0.05 = 0.0491 T0.01 = 0.0616
Adidas vs. New Balance 0.0046 NS NS* Adidas vs. Nike 0.0646 * ** Adidas vs. Wilson 0.0879 * ** New Balance vs. Nike 0.0692 * ** New Balance vs. Wilson 0.0833 * ** Nike vs. Wilson 0.1525 * **
* - significant difference, NS – no significant difference at α = 0.05 ** - significant difference, NS* – no significant difference at α = 0.01
51 Chapter 5
DISCUSSION
There is a debate about the safety of the Rebound Ace™ tennis court surface at
high temperatures. This study was conducted to gather important knowledge about the
COF between the surface and shoe as the surface temperature increases, thus determining if the surface is less safe at higher temperatures.
The average combined COF for all shoes across all temperatures was 1.60 ±
0.003. This value corresponds to COF values for sport surfaces found in various studies.
Winterbottom’s (1985) study on artificial turf found the COF to be 1.50 to 2.20. McNitt and Petrunak (2006) had similar results of 1.17 to 2.43 when examining different synthetic turf surfaces. Baker (1990) conducted a study on natural turf and found the
COF to be between 1.38 and 1.96.
Previous studies indicate a safe COF in active sports should fall within a range of
1.1 to 2.0. COF values from the present study fall within this range indicating Rebound
Ace™ is safe at all temperatures, but the limits on the COF were not addressed in any of
the previous studies. Barrett (1956) suggested a minimum COF of 1.1 was required for
running while Van Gheluwe, Deporte, and Hebbelink (1983) found a COF of 1.2 gave enough traction for all maneuvers in soccer and a COF below 1.0 caused slips and falls.
Player evaluations performed by Bell and Holmes (1988) established that the minimum
preferred COF was 1.37. No studies have set a maximum COF, but the AstroTurf researchers at Southwest Recreational Industries (Player traction on synthetic turf, n.d.) believe that values over 2.0 do not add to the performance of the game and may increase 52 the exposure to injury. If the values from these studies are taken as the safe limit, the
COF values in the current study indicate that the Rebound Ace™ surface is safe at all
temperatures.
Effect of Temperature on the COF
Statistical analysis revealed that temperature was a significant factor in the COF
of the Rebound Ace™ surface, thus failing to reject the hypothesis that there is a
significant difference in the COF with a change in temperature. This finding indicates
that the surface properties of Rebound Ace™ may change with the temperature. Ideally,
the COF is dependent on the material of the two contact surfaces and independent of the
weight and surface area (Nigg, 1990). Therefore, if the surfaces are kept constant, there
must be another factor that causes the COF to change. From the results of this study,
temperature appears to be a factor that affects the COF of the Rebound Ace™ surface.
However, the results show that the highest temperature (155 degrees F) actually had the lowest COF at 1.45 ± 0.008 while the COF at room temperature (the lowest temperature) was the highest at 1.68 ± 0.002. In addition, the second and third highest temperatures (140 and 130 degrees F) were the third and fourth lowest COF values, respectively. Further, the study shows that the COF at the average maximum temperature of Melbourne in January (79 degrees F, estimated by the results at 85 degrees F) is higher than the COF at the maximum court temperature at the Australian Open (140 degrees F).
Another interesting result is the drop in the overall mean COF at 85 and 110 degrees F.
Without this occurrence, the COF would continually decrease as the temperature increased thus rejecting the hypothesis that the COF will increase as the temperature
53 increases. These findings contradict the research by Torg, Stilwell, and Rogers (1996)
that found the release coefficient of artificial football turf increased with the temperature.
However, their research examined rotational friction and Nigg and Yeadon (1987) stated
that transitional and rotational COF’s usually show different results.
The mainly decreasing trend in the COF as the surface temperature increases may
indicate the surface properties change as the surface gets warmer. One possibility is that
the surface liquefies at higher temperatures. This would cause the surface to then behave
more like a fluid surface, such as sand, clay, or gravel, that has been shown by Bastholt
(2000), Briner and Ely (1999), Bjordal et al. (1997), and Nigg and Segesser (1988) to
have a lower COF. The drop in the COF at higher temperatures leads one to believe that
the surface is actually safer at higher temperatures. This may also suggest that any safety
issues related to the Rebound Ace™ surface are contributed to other properties of the
materials.
One issue is the possibility of the court surface becoming sticky as hot tar does when the temperature rises. This occurrence may actually cause the shoe to adhere or stick to the surface. This would alter the frictional properties of the surface-shoe interface
as the interface may become adhesive, and Coulomb’s law of friction would no longer be
a valid approximation (Nigg & Segesser, 1988). This assumption is supported by Brown
(1987) as he described how the classical friction laws usually do not hold true for sports shoe-artificial surface interfaces. Instead, the COF may also be dependent on the area of contact (Friction, 2006). The magnitude of the normal force may also become an issue since the greater the normal force, the more the two surfaces would adhere to each other.
Nigg (1987) and Torg, Quedenfeld, and Landau (1974) found through their respective
54 studies on football cleats that the COF increased as the normal force increased. Thus,
Nigg (1990) has suggested the normal force has an influence on sports shoe-artificial
surface interfaces. The physical interface between the shoe and surface is one suggested
reason why the surface area and normal force can influence the COF during sport
activities (Brown, 1987).
Effect of Shoe on the COF
Statistical analysis also showed that shoe is a significant factor in the shoe-surface
COF. This finding rejects the hypothesis that there is no significant difference in the COF
among the different shoe brands and indicates that certain shoe soles may be safer on
Rebound Ace™ than others. Although some may believe shoes are the same, previous
studies support the results of this study on the significance of the shoe in the COF. At first full contact of the foot, Van Gheluwe and Deporte (1992) observed a significant difference in the COF among tennis shoes. Studies on football shoes and different turf surfaces also indicate that the shoe is an important factor in the COF. Lambson, Barnhill, and Higgins (1996) suggested that the cleat design on football cleats have a significant influence on the COF and risk for serious knee injury. Torg, Stilwell, and Rogers (1996) found that the release coefficient of football shoes on artificial turf varied with cleat design and sole material. The findings of Scranton et al. (1997) indicate that the type of shoe is an important factor in noncontact ACL injuries that are believed to be linked to the COF.
Three of the shoes, Adidas Barricade, New Balance 650, and Wilson DST, had an
average COF across all temperatures of above 1.60 (1.60 ± 0.0008, 1.64 ± 0.005, and
55 1.67 ± 0.006, respectively). While these values are within the range found in earlier
studies (Baker, 1990; McNitt & Petrunak, 2006; Winterbottom, 1985), the Nike Air
Zoom Thrive had a significantly lower average COF of 1.49 ± 0.004. This may lead one
to believe that the Nike is the best shoe for play on Rebound Ace™ since its COF is
closer to the preferred value of 1.37 stated by Bell and Holmes (1988). The difference in
the COF for the Nike shoe may be related to the sole material or the distribution of the
Herringbone pattern. Since this study did not examine the individual shoe soles in terms
of material or pattern distribution, no definite conclusions can be made on these factors.
However, the studies of Lambson, Barnhill, and Higgins (1996), Torg, Stilwell, and
Rogers (1996), and Van Gheluwe and Deporte (1992) lead one to believe these factors are critical in the COF.
Temperature Comparisons
The significance of temperature on the COF was also evident through Tukey’s pairwise comparison of the means. Since shoe was also found to be a significant factor,
the comparisons were conducted for each shoe. The results indicate that the influence of
temperature may be different for each shoe. There was no significant difference in the
COF for the Adidas Barricade at 85, 100, and 115 degrees F. This indicates that at these
temperatures, the properties of the shoe-surface interface are similar. These temperatures
are in the normal temperature range for the Australian Open, which would imply that no
changes occur in the COF while playing under normal conditions. Results from the
Adidas also showed no difference in the COF at 110 and 130 degrees F for α = 0.05 and
0.01 and no difference at 115 and 130 for α = 0.01. No significant difference at these
56 temperatures points to the conclusion that Rebound Ace™ may change properties at higher temperatures, causing the COF to drop. The Nike Air Zoom Thrive showed
similar results with no differences found between 72 and 100 degrees F, 85 and 115
degrees F, and 110, 140, and 155 degrees F. The non-significant differences between the
lower and higher temperatures for the Nike indicate the shoe is safer than the others are
across all temperatures.
The COF for the New Balance 650 was almost constant across the temperatures
72 to 140 degrees F with a peak at 100 degrees F. The only main difference in the COF is
at 155 degrees F, which is 0.19 lower than the next lowest COF. Similar results are seen
for the Wilson DST shoe. The COF’s for the Wilson shoe are reasonably constant over all
temperatures except 110, 140 and 155 degrees F. No significant difference was found
between the COF at 110 and 155 degrees F, while these temperatures had a significant
difference between all other temperatures. The COF for 140 degrees F was significantly
different from all other temperatures. The consistency in the COF for the New Balance
and Wilson shoes indicate that there would be no noticeable difference in the COF over a
wide temperature range. This characteristic may be more beneficial to the player since
they would not have to adjust their play according to the surface temperature. Players
usually make adjustments to adapt to the playing environment, and any small change
could lead to over or under adjustment (Nigg, 1986). Consistency in the COF of the shoe-
surface interface would not force the player to change their play constantly as the
temperature increases.
57 Shoe Comparisons
Tukey’s pairwise comparisons on the mean COF for shoe brand supported the
findings that the shoe had a significant effect on the COF. The Nike Air Zoom Thrive
was significantly different from all other shoes at every temperature. As stated earlier, the
Nike shoe had both the lowest overall COF and COF at every temperature. The findings
from Tukey’s analysis further backs the conclusion that the Nike may be best suited for
playing on the Rebound Ace™ surface. Only five non-significant comparisons were
found among the other three shoes for α = 0.05 with one additional non-significant
comparison for α = 0.01. The pairs with a non-significant difference were spread across
five different temperatures, and each of the three shoes was paired with the other two at least once. Thus, no two shoes can be said to be the same. The rarity of non-significant differences among shoes stresses the findings that the shoe has a significant effect on
COF and supports the findings of Scranton et al. (1997) and Torg, Stilwell, and Rogers
(1996).
58 Chapter 6
CONCLUSIONS AND RECOMMENDATIONS
This study was conducted to gather important knowledge about how the COF
between the surface and shoe changes as the Rebound Ace™ surface temperature increases. The research indicated that:
1. Temperature was found to be a significant factor in the COF.
2. A significant difference was found in the COF among the different shoe brands.
3. The COF varied as the surface temperature of the Rebound Ace™ tennis court
increased.
4. The COF decreased over the higher temperatures (115 to 155 degrees F).
5. The lowest COF was found at the highest temperature of 155 degrees F.
6. The Nike Air Zoom Thrive had the lowest COF across all temperatures.
The results of this study indicate that the Rebound Ace™ tennis court surface may
not have a safety issue at higher temperatures within the studied temperature range.
However, there were limitations in the study with respect to sample size, normal load amount, measurement of rotational friction, possible adhesion and stiction effects, and measurements in the field that prevent this conclusion from be generalized for real-life
playing situations.
59 Recommendations for Future Research
Two easy modifications would improve upon the study by allowing for a larger sample size. The first is to include a greater number of shoes in the study. With duplicates of each brand of shoe, the effect of the shoe brand on the COF can be further explored by comparing the COF within each brand. This would also reduce the chance of wearing the sole during multiple slides. Other brands and styles should also be examined to include a better sample of the tennis shoe population worn by players. The second modification would be to include multiple samples of the Rebound Ace™ surface. This would ensure consistency of the measurements since the surface could have small variations that might affect the COF.
There are also adjustments that could be made to replicate real-life situations.
Some of these changes would be more difficult to incorporate, but would allow the results to be applied to actual playing conditions. One change would be to heat both the surface and shoe sole. This would replicate the actual interface better due to the fact that the shoe sole also heats due to contact with the heated surface. Another change would be to conduct the tests outdoors in the actual environment rather than indoors with artificial light sources. The reason this was not already done for this study was that the climatic conditions of the test site were not similar to the conditions of the site in question. To simulate further a real-life playing situation, actual players should be used in the testing.
This would require the use of a force plate covered in the surface material. The use of the force plate would not only allow for the actual forces between the foot and surface, but also account for differences in the normal forces among the athletes since the normal
60 force has been shown by Brown (1987) and Nigg (1990) to be a factor in the COF in
sport activities.
Rotational friction and dynamic COF measurements should be studied to analyze
all movement aspects of the game. This study was limited to static translational COF
measurements, but the game of tennis is a dynamic game played in all directions. By
examining both static and dynamic COF for the translational and rotational movements,
any possible conflicts between the types of motions could be recognized. Other frictional
properties, such as adhesion and stiction, should also be examined since the shoe-surface
interface does not appear to follow Coulomb’s law of friction. This addition would
include studying the possibility of the surface liquefying at higher temperatures as well as
the affect of sole material and pattern distribution on the COF. Compression of the
surfaces should also be examined to explore the ideas of adhesion and stiction. As the
two surfaces compress or flatten, the actual contact area changes, which might affect how
much the two surfaces adhere to each other and in turn affect the COF. The use of the
English XL Tribometer would help eliminate the issues of constant pull, adhesion, and
stiction.
Since the debate on the Rebound Ace™ surface relates to its safety at higher temperatures, an assessment of the forces through the body would help determine if the
friction forces created are great enough to cause damage to the ankle and knee. Due to the
lack of a consensus safe COF range for athletic surfaces, the COF values alone do not
confirm or deny the claim that Rebound Ace™ is unsafe at high temperatures. By
including an evaluation of the actual forces experienced by the athlete’s body, the COF
values found in this study can be used to help find an ideal safe COF range.
61
List of References
62 List of References
AFP (Agence France-Presse). (2006). Blake joins court critics. Retrieved July 21, 2006, from ABC Sport: Tennis, http://www.abc.net.au/sport/content/200601/ s1550599.htm
ASTM (American Society for Testing and Materials). (1996). Standard C 1028: Standard test method for determining the static coefficient of friction of ceramic tile and other like surfaces by the horizontal dynamometer pull-meter method. Annual Book of ASTM Standards. ASTM: West Conshohocken, PA.
Asperities. (n.d.). Retrieved November 7, 2006, from Nano-World, http://www.nano world.org/frictionmodule/content/0200makroreibung/0500asperities/?lang=en
Baker, S.W. (1990). Standards for the playing quality of artificial turf for Association Football. In R.C. Schmidt, E.F. Hoerner, E.M. Milner, & C.A. Morehouse (Eds.) Natural and Artificial Playing Fields: Characteristics and Safety Features, ASTM STP 1073 (pp. 48-57). Philadelphia, PA: The American Society for Testing and Materials.
Baker, S.W. (1991). Temporal variation of selected mechanical properties of natural turf football pitches. Journal of the Sports Turf Research Institute, 67, 53-65.
Barrett, G.F.C. (1956). Polymeric Surfaces for Sports and Recreation. Barking, Essex: Applied Science Publishers Ltd.
Bastholt, P. (2000). Professional tennis (ATP tour) and number of medical treatments in relation to type of surface. Medicine and Science in Tennis, 5(2). Retrieved August 23, 2006, from Society for Tennis Medicine and Science, http://www.stms.nl
Bell, M.J., Baker, S.W., & Canaway, P.M. (1985). Playing quality of sports surfaces: A review. Journal of the Sports Turf Research Institute, 61, 26-45.
Bell, M.J. & Holmes, M.J. (1988). The playing quality of Association Football pitches. Journal of the Sports Turf Research Institute, 64, 19-47.
Bhushan, B. (2003). Adhesion and stiction: Mechanisms, measurement techniques, and methods for reduction. The Journal of Vacuum Science and Technology B, 21(6), 2262-2296.
63 Bjordal, J.M., Arnly, F., Hannestad, B., & Strand, T. (1997). Epidemiology of anterior cruciate ligament injuries in soccer. American Journal of Sports Medicine, 25(3), 341-345.
Bowers, R. (2002). Between the lines: Australian retrospect 2002. Retrieved July 21, 2006, from Tennis Server, http://www.tennisserver.com/lines/lines_02_02_09. html
Bramwell, S.T., Requa, R.K., & Garrick, J.G. (1972). High school football injuries: A pilot comparison of playing surfaces. Medicine and Science in Sports, 4(3), 166- 169.
Briner, W. & Ely, C. (1999). Volleyball injuries at the 1995 United States Olympic Festival. International Journal of Volleyball Research, 1(1), 7-11.
Brown, R.P. (1987). Performance tests for artificial sports surfaces. Polymer Testing, 7(44), 279-292.
BSI (British Standards Institution). (1990). BS 7044: Artificial sports surfaces. Part 1: Classification and general introduction. Paramus, NJ: ILI.
Buczek, F.L. Jr., & Banks, S.A. (1996). High-Resolution Force Plate Analysis of Utilized Slip Resistance in Human Walking. Journal of Testing and Evaluation, 24, 353-358.
Canaway, P.M. (1979). Studies on turfgrass abrasion. Journal of the Sports Turf Research Institute, 55, 107-120.
Coefficient of friction. (2006). Retrieved July 23, 2006, from Wikipedia, http://en.wikipedia.org/wiki/Coefficient_of_friction
Courtney, T.K., Chang, W., Gronqvist, R., & Redfern, M.S. (2001). The measurement of slipperiness-an international scientific symposium. Ergonomics, 44(13), 1097- 1101.
Cross, R. (2004). Physics of ankle injuries. Medicine and Science in Tennis, 9(1). Retrieved August 23, 2006, from Society for Tennis Medicine and Science, http://www.stms.nl
D’Ambrosia, R.D. (1985). Orthotic devices in running injuries. Clinics in Sports Medicine, 4(4), 611-618.
Dixon, S.J., Batt, M.E., & Collop, A.C. (1999). Artificial playing surfaces research: A Review of medical, engineering, and biomedical aspects. International Journal of Sports Medicine, 20, 209-218.
64 Ekstrand, J. &. Nigg, B.M. (1989). Surface-related injuries in soccer. Sports Medicine, 8(1), 56-62.
Fast Car: Complete technical description. (2004). Retrieved July 23, 2006, from Oxford Dynamics, http://www.oxforddynamics.co.uk/complete_technical_description.htm
Friction. (2006). Retrieved July 23, 2006, from Wikipedia, http://en.wikipedia.org/wiki/Friction
Garrick, J.G. & LaVigne, A.B. (1972). Frictional properties of a synthetic turf. Journal of Safety Research, 4(1), 26-30.
The Grand Slam tennis surface. (n.d.). Retrieved July 22, 2006, from Ace Surfaces North America, http://www.reboundace.com/
Heidt, R.S. Jr., Dormer, S.G., Scranton, P.E. Jr, Losse, G., & Howard, M. (1996). Differences in friction and torsional resistance in athletic shoe-turf surface interfaces. American Journal of Sports Medicine, 24(6), 834-842.
How hot is hot? How safe is safe? (1998). Retrieved July 21, 2006, from The Zunis Foundation, http://www.zunis.org/how_hot_is_hot_how_safe_is_safe.htm
How to buy a tennis shoe. (n.d.). Retrieved September 3, 2002, from http://www.a1tennis.comtennis%20shoe.htm
Ichii, S. (1987). Relation of running injuries to surfaces and shoes. Japanese Journal of Sports Sciences, 6, 554-561.
Lambson, R.B., Barnhill, B.S., & Higgins, R.W. (1996). Football cleat design and its effect on anterior crucciate ligament injuries: A three-year prospective study. The American Journal of Sports Medicine, 24(2), 155-159.
Leclercq, S. (1999). The prevention of slipping accidents: a review and discussion of work related to the methodology of measuring slip resistance. Safety Science, 31(2), 95-125.
Luethi, S. & Nigg, B.M. (1985). The influence of different shoe constructions on discomfort and pain in tennis. In D.A. Winter, R.W. Norman, R.P. Wells, K.C. Hayes, & A.E. Patla (Eds.) Biomechanics IX-B (149-153). Champaign, IL: Human Kinetics Publishers.
McClements, I. & Baker, S.W. (1994). The playing quality of rugby pitches. Journal of the Sports Turf Research Institute, 70, 29-43.
65 McNitt, A.S. & Petrunak, D. (2006). Evaluation of playing surface characteristics of various in-filled systems. Retrieved August 3, 2006, from Penn State Department of Crop and Soil Sciences, http://cropsoil.psu.edu/mcnitt/infill.cfm
Milburn, P.D. & Barry, E.B. (1998). Shoe-surface interface and the reduction of injury in rugby union. Sports Medicine, 25(5), 319-327.
Milner, E. (1972). The shoe-turf interface. Paper presented to the Football Rules Committee of the National Collegiate Athletic Association, Sarasota, FL.
Morehouse, C.A. & Morrison, W.E. (1975). The artificial turf story: A research review. Penn State University HPER Series No 9.
Newton, R., Doan, B., Meese, M., Conroy, B., Black, K., Sebstianelli, W., & Kramer, W. (2002). Interaction of wrestling shoe and competition surface: Effects on coefficient of friction with implications for injury. Sports Biomechanics, 1(2), 157-166.
Nigg, B.M. (1986). Experimental techniques used in running shoe research. In B.M. Nigg (Ed.) Biomechanics of Running Shoes (27-61). Champaign, IL: Human Kinetics Publishers.
Nigg, B.M. (1987). Results of a test battery for artificial turf. Technical Report for the Toronto SkyDome Surfaces. Toronto, Canada.
Nigg, B.M. (1990). The validity and relevance of tests used for the assessment of sports surfaces. Medicine and Science in Sports and Exercise, 22(1), 131-139.
Nigg, B.M. & Segesser, B. (1988). The influence of playing surfaces on the load on the locomotor system and on football and tennis injuries. Sports Medicine, 5(6), 375- 385.
Nigg, B.M. & Segesser, B. (1992). Biomechanical and orthopedic concepts in sport shoe construction. Medicine and Science in Sports and Exercise, 24(5), 595-602.
Nigg, B.M. & Yeadon, M.R. (1987). Biomechanical aspects of playing surfaces. Journal of Sports Sciences, 5(2), 117-145.
Nirshl, R.P. & Sobel, N.P. (1994). Injuries in tennis. In P. Renström (Ed.) Clinical Practice of Sports Injury: Prevention and Care (460-474). Oxford, England: Blackwell Scientific.
Olsen, O.E., Myklebust, G., Engebretsen, L., Holme, I., & Bahr, R. (2003). Relationship between floor type and risk of ACL injury in team handball. Scandinavian Journal of Medicine & Science in Sports, 13(5), 299-304.
66 Orchard, J. (2001). The AFL penetrometer study: Work in progress. Journal of Science and Medicine in Sport, 4(2), 220-232.
Orchard, J. (2002). Is there a relationship between ground and climatic conditions and injuries in football? Sports Medicine, 32(7), 419-432.
Orchard, J., Seward, H., McGivern, J., & Hood, S. (1999). Rainfall, evaporation and the risk of non-contact anterior cruciate ligament injury in the Australian Football League. The Medical Journal of Australia, 170(7), 304-306.
Pearce, L. (2004). Injury probe on Rebound Ace. The Age. Retrieved July 21, 2006, from http://www.theage.com.au/articles/2004/04/15/1081998295352.html?from=story hs
Player traction on synthetic turf. (n.d.). Retrieved November 20, 2002, from Southwest Recreational Industries, http://www.astroturf.com/playertraction.htm
Powell, J.W. & Schootman, M.. (1992). A multivariate risk analysis of selected playing surfaces in the National Football League: 1980 to 1989. American Journal of Sports Medicine, 20(6), 686-694.
Renström, A.F.H. (1995). Knee pain in tennis players. Clinics in Sports Medicine, 14(1), 163-175.
Schlink, L. (2006). Open courts a ‘major danger’. Retrieved July 21, 2006, from Fox Sports News, http://foxsports.news.com.au/story/0,8659,17847788-5002324,00.html
Scranton, P.E. Jr., Whitesel, J.P., Powell, J.W., Dormer, S.G., Heidt, R.S., Losse, G., & Cawley, P.W. (1997). A review of selected noncontact anterior cruciate ligament injuries in the National Football League. Foot & Ankle International, 18(12), 772-776.
Sheppard, W. (2006). Rebound Ace claims another top seed. Retrieved July 21, 2006, from ESPN: Australian Open, http://sports.espn.go.com/sports/tennis/aus06/news/ story?id=2306726
Stadiums and the heat island phenomenon. (2002). Retrieved July 21, 2006 from G- ForSE (Global Forum for Sports and Enviroment), http://www.g-forse.com/ archive/news165_e.html
Stanitski, C.L., McMaster, J.H., & Ferguson, R.J. (1974). Synthetic turf and grass: a comparative study. The Journal of Sports Medicine, 2(1), 22-26.
67 Torg, J.S. (1982). Athletic footwear and orthotic appliances. Clinics in Sports Medicine, 1(1), 157-175.
Torg, J.S., Quedenfeld, T.C., & Landau, B.S. (1974). The shoe-surface interface and its relationship to football knee injuries. The Journal of Sports Medicine, 2(5), 261- 269.
Torg, J.S., Stilwell, G., Rogers, K. (1996). The effect of ambient temperature on the shoe- surface interface release coefficient. American Journal of Sports Medicine, 24(1), 79-82.
Valiant, G.A. (1990). Traction characteristics of outsoles for use on artificial playing surfaces. In R.C. Schmidt, E.F. Hoerner, E.M. Milner, & C.A. Morehouse (Eds.) Natural and Artificial Playing Fields: Characteristics and Safety Features, ASTM STP 1073 (pp. 61-68). Philadelphia, PA: The American Society for Testing and Materials.
Van Gheluwe, B. & Deporte, E. (1992). Friction measurement in tennis on the field and in the laboratory. International Journal of Sports Biomechanics, 8, 48-61.
Van Gheluwe, B., Deporte, E., & Hebbelink, M. (1983). Frictional forces and torques of soccer shoes on artificial turf. Proc. Int. Symposium – Biomechanical Aspects of Sport Shoes and Playing Surfaces.
Victoria Climate Centre. (2006). Monthly Climate Summary – Victoria: 3rd warmest January in Victoria. Retrieved July 21, 2006, from Australian Government Bureau of Meteorology, http://www.bom.gov.au/announcements/media_releases/ vic/20060201.shtml
Winterbottom, W. (1985). Artificial Grass Surfaces for Association Football. London: Sports Council.
A world leader. (2005). Retrieved July 22, 2006, from Rebound Ace Sports Surfaces, http://www.reboundace.com.au/Default.aspx?tabid=325
68
Appendix
69 Table A.1. Minimum Sample Size Needed.
Maximum Standard Sample Size Needed Factor Levels Difference Deviation at Each Level Power
Temperature 8 0.33 0.00296 2 1
Shoe 4 0.23 0.00296 2 1
Temp*Shoe 32 0.33 0.00296 2 1
Calculated using Minitab 14.0
99.99
99 95
80
50
Percent 20
5 1
0.01 2.12.01.91.81.71.61.51.41.31.2 COF
Mean: 1.602 Anderson-Darling Test StDev: 0.1139 AD: 6.102 N: 768 p-value: < 0.005
Figure A.1. Normal Probability Plot of COF.
70 99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 Adidas COF
Mean: 1.606 Anderson-Darling Test StDev: 0.1062 AD: 0.511 N: 192 p-value: 0.194
Figure A.2. Normal Probability Plot of Adidas Barricade COF.
99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.3 1.4 1.5 1.6 1.7 1.8 1.9 New Balance COF
Mean: 1.637 Anderson-Darling Test StDev: 0.08781 AD: 9.240 N: 192 p-value: < 0.005
Figure A.3. Normal Probability Plot of New Balance 650 COF. 71 99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Nike COF
Mean: 1.494 Anderson-Darling Test StDev: 0.08799 AD: 1.773 N: 192 p-value: < 0.005
Figure A.4. Normal Probability Plot of Nike Air Zoom Thrive COF.
99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.4 1.5 1.6 1.7 1.8 1.9 2.0 Wilson COF
Mean: 1.670 Anderson-Darling Test StDev: 0.08794 AD: 6.953 N: 192 p-value: < 0.005
Figure A.5. Normal Probability Plot of Wilson DST COF. 72 99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.4 1.5 1.6 1.7 1.8 1.9 72 F
Mean: 1.684 Anderson-Darling Test StDev: 0.07353 AD: 0.643 N: 96 p-value: 0.091
Figure A.6. Normal Probability Plot of 72 degrees F COF.
99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.4 1.5 1.6 1.7 1.8 1.9 85 F
Mean: 1.639 Anderson-Darling Test StDev: 0.07057 AD: 1.811 N: 96 p-value: < 0.005
Figure A.7. Normal Probability Plot of 85 degrees F COF. 73 99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.4 1.5 1.6 1.7 1.8 1.9 100 F
Mean: 1.671 Anderson-Darling Test StDev: 0.07426 AD: 0.806 N: 96 p-value: 0.036
Figure A.8. Normal Probability Plot of 100 degrees F COF.
99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 110 F
Mean: 1.546 Anderson-Darling Test StDev: 0.08851 AD: 1.929 N: 96 p-value: < 0.005
Figure A.9. Normal Probability Plot of 110 degrees F COF. 74 99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.4 1.5 1.6 1.7 1.8 1.9 115 F
Mean: 1.649 Anderson-Darling Test StDev: 0.07996 AD: 1.181 N: 96 p-value: < 0.005
Figure A.10. Normal Probability Plot of 115 degrees F COF.
99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 130 F
Mean: 1.619 Anderson-Darling Test StDev: 0.09963 AD: 0.963 N: 96 p-value: 0.015
Figure A.11. Normal Probability Plot of 130 degrees F COF. 75 99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 140 F
Mean: 1.557 Anderson-Darling Test StDev: 0.1121 AD: 1.002 N: 96 p-value: 0.012
Figure A.12. Normal Probability Plot of 140degrees F COF.
99.9
99
95 90 80 70 60 50 40 30 Percent 20 10 5
1
0.1 1.2 1.3 1.4 1.5 1.6 1.7 155 F
Mean: 1.447 Anderson-Darling Test StDev: 0.08386 AD: 0.318 N: 96 p-value: 0.531
Figure A.13. Normal Probability Plot of 155 degrees F COF. 76 Vita
Denise Helen Bauer was born and raised in Little Rock, AR where she graduated from Parkview Arts and Science Magnet High School. She attended the University of
Tennessee, Knoxville and received a B.S. in Engineering Science with a concentration in
Biomedical Engineering in 2001. Denise was then a Graduate Teaching Assistant in the
Engage Engineering Fundamentals Program while pursing a M.S. in Industrial
Engineering with a concentration in Human Factors Engineering. She received her M.S.
in 2006.
Denise is currently pursing a Ph.D. in Industrial Engineering at the Pennsylvania
State University in State College, PA. Her current research is related to children and
backpack use.
77