School of Civil Engineering Sydney NSW 2006 AUSTRALIA http://www.civil.usyd.edu.au/ Centre for Advanced Structural Engineering

Scaffold Cuplok Joint Tests Research Report No R893

Tayakorn Chandrangsu BSc MSc Kim JR Rasmussen MScEng PhD

December 2008 ISSN 1833-2781

School of Civil Engineering Centre for Advanced Structural Engineering http://www.civil.usyd.edu.au/

Scaffold Cuplok Joint Tests Research Report No R893 Tayakorn Chandrangsu, BSc, MSc Kim JR Rasmussen, MScEng, PhD

December 2008

Abstract: This report describes the setup, procedure, and results of scaffold Cuplok joint tests. The aims of the tests are to investigate the joint stiffness for rotations about vertical and horizontal axes in various joint configurations, and carry out statistical analyses of the experimental results. The tests were performed in the laboratory of the School of Civil Engineering at the University of Sydney from second-hand Cuplok scaffold parts provided by Boral Formwork and Scaffolding Pty Ltd. A total of 172 tests were carried out on various joint configurations, bending axes, loading directions, types of material (galvanised or painted components), and degree of tightening of the joints. The results are shown graphically in terms of momentrotation curves. Since there was substantial variation in Cuplok joint stiffness, a statistical analysis on the results was performed. These experimental studies are especially useful for modelling and performing probabilistic analysis on the Cuplok scaffold systems.

Keywords: Joint tests, Joint stiffness, Probabilistic analysis, Support scaffold systems, Falsework

Scaffold Cuplok Joint Tests

December 2008

Copyright Notice School of Civil Engineering, Research Report R893 Scaffold Cuplok Joint Tests © 2008 Tayakorn Chandrangsu and Kim JR Rasmussen [email protected] and [email protected] ISSN 1833-2781 This publication may be redistributed freely in its entirety and in its original form without the consent of the copyright owner. Use of material contained in this publication in any other published works must be appropriately referenced, and, if necessary, permission sought from the author.

Published by: School of Civil Engineering The University of Sydney Sydney NSW 2006 AUSTRALIA December 2008 This report and other Research Reports published by the School of Civil Engineering are available on the Internet: http://www.civil.usyd.edu.au

School of Civil Engineering Research Report No R893

2

Scaffold Cuplok Joint Tests

December 2008

Table of Contents 1. Introduction.......................................................................................................5 2. Test Setup..........................................................................................................6 3. Test Materials....................................................................................................8 4. Test Series .........................................................................................................9 5. Test Procedure.................................................................................................11 6. Test Results .....................................................................................................12 7. Discussion .......................................................................................................21 8. Conclusions.....................................................................................................24 Acknowledgement ..............................................................................................24 References ...........................................................................................................24 Appendix A .........................................................................................................25 Appendix B .........................................................................................................39

School of Civil Engineering Research Report No R893

3

Scaffold Cuplok Joint Tests

School of Civil Engineering Research Report No R893

December 2008

4

Scaffold Cuplok Joint Tests

December 2008

1. Introduction Cuplok scaffold systems are widely used in the construction industry for providing general access and supporting vertical loads. For example, a typical Cuplok system is used to support formwork as shown in Figure 1. The system is fast to erect and easy to disassemble with the well-known Cuplok joint providing flexibility and ease of construction. Cuplok components consist of a lower cup, which is welded on to the standard (vertical) at 500 mm intervals, and a sliding upper cup, as shown in Figure 2.

Figure 1: Typical Cuplok system The locking mechanism of Cuplok joints uses no nuts, bolts or wedges; instead the method of connecting the ledger (horizontal) to the standard (vertical) is by simply positioning the end blades of up to 4 ledgers at desired angles into the lower cup, moving the upper cup down, and rotating it clockwise by hammer blows until it locks up against a locking bar to achieve a tight connection. Figure 3 shows the components of a Cuplok joint. Conversely, disassembling of the Cuplok joint is done by applying hammer blows in the counter clockwise direction until the upper cup can be lifted up to allow the ledgers to be taken out.

Figure 2: Typical Cuplok components School of Civil Engineering Research Report No R893

5

Scaffold Cuplok Joint Tests

December 2008

Cuplok joints can be assembled into 2-, 3- and 4-way connections at any angles providing versatility of application. Cuplok parts are made of steel, and depending on the manufacturer they are usually either galvanised or painted for corrosion resistance, durability, and better handling. These components are commonly reused in construction practice.

Figure 3: Locking mechanism of Cuplok Little information on Cuplok joint tests is available in the literature; only Godley and Beale provide some details and results of Cuplok joint tests performed in 1990 [1]. Due to the lack of available joint test data, this report presents tests on Cuplok joint and discusses the semirigid joint behaviour observed from the tests as well as the joint stiffness derived from the test data. Particular attention is paid to the nonlinear moment-rotation curve. The slope of this curve is a direct measure of the stiffness of the joint, which can be incorporated in the numerical modelling of Cuplok scaffold system. Furthermore, since the Cuplok joint tests exhibit considerable variability in joint stiffness, statistical analyses of the joint test data have been carried out in preparation for numerical modelling.

2. Test Setup The setup was adapted from a multi-purpose test rig described by Lightfoot and Bhula [2, 3]. Figure 4 shows schematically the Cuplok joint test setup. The test rig was specifically designed and built for the Cuplok joint tests, consisting of a 310UC158 column mounted on to a strong floor and a pivoted U-shaped clamp made from three lengths of 150x50x5 RHS section, attached to the column by M20 bolts and a rotatable pin. The clamp was designed to be able to rotate via the pin so that the specimen can be tested in any loading direction. Two 50 mm thick plates were attached to the inside of the top and bottom tips of the U-shaped clamp to grip the test specimen. Each plate featured a hole slightly smaller than the nominal diameter of the standard and two M12 bolts for tightening the grip, as shown in Figure 4. A hydraulic jack, capable of producing up to 32-kN of load and attached with a load cell and a half-circular loading plate, was mounted on to the rail beams of the strong floor to apply loading. Five LVDTs, clamped on external posts, were used to read the displacements at different locations along the specimen, as shown in Figure 4. Flat and smooth plates were attached with pipe clamps to the specimen at those locations to ensure accurate readings of the LVDTs. Figure 5 shows a typical test setup.

School of Civil Engineering Research Report No R893

6

Scaffold Cuplok Joint Tests

December 2008

Note: LVDTs are clamped on to movable posts. Plate attached to a pipe clamp to ensure complete contact to the specimen during loading.

150x50x5 RHS

620 500

Plan view M12

2000

Specimen

M20 600

32 kN Hydraulic jack with load cell in stroke control mode

Pivot M12 150x50x5 RHS 784

504

857

583

M20

Strong floor

Unit: mm

Elevation view

Rail beam

Figure 4: Schematic of Cuplok joint test setup The jack was connected to a hydraulic pump and an MTS controller which operated the actuator in stroke control mode. A data logger was used for taking measurements of the LVDTs, load, and stroke. The data logger was connected to a computer with StrainSmart software and configured to take readings every second. Figure 6 shows details of the data logging equipment.

Figure 5: Typical Cuplok joint test setup School of Civil Engineering Research Report No R893

7

Scaffold Cuplok Joint Tests

December 2008

Figure 6: Test equipment setup

3. Test Materials The materials for testings were provided by Boral Formwork and Scaffolding Pty Ltd from available stocks of used components. It was decided that all test specimens should be sourced from secondhand materials in order to present the real stiffness and strength of the Cuplok joints being employed in construction practice. The materials supplied by Boral consisted of 12 Cuplok open ended 2.80 m standards and 27 Cuplok 1.83 m ledgers. Figure 7 shows details of those components. The nominal tubular cross-section dimensions and yield stress of the Cuplok standards were 48 mm x 4 mm and 450 MPa respectively. Also, the Cuplok ledgers were of nominal tubular dimensions of 48 mm x 3.2 mm with a nominal yield stress of 350 MPa. Some of the standards and ledgers showed visible out-of-straightness and some Cuplok joints showed signs of wear. These materials were cut into specific lengths to fit in the test rig. Each standard was cut into four usable Cuplok joint specimens with an approximate length of 500 mm and each ledger was cut into 2 usable ledger specimens with an end blade at one end. The lengths of the ledger specimens were 300 mm for connection elements and 600 mm for loading elements, as shown in Figure 8.

School of Civil Engineering Research Report No R893

8

Scaffold Cuplok Joint Tests

December 2008

Figure 7: Typical Cuplok standard and ledger

4. Test Series The objectives of these experiments were to investigate the stiffness and strength of Cuplok joints in the vertical direction (rotation about the z-axis) and horizontal direction (rotation about the y-axis), see Figure 8. To clearly distinguish the factors affecting the stiffness, the tests were categorised based on the axis of bending, joint configuration, and loading direction (up or down and left or right). Moreover, the type of finish (galvanised or painted), and the number of hammer blows exerted on the cup during assembly were recorded for each test to study the influence of these factors. Figure 8 shows the bending axes and Figure 9 shows different joint configurations and loading directions being considered. Types A, B and C, and D represent 4-way, 3-way, and 2-way connections respectively. The loading directions consist of up or down for bending about the z-axis and left or right for bending about the yaxis. To be able to differentiate the test results, different labels for each test configuration are used, as shown in Table 1.

Figure 8: Joint stiffness axes

School of Civil Engineering Research Report No R893

9

Scaffold Cuplok Joint Tests

December 2008

Figure 9: Joint configurations and loading directions in top view

School of Civil Engineering Research Report No R893

10

Scaffold Cuplok Joint Tests

December 2008

Table 1: Labels for test configurations Label KzA1 KzB1 KzC1 KzD1 KzA2 KzB2 KzC2 KzD2 KyA1 KyB1 KyC1 KyD1 KyC2 KyD2

Bending axis z z z z z z z z y y y y y y

Joint type A (4-way) B (3-way ) C (3-way ) D (2-way) A (4-way) B (3-way ) C (3-way ) D (2-way) A (4-way) B (3-way ) C (3-way ) D (2-way) C (3-way ) D (2-way)

Loading direction 1(down) 1(down) 1(down) 1(down) 2(up) 2(up) 2(up) 2(up) 1(right) 1(right) 1(right) 1(right) 2(left) 2(left)

5. Test Procedure Prior to testing, the five LVDTs were calibrated based on their displacement recordings from StrainSmart software against displacement readings obtained using a digital vernier caliper with an accuracy of 0.01 mm. The load cell was also calibrated prior to testing. The Cuplok joint was assembled based on the required testing joint configuration and then clamped onto the test rig and rotated via the pivot to the desired loading position. Afterwards, the clamp was fastened to the column by M20 bolts, and the LVDTs were fitted at five locations along the specimen, as shown in Figure 10.

620 315

100

LVDT

85

3

LVDT

LVDT

4 LVDT

2

1

Specimen

270 5 Loading

LVDT

48

Figure 10: Typical numberings and location of LVDTs on test specimen

School of Civil Engineering Research Report No R893

11

Scaffold Cuplok Joint Tests

December 2008

The hydraulic jack was programmed to run at the rate of 3 mm/min in stroke control mode and the data logger was set to record data at 1 scan/second. The half-circular loading plate on the hydraulic jack was then moved toward the specimen until barely touching before zeroing all recording values. The recording of data and the movement of the jack were started simultaneously at the commencement of the test. Some of the tests of each configuration were kept in the elastic range and the Cuplok joints were retested by varying the number of hammer blows (3-to-7) applied to tighten the cup. These tests were carried out to compare the initial stiffness of the Cuplok joints. Also, for each configuration two types of finish (galvanised or painted components) were tested to see if the finish resulted in significant differences in joint stiffness. Finally, some of the tests of each configuration were continued into plastic range to obtain the joint strength.

6. Test Results From the data recorded, the joint behaviour was investigated by determining relationship between moment and rotation. The moment, M, is calculated as M = F×L

(1)

where F is the applied load and L is the perpendicular distance between load application point and the centre of the joint. The rotation, θ, is given as  ∆2 − ∆3    ∆ − ∆4   −  arctan 5     d d + d 2 −3 4      5

θ = arctan

(2)

where ∆2, ∆3, ∆4, and ∆5 are the displacement of LVDT2, LVDT3, LVDT4 and LVDT5 respectively, d2-3 is the distance between LVDT2 and LVDT3, d4 is the perpendicular distance between LVDT4 and the centre of the joint, and d5 is the perpendicular distance between LVDT5 and centre of the joint. The LVDT numbering system is shown in Figure 10. LVDT1 and stroke readings are used only for calibration. After calculation of moment and rotation, moment-rotation curves were plotted for each joint configuration, as categorised in Table 1. Figures 11, 13, 15, and 17 show the results for joints bending downward about the z-axis for different joint configurations. Figures 12, 14, 16, and 18 present the results for joints bending upward about the z-axis with different joint configurations. Figures 19, 20, 21 and 23 illustrate the results for joints bending to the right about the y-axis for four different joint configurations. Figures 22 and 24 present the results for joints bending to the left about the y-axis for two different joint configurations. For each configuration, a number of tests are shown correspondingly to different combinations of the number of hammer blows and type of finish (galvanised or painted components). Test 1 to test 5 and test 6 to test 10 corresponded to 7 hammer blows and 3 hammer blows respectively, and test 11 to test 14 were all tightened using 5 blows. Galvanised components were used in test 1 to test 5, test 11 and test 13. Painted components were used in test 6 to test 10, test 12 and test 14.

School of Civil Engineering Research Report No R893

12

Scaffold Cuplok Joint Tests

December 2008

4.5

0.16

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10 Test 11 Test 12 Test 13 Test 14

0.2

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10 Test 11 Test 12 Test 13 Test 14

4

Moment (kNm)

3.5 3 2.5 2 1.5 1 0.5 0 0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

Rotation (rad)

Figure 11: Cuplok test results for KzA1

4.5 4

Moment (kNm)

3.5 3 2.5 2 1.5 1 0.5 0 0

0.05

0.1

0.15

Rotation (rad)

Figure 12: Cuplok test results for KzA2

School of Civil Engineering Research Report No R893

13

Scaffold Cuplok Joint Tests

December 2008

4.5

0.16

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10 Test 11 Test 12 Test 13 Test 14

0.2

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10 Test 11 Test 12 Test 13 Test 14

4

Moment (kNm)

3.5 3 2.5 2 1.5 1 0.5 0 0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

Rotation (rad)

Figure 13: Cuplok test results for KzB1

4.5 4

Moment (kNm)

3.5 3 2.5 2 1.5 1 0.5 0 0

0.05

0.1

0.15

Rotation (rad)

Figure 14: Cuplok test results for KzB2

School of Civil Engineering Research Report No R893

14

Scaffold Cuplok Joint Tests

December 2008

5 4.5

0.16

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10 Test 11 Test 12 Test 13 Test 14

0.16

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10 Test 11 Test 12 Test 13 Test 14

4 Moment (kNm)

3.5 3 2.5 2 1.5 1 0.5 0 0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

Rotation (rad)

Figure 15: Cuplok test results for KzC1

4 3.5

Moment (kNm)

3 2.5 2 1.5 1 0.5 0 0

0.02

0.04

0.06

0.08

0.1

0.12

Rotation (rad)

Figure 16: Cuplok test results for KzC2

School of Civil Engineering Research Report No R893

15

0.14

Scaffold Cuplok Joint Tests

December 2008

3.5

0.16

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10 Test 11 Test 12 Test 13 Test 14

0.16

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10 Test 11 Test 12 Test 13 Test 14

3

Moment (kNm)

2.5 2 1.5 1 0.5 0 0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

Rotation (rad)

Figure 17: Cuplok test results for KzD1

4 3.5

Moment (kNm)

3 2.5 2 1.5 1 0.5 0 0

0.02

0.04

0.06

0.08

0.1

0.12

Rotation (rad)

Figure 18: Cuplok test results for KzD2

School of Civil Engineering Research Report No R893

16

0.14

Scaffold Cuplok Joint Tests

December 2008

0.7 0.6

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10

Moment (kNm)

0.5 0.4 0.3 0.2 0.1 0 0

0.02

0.04

0.06

0.08

0.1

0.12

Rotation (rad)

Figure 19: Cuplok test results for KyA1

0.7 0.6

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10

Moment (kNm)

0.5 0.4 0.3 0.2 0.1 0 0

0.02

0.04

0.06

0.08

Rotation (rad)

Figure 20: Cuplok test results for KyB1

School of Civil Engineering Research Report No R893

17

0.1

Scaffold Cuplok Joint Tests

December 2008

0.6

Moment (kNm)

0.5

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10

0.4 0.3 0.2 0.1 0 0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

Rotation (rad)

Figure 21: Cuplok test results for KyC1

0.3

Moment (kNm)

0.25

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10

0.2 0.15 0.1 0.05 0 0

0.005

0.01

0.015

0.02

0.025

0.03

Rotation (rad)

Figure 22: Cuplok test results for KyC2

School of Civil Engineering Research Report No R893

18

0.035

0.04

Scaffold Cuplok Joint Tests

December 2008

0.6

Moment (kNm)

0.5

Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10

0.4 0.3 0.2 0.1 0 0

0.05

0.1

0.15

0.2

Rotation (rad)

Figure 23: Cuplok test results for KyD1

0.25 Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8 Test 9 Test 10

Moment (kNm)

0.2

0.15

0.1

0.05

0 0

0.02

0.04

0.06

0.08

0.1

Rotation (rad)

Figure 24: Cuplok test results for KyD2

The Cuplok joint stiffness is determined from the initial slope of the moment-rotation curves shown in Figures 11-24. Table 2 summarises the Cuplok joint stiffness for each test for the various configurations. The tests show that approximately 30% of the joints exhibit looseness School of Civil Engineering 19 Research Report No R893

Scaffold Cuplok Joint Tests

December 2008

at the beginning of loading, notably among the joints loaded in vertical bending (rotation about the z-axis). A statistical analysis of the coordinates of the moment-rotation curves corresponding to the end of the initial looseness range for those joints exhibiting looseness shows that looseness is overcome when the moment reaches 0.4 kNm with a corresponding average rotation of 0.01 rad. The ultimate moment capacity of the joints bent about the z-axis is about 3.5 kNm on average. In contrast, joints bent about the y-axis have a very small moment capacity of 0.4 kNm on average and much smaller initial stiffness than the joints bent about the z-axis. Most of the Cuplok joints have a large plastic range associated with large rotation, except a few of the joints that failed abruptly resulting from fracture at the end blade of the ledger. Figure 25 shows a Cuplok component with plastic deformations and minor cracks after being tested. Table 2: Summary of initial Cuplok joint stiffness

Test 1 2 3 4 5 6 7 8 9 10 11 12 13 14

KzA1 110 109 112 124 150 100 160 105 102 90 85 75 80 105

Initial Cuplok joint stiffness (kNm/rad) KzB1 KzC1 KzD1 KzA2 KzB2 100 96 82 113 105 92 90 90 130 63 95 91 99 120 78 96 95 100 120 78 101 104 97 113 80 60 75 65 113 80 70 100 60 135 115 80 120 53 118 118 75 118 54 133 117 73 100 59 138 125 79 83 60 90 79 68 85 55 88 80 75 92 88 113 84 100 70 75 115 100

KzC2 113 135 128 115 117 96 117 120 120 125 90 65 60 72

Test 1 2 3 4 5 6 7 8 9 10

Initial Cuplok joint stiffness (kNm/rad) KyA1 KyB1 KyC1 KyD1 KyC2 22 12 10 2 13 20 12 9 3 10 21 15 12 3 11 16 17 12 2 20 18 13 9 3 12 10 12 10 3 15 13 10 15 8 22 30 12 17 9 20 17 11 8 10 21 29 13 12 5 16

KyD2 2 3 4 4 5 1 3 4 5 8

School of Civil Engineering Research Report No R893

20

KzD2 88 94 55 75 99 94 90 83 85 79 85 100 68 80

Scaffold Cuplok Joint Tests

December 2008

Figure 25: Cuplok components after being tested

7. Discussion By examining the test results, the number of hammer blows applied to tighten the cup and the type of finish (galvanised or painted) have no substantial influence on the joint stiffness. According to construction practice, the hammer blows on the cup are usually a minimum of three strong blows suggesting that the Cuplok joint is completely fastened. More blows, up to seven as investigated herein, do not increase the joint stiffness. Moreover, galvanised or painted components have insignificant effects on the joint stiffness. The comparisons of these results based on initial joint stiffness are shown in Appendix A. Nonetheless, the joint stiffness depends greatly on the axis of bending. A statistical analysis is presented in Table 3 for each test configuration based on different tested joints. Table 3: Statistical results for initial Cuplok joint stiffness Initial Cuplok joint stiffness (kNm/rad) KzA1

KzB1

KzC1

KzD1

KzA2

KzB2

KzC2

KzD2

Mean 96.23 81.73 87.97 71.62 108.54 88.99 87.32 83.07 Standard deviation 23.28 13.29 13.62 17.46 14.25 18.99 23.48 11.86 Coefficient of variation 0.24

0.16

0.15

0.24

0.13

0.21

0.27

Initial Cuplok joint stiffness (kNm/rad) KyA1

KyB1

KyC1

KyD1

KyC2

KyD2

Mean 19.50 12.65 11.38

4.77

15.95

3.87

Standard deviation 6.09

2.06

2.71

2.91

4.28

1.82

Coefficient of variation 0.31

0.16

0.24

0.61

0.27

0.47

School of Civil Engineering Research Report No R893

21

0.14

Scaffold Cuplok Joint Tests

December 2008

The statistical results show that Cuplok joints bent in the vertical plane (rotation about the zaxis) are much stiffer than when bent in the horizontal plane (rotation about the y-axis), and that the 4-way joint connection is the stiffest among all other configurations, particularly the 2-way joint connection which has the lowest stiffness. Also, both types of 3-way connections provide similar stiffness for the same bending axis. The difference in vertical stiffness between 4-way and 2-way joints is about 30%. Besides, the upward bending stiffness tends to be slightly higher than the downward bending stiffness which is rational since the degree of fixity in the welded lower cup is higher than that of the locking upper cup and hence less joint movement is possible under upward bending. The coefficient of variation values shown in Table 3 suggest that the variation in Cuplok joint stiffness is substantial and hence may significantly influence the ultimate strength of scaffold systems. The relation between the moment and rotation of the Cuplok connections can be modelled by a tri-linear curve, as illustrated in Figure 26. The parameters that describe the tri-linear curve (k1, k2, k3, β1, β2, β3) were obtained from the joint test data herein. The mean values for k1, k2, and k3 for different joint configurations and bending axes are presented in Table 4. The mean values of k1 are presented for joint test data considering looseness alone (moment-rotation curves that show looseness only), and with and without looseness. The mean values for β1, β2 and β3 are presented for different joint configurations and bending axes in Table 5. Since probabilistic assessment of the strength of scaffold systems depends mainly on the joint bent about the horizontal axis, but not about the vertical axis, thus only variation in the vertical bending stiffness may be considered, and modelling of the horizontal bending stiffness can be assumed to be deterministic using the mean values for the tri-linear curve, as presented in Tables 4-5 for bending about the vertical axis. For practicality in probabilistic modelling, the values of β1, β2 and β3 are assumed to be deterministic, taken as their mean values; however, the variations of k1, k2, and k3 in bending about the horizontal axis are taken into account. The coefficient of variation values of the joint stiffness for k1, k2, and k3 in bending about the horizontal axis for different joint configurations are presented in Table 6. The coefficient of variation values of k1 are presented for joint test data considering looseness alone, and with and without looseness. It is expected that probabilistic modelling of joint stiffness with k1 considering looseness alone will produce conservative results of the strength of scaffold systems. The data of the joint stiffness (k1, k2, k3) in bending about the horizontal axis is normalised with the mean value of its configuration and fitted to normal distribution for probabilistic modelling, as shown in Appendix A with corresponding statistical functions. In the probabilistic model of joint stiffness, normal distribution can be applied for k1, k2, and k3 in bending about the horizontal axis for each joint configuration as random variables by generating normal random values with corresponding mean and coefficient of variation, shown in Appendix B, and multiplying by their corresponding mean joint stiffness values from Table 4 for each joint configuration.

School of Civil Engineering Research Report No R893

22

Scaffold Cuplok Joint Tests

December 2008

Moment

k3

k2

k1 Rotation β1

β3

β2

Figure 26: Tri-linear moment-rotation for the Cuplok joints

Table 4: Mean Cuplok joint stiffness (kNm/rad) Bending about horizontal axis

Bending about vertical axis

k1 Joint configuration

Looseness alone

With and without looseness

k2

k3

k1

k2

k3

4-way

39

80

102

5.3

15

7.5

0.8

3-way

36

75

87

5.1

14

7

1

2-way

41

70

77

4.6

7.5

5

1.5

Table 5: Mean rotation for Cuplok joints (rad) Bending about horizontal axis

Bending about vertical axis

Joint configuration

β1

β2

β3

β1

β2

β3

4-way

0.014

0.036

0.16

0.02

0.04

0.1

3-way

0.012

0.036

0.16

0.02

0.04

0.1

2-way

0.007

0.036

0.16

0.02

0.04

0.1

School of Civil Engineering Research Report No R893

23

Scaffold Cuplok Joint Tests

December 2008

Table 6: Coefficient of variation of Cuplok joint stiffness

Joint configuration

Bending about horizontal axis k1 With and k2 Looseness without alone looseness

k3

4-way

0.22

0.35

0.18

0.30

3-way

0.38

0.37

0.21

0.37

2-way

0.35

0.27

0.20

0.46

8. Conclusions This report investigates the joint stiffness of Cuplok scaffold systems. Using secondhand Cuplok components provided by Boral Formwork and Scaffolding Pty Ltd, Cuplok joints were assembled and tested in a specially designed rig in the structures laboratory of the School of Civil Engineering at the University of Sydney. The joint tests featured bending about two distinct axes, different loading directions, four types of joint configuration, two finishes (galvanised or painted components), and different numbers of hammer blows to tighten the cup. The results are presented in the form of moment-rotation curves, the initial slopes of which were used to determine the joint stiffness. The joint stiffness for each test configuration was tabulated and a statistical analysis was performed. The stiffness for bending in the vertical plane was found to be much higher than the stiffness for bending in the horizontal plane, and 4-way connections provided greater stiffness than other joint configurations. However, the surface finish (galvanised or painted components) and the number of hammer blows used to tighten the connection had insignificant effect on the joint stiffness. The joint stiffness data was fitted to normal distributions for probabilistic modelling. The results from the investigation of Cuplok joint behaviour are useful for modelling and performing probabilistic analyses of the strength of Cuplok scaffold systems.

Acknowledgement The authors would like to thank Boral Formwork & Scaffolding Pty Ltd for providing testing materials and support through linkage grant on this research project.

References [1] Godley MHR, Beale RG. Sway stiffness of scaffold structures. Structural Engineer 1997;75(1):4-12. [2] Lightfoot E, Bhula D. The idealization of scaffold couplers for performance tests and scaffold analysis 1977;10(57):159-168. [3] Lightfoot E, Bhula D. A test rig for scaffold couplers. Materials and Structures 1977;10(3):168-173.

School of Civil Engineering Research Report No R893

24

Scaffold Cuplok Joint Tests

December 2008

Appendix A

Initial joint stiffness (kNm/rad)

The comparisons are shown for initial Cuplok joint stiffness in different joint configurations based on the number of hammer blows (Figures A1-A14) and the type of finish (Figures A15-A28). The trend lines are added, passing through the mean values for better comparisons.

17 0 15 0 13 0 11 0 9 0 7 0 5 0 1

2

3

4 5 Hammer blows

6

7

8

Initial joint stiffness (kNm/rad)

Figure A1: Initial Cuplok joint stiffness for KzA1 based on the number of hammer blows applied to tighten the cup

170 150 130 110 90 70 50 1

2

3

4

5

6

7

8

Hammer blows Figure A2: Initial Cuplok joint stiffness for KzB1 based on the number of hammer blows applied to tighten the cup

School of Civil Engineering Research Report No R893

25

Initial joint stiffness (kNm/rad)

Scaffold Cuplok Joint Tests

December 2008

170 150 130 110 90 70 50 1

2

3

4

5

6

7

8

Hammer blows

Initial joint stiffness (kNm/rad)

Figure A3: Initial Cuplok joint stiffness for KzC1 based on the number of hammer blows applied to tighten the cup

170 150 130 110 90 70 50 1

2

3

4

5

6

7

8

Hammer blows Figure A4: Initial Cuplok joint stiffness for KzD1 based on the number of hammer blows applied to tighten the cup

School of Civil Engineering Research Report No R893

26

Initial joint stiffness (kNm/rad)

Scaffold Cuplok Joint Tests

December 2008

170 150 130 110 90 70 50 1

2

3

4

5

6

7

8

Hammer blows

Initial joint stiffness (kNm/rad)

Figure A5: Initial Cuplok joint stiffness for KzA2 based on the number of hammer blows applied to tighten the cup

170 150 130 110 90 70 50 1

2

3

4

5

6

7

8

Hammer blows Figure A6: Initial Cuplok joint stiffness for KzB2 based on the number of hammer blows applied to tighten the cup

School of Civil Engineering Research Report No R893

27

Initial joint stiffness (kNm/rad)

Scaffold Cuplok Joint Tests

December 2008

170 150 130 110 90 70 50 1

2

3

4

5

6

7

8

Hammer blows

Initial joint stiffness (kNm/rad)

Figure A7: Initial Cuplok joint stiffness for KzC2 based on the number of hammer blows applied to tighten the cup

170 150 130 110 90 70 50 1

2

3

4

5

6

7

8

Hammer blows Figure A8: Initial Cuplok joint stiffness for KzD2 based on the number of hammer blows applied to tighten the cup

School of Civil Engineering Research Report No R893

28

Scaffold Cuplok Joint Tests

December 2008

Initial joint stiffness (kNm/rad)

30 25 20 15 10 5 0 1

2

3

4

5

6

7

8

Hammer blows Figure A9: Initial Cuplok joint stiffness for KyA1 based on the number of hammer blows applied to tighten the cup

Initial joint stiffness (kNm/rad)

30 25 20 15 10 5 0 1

2

3

4

5

6

7

8

Hammer blows Figure A10: Initial Cuplok joint stiffness for KyB1 based on the number of hammer blows applied to tighten the cup

School of Civil Engineering Research Report No R893

29

Scaffold Cuplok Joint Tests

December 2008

Initial joint stiffness (kNm/rad)

30 25 20 15 10 5 0 1

2

3

4

5

6

7

8

Hammer blows

Initial joint stiffness (kNm/rad)

Figure A11: Initial Cuplok joint stiffness for KyC1 based on the number of hammer blows applied to tighten the cup

30 25 20 15 10 5 0 1

2

3

4

5

6

7

8

Hammer blows Figure A12: Initial Cuplok joint stiffness for KyD1 based on the number of hammer blows applied to tighten the cup

School of Civil Engineering Research Report No R893

30

Initial joint stiffness (kNm/rad)

Scaffold Cuplok Joint Tests

December 2008

30 25 20 15 10 5 0 1

2

3

4

5

6

7

8

Hammer blows

Initial joint stiffness (kNm/rad)

Figure A13: Initial Cuplok joint stiffness for KyC2 based on the number of hammer blows applied to tighten the cup

30 25 20 15 10 5 0 1

2

3

4

5

6

7

8

Hammer blows Figure A14: Initial Cuplok joint stiffness for KyD2 based on the number of hammer blows applied to tighten the cup

School of Civil Engineering Research Report No R893

31

Initial joint stiffness (kNm/rad)

Scaffold Cuplok Joint Tests

December 2008

170 150 130 110 90 70 50 Galvanised

Painted

Materials

Initial joint stiffness (kNm/rad)

Figure A15: Initial Cuplok joint stiffness for KzA1 based on the type of finish (galvanised or painted)

170 150 130 110 90 70 50 Galvanised

Painted

Materials Figure A16: Initial Cuplok joint stiffness for KzB1 based on the type of finish (galvanised or painted)

School of Civil Engineering Research Report No R893

32

Initial joint stiffness (kNm/rad)

Scaffold Cuplok Joint Tests

December 2008

170 150 130 110 90 70 50 Galvanised

Painted

Materials

Initial joint stiffness (kNm/rad)

Figure A17: Initial Cuplok joint stiffness for KzC1 based on the type of finish (galvanised or painted)

170 150 130 110 90 70 50 Galvanised

Painted

Materials Figure A18: Initial Cuplok joint stiffness for KzD1 based on the type of finish (galvanised or painted)

School of Civil Engineering Research Report No R893

33

Initial joint stiffness (kNm/rad)

Scaffold Cuplok Joint Tests

December 2008

170 150 130 110 90 70 50 Galvanised

Painted

Materials

Initial joint stiffness (kNm/rad)

Figure A19: Initial Cuplok joint stiffness for KzA2 based on the type of finish (galvanised or painted)

170 150 130 110 90 70 50 Galvanised

Painted

Materials Figure A20: Initial Cuplok joint stiffness for KzB2 based on the type of finish (galvanised or painted)

School of Civil Engineering Research Report No R893

34

Initial joint stiffness (kNm/rad)

Scaffold Cuplok Joint Tests

December 2008

170 150 130 110 90 70 50 Galvanised

Painted

Materials

Initial joint stiffness (kNm/rad)

Figure A21: Initial Cuplok joint stiffness for KzC2 based on the type of finish (galvanised or painted)

170 150 130 110 90 70 50 Galvanised

Painted

Materials Figure A22: Initial Cuplok joint stiffness for KzD2 based on the type of finish (galvanised or painted)

School of Civil Engineering Research Report No R893

35

Initial joint stiffness (kNm/rad)

Scaffold Cuplok Joint Tests

December 2008

30 25 20 15 10 5 0 Galvanised

Painted

Materials

Initial joint stiffness (kNm/rad)

Figure A23: Initial Cuplok joint stiffness for KyA1 based on the type of finish (galvanised or painted)

30 25 20 15 10 5 0 Galvanised

Painted

Materials Figure A24: Initial Cuplok joint stiffness for KyB1 based on the type of finish (galvanised or painted)

School of Civil Engineering Research Report No R893

36

Initial joint stiffness (kNm/rad)

Scaffold Cuplok Joint Tests

December 2008

30 25 20 15 10 5 0 Galvanised

Painted

Materials

Initial joint stiffness (kNm/rad)

Figure A25: Initial Cuplok joint stiffness for KyC1 based on the type of finish (galvanised or painted)

30 25 20 15 10 5 0 Galvanised

Painted

Materials Figure A26: Initial Cuplok joint stiffness for KyD1 based on the type of finish (galvanised or painted)

School of Civil Engineering Research Report No R893

37

Initial joint stiffness (kNm/rad)

Scaffold Cuplok Joint Tests

December 2008

30 25 20 15 10 5 0 Galvanised

Painted

Materials

Initial joint stiffness (kNm/rad)

Figure A27: Initial Cuplok joint stiffness for KyC2 based on the type of finish (galvanised or painted)

30 25 20 15 10 5 0 Galvanised

Painted

Materials Figure A28: Initial Cuplok joint stiffness for KyD2 based on the type of finish (galvanised or painted)

School of Civil Engineering Research Report No R893

38

Scaffold Cuplok Joint Tests

December 2008

Appendix B The data of joint stiffness (k1, k2, k3) for different configurations is normalised with the mean value of its configuration and fitted to normal distribution.

Relative frequency

0.3

Normal mean = 1.00 cov = 0.33

0.25 0.2 0.15 0.1 0.05 0 0.4

0.6

0.8

1

1.2

1.4

1.6

1.8 More

Normalised k1

Relative frequency

Figure B1: Fitted normal distribution of normalised Cuplok joint stiffness k1 (looseness alone) in bending about the horizontal axis

0.4 0.35

Normal mean = 1.07 cov = 0.2

0.3 0.25 0.2 0.15 0.1 0.05 0 0.6

0.8

1

1.2

1.4

1.6

More

Normalised k2

Figure B2: Fitted normal distribution of normalised Cuplok joint stiffness k2 in bending about the horizontal axis

School of Civil Engineering Research Report No R893

39

Scaffold Cuplok Joint Tests

December 2008

Relative frequency

0.25 0.2

Normal mean = 0.99 cov = 0.38

0.15 0.1 0.05 0 0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6 More

Normalised k3 Figure B3: Fitted normal distribution of normalised Cuplok joint stiffness k3 in bending about the horizontal axis

School of Civil Engineering Research Report No R893

40