Scholarly article on topic 'Experimental study of stiffening rings reinforced tubular T-joint with precompression chord'

Experimental study of stiffening rings reinforced tubular T-joint with precompression chord Academic research paper on "Materials engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia Engineering
OECD Field of science
Keywords
{"tubular T-joint" / impact / "compression load" / "stiffening ring" / "impact resistance"}

Abstract of research paper on Materials engineering, author of scientific article — Xing Su, Kang Gao, Hui Qu

Abstract This paper examines the dynamic behaviour of stiffening rings reinforced tubular T-joints with precompression chord by means of experimental studies. An experimental programme involved one test on unstiffened tubular T-joint and three on stiffened configurations is carried out through a high-performance drop hammer machine. A series of disc springs is installed to apply axial compression to the chord. The dynamic response of four tubular T-joints are described and discussed with emphasis on the dimensional parameters of stiffening rings. Based on the experimental results, the key behavioural patterns including the development of impact force, deformation and strain, as well as deformation modes are identified. In general, the presence of stiffening rings significantly improves the impact resistance of tubular T-joints.

Academic research paper on topic "Experimental study of stiffening rings reinforced tubular T-joint with precompression chord"

ELSEVIER

Available online at www.sciencedirect.com

ScienceDirect

Engineering

Procedía

Procedía Engineering 210 (2017) 278-285

www.elsevier.com/locate/procedia

6th International Workshop on Performance, Protection & Strengthening of Structures under Extreme Loading, PROTECT2017, 11-12 December 2017, Guangzhou (Canton), China

Experimental study of stiffening rings reinforced tubular T-joint with

precompression chord

This paper examines the dynamic behaviour of stiffening rings reinforced tubular T-joints with precompression chord by means of experimental studies. An experimental programme involved one test on unstiffened tubular T-joint and three on stiffened configurations is carried out through a high-performance drop hammer machine. A series of disc springs is installed to apply axial compression to the chord. The dynamic response of four tubular T-joints are described and discussed with emphasis on the dimensional parameters of stiffening rings. Based on the experimental results, the key behavioural patterns including the development of impact force, deformation and strain, as well as deformation modes are identified. In general, the presence of stiffening rings significantly improves the impact resistance of tubular T-joints.

© 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the scientific committee of the 6th International Workshop on Performance, Protection & Strengthening of Structures under Extreme Loading

Keywords: tubular T-joint, impact, compression load, stiffening ring, impact resistance.

1. Introduction

Tubular structures have been widely used in industrial building, railway station, offshore platform, bridge and breakwater due to their inherent architectural and structural advantages, but they are likely to be subjected to impact loads in the events of explosion and fire during their service period.

Corresponding author. Tel.: +86-153-1863-9025; fax: +86-0535-6902606. E-mail address: quhuiytu@gmail.com

1877-7058 © 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the scientific committee of the 6th International Workshop on Performance, Protection & Strengthening of Structures under Extreme Loading. 10.1016/j.proeng.2017.11.078

Xing Sua, Kang Gaoa, Hui Qua*

College of Civil Engineering, Yantai University, Yantai 264005, China

Abstract

During the past decades, a large number of studies have been carried out on tubular truss under both static and impact loading. Guedes Soares and Soreide [1], Bai and Pedersen [2], Cerik et al. [3], Buldgen et al. [4] and Cerik et al. [5] carried out experimental or numerical studies on the behaviour tubular frame, where the global behaviour of the platform under multiple impact points and various collision angles as well as the ship inelastic deformations were taken into account.

The tubular joints are the critical structural components to transfer loads in the tubular structure. Yu et al. [6] and Qu et al. ([7], [8]) carried out a series of experimental and numerical studies on the dynamic behaviours of tubular T/K-joint under different impact loading conditions to investigate the deformation modes and impact resistance of

tubular joints.

At present, various approaches are available to enhance the strength of tubular joints, such as the addition of ring stiffeners, double plate, collar plate. However, most of the researches focused on the static performance of different joint configurations (Choo et al. [9], Gaoetal. [10], Chen et al. [11]). Few reports were presented to investigate the dynamic performance of reinforced tubular joint under impact loading.

Moreover, it is evident that the offshore legs and building columns always bear axial load induced by live and dead load of slab or deck simultaneously during the impact accidents. Zeinoddini et al. [12-14], Wang et al. [15] and Yousuf et al. [16] investigated the impact performance of pre-loaded mild or stainless steel tube subjected to lateral impact by means of experimental and finite element methods. However, in the available literatures, limited research lias emphasized on the impact performance of tubular joints considering the axial compression in the chord.

To this end, this paper examines dynamic behaviour of collar plate reinforced tubular T-joints with precompression chord by means of experimental studies. An experimental programme involved one test on unstiffened tubular T-joint and three on stiffened configurations is carried out through a high-performance drop hammer machine. The dynamic response of four tubular T-joints is investigated with emphasis on the effect of dimensional parameters of stiffening

2. Experiment investigation

2.1. Specimen details

According to the geometry requirement recommended by Choo et al. [9] and design code of API [17], three tubular T-joints reinforced by stiffening rings and one without reinforcement were designed to experimentally study their behaviour under impact loading. Figure 1(a) and (b) illustrates the configurations of the tubular T-joints with or without stiffening rings, respectively. The geometric details of each specimen are listed in Table 1, where T in the reference number denotes an unstiffened tubular T-joint, while CT represents a collar plate reinforced tubular T-joint. D, T, L are the outer diameter, thickness and effective length of the chord, respectively, while d, t, 1 are the outer diameter, thickness and length of the brace, respectively, rw, rt and n are the width thickness and amount of the stiffening rings. In addition the testing information was also summarized in Table 1, where m and v represent the weight and impact velocity of drop hammer. No is the precompressive load in the chord, and A and 5 represent the indentation depth and expansion width of the impact cross-section after testing, which are measured from chord mid-span section after testing, respectively.

Figure 1 Configuration of K-j oints (a) Stiffened T-joint (b) Unstiffened T-joint

The chords and braces of T-joints were made of hot-rolled seamless steel tubes, and they were connected by using partial penetration welds. Strip coupons cut from of the steel tubes were tested in tension. The obtained yield strength (fy) of 6 mm thick chord and brace is 349 MPa, and the ultimate strength (fu) is 539 MPa, while the measured yield strengths (fy) of 4 mm and 6 mm thick stiffening ring are 325 MPa and 303 MPa, and the ultimate strengths (fu) are 490 MPa and 470 MPa, respectively.

Table 1 Details of testing specimens

Chord Brace Stiffening ring

(mm) No A 5

Reference Type DXTXL (mm) rfx/x/ (mm) Fw rt _ v (m/s) (kg) (kN) (mm) (mm)

mm mm n

T-l Unstiffened 180x6x1970 68x6x600 - - 445 8 222 50 207

RT-1 stiffened 180x6x1970 68x6x600 30X4 2 445 8 222 27 195

RT-2 stiffened 180x6x1970 68x6x600 15X6 2 445 8 222 38 200

RT-3 stiffened 180x6x1970 68x6x600 30X6 2 445 8 222 18 191

2.2 Testing setup

The impact tests were conducted by employing the drop hammer machine at the Center for Integrated Protection Research of Engineering Structures (CIPRES) in Hunan University, and the details of this testing machine were described elsewhere [13]. Figure 2 shows the general view of impact test set-up, including a drop hammer facility, a U-shaped rigid base frame and the self-reacting system of disc springs. The supports at both ends of the chord provided rigid boundary conditions, but allowed chord to freely translate in the axial direction at one end. The axial springs were employed to impose a 222kN axial load to the specimen and keep it as a constant during the impact loading, as shown in Figure 2.

Figure 2 Test setup and corresponding details

In addition, a piezoelectric force transducer was installed in the hammer to measure the impact force, and some strain gauges and displacement transducers are located around the specimens, as depicted in Figure 3. Ten strain gauges were mounted onto the chord, brace and brace-to-chord intersection zone in order to monitor their strain development during impact loading, while four displacement transducers (Dl, D2, D3 and D4) were placed on the impacted cross-section in the middle of chord to record the vertical displacement of the brace near the joint zone, the central-axis horizontal and vertical deformation of mid-span chord, and the vertical deformation of chord bottom surface, respectively. When the desired compression load was reached, a drop hammer of 445 kg weight was lifted to the designed height and then released to impact the top end of brace with the initial velocity of 8 m/s.

3. Deformation patterns

Figure 4 shows the deformation patterns of four testing specimens. It can be observed from Figure 4 that no bending deformation occurred on the braces of four specimens subjected to impact load, and no evident bulge deformation can be seen at the end of brace near the joint zone. For unstiffened specimen, it failed due to the indentation of chord top surface along the intersection line between chord and brace. However, for stiffening rings reinforced specimens, they failed attributed to the indentation of chord top surface along the intersection line between chord and brace together with buckling of stiffening rings.

leading-out plate

(a) (b)

Figure 3 Layout of measure equipment (a) Strain gauges (b) Displacement transducers

Comparing the deformation patterns of unstiffened (Specimen T-l) and stiffening rings reinforced specimens (Specimens RT-1, RT-2 andRT-3) as shown inFigure 4, it is clearly observed that not only the local indentation depth and area of chord top surface but also the transverse bulge of chord for stiffened specimens is much smaller than that for unstiffened specimen. For specimens RT-1 and RT-2 within the same width but different thickness of stiffening rings as shown in Figure 4 (b) and (d), it is found that the thinner the thickness of stiffening ring was, the larger the indentation depth of chord top surface was, the flatter the impacted cross-section of chord was. For Specimens RT-2 and RT-3 (as shown in Figure 4 (c) and (d)) within the same thickness but different width of stiffening rings, there is no evident difference between the deformation patterns of these two joints except that the indentation of chord top surface increased with the increasing of the width of stiffening rings. Additionally, it can be observed from Figure 4 that, for all reinforced joints, the stiffening rings buckled at the upper part connected with the chord. .

As illustrated in Figure 4 (b) and Table 1, it is found that the enhancement of stiffening rings for Specimens (RT-1, RT-2 and RT-3) resulted in reduction of the deformation in the chord bottom surface compared with that of unstiffened joint (T-l). The indentation of chord top surface along the longitudinal section decreased with the increase of thickness and width of stiffening rings. For the unstiffened Specimen (T-l), the maxim indentation occurred in the mid span of impacted chord, however, for all stiffened Specimens (RT-1, RT-2 and RT-3), the maximum indentation were all located adjacent to the stiffening rings. Moreover, as illustrated in Table 1, the expansion width of the impacted chord cross-section decreased due to the use of collar plate for the tubular T-joint.

(b) (c) (d)

Figure 4 Tested and modeled deformed shapes of steel tubular K-joints witli/without internal ring (a)T-l (b) RT-1 (c) RT-2. (d) RT-3

4. Test results

4.1 Time history of impact force and displacement

The time history curves of impact force and displacement for all specimens are presented in Figure 5. In addition, on the basis of the generally accepted definitions for local indentation damage and global bending damage, the local indentation depth (A) can be defined approximately as the difference between the displacements of Transducer D1 and Transducer D3, and the bending deflection of chord can be obtained directly from the measurement of Transducer D2.

In order to clearly demonstrate the process of impact loading, the curves were divided into four stages on the base of the development characteristics of impact force: (1) for both unreinforced and reinforced specimens, Stage I is impact stage, where the impact force increased rapidly to a peak value, and then gradually reduced from its peak value to the smallest, but little displacement developed during the period of impact oscillation. (2) Stage II is named as fluctuant stage. The impact force fluctuantly increased to another peak point as a result of the increasing contact area between the drop hammer and the brace, and the inertial force plays an important role in this stage. The top surface of chord dented under impact loading which lead the Displacement transducer (D1) to move downward. The bulge of chord resulted in the horizontal displacement of the Displacement transducer (D2) increasing and the bottom surface moving upward, so there is an obvious slip stage on the displacement time history curve of Displacement transducer (D3). The Displacement transducer (D4) deformed downward under impact inertial force at the beginning of the impact loading. Due to the accuracy of test equipment, the time history curve of local dent, which is denoted as the displacement difference between Displacement transducer (D1) and Displacement transducer (D4), fluctuated slightly around original point. (3) Stage III, stable stage, where impact force tended to be stable, and the impact load-time curve appeared to be flat, while the vertical displacement of the Displacement transducer (D1, D3 and D4) incrementally reached to its maximum value. It can be clearly noticed that the local dent on the top surface of chord (the displacement difference between the Displacement transducer (D1) and (D4)) was obviously more than that of the Displacement transducer (D3), which means that the local dent on the top surface of chord plays an important roles in the impact loading. The tension member force occurred in the j oint zone made the impact force increase slowly to its maxim value. (4) Stage IV, decaying stage, where the impact force reduced to zero, and the displacement also slightly decreased to some level.

For the stiffening rings reinforced specimens, with the increasing of the width and thickness of the stiffening rings, the time duration decreased evidently, and the impact loading can also be divided into four stages. Compared with the unreinforced specimen, only one or two big fluctuations appeared on the time history curve of impact force in stage II for the stiffening rings in the joint zone buckled successively. The vertical displacements of the Displacement transducer (D1, D3 and D4) were much less than those of unreinforced specimen. On the contrary of the unreinforced specimen (T-1), no member tension force appeared in the dented joint zone, and the impact force decreased slowly with the development of deformation after the impact fore reached to its peak load in Stage III until the drop hammer left the end of brace.

For Specimen (RT-1 and RT-3) with the same width but different thickness of stiffening rings, it is demonstrated in Figure 5(b and d) that the thicker the stiffening rings are, the bigger the stiffness of joint is, and the higher the impact force is, and the smaller the deformation is. Although there are some the acquisition failure in Stage II and III, the bending deformation of both specimens are all bigger than their local indentation on the top surface of chord, which is as opposite to the unreinforced specimen (T-1). For Specimen (RT-2 and RT-3) with the same thickness but different width of stiffening rings, it is depicted in Figure 7(c and d) that, for specimen (RT-2), when the width of stiffening ring is about D/10 (D denotes the diameter of the chord), the deformation is a little smaller and the impact force is 10% higher than those of unreinforced specimen (T-1). The local indentation on the top surface of chord is bigger than the bending deflection, which is the same as the deformation development of unreinforced specimen (T-1). For specimen (RT-2), when the width of stiffening ring is about D/6, the global stiffness increased evidently. Compared to unreinforced specimen (T-1), the impact deformation decreased about 33%, and the impact force increased about 3 3 %. The bending deflection is bigger than the local indentation on the top surface of chord. It can be concluded that changing the width of stiffening ring can increased the impact performance evidently than changing the thickness of stiffening ring, but the different width of stiffening ring corresponds to different failure mode.

lii \

• I s-

i -»rF ii

;i i: :11

(b) (C)

Figure 7 Time history curves of impact force and displacement for all T-joints (a)Specimen T-l (b) Specimen RT-l(c) Specimen RT-2 (d) Specimen RT-3

4.2 Time history of strain and strain rate

The time history curves of strain and strain rate are demonstrated in Figure 6. It can be observed from Figure 6 (ai, bi, ci, and di) that either unreinforced specimen or reinforced specimens, the deformations mainly happened in Stage I , Stage II and the beginning of Stage III.

It is known that the yield stress of steel is sensitive to strain rate [18], and the yield stress of steel increases with the increase of strain rate. The Cowper-Symonds constitutive equation is used in this paper to simulate the strain-rate sensitive behaviour of steel. The Cowper-Symonds constitutive equation is given as follow:

DIF = l*L=l + (±y,

o dy is the dynamic yield stress at a uniaxial plastic strain rate, " y is the

Where DIF is dynamic amplification factor, corresponding static yield stress, and e is strain rate for steel, D is a rate-dependent constant and n is a positive dimensionless parameter. According to previous study elsewhere (Abramowicz and Jones [19), D is set as 40.4 s1, and n is taken as 5 for mild steel. Based on the calculated dynamic yield stress and elastic modulus, the yield strain can be easily obtained. It canbe clearly known from Figure 8(aii, bii, cii, and dii) that most strain rate of four specimens are about±5 s1 and below ±20 s1. According to Equation (1), it is found that when e = ±5 s1, DIF is equal to 1.66 and the dynamic yield strain is 2800 p:. Similarly, when e = ±20 s"1, DIF is evaluated as 1.87 and the dynamic yield strain is equal to 3150 p:.

It canbe seen from Figure 6 (ai) that, for unreinforced tubular joint (T-l), the plastic deformation mainly focused on the joint zone between chord and brace, and the steel plate away from the joint zone was not yield, and the circumferential deformation of Strain gauge (S3 and S4) mounted on the top surface of chord around the joint zone was more than that of Strain gauge (S7 and S8) mounted on the side surface of chord, but the longitudinal deformation of Strain gauge (S10) mounted on the bottom surface of chord was in elastic, but it is nearly yield. For the reinforced tubular joints (RT-1, RT-2, and RT-3), except for the Strain gauges (S2, S7 and S8), the longitudinal deformation of Strain gauge (S10) was more than the yield strain, and the deformation increased with the increasing of the width and thickness of stiffening rings. Based on the phenomena depicted above, it can be concluded that, under the same impact load, the bending deformation of impact chord played important role in the deformation development of stiffening rings reinforced tubular joint, however, the failure mode of unreinforced tubular joint would be the combination of the local indentation on the top surface of chord and bending deflection of the chord.

5. Conclusions

The impact behaviour of collar plate reinforced tubular T-joints has been examined by means of experimental studies. One impact test on unreinforced tubular T-joints and three on reinforced configurations have been described in details. The key behavioural patterns including the development of impact force, deformation and strain were discussed.

Changing the width of stiffening ring can increased the impact performance evidently than changing the thickness of stiffening ring, but the different width of stiffening ring corresponds to different failure mode. Under the same impact load, the bending deformation of impact chord played important role in the deformation development of stiffening rings reinforced tubular joint, however, the failure mode of unreinforced tubular joint would be the combination of the local indentation on the top surface of chord and bending deflection of the chord.

It was found that the time history of impact loading and displacement for all the cases can be divided into four stages: impact stage, fluctuant hardening stage, stable stage and decaying stage. The presence of stiffening rings significantly improved the impact resistance of tubular T-joints.

Acknowledge

The work was funded by National Natural Science Foundation project (51478407), and particular thanks to the Center for Integrated Protection Research of Engineering Structures (CIPRES), Ministry of Education Key Laboratory of Building Safety and Efficiency of the Hunan University.

References

[1] C. Guedes Soares, T.H. Soreide. Plastic analysis of laterally loaded circular tubes. J. Struct Eng 1983; 109: 451-67.

[2] Y. Bai, P.T. Pedersen. Elastic-plastic behavior of offshore steel structures under impact loads. Int. J. Impact Eng 1993;13:99-115.

[3] Cerik BU, Shin HK, Cho S-R. On the resistance of steel ring-stiffened cylinders subjected to low-velocity mass impact. Int. J. Impact Eng 2015; 84:108-23.

[4] L. Buldgen, H. Le Sourne, T. Pire. Extension of the super-elements method to the analysis of a jacket impacted by a ship. Mar Struct 2014; 38: 44-71.

[5] B.C. Cerik, H.K. Shin, S.R. Cho. A comparative study on damage assessment of tubular members subjected to mass impact. Mar Struct 2016; 46: 1-29.

[6] W.J. Yu, J.C. Zhao, H.X. Luo, et al. Experimental study on mechanical behavior of an impacted steel tubular T-joint in fire. J.ConstrSteel Res

2011; 67(9): 1376-85.

[7] H. Qu, J.S. Huo, C. Xu, et L. Numerical studies on dynamic behavior of tubular T-joint subjected to impact loading. Int. J. Impact Eng. 2014,

67:12-26.

[8] H. Qu, Y.F. Hu, J.S. Huo, et al. Experimental study on tubular K-joints under impact loadings. J. Construct. Steel Res 2015; 122(9): 22-9.

[9] Y.S. Choo, J.X. Liang, G.J. Van der Vegte, et al. Static strength of collar plate reinforced CHS X-joints loaded by in-plane bending. J Construct

Steel Res 2004, 60(5): 1745-60.

[10] F. Gao, X. Guan, H. Zhu, X. Liu. Fire resistance behaviour of tubular T-joints reinforced with collar plates. J Construct Steel Res 2015, 115(12):106-20.

[11] C. Chen, Y. Shao, J. Yang. Study on fire resistance of circular hollow section (CHS) T-joint stiffened with internal rings. Thin Wall Struct 2015. 92(7): 104-14.

[12] M. Zeinoddini, J.E. Harding, G.A.R. Parke. Dynamic behaviour of axially pre-loaded tubular steel members of offshore structures subjected to impact damage. Ocean Eng 1999; 26(10): 963-78.

[13] M. Zeinoddini, G.A.R. Parke, J.E. Harding. Axially pre-loaded steel tubes subjected to lateral impacts: An experimental study. Int. J. Impact Eng 2002; 27(6):669-90.

[14] M. Zeinoddini, J.E. Harding, G.A.R. Parke. Axially pre-loaded steel tubes subjected to lateral impacts (a numerical simulation). Int. J. Impact Eng 2008; 35(11): 1267-79.

[15] R Wang, L.H. Han, C.C. Hou. Behavior of concrete filled steel tubular (CFST) members under lateral impact: Experiment and FEA model. J. Construct Steel Res 2013; 80(1): 188-201.

[16] M. Yousuf, B. Uy, Z. Tao, et al. Impact behavior of pre-compressed hollow and concrete filled mild and stainless steel columns. J. Construct Steel Res 2014, 96(5): 54-68.

[17] American Petroleum Institute (API), Recommended Practice for Planning , Designing and Constructing Fixed Offshore Platforms- Working Stress Design, 21st Ed., API Recommended Practice 2A WSD (RP 2A WSD), Washing ton, DC, 2000.

[18] L.J. Malvar, C.A. Ross. Review of strain rate effects for concrete in tension. ACI Mater J 1998; 95(6): 735-39.

[19] W. Abramowicz, N. Jones. Dynamic axial crushing of square tubes. Int. J. Impact Eng 1984; 2(2): 179-208.