The Eighth Asia-Pacific Conference on Wind Engineering, December 1 14, 13, Chennai, India Experimental Studies on Bolted Joint Damage Assessment due to Wind Load R. Balagopal 1, Ananth Ramaswamy 2, G. S. Palani 3 and N. Prasad Rao 4 1 Scientist, CSIR-SERC, Taramani, Chennai and Ph.D Scholar, IISc Bangalore, India, bala@serc.res.in 2 Professor, Department of Civil Engineering, Indian Institute of Science, Bangalore, India, ananth@civil.iisc.ernet.in 3 Sr. Principal Scientist and Head, Steel Structures Research Facility, CSIR SERC, Chennai, pal@serc.res.in 4 Principal Scientist, Tower Testing and Research Station, CSIR SERC, Chennai, nprao@serc.res.in ABSTRACT Bearing type bolted connections are normally used in transmission line towers. Damage in these bolted connections occur due to loosening of bolts because of reversal of stress under wind loads. Damage detection in these bolted joints is studied experimentally at component level and at structural level on a full scale truss. In the present study, element level experimental studies are carried out separately on bolted lap joint with steel plates of size 6 8 mm and steel angles of size 6 mm with 16mm diameter bolts. Damage is inflicted in these joints through variation in applied torque to the bolts. The slip load in the bolted joint increases by about %, for an increase in the torque from 3 to Nm applied to the bolts. From the experimental investigation on the full scale truss, change in stiffness ranging from 18 to 28% is noticed due to the damage inflicted in the bolted joint due to variations of to 1Nm in torque applied to the bolts. Keywords: Transmission Tower, Bolted joint, Damage Assessment, Torque, Wind load Introduction Transmission line (TL) towers are a lattice type structure consisting of leg, primary and secondary bracing, redundant and cross arm members. Bearing type bolted connections having nominal bolt-hole clearance of 1.mm is used for connections in TL towers. The joints are subjected to reversal of stress due to wind load. Damage in the bolted joints occur by loosening of bolts due to reversal of stresses. The leg members are connected through single or double cover bolted butt joint whereas gusset plates may or may not be used in the leg to bracing member connections. Extensive research has been carried out to study the influence of bolt slip on the tower deformation and its influence on ultimate strength behavior of TL towers (Kitipornchai et al. 1994). In contrast little attention has been given to study the influence of joint damage in TL towers. The damage in bolted joints is due to loosening or removal of bolts in the joints. In the present study, extensive research work has been carried out to study damage detection of bolted joints in TL towers. Element level experimental studies on lap joint made with steel plates of size 6mm 8mm and steel angles of size 6 mm with 16mm diameter bolts is carried out. The joint is subjected to tensile axial load and the loosening of bolts is simulated through a variation in the applied torque. The variation in bolt axial force for different applied torque (in the range of 1 to 11 Nm) is also recorded. It has been observed in tests that the slip and rotation of bolted joints occur, when the bolt clamping force is exceeded by the axial force in the joint, i.e. the frictional contact forces have been overcome. The clamping force in the bolted joint is calculated by Proc. of the 8th Asia-Pacific Conference on Wind Engineering Nagesh R. Iyer, Prem Krishna, S. Selvi Rajan and P. Harikrishna (eds) Copyright c 13 APCWE-VIII. All rights reserved. Published by Research Publishing, Singapore. ISBN: 978-981-7-811-1 doi:1.38/978-981-7-812-8 326 78
multiplying the axial force in the bolt with friction co-efficient times the number of bolts in the joint. Prasad Rao et al. (12) studied experimentally the variation in axial force for different applied torque to the bolts. The slip load varies with difference in applied torque to the bolts. Further, full scale experimental investigation is conducted on king post truss of m span with gusset plate bolted joints. The damage is inflicted through bolt loosening through variation in applied torque to the bolts. The change in the stiffness of the bolted joint due to variation in applied torque to the bolts is studied. Literature Survey Kirkegaard et al (1993), studied experimentally the damage in a steel lattice mast due to removal of lower diagonal member in natural wind load condition. The sensitivity of modal parameters to environmental wind load is investigated using ARMA model and found that the modal damping ratio vary by % and the torsional frequency vary by 1% for undamaged and damage state of the mast. From these experimental investigation it was concluded that the damage due to natural excitation by wind can be detected using effective system identification techniques. Kitipornchai et al. (1994), have investigated theoretically the effect of bolt slippage on the deflection and ultimate-strength response of lattice structures. Two idealized bolt-slippage models were presented. The effect of bolt slippage was incorporated into their nonlinear finite element computer program called as AK Tower. Two lattice structures are used to study the influence of bolt slippage. Results of this study indicated that while the slippage of bolts may have some effect on deflection, it does not significantly influence the ultimate strength of lattice structures. The study also showed that bolt slippage adds to the uncertainties in estimating structural deflection. Ungkurapinan et al. (3), have studied the joint slip in steel transmission towers. Joint slip is the relative displacement of a bolted joint under shear. Thirty-six joint tests were conducted and mathematical expressions were developed to describe the effect of bolt slip on the load-deformation behaviour of the bolted joints. Ju et al. (4) studied the behavior of butt jointed steel specimens. The bolt force was found be linear despite the steel plate in the joint reaching a non-linear state and the bolt failure was found to depend on the thickness of the connected plate. From the finite element analysis, it was concluded that linear elastic fracture mechanics can be used for the analysis of bolted joints. Prasad Rao et al. (12), studied experimentally the variation in bolt force with the applied torque and bolt slip on butt jointed specimen. Based on the bolt slip study conducted on seven transmission towers, a rotation factor was suggested for bolted joints. The rotation factor of.7 o at stub level and.34 o at all other levels of the bolted joints in the tower was suggested to modify the analytical deformation and have them match the corresponding experimental deformation values. Present Study Element level experimental studies have been conducted on lap joint made with steel plates of size 6 8 mm and steel angles of size ISA L 6 mm with single 16mm diameter bolt with 1.mm bolt-hole clearance. The schematic view of both the lap joint is shown in Fig.1. The joint is fixed at one end and is subjected to tensile load through manually operated mechanized hydraulic jack at the other end. The applied tensile load is measured 79
through digital wigameter. The lateral movement of the joint is measured through digital dial gauges. Typical experimental setup in the calibration test rig available is shown in Fig. 2. The bolts are aligned in the bolt hole with sufficient clearance in the hole and the bolt is tightened by using torque wrench to the required torque. The lap joint is subjected to tensile load upto 4 kn and the load vs lateral deflection is recorded. The bolt is loosened after this test, again placed in the bolt hole and then the bolt is tightened to the next level torque and the lateral deflection of the bolt for this applied torque is recoded. The variation in the slip load for the two different applied torque is shown in Fig 3 for bolts in the lap joint. From the load vs lateral deflection graph, it is observed that the slip load increases with increase in tightening of bolt. The slip in the joint occurs when the bolt clamping frictional force is exceeded by the axial force in the joint. Full scale experimental investigation is conducted on a king post truss of m span. The bottom and top chord members are ISA 6 mm double angle and all other members are ISA 4 4 mm double angle sections. The schematic view of the truss is shown in Fig. 4. The wind load is calculated based on IS 87 and the load points are designated as LP1 to LP. The boundary condition is chosen as hinged at one end and slotted end at other end. The truss is fabricated and assembled in the test pad as shown in Fig.. The load is applied in the horizontal direction through 1 kn load cell. The load is applied in steps of 1% till the design load of 2 kn is reached at load points 1,2,3 and 12. kn at load points 4 and through closed loop servo controlled hydraulic system. The joint deflection is measured using digital dial gauges and LVDT s at joints 1 to 6 at every stage of loading. The damage in the bolted joint is inflicted through bolt tightening at different torque levels. The objective of the investigation is to study the deflection behaviour of the truss for different bolt tightening conditions. All the bolts in the joints are tightened to Nm torque and the design load is applied at all load points. The load vs deflection is recorded. After this test, all bolts are loosened and aligned in the centre of bolt holes and all the bolts in the joints are tightened to next levels of torque such as 6, 8 and 1 Nm and the corresponding deflection is recorded for all the test conditions. Typical load vs deflection for all the joints is shown in Fig. 6 to Fig. 11 respectively. Results and Discussion The truss is subjected to design wind load of 4 kn at each of the joints and the change in joint stiffness for each level of bolt tightening is recorded. The joint stiffness increases in the order of 18 to 28% for all the bolted joints in the truss by varying the torque from Nm to 1 Nm. About % variation in stiffness is noticed near the hinged end of the truss in comparison to the joint near the slotted end. The bolted joint near the hinged end is observed to be stiffer than slotted end. From the component level study on bolted lap joint subjected to axial load of 4 kn, the slip load in the bolted joint increases by %, while increasing the torque from 3 Nm to Nm to the bolts in lap joint with steel plates and angles. This is due to the increase in the in-plane frictional force with the applied clamping force in the bolted joints. Once the clamping frictional force is exceeded the axial force in the joint, the slip and rotation of the bolts in the lap joint occurred. 76
Acknowledgements This paper is being published with the kind permission of Director, CSIR-Structural Engineering Research Centre, Chennai, INDIA. Fig.1 Schematic view of lap joint with plate and angles Fig.2 Typical experimental set up for lap joint 4 4 4 4 3 3 3 3 Load (kn) 2 Load (kn) 2 1 3 Nm Torque 1 Nm Torque Nm Torque 3 Nm Torque. 1 1. 2 2. 3 3. 4. 1 1. 2 2. 3 3. 4 Deflection (mm) Deflection (mm) Fig.3 Load vs lateral deflection plot for lap with plate and angles Fig.4 Schematic view of the experimental setup of full scale truss 761
Fig. Experimental set up for full scale truss 2 1 Nm Torque 4 Nm Torque 6 Nm Torque 1 Nm Torque 2 4 6 8 1 12 14 Deflection in mm Fig. 6 Load vs deflection of joint 1 (bottom chord near hinged end) 762
2 1 Nm Torque 6 Nm Torque 8 Nm Torque 1 Nm Torque 2 4 6 8 1 12 14 Deflection (mm) Fig. 7 Load vs deflection of joint 2 (mid span of bottom chord) 2 1 Nm Torque 6 Nm Torque 8 Nm Torque 1 Nm Torque 2 4 6 8 1 12 Deflection in mm Fig. 8 Load vs deflection of joint 3 (bottom chord near slotted end) 763
2 Nm Torque 6 Nm Torque 8 Nm Torque 1 Nm Torque 1-12 -1-8 -6-4 -2 Deflection in mm Fig. 9 Load vs deflection of joint 4 (top chord near hinged end) 2 Nm Torque 6 Nm Torque 8 Nm Torque 1 Nm Torque 1-14 -12-1 -8-6 -4-2 Deflection in mm Fig. 1 Load vs deflection of joint (mid span of top chord) 764
2 Nm Torque 6 Nm Torque 1 Nm Torque 8 Nm Torque 1-12 -1-8 -6-4 -2 Deflection (mm) Fig. 11 Load vs deflection of joint 6 (top chord near slotted end) References Kirkegaard P. H and Rytter. A., (1993), An Experimental Study of a Steel Lattice Mast under Natural Excitation, Paper No: 46, IUTAM Symposium on Identification of Mechanical Systems, Wuppertal, Germany. Kitipornchai S., AI-Bermani F.G.A. and Peyrot A.H. (1994), Effect of bolt slippage on ultimate behavior of lattice structures, Journal of Structural Engineering, Vol. 1, No. 8, pp. 2281-2287. Ungkurapinan N., De S.R., Chandrakeerthy S., Rajapakse R.K.N.D. and Yue S.B. (3), Joint slip in steel electric transmission towers, Engineering Structures, Vol. 2, pp.779-788. Ju S.H., Fan C.Y. and Wu G.H. (4) Three dimensional finite element analysis of steel bolted connections, Engineering Structures, Vol. 26, pp. 43-413. Prasad Rao N., Samuel Knight G. M. Lakshmanan N and Nagesh. R. Iyer (12), Effect of Bolt Slip on Tower Deformation, Journal of Practice Periodicals on Structural Design and Construction (ASCE), Vol. 17, No. 2, pp 6-73 -------- 76