230 R.Q. Bridge strength and mode of failure of the pin or plate. The results have been compared with design values from steel design standards and modifications to the design procedures have been proposed. TEST PROGRAMME The first series of test specimens consisted of a snug-fit single pin loaded in double shear by an interior plate between two exterior cover plates as shown in Figure 1. The lower half of the specimen was bolted and was designed to have a greater capacity than the pin. The specimens were tested to failure under load control in a 580 kN capacity tensile testing machine. Deformations of the interior pin plate relative to the exterior cover plates were measured. The main variables tested were the pin diameter df (10, 16 and 27mm), the interior plate thickness tp (3, 6, 10, 16 and 20mm) and the material properties of the pin. The pins were cut from two types of commercially available steel rod: black rod with a high ductility and low ratio of yield strength fyp to ultimate tensile strength fuf; and bright rod with a lower ductility. Typical stress-strain curves are show in Figure 2. The plate steel had a similar behaviour to the black pin. The distance from the pin to the edge of the plate in the direction of loading was kept generally within the limits of AS 4100-1998 for end plate tear-out to limit this mode of failure. '.] t ~- "1 100 50 95 T Top interior test plate Width = 100mm Test pin Cover plates t = 12mm 50 70 [ [ [ 1[ ~ Two M20 8.8 bolts 50 100 Bottom interior plate All dimension in mm Figure 1. Test specimen for double shear 600 The Design of Pins in Steel Structures 400 r~ \ 16mm bright pin m 200 16mm black pin 0.0 1.0 2.0 3.0 Strain % 231 Figure 2. Stress-strain curves for the two types of steel Hayward and van Ommen (1992) conducted the tests. The geometrical properties, material properties, test results and modes of failure for the 18 test specimens are shown in Table 1. The primary mode of failure was either shearing of the pin (shear deformation of the pin generally being 25% of the pin diameter or greater) or large bearing deformations of the plate (63% of the pin diameter or greater). In some cases, large plate bearing deformations were observed prior to final shearing of the pin. In other cases, fracture of the plate occurred at the cross-section through the pin. These failures have been labelled as secondary modes of failure. Pin bearing failures were not observed. TABLE 1 THE GEOMETRICAL AND MATERIAL PROPERTIES AND THE TEST RESULTS OF THE PIN TEST SPECIMENS Test Pin No. df Innl 1 10.06 2 10.04 3 10.06 4 16.13 5 16.14 6 16.13 7 26.95 8 26.95 9 26.95 10 9.97 11 10.09 12 10.00 13 15.97 14 15.97 15 15.97 16 26.90 17 26.90 18 26.90 Pin Pin Plate Plate Plate ~f ~f thick f~ ~p MPa MPa mm MPa MPa 250 455 3.12 360 496 250 455 5.97 310 469 250 455 9.85 260 485 300 499 3.23 360 496 300 499 10.05 260 485 300 499 15.86 250 460 270 485 3.12 360 496 270 485 9.9 260 485 270 485 19.93 250 446 480 558 3.14 360 496 480 558 6.12 310 469 480 558 10.11 260 485 460 523 3.16 360 496 460 523 9.85 260 485 460 523 15.9 250 460 450 524 3.12 360 496 450 524 10.17 260 485 450 524 19.87 250 446 *Pin also sheared 25% of diameter Max. Hole Load Elong. kN 53.6 54.0 54.3 97.0 150.8 146.5 113.0 346.0 344.0 53.6 56.8 56.4 92.5 137.0 131.0 110.0 352.0 350.0 Primary Failure % 63.4 Pin shear 7.9 Pin shear 0.4 Pin shear 126.4 Plate bearing 7.5 Pin shear 2.3 Pin shear 69.7 Plate bearing 68.6 Pin shear 4.6 Pin shear 72.3 Pin shear 8.9 Pin shear 1.5 Pin shear 167.6 Plate bearing 7.0 Pin shear 9.2 Pin shear 78.7 Plate bearing 125.7 Plate bearing 5.7 Pin shear Secondary Failure Plate beating Plate fracture Plate bearing Plate bearing Plate fracture Plate fracture* 232 R.Q. Bridge Typical load-deformation curves are shown in Figure 3 for 10mm and 16mm diameter bright pins in three different thicknesses of plate. The 16mm diameter pin in 3mm plate (specimen 13) exhibited a primary plate bearing failure whereas the 10mm pin in 3mm plate (specimen 10 exhibited a secondary bearing failure. With plate bearing failures, hole elongations in excess of 60% of the hole diameter are attained. The other specimens shown in Figure 3 exhibited pin shear failures. Pin shear failures are associated with shear deformation through the pin itself of 25% of the pin diameter or more prior to failure, even for the pins manufactured from bright steel with a lower ductility than the black steel. 150 lOO o ~f 50 "- 9 AI m r 3mm piate, 16mm pin - 10mm plate, 16mm pin x 16mm plate, 16mm pin = 3mm plate, 10mm pin o 6mm plate, 10mm pin ,,- - 10mm plate, 10mm pin 0 0 5 10 15 Deformation (ram) Figure 3. Typical load-deformation behaviour for pin shear and plate bearing failures The second test series under deformation control was conducted by Sukkar (1998). This examined the effect of the shape of eye-bars (see Figure 4), typically used at the end of tension members, on end tear- out failures. Standards such as AS4100-1998, BS5950-1990 and Eurocode 3-1992 require an elongated end on the eye-bar (D3 > D2) whereas AISC-1993 permits a simpler circular eye bar end (D3 = D2). The eye-bar dimensions, material properties and test results are shown in Table 2. D1 to Figure 4. Typical shape of eye-bars at end of pinned tension members The Design of Pins in Steel Structures TABLE 2 DIMENSIONS AND MATERIAL PROPERTIES AND TEST RESULTS FOR EYE-BAR SPECIMENS 233 Test Head Pin Plate No. Type df tp ITlln mnl 19 Elong. 20.00 5.0 20 Circ. 20.00 5.0 21 Elong. 20.00 6.0 22 Circ. 20.00 6.0 23 Elong. 20.00 8.0 24 Circ. 20.00 8.0 25 Elong. 27.00 5.0 26 Circ. 27.00 5.0 27 Elong. 27.00 6.0 28 Circ. 27.00 6.0 29 Elong. 27.00 8.0 30 Circ. 27.00 8.0 *Failure not reached. Instron Eye-bar dimensions Pin Pin Plate Plate Max. D1 DE O3 fyf fuf fyp fup Load mm mm mm MPa MPa MPa MPa kN 22.5 15.0 22.5 730 870 280 440 46.9 22.5 15.0 15.0 730 870 280 440 44.4 22.5 15.0 22.5 730 870 280 440 52.6 22.5 15.0 15.0 730 870 280 440 47.0 22.5 15.0 22.5 730 870 280 440 53.6 22.5 15.0 15.0 730 870 280 440 52.7 30.0 20.0 30.0 730 870 280 440 51.4 30.0 20.0 20.0 730 870 280 440 53.0 30.0 20.0 30.0 730 870 280 440 54.4 30.0 20.0 20.0 730 870 280 440 54.0 30.0 20.0 30.0 730 870 280 440 62.2 30.0 20.0 20.0 730 870 280 440 63.1 6027 testing machine disabled by frame error Failure mode Tear-out Tear-out Tear-out Tear-out Deform* Deform* Tear-out Tear-out Tear-out Tear-out Deform* Deform* under deformation control. COMPARISON WITH DESIGN METHODS The possible of modes of failure considered by most design codes are shown in Figure 5. Figure 5. Modes of failures in pin connections. The dimensions of eye-bars and the design strengths for the conditions of pin shear, pin bearing and plate bearing according to Australian, European, British and American practice are listed in Table 3. TABLE 3 COMPARISON OF DESIGN STRENGTHS IN STEEL CODES AND SPECIFICATIONS Steel code tp D2 D3 D4 Pin shear Pin bearing Plate bearing AS4100-1998" _>0.25D2 20.67D, 2 1.OD1 _> 1.ODI VI = 0.62fyiA I Vb = 1.4frrdrt p Vb = 3.2f, pdrt p Eurocode 3-1992 n.a. _>0.75dp _> 1.1dp 21.1dp Vr = 0.60furAr Vb = 1.5f~rdrt p Vb = 1.5fypdrt p BS5950-1990 !_>0.25D2 20.67D~ 2 1.OD~ _> 1.OD~ Vr = 0.60frrAr Vb = 1.2fedrtp Vb - 1.2fypdrt p AISC-1993 20.12D~ 20.67D~ =l.OD2 n.a. Vt = 0.60feAr Vb = 1.4f~rdrt p Vb = 1.4fypdtt p *In addition, AS4100-1998 requires Vb -f, paetp for plate tear-out where ae is the clear distance from the pin to the edge of the plate in the direction of loading. 234 R.Q. Bridge In Table 3, Ay is the cross-sectional area of the pin, df is the diameter of the pin, frY is the yield stress of the steel in the pin, fuy is the ultimate tensile strength of the steel in the pin, tp is the thickness of the load-bearing plate, fyp is the yield stress of the steel in the plate, and f,p is the ultimate tensile strength of the steel in the plate. Most codes are similar with two major exceptions: Eurocode 3 uses the ultimate tensile strength of the pin in calculating the shear strength of the pin (similar to that for bolt strength in most steel codes); and AS4100-1998 uses the ultimate tensile strength of the plate (and a large factor of 3.2) in calculating the bearing strength of the plate. Therefore only AS4100-1998 and Eurocode 3-1992 are considered in the following comparisons of codes with the test strengths. TABLE 4 COMPARISON OF TEST RESULTS WITH DESIGN VALUES PREDICTED BY AS4100-1998. Test Max Vf Load/Vf Vb Load/Vb Vb Load/Vb Vb Load/Vb No. Load Pin Pin Pin§ Pin§ Plate Plate Tear-out Tear-out , kN , kN , kN kN ~ , kN , 1 53.6 24.6 2.18" 11.0 4.88 49.8 1.08 139.2 0.38 2 54.0 24.5 2.20* 21.0 2.57 90.0 0.60 251.9 0.21 3 54.3 24.6 2.20* 34.7 1.57 153.8 0.35 429.8 0.13 4 97.0 76.0 1.28" 21.9 4.43 82.7 1.17 139.3 0.70 5 150.8 76.1 1.98" 68.1 2.21 251.7 0.60 423.7 0.36 6 146.5 76.0 1.93" 107.4 1.36 376.6 0.39 634.2 0.23 7 113.0 191.0 0.59 31.8 3.56 133.5 0.85* 126.2 0.90 8 346.0 191.0 1.81" 100.9 3.43 414.1 0.84 391.4 0.88 9 344.0 191.0 1.80" 203.0 1.69 766.6 0.45 724.7 0.47 10 53.6 46.5 1.15" 21.0 2.55 49.7 1.08 140.2 0.38 11 56.8 47.6 1.19" 41.5 1.37 92.7 0.61 258.2 0.22 12 56.4 46.7 1.21" 67.9 0.83 156.9 0.36 i 441.3 0.13 13 92.5 114.3 0.81 32.5 2.85 80.1 1.15" 136.4 0.68 14 137.0 114.3 1.20" 101.3 1.35 244.1 0.56 415.7 0.33 15 131.0 114.3 1.15" 163.5 0.80 373.8 0.35 636.4 ! 0.21 16 110.0 317.1 0.35 52.9 2.08 133.2 0.83* 126.2 0.87 17 352.0 317.1 1.11" 172.4 2.04 424.6 0.83 402.2 0.88 18 350.0 317.1 1.10" 336.7 1.04 762.8 0.46 722.7 0.48 19 46.9 284.4 0.16 102.2 0.46 140.8 0.33 49.5 0.95* 20 44.4 284.4 0.16 102.2 0.43 140.8 0.32 33.0 1.35" 21 52.6 284.4 0.18 122.6 0.43 169.0 0.31 59.4 0.89* 22 47.0 284.4 0.17 122.6 0.38 169.0 0.28 39.6 1.19" 23 53.6 284.4 0.19 163.5 0.33 225.3 0.24 79.2 0.68* 24 52.7 284.4 0.19 163.5 0.32 225.3 0.23 52.8 1.00" 25 51.4 518.3 0.10 138.0 0.37 190.1 0.27 66.0 0.78* 26 53.0 518.3 0.10 138.0 0.38 190.1 0.28 44.0 1.20" 27 54.4 518.3 0.10 165.6 0.33 228.1 0.24 79.2 0.69* 28 54.0 518.3 0.10 165.6 0.33 228.1 0.24 52.8 1.02" 29 62.2 518.3 0.12 220.8 0.28 304.1 0.20 105.6 0.59* 30 63.1 518.3 0.12 220.8 0.29 1304.1 0.21 70.4 0.90* * Asterisk indicates mode of failure predicted by the AS4100-1998 + Pin bearing ignored in predicting failure as none was observed in tests The predicted failure modes from Table 3 compare well with actual failure modes in Table 1. However, for test specimens 1, 2, 3, 5, 6, 8, 9, 10, 1 l, 12, 14, 15 and 18 where the actual primary failure was by pin shear, the strength of the pin in shear predicted by AS4100-1998 was markedly lower than the test strengths, particularly for the ductile pins made from black steel rod. For the test specimens l, 4, 7, 8, 10, 13, 16 and 17 where the primary or secondary failure was by bearing of the The Design of Pins in Steel Structures 235 plate, the strength predicted by AS4100-1998 was close to the actual test strengths. The predicted bearing strengths for specimens 7, 16, and 17 appear a little high because the full bearing strength of the plate was not attained in the test due to premature fracture of the plate adjacent to the hole. For the eye-bars where plate tear-out was both the predicted and the actual mode of failure, AS4100-1998 provided a reasonable estimate of the test strength taking the edge distance a3 = D3. However, it is interesting to note that the elongation of the eye-bar as used by AS4100-1998, BS5950-1990 and Eurocode-1992 did little to improve the strength of the eye-bar and its use should be questioned. When Eurocode 3-1992 is compared with the test results in a similar manner to that shown in Table 4 for AS4100-1998, it is found that the failure modes predicted by the code strengths do not compare well with actual test failure modes shown in Table 1. However, for the test specimens 1, 2, 3, 5, 6, 8, 9, 10, 11, 12, 14, 15 and 18 where the primary failure was by pin shear, the strength of the pin in shear predicted by Eurocode 3-1992 was close to the actual test strengths. For the test specimens 1, 4, 7, 8, 10, 13, 16 and 17 where the primary or secondary failure was by bearing of the plate, the strength predicted by Eurocode 3-1992 was significantly lower than the actual test strengths. DESIGN RECOMMENDATIONS From the comparisons with AS4100-1998 and Eurocode 3-1992, it was clear that the AS4100-1998 provided the best model for plate bearing strength based on the ultimate strength of the steel in the plate whereas Eurocode 3-1992 provided the best model for the pin shear strength, again based on the ultimate strength of the steel in the pin. It is therefore proposed that the strength Vf of a pin in shear should be given by Vf = 0.62fqAf (1) This is similar to the strength of a bolt in shear as given in AS4100-1998. The shear factor of 0.62 on the ultimate tensile strength is used to give the shear strength of the steel in the pin. In the tests, the mean value of this factor for the ductile black steel pins was 0.71 with a coefficient of variation of 0.08 with factors ranging from 0.74 for the 10mm diameter pins to 0.62 for the larger 27mm diameter pins. For the lower ductility bright steel pins, the mean value of the factor was 0.63 with a coefficient of variation of 0.03 with factors ranging from 0.65 for the 10mm diameter pins to 0.59 for the larger diameter 27mm pins. It is also proposed that the strength of the plate in bearing should be given by v~ = 3.2Lpdjtp (2) This is identical to the current requirements in AS4100-1998 for both pins and bolts. The bearing factor of 3.2 on the ultimate tensile strength is used to give the bearing strength of the steel in the plate. In the two tests that had primary bearing failures without plate fracture, the mean value of the factor was 3.74. In the other three bearing failure tests where premature plate fracture occurred, the mean value of this factor was still 2.67, a value close to 3.2. It is proposed that a new serviceability condition for plate bearing be included in design codes. As shown in Figure 2 for the 3mm plate that failed in bearing, the bearing deformations of the plate at maximum load are very large and typically exceed 60% of the hole diameter. Using a proof load at 2% of the hole diameter as the basis to define the maximum service load Vs that can be sustained prior to the onset of large plate bearing deformations, a mean design value of bearing strength Vb~ for serviceability conditions has been determined as Vbs = 1.6fypdftp (3) 236 R.Q. Bridge The value of the factor 1.6 was derived from the eight tests that had primary and secondary bearing failures. It is close to the factors shown in Table 3 for plate bearing strengths based on the yield strength, indicating that this should be a serviceability condition, not a strength condition. Comparisons of the proposals with the test results are given in Table 5. Values are shown for both primary and secondary failure modes and indicate reasonable agreement over the test range. TABLE 5. COMPARISON OF TEST RESULTS WITH DESIGN VALUES PREDICTED BY MODIFICATIONS TO CODES Test Max Vf Load/Vf Load/Vb Service Vb~ VflVbs Load Pin Pin Plate Load Vs Plate kN kN kN i i 1 53.6 44.8 1.20 1.08 20 18.1 1.11 2 54.0 44.7 1.21 3 54.3 44.8 1.21 4 97.0 126.4 1.17 30 30.0 1.00 5 150.8 ! 126.6 1.19 6 146.5 126.4 1.16 7 113.0 343.1 0.85 38 48.4 0.78 8 346.0 343.1 1.01 0.84 125 111.0 1.13 9 344.0 343.1 1.00 10 53.6 54.0 0.99 1.08 20 18.0 1.11 11 56.8 55.3 1.03 12 56.4 54.3 1.04 13 92.5 129.9 1.15 i 29 29.1 1.00 14 137.0 129.9 1.05 15 131.0 i 129.9 1.01 16 110.0 369.3 0.83 42 48.3 0.87 17 352.0 369.3 0.95 0.83 130 113.8 1.14 18 350.0 369.3 0.95 *Plate bearing (+pin shear) was a secondary failure mode in the tests. Predicted Failure Pin shear* Pin shear Pin shear Plate bearing Pin shear Pin shear Plate bearing Pin shear* Pin shear Plate bearing Pin shear Pin shear Plate bearing Pin shear Pin shear Plate bearing Pin shear *+ Pin shear CONCLUSIONS Tests have highlighted some deficiencies in current codes that are used to predict the strength of pins in plated structures. Modifications have been proposed that better model the modes of failure. Plate tear- out is an important consideration. A new serviceability condition is proposed. Bearing of the pin was not identified as a mode of failure and this aspect needs further examination. REFERENCES American Institute of Steel Construction AISC-1993, Load and resistance factor design specification for structural steel buildings - Second edition. American Institute of Steel Construction, Chicago. Australian Standard AS4100-1990. Steel structures, Standards Australia, Sydney. British Standard BS5950-1990. Structural use of steel in buildings, British Standards Institution, London. Eurocode 3-1992. ENV 1993-1-1 Design of steel structures- Partl.l: General rules and rules for buildings, European Committee for Standardization, Brussels. Hayward, I.G.and Van Ommen, M.(1992). Pins in steel structures. B.E. Thesis, University of Sydney. Sukkar, T. (1998). Pins in steel Structures. B.E.Thesis, University of Western Sydney, Nepean. FINITE ELEMENT MODELLING OF EIGHT-BOLT RECTANGULAR HOLLOW SECTION BOLTED MOMENT END PLATE CONNECTIONS A. T. Wheeler 1, M. J. Clarke 2 and G. J. Hancock 2 1 Department of Civic Engineering and Environment, The University of Western SydneymNepean, Kingswood, N.S.W., 2747, Australia 2 Department of Civil Engineering, The University of Sydney, Sydney, N.S.W. 2006, Australia ABSTRACT This paper describes the finite element modelling philosophy employed to analyse bolted moment end plate connections joining square and rectangular hollow sections which are subjected to pure bending. The ABAQUS finite element package (HKS, 1995) is used to simulate the experimental be- haviour observed in tests performed at the University of Sydney. The parameters varied in both the experiments and the ABAQUS simulations include the end plate thickness, the section shape (square or rectangular), and the position of the bolts. The results obtained from the finite element analyses are evaluated and the appropriateness of the model assessed by comparing the numerically predicted ul- timate loads and moment-rotation responses with those of the corresponding tests. Overall, it is con- cluded that the numerical analysis is effective in modelling the behaviour of the connections, al- though there are some failure modes observed experimentally which could not be directly reproduced in the finite element models. KEYWORDS Bolted connections, end plate connections, tubular sections, moment-rotation behaviour, ABAQUS. INTRODUCTION The increase in the use of rectangular hollow sections in mainstream structures coupled with the eco- nomics of prefabrication have highlighted the need for simple design methods that produce economi- cal tubular connections. Although tubular connection design handbooks have been published recently (Syam and Chapman, 1996; AISC, 1997), the eight-bolt moment end plate connection described in this paper is one configuration for which a design model is not widely available. A suitable model is described in the companion paper by Wheeler et al. (1999). The eight-bolt connection described in this paper and depicted in Figure la represents one of two fundamental bolting arrangements studied by Wheeler (1998). The other bolting arrangement utilises four bolts, as shown in Figure lb. The eight-bolt detail described in this paper is superior to the four-bolt variant from the point of view of connection strength and stiffness, but is nevertheless more costly. 237 238 A.T. Wheeler et al. Figure 1: Typical applications of bolted moment end plate connections using RHS Realistic modelling of bolted end plate connections is highly complex because the problems are three-dimensional in nature, and involve the added complications of geometric and material nonline- arities, and contact/separation between various components (Bursi and Jaspart, 1997a, 1997b). Bursi and Jaspart (1997a, 1997b) also highlight the importance of correct element selection to obtain accu- rate solutions, and have endeavoured to establish benchmarks that can be used to calibrate finite ele- ment models. This paper describes the finite element modelling philosophy employed to analyse eight-bolt moment end plate connections joining square and rectangular tubes subjected to pure bending. The ABAQUS finite element package (HKS, 1995) is used to simulate the behaviour observed in tests performed at the University of Sydney (Wheeler et al., 1995). The parameters varied in both the experiments and the ABAQUS simulations include the end plate thickness, the section shape (square or rectangular), and the position of the bolts. The results obtained from the finite element analyses are evaluated and the appropriateness of the model assessed by comparing the numerically predicted ultimate loads and moment-rotation responses with those of the corresponding tests. DEVELOPMENT OF FINITE ELEMENT MODEL Overview The generation of a three-dimensional finite element model of the bolted tubular end plate connection was carried out using the PATRAN pre-processor (PDA Engineering, 1994). The connections were analysed using the ABAQUS finite element software package (HKS, 1995). The analysis incorpo- rated the effects of both material and geometric nonlinearities. The finite element model of a typical eight-bolt end plate connection is shown in Figure 2, with the vertical axis of symmetry along the beam length being utilised to reduce the size of the model. To aid the model verification process, the connection was divided into five individual sub-models, each of which represents a specific component of the connection. These components are labelled in Figure 2. The model employed solid three dimensional brick elements for each of the components, with addi- tional interface elements used to model the contact/separation between various surfaces. The material properties used for the various components of the model were determined from the engineering stress-strain curves obtained through tensile tests (Wheeler et al., 1995). It should be noted that the incorporation of material nonlinearity in an ABAQUS model requires the use of the true stress (Otrue) versus the logarithmic plastic strain ( el pl ) relationship. Finite Element Modell&g of Bolted Moment End Plate Connections 239 Beam Section To model the beam section, eight-noded linear brick elements were utilised as shown in Figure 2. These elements are of type C3D8 in ABAQUS terminology. Two section sizes were employed in the tests: a square hollow section (SHS) of nominal dimensions 150x150• mm, and a rectangular hollow section (RHS) of nominal dimensions 200•215 mm. Figure 2: Finite element model of eight-bolt connection The tubular sections employed in the bolted end plate connections were manufactured using a cold- forming process. As a result, the material in the comers of the section was of higher strength than the material in the fiats. Consequently, different material properties were assigned to the comer and fiat regions of the sections in the finite element analysis. The relevant stress-strain curves are depicted in Figure 3. Figure 3: Typical section material properties End Plate The general layout and the corresponding dimensions of the end plates are given in Figure 1 and Ta- ble 1 of the companion paper (Wheeler et al., 1999). For all tests, the edge distance (ae) to the centre of the bolt holes was 30 mm, and the diameter of the holes was 22 mm for M20 bolts. In the finite element simulations, the end plate was modelled using eight-noded linear hybrid bricks, corresponding to element type C3D8H in ABAQUS. This element type was selected to prevent pos- sible problems of volume strain locking, which can occur in the C3D8 linear elements (HKS, 1995). Following a convergence study, it was decided to use four elements through the thickness of the end plate for all analyses. A typical layout is shown in Figure 2. The measured stress-strain relationships of the end plates follow the classic elastic-plastic-strain hardening pattem. Since the measured yield stresses (fy) and ultimate tensile strengths (fu) for the dif- . 63.4 Pin shear 7.9 Pin shear 0.4 Pin shear 126. 4 Plate bearing 7.5 Pin shear 2.3 Pin shear 69.7 Plate bearing 68.6 Pin shear 4.6 Pin shear 72.3 Pin shear 8.9 Pin shear 1.5 Pin shear. BS595 0-1 990. Structural use of steel in buildings, British Standards Institution, London. Eurocode 3-1 992. ENV 199 3-1 -1 Design of steel structures- Partl.l: General rules and rules for buildings,. *Plate bearing (+pin shear) was a secondary failure mode in the tests. Predicted Failure Pin shear* Pin shear Pin shear Plate bearing Pin shear Pin shear Plate bearing Pin shear* Pin shear