High temperature tests of cold-formed stainless steel double shear bolted connections

Journal of Constructional Steel Research 104 (2015) 49–63

Contents lists available at ScienceDirect

Journal of Constructional Steel Research

High temperature tests of cold-formed stainless steel double shear bolted connections Yancheng Cai, Ben Young ⁎ Department of Civil Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong

a r t i c l e

i n f o

Article history: Received 29 November 2013 Accepted 20 September 2014 Available online xxxx Keywords: Double shear bolted connection Elevated temperatures Failure modes Stainless steel Steady state tests Transient state tests

a b s t r a c t There is currently no design rule on bolted connections of cold-formed stainless steel structures at elevated temperatures. In this study, a total of 194 double shear bolted connection specimens with three different grades of stainless steel were tested, where 106 specimens were tested by steady state test method and 88 specimens were tested by transient state test method. The three different grades of stainless steel are austenitic stainless steel EN 1.4301 (AISI 304) and EN 1.4571 (AISI 316Ti having small amount of titanium) as well as lean duplex stainless steel EN 1.4162 (AISI S32101). The connections were designed with different bolt diameters, number of bolts and arrangement of the bolts. Bearing failure and net section tension failure modes were observed in the double shear bolted connection tests. The test results were compared with the nominal strengths calculated from the design rules in the American Specification, Australian/New Zealand Standard and European Codes for stainless steel structures. In calculating the nominal strengths of the connections, the material properties at elevated temperatures were used in the design equations for room temperature. It is shown that the strengths of the cold-formed stainless steel double shear bolted connections obtained from the specifications are generally conservative at elevated temperatures. The connection strengths decrease as the temperature increases in the similar manner for the steady state tests and the transient state tests as well as the material coupon tests. It is also found that the austenitic stainless steel type EN 1.4571 generally has better resistance than the stainless steel types EN 1.4301 and EN 1.4162 for double shear bolted connections at elevated temperatures. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction The desirable characteristics of stainless steel, such as attractive appearance, corrosion resistance, better fire resistance as compared to carbon steel and low maintenance cost, are able to provide attractive usage in certain applications, where the achievable benefits may compensate for the extra material cost compared to carbon steel [1,2]. Significant progress has been made in developing design rules for stainless steel structures at room (ambient) temperature in recent years, while the performance of fire resistance of stainless steel structures has received little attention [3]. Previous research has shown that the strength and stiffness retention of austenitic stainless steel (EN 1.4301) at elevated temperatures is superior to those of carbon steel [4]. Bolted connections are commonly used in construction for both carbon steel and stainless steel structures. Currently, the design rules of stainless steel bolted connections are available in different specifications, including the American Society of Civil Engineers Specification (ASCE) [5], Australian/New Zealand Standard (AS/NZS) [6] and Eurocode 3 Part 1-4 (EC3-1.4) [7]. Zadanfarrokh [8] and Rogers and Hancock [9–11] conducted a series of tests on carbon steel bolted connections at room temperature, whereas tests of stainless steel ⁎ Corresponding author. Tel.: +852 2859 2674; fax: +852 2559 5337. E-mail address: [email protected] (B. Young).

http://dx.doi.org/10.1016/j.jcsr.2014.09.015 0143-974X/© 2014 Elsevier Ltd. All rights reserved.

bolted connections conducted by Bouchaïr et al. [12] and Cai and Young [13] were also at room temperature. However, there are presently few investigations on the behavior of stainless steel bolted connections at elevated temperatures. The material properties of stainless steel at elevated temperatures have been investigated by Gardner and Baddoo [3], Sakumoto et al. [14], Makelainen and Outinen [15], Chen and Young [16] and Brnic et al. [17,18]. The behavior of stainless steel columns and laterally restrained beams at elevated temperatures has been investigated by Ng and Gardner [19]. To and Young [20] numerically studied the performance of cold-formed stainless steel tubular columns at elevated temperatures. Yan and Young [21,22] studied the structural behavior of single shear bolted connections of thin sheet carbon steels at elevated temperatures by steady state and transient state test methods. Furthermore, Yan and Young [23] also conducted a series of tests on double shear bolted connections of carbon steel at elevated temperatures. It should be noted that investigation of stainless steel bolted connections at elevated temperatures is limited up-to-date. Cai and Young [24] recently conducted 100 tests of cold-formed stainless steel single shear bolted connections at high temperatures using steady state test method. It should be noted that the current design rules [5–7] on stainless steel bolted connections are only applicable at room temperature, and these design rules do not cover elevated temperatures.

50

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

Table 1 Coupon test results at elevated temperatures [24]. Nominal temperature (°C)

Stainless steel series

22

200

350

450

500

550

650

800

950

Coupon specimen temperature (°C)

A T L A T L A T L

22 22 22 1.00 (474a) 1.00 (463a) 1.00 (724a) 1.00 (759a) 1.00 (677a) 1.00 (862a)

205 206 206 0.81 0.84 0.78 0.69 0.77 0.82

351 356 356 0.71 0.80 0.70 0.66 0.77 0.81

– 449 – – 0.79 – – 0.78 –

496 498 501 0.67 0.75 0.62 0.62 0.75 0.73

544 548 553 0.65 0.72 0.54 0.58 0.71 0.60

648 645 652 0.54 0.67 0.42 0.46 0.62 0.42

800 800 795 0.33 0.53 0.16 0.21 0.38 0.16

950 950 948 0.13 0.20 0.02 0.09 0.14 0.03

f0.2,T/f0.2,N

fu,T/fu,N

a

Test results obtained at room temperature in the unit of MPa shown in Ref. [13].

In this study, a total of 194 double shear bolted connection specimens with three different grades of cold-formed stainless steel were tested, where 106 specimens were tested by steady state test method and 88 specimens were tested by transient state test method. In the steady state tests, the stainless steel bolted connections were investigated in the temperature ranged from 200 to 950 °C, while in the transient state tests, the connections were tested under 3 different load levels, namely 0.25, 0.50 and 0.75 of the failure load at room temperature. The three different types of stainless steels are austenitic stainless steel EN 1.4301 (AISI 304) and EN 1.4571 (AISI 316Ti having small amount of titanium) as well as lean duplex stainless steel EN 1.4162 (AISI S32101). The investigation of the double shear bolted connections involved different bolt diameters, number of bolts and arrangement of the bolts. A total of 15 series of specimens were considered. The observed failure modes involved bearing failure and net section tension failure. The ultimate strengths of stainless steel double shear bolted

connection tests at elevated temperatures were compared. It is found that the stainless steel type EN 1.4571 generally has better resistance than the stainless steel types EN 1.4301 and EN 1.4162 at elevated temperatures. It is also shown that the strengths of the cold-formed stainless steel double shear bolted connections predicted by the specifications are generally conservative at elevated temperatures. The connection strengths decrease as the temperature increases in the similar manner for the steady state tests and the transient state tests as well as the material coupon tests. 2. Current design rules of cold-formed stainless steel bolted connections The current stainless steel design specifications for bolted connections are applicable at room temperature only [5–7]. In the current design specifications, different failure modes for bolted connections

(a) One-bolted M12 of series A, T and L

(b) Two-parallel bolted M8 of series A, T and L

(c) Two-perpendicular bolted M8 of series A, T and L

(d) Three-bolted M8 of series A, T and L

(e) Four-bolted M6 of series A and T

(f) Four-bolted M8 of Series L

Fig. 1. Nominal dimension of double shear (internal plate) bolted connection specimens: (a) One-bolted M12 of series A, T and L; (b) Two-parallel bolted M8 of series A, T and L; (c) Two-perpendicular bolted M8 of series A, T and L; (d) Three-bolted M8 of series A, T and L; (e) Four-bolted M6 of series A and T; (f) Four-bolted M8 of series L.

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

51

are specified. These failure modes include the bolt in shear, bearing, net section tension, bolt in tension and bolt subject to combined shear and tension. In this study, two of these failure modes were observed, and they are the bearing failure and net section tension failure. The characteristics of different failure modes of carbon steel bolted connections are illustrated in Yan and Young [21]. In this study, the American Society of Civil Engineers Specification (ASCE) [5] for the design of cold-formed stainless steel members, the Australian/New Zealand Standard (AS/NZS) [6] for cold-formed stainless steel structures, Eurocode 3 – Design of steel structures — Part 1-4: General rules — Supplementary rules for stainless steels (EC3-1.4) [7] and Eurocode 3 – Design of steel structures – Part 1-8: Design of joints (EC3-1.8) [25] were used. It should be noted that the stainless steel bolted connection design equations in the ASCE Specification [5] are identical to those in the AS/NZS Standard [6].

3. Double shear bolted connection tests 3.1. Coupon tests The double shear bolted connection test specimens were fabricated from the same batch of stainless steels as those used in the bolted connection test specimens reported by Cai and Young [13,24]. The stainless steel grades are austenitic stainless steel EN 1.4301 (AISI 304) and EN 1.4571 (AISI 316Ti having small amount of titanium) as well as lean duplex stainless steel EN 1.4162 (AISI S32101). In this study, the three types of stainless steels of austenitic EN 1.4301 (AISI 304) and EN 1.4571 (AISI 316Ti) as well as lean duplex EN 1.4162 (AISI S32101) are labeled as A, T and L in the context of this paper, respectively. An MTS 810 Universal testing machine with the heating device of MTS model 653.04 high temperature furnace that contains three independent-controlled heating chambers with a maximum temperature up to 1400 °C was used for the coupon tests. Cai and Young [13,24] reported the elastic modules (EN and ET), 0.2% proof stresses (f0.2,N and f0.2,T), tensile strengths (fu,N and fu,T) and ultimate strains (εu,N and εu,T) of the stainless steel coupon test results conducted at room temperature of 22 °C and elevated temperatures up to 950 °C using steady state test

Fig. 2. Test set-up of double shear bolted connection at elevated temperatures.

Fig. 3. Schematic view of transient state tests and location of external thermocouple for double shear bolted connections.

method. Table 1 shows the reduction factors of the 0.2% proof stresses and tensile strengths due to the deterioration of the materials at elevated temperatures. The details of the coupon tests and the test results are reported in Cai and Young [13,24].

3.2. Specimen design The cold-formed stainless steel double shear bolted connection specimens were designed to avoid end-tear out failure and bolt shear failure. The type of stainless steel, the size of bolt as well as the number and arrangement of bolts were considered in the connection specimens. A total of 15 test series of double shear bolted connection specimens were tested at elevated temperatures. Fig. 1 shows the detail specimen dimensions of internal plate for each test series. The specimens were cut from stainless steel rectangular hollow sections. The overall length of each part of the specimen was ranged from 390 to 404 mm, and the

Fig. 4. Comparison of specimen temperature (T-D-4-6 at 0.25Pu,N) and ISO fire curve.

52

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

Table 2 Comparison of test results with predicted values for double shear one-bolted connections. Specimen series

Series A-D-1-12

Series T-D-1-12

Series L-D-1-12

Temperature (°C) Nominal

Coupon

Specimen

22 200 350 500 650 800 950

22 205 351 496 648 800 950

22 205 360 507 652 803 948

22

22

200 350 500 650 800 950

206 356 498 645 800 950

22

22

200 350 500 650 800 950

206 356 501 652 795 948

22 22 206 359 508 652 806 950

22 22 206 358 509 651 800 951

Pu,N or Pu,T

Pu,T/Pu,N

Pu,N/PASCE or Pu,T/PASCE

Pu,N/PEC or Pu,T/PEC

Failure mode ASCE and AS/NZS

EC

Test

38.5 26.6 25.6 24.1 18.1 8.5 3.1

1.00 0.69 0.66 0.63 0.47 0.22 0.08

1.56 1.32 1.44 1.45 1.34 1.21 1.11

1.28 1.20 1.25 1.25 1.22 1.19 1.09

NS – – – – – –

B – – – – – –

B B B B B B B

Mean COV

1.35 0.114

1.21 0.051

1.00 0.94 0.78 0.75 0.73 0.59 0.39 0.13

1.45 1.36 1.63 1.36 1.41 1.30 1.22 1.11

1.26 1.18 1.50 1.22 1.23 1.18 1.20 1.06

NS NS – – – – – –

B B – – – – – –

B B B B B B B B

Mean COV

1.36 0.115

1.23 0.101

1.00 0.96 0.79 0.77 0.67 0.43 0.16 0.03

1.24 1.17 1.26 1.36 1.34 1.32 1.28 1.26

1.37 1.30 1.21 1.23 1.20 1.29 1.26 1.14

NS NS – – – – – –

B B – – – – – –

B B B B B B B B

Mean COV

1.28 0.048

1.25 0.057

35.3 33.3 27.5 26.5 25.7 21.0 13.6 4.5

47.1 45.0 37.4 36.2 31.6 20.3 7.6 1.2

Note: B = Bearing failure; NS = Net section tension failure. Table 3 Comparison of test results with predicted values for double shear two-parallel bolted connections. Specimen series

Temperature (°C) Nominal

Series A-D-2Pa-8

Series T-D-2Pa-8

Series L-D-2Pa-8

Coupon

22

22

200 350 500 650 800 950

205 351 496 648 800 950

22 200 350 500 650 800 950

22 206 356 498 645 800 950

22

22

200 350 500 650 800 950

206 356 501 652 795 948

Pu,N or Pu,T

Pu,T/Pu,N

Pu,N/PASCE or Pu,T/PASCE

Pu,N/PEC or Pu,T/PEC

Failure mode ASCE and AS/NZS

EC

Test

42.1 41.4 28.9 28.5 27.4 20.2 9.9 3.6

1.00 0.98 0.69 0.68 0.65 0.48 0.24 0.09

1.49 1.47 1.26 1.41 1.44 1.31 1.28 1.17

1.23 1.20 1.12 1.19 1.22 1.17 1.24 1.13

NS NS – – – – – –

NS NS – – – – – –

B B B B B B B B

Mean COV

1.35 0.085

1.19 0.038

1.00 0.80 0.76 0.75 0.66 0.40 0.13

1.42 1.35 1.34 1.41 1.40 1.27 1.09

1.21 1.23 1.17 1.20 1.26 1.22 1.05

NS – – – – – –

B – – – – – –

B B B B B B B

Mean COV

1.33 0.088

1.19 0.057

1.00 0.98 0.81 0.80 0.69 0.44 0.17 0.03

1.29 1.27 1.27 1.38 1.34 1.35 1.33 1.30

1.24 1.22 1.22 1.23 1.17 1.30 1.28 1.14

NS NS – – – – – –

NS NS – – – – – –

B B B B B B B B

Mean COV

1.32 0.030

1.23 0.043

Specimen 22 22 209 352 502 649 799 950

22 204 354 504 651 803 954

22 22 207 355 505 648 800 951

Note: B = Bearing failure; NS = Net section tension failure.

39.2 31.4 29.8 29.3 25.9 15.5 5.0

52.2 51.3 42.5 41.9 35.9 22.8 8.7 1.4

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

total assembled specimen length was maintained at 690 mm for each specimen. The lapped connection part of each specimen was always located at the center position of the furnace. The nominal width of the connection specimen was 50 mm, and the nominal thickness of the specimen was 1.5 mm. Stainless steel washers were used in both sides of the bolt. The double shear bolted connection specimens have the same configuration as those specimens tested at room temperature [13]. A torque of approximately 10 Nm was applied to each bolt by hand that allowed the connection to slip between the bolt holes with a small load. Three different sizes of A4 stainless steel bolts [26] were used. The sizes of the bolts were M6, M8 and M12, and the corresponding sizes of stainless steel washers and nuts were also used. The standard size of bolt hole (do) was adopted in accordance with the ASCE Specification [5] and AS/NZS Standard [6]. The size of bolt hole (do) is 1 mm larger than the nominal bolt diameter (d) only if d is smaller than 12 mm, but do is 2 mm larger than the nominal bolt diameter when d is greater than or equal to 12 mm. Generally, the spacing in the connected parts of the specimens satisfied the minimum requirements of the specifications [5,6,25]. However, for specimens containing two bolt holes or more that are perpendicular to the direction of loading, as shown in Fig. 1(c), (d) and (f), the spacing in the connected parts could not satisfy the minimum requirements of the ASCE Specification [5] and AS/NZS Standard [6], but satisfy the minimum requirements of EC3-1.8 [25].

53

3.3. Specimen labeling The connection specimens are divided into three groups (A, T and L) according to the grades of stainless steel. Each group of bolted connections contains five different cases based on the bolt diameter, bolt number and bolt arrangement as illustrated in Fig. 1. The specimen label contains four segments in order to identify the type of stainless steel, the connection type, the bolt number and arrangement as well as the bolt size. For examples, the labels “A-D-4-6” and “L-D-2Pa-8” define the following specimens: • The first letter indicates the type of stainless steel, where “A” is referred to EN 1.4301 (AISI 304) and “L” is referred to EN 1.4162 (AISI S32101). • The second letter represents the connection type, where “D” means the double shear bolted connection. • The third segment of the label is referred to the number of bolt used in the connection specimen, where “4” means four bolts were used in the specimen. There were two different bolt arrangements for the twobolted connections. In the label, the letters “Pa” refer to the bolts arranged parallel to the loading direction, while the letters “Pe” refer to bolts perpendicular to the loading direction. The “2Pa” means there are two-parallel bolts in the specimen. • The fourth part of the label means the nominal diameter of the bolts. The number “6” represents the bolt diameter of 6 mm, while “8” stands for 8 mm.

Table 4 Comparison of test results with predicted values for double shear two-perpendicular bolted connections. Specimen series

Series A-D-2Pe-8

Series T-D-2Pe-8

Series L-D-2Pe-8

Temperature (°C) Nominal

Coupon

Specimen

22 200 350

22 205 351

500

496

650 800 950

648 800 950

22 203 353 353 502 502 650 802 948

22 200 350

22 206 356

500

498

650 800 950

645 800 950

22

22

200 350

206 356

500

501

650 800 950

652 795 948

22 208 353 351 503 502 650 801 949

22 22 207 356 353 507 502 653 803 950

Note: B = Bearing failure; NS = Net section tension failure.

Pu,N or Pu,T

Pu,T/Pu,N

Pu,N/PASCE or Pu,T/PASCE

Pu,N/PEC or Pu,T/PEC

Failure mode ASCE and AS/NZS

EC

Test

32.5 24.7 23.1 23.7 24.0 24.8 18.4 8.5 3.2

1.00 0.76 0.71 0.73 0.74 0.76 0.57 0.26 0.10

1.47 1.38 1.46 1.50 1.62 1.67 1.53 1.17 1.12

0.94 1.01 0.99 1.02 1.09 1.13 1.12 1.12 1.05

NS – – – – – – – –

NS – – – – – – – –

B B B B B B B B B

+ + + + + +

NS NS NS NS NS NS

Mean COV

1.44 0.130

1.05 0.064

1.00 0.84 0.80 0.80 0.85 0.88 0.75 0.44 0.15

1.40 1.42 1.42 1.42 1.60 1.65 1.60 1.18 1.03

0.96 1.06 1.01 1.02 1.10 1.14 1.17 1.13 0.99

NS – – – – – – – –

NS – – – – – – – –

B B B B B B B B B

+ + + + + +

NS NS NS NS NS NS

Mean COV

1.41 0.143

1.06 0.070

1.00 1.00 0.87 0.78 0.82 0.78 0.77 0.51 0.19 0.04

1.18 1.20 1.34 1.34 1.42 1.53 1.50 1.47 1.38 1.90

1.00 1.00 1.07 0.98 1.04 1.10 1.07 1.25 1.19 1.37

NS NS – – – – – – – –

NS NS – – – – – – – –

B B B B B B B B B B

+ + + + + + +

NS NS NS NS NS NS NS

Mean COV

1.43 0.142

1.11 0.114

30.7 25.7 24.6 24.7 26.0 26.9 23.0 13.6 4.5

40.9 40.7 35.4 31.9 33.7 32.1 31.3 20.9 7.7 1.6

54

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

3.4. Test set-up and procedure

the furnace for the connection tests was approximately 40–60 °C/min, depending on the specified temperature level. Higher heating rate was used as the temperature increases. The temperature was hold for 8 to 15 min once the pre-selected temperature was reached. Hence, this allows the temperature to stabilize. Finally, the bottom end of the specimen was gripped. Displacement control with the loading rate of 1.5 mm/min was used in the tests until the specimen fails. Six different elevated temperature levels of 200, 350, 500, 650, 800 and 950 °C were considered in the steady state tests. In the transient state test, the specimen is subjected to a constant load when the temperature rises. The test specimen was firstly loaded up to a pre-selected load level in 1–3 min. The pre-selected constant load was maintained for approximately 1 min to ensure the expected load level stabilizes. The temperature was then raised according to the ISO standard fire curve as specified in EC1-1.2 [27] until the specimen fails. The force control procedure was used during the entire process in order to maintain the load constant. Three different load levels were adopted in the transient state tests, namely 0.25, 0.50 and 0.75 of the failure load (Pu,N) at room temperature. It means that the loading on each specimen was chosen as 0.25Pu,N, 0.50Pu,N, and 0.75Pu,N, respectively. The test strengths (Pu,N) of the stainless steel double shear bolted connections at room temperature are reported in Cai and Young [13]. The ISO standard fire curve as specified in EC1-1.2 [25] is given as θ = 20 + 345log10(8t + 1), where θ is the gas temperature in fire compartment, and t is the time in minute. It should be noted that the

The double shear bolted connection tests were conducted using the same MTS machine for the single shear bolted connection tests [13,24]. Fig. 2 shows the test set-up of a cold-formed stainless steel double shear bolted connection at elevated temperatures. The same MTS Universal testing machine as the coupon tests was used for the bolted connection tests. The test specimen was assembled on a pair of gripping apparatus, which was specially fabricated in order to provide the pin end boundary condition. The details of the gripping apparatus are shown in Cai and Young [13]. The actual specimen temperature was measured by an external thermocouple, as shown in Fig. 3. It should be noted that previous research using the same furnace has shown that the temperature distribution in the specimen is quite uniform inside the furnace [21]. A data acquisition system was used to record the applied load, temperatures from the external thermocouple and the furnace air temperature at regular intervals. In the steady state test, the specimen was clamped at the top end, while the bottom end was not fixed. The external thermocouple was inserted inside the furnace and contacted on the surface of the middle part of the specimen. The specimen temperature obtained by the external thermocouple was used. The furnace was then closed. The temperature was raised to a pre-selected level. It should be noted that the thermal expansion of the specimen was allowed by the free bottom end of the specimen during the heating process. The heating rate of

Table 5 Comparison of test results with predicted values for double shear three-bolted connections. Specimen series

Series A-D-3-8

Series T-D-3-8

Series L-D-3-8

Temperature (°C) Nominal

Coupon

Specimen

22 200 350

22 205 351

500

496

650 800

648 800

950

950

22 206 357 353 505 509 656 800 801 955

22 200 350

22 206 356

500

498

650 800 950

645 800 959

22

22

200

206

350 500 650 800 950

356 501 652 795 948

22 204 356 353 506 502 654 804 951

22 22 207 204 358 507 650 800 956

Note: B = Bearing failure; NS = Net section tension failure.

Pu,N or Pu,T

Pu,T/Pu,N

Pu,N/PASCE or Pu,T/PASCE

Pu,N/PEC or Pu,T/PEC

Failure mode ASCE and AS/NZS

EC

Test

33.4 25.1 23.1 23.4 24.0 21.9 17.5 9.0 9.3 3.3

1.00 0.75 0.69 0.70 0.72 0.66 0.52 0.27 0.28 0.10

1.50 1.41 1.46 1.48 1.62 1.48 1.46 1.23 1.28 1.16

0.96 1.02 0.99 1.00 1.09 1.00 1.07 1.18 1.22 1.09

NS – – – – – – – – –

NS – – – – – – – – –

B B B B B B B B B B

+ + + + + +

NS NS NS NS NS NS

Mean COV

1.41 0.100

1.06 0.080

1.00 0.82 0.77 0.78 0.82 0.83 0.70 0.43 0.15

1.46 1.45 1.40 1.43 1.61 1.62 1.55 1.18 1.12

1.00 1.08 1.01 1.02 1.10 1.12 1.14 1.13 1.08

NS – – – – – – – –

NS – – – – – – – –

B B B B B B B B B

+ + + + + +

NS NS NS NS NS NS

Mean COV

1.42 0.122

1.08 0.050

1.00 1.00 0.78 0.82 0.78 0.75 0.46 0.20 0.04

1.27 1.24 1.27 1.32 1.41 1.54 1.38 1.51 1.79

1.06 1.04 1.01 1.05 1.03 1.10 1.17 1.30 1.28

NS NS – – – – – – –

NS NS – – – – – – –

B B B B B B B B B

+ + + + + +

NS NS NS NS NS NS

Mean COV

1.41 0.125

1.12 0.098

31.8 26.2 24.4 24.8 26.1 26.4 22.3 13.6 4.9

42.8 42.6 33.4 34.9 33.4 32.3 19.6 8.4 1.5

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

first 5–6 min of the tests was not able to follow the ISO standard fire curve in this study. This is due to the limitation of the heating rate in the furnace, and the maximum heating rate of the furnace is only 100 °C/min. Therefore, the heating rate as specified by Yan and Young [21] was adopted in this study. The maximum heating rate of 100 °C/min of the furnace was used for the first 5 min, and the temperature was raised from 20 to 520 °C. Subsequently, the heating rate was adjusted in order to follow the ISO standard fire curve. The temperature of specimen T-D-4-6 at the load level of 0.25Pu,N is plotted together with the ISO fire curve as well as the actual heating rate of the furnace, as shown in Fig. 4. The horizontal axis represents the testing time in minute, while the vertical axis plotted the temperature in degrees Celsius (°C). After the temperature goes beyond 600 °C, the specimen temperature, the ISO fire curve and the actual heating rate of the furnace are almost identical as shown in Fig. 4.

55

except for specimen A-D-3-8 at the temperature 500 °C that had a difference of 8.8%. Fig. 5 shows the test curves of one-bolted connection specimens A-D-1-12, T-D-1-12 and L-D-1-12 for different temperatures, in which “R” represents the repeated test. The displacement of bolt slip during the initial loading stage was shifted in all the curves. In the transient state tests, the critical temperatures for the double shear bolted connection specimens can be determined by the loadtemperature curves for which the load drops 5% of the pre-selected load level. The critical temperature is defined as the temperature that causes the specimen to fail. The definition of the critical temperature of a test specimen with the corresponding displacement is illustrated in Fig. 6. The load–temperature curves of the stainless steel double shear one-bolted connections at three different loading levels of 0.25Pu,N, 0.50Pu,N and 0.75Pu,N are shown in Fig. 7(a)–(c), respectively. The horizontal axis of these graphs plotted the specimen temperatures and the vertical axis plotted the applied load on the specimens. Fig. 7(a)–(c) shows the three different grades of stainless steel that compared under the same load level. Fig. 8(a)–(c) shows the corresponding load–displacement curves of these specimens. The displacement was plotted using the hydraulic actuator movement, and subsequently determined by the 5% drop as mentioned earlier. The critical temperature and the corresponding displacement of the stainless steel double shear bolted connections in transient state tests are listed in Tables 7–11. Repeated tests were carried out in each test series that covers all the specimens at the load level of 0.50Pu,N. It was found that the repeated test results are quite close to their first test results as shown in Tables 7–11, especially for specimens under the load

4. Test results In the steady state tests, the test strengths (Pu,N and Pu,T) of the coldformed stainless steel double shear bolted connection specimens at room and elevated temperatures are presented in Tables 2–6. The test results of the stainless steel double shear bolted connections at room temperature were already reported by Cai and Young [13]. It is shown that the specimen temperatures measured in the connection tests were quite close to those obtained from the coupon tests. Some repeated connection tests were conducted. The repeated tests were generally within few percentage difference compared with their first test values,

Table 6 Comparison of test results with predicted values for double shear four-bolted connections. Specimen series

Series A-D-4-6

Series T-D-4-6

Series L-D-4-8

Pu,N or Pu,T

Temperature (°C) Nominal

Coupon

22

22

200 350 500

205 351 496

650 800 950

648 800 950

22 200 350

22 206 356

500

498

650 800 950

645 800 950

22 200

22 206

350 500 650 800 950

356 501 652 795 948

Pu,T/Pu,N

Pu,N/PASCE or Pu,T/PASCE

Pu,N/PEC or Pu,T/PEC

Failure mode ASCE and AS/NZS

EC

Test

1.00 0.98 0.72 0.67 0.67 0.67 0.51 0.25 0.09

1.59 1.56 1.41 1.49 1.60 1.60 1.49 1.21 1.06

1.14 1.13 1.03 1.07 1.15 1.15 1.09 1.16 1.00

NS NS – – – – – – –

NS NS – – – – – – –

B B B B B B B B B

+ + + + + +

NS NS NS NS NS NS

Mean COV

1.45 0.132

1.10 0.053

1.00 0.83 0.76 0.78 0.80 0.81 0.70 0.43 0.15

1.43 1.43 1.37 1.41 1.55 1.56 1.52 1.17 1.06

1.03 1.07 0.98 1.02 1.11 1.13 1.11 1.11 1.02

NS – – – – – – – –

NS – – – – – – – –

B B B B B B B B B

+ + + + + +

NS NS NS NS NS NS

Mean COV

1.39 0.123

1.06 0.050

1.00 0.78 0.77 0.75 0.69 0.42 0.19 0.03

1.32 1.31 1.29 1.39 1.45 1.31 1.53 1.79

1.11 1.04 1.02 1.02 1.04 1.11 1.32 1.28

NS – – – – – – –

NS – – – – – – –

B B B B B B B B

+ + + + +

NS NS NS NS NS

Mean COV

1.42 0.119

1.12 0.106

Specimen 22 22 202 360 504 504 651 801 951

22 207 355 354 506 499 651 799 949

22 204 204 359 504 655 803 952

Note: B = Bearing failure; NS = Net section tension failure.

39.5 38.7 28.4 26.4 26.6 26.6 20.1 9.9 3.4

35.3 29.2 26.7 27.6 28.3 28.6 24.6 15.1 5.2

44.3 34.6 33.9 33.1 30.4 18.6 8.5 1.5

56

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

levels of 0.25Pu,N and 0.50Pu,N, in which the maximum difference of the critical temperature between the first and repeated test results is 2.7%. Fig. 9 plotted the connection test results at elevated temperatures for steady state and transient state test methods. The vertical axis plotted the test strengths normalized with the test strength at room temperature (Pu,T/Pu,N), while the horizontal axis plotted against the specimen temperatures obtained by the external thermocouple. It was found

(a) Series A-D-1-12

that the ultimate strengths of the double shear connections Pu,T dropped rapidly when the temperature exceeded 500 °C. Generally, the stainless steel type T (EN 1.4571 or AISI 316Ti) has better resistance than the stainless steel type A (EN 1.4301 or AISI 304) and type L (EN 1.4162 or AISI S32101) at elevated temperatures, especially when the temperature exceeded 500 °C. The stainless steel type L generally has better resistance than stainless steel type A for the temperature ranged from 22 to 500 °C, but type A has a slightly better resistance than type L for the temperature ranged from 650 to 950 °C. The same observation was found for cold-formed stainless steel single shear bolted connections at elevated temperatures [24]. The stainless steel double shear bolted connections in transient state tests clearly showed that when the load level of the test specimen increases, the critical temperature decreases accordingly. In Tables 7–11, it is shown that the specimens of cold-formed stainless steel grade T (AISI 316Ti having small amount of titanium) generally have better resistance than the other two grades of stainless steels under the same load levels as higher critical temperatures were obtained for specimens of stainless steel grade T. This finding is similar to the steady state test results, as shown in Fig. 9. For examples, the average critical temperatures at the load level of 0.25Pu,N for specimens A-D-112, T-D-1-12 and L-D-1-12 were approximately 808, 885 and 769 °C, respectively; the average critical temperatures at the load level of 0.50Pu,N for specimens A-D-2Pa-8, T-D-2Pa-8 and L-D-2Pa-8 were approximately 680, 757 and 663 °C, respectively; the average critical temperatures at the load level of 0.75Pu,N for specimens A-D-2Pe-8, T-D-2Pe-8 and L-D-2Pe-8 were approximately 405, 615 and 495 °C, respectively. Generally, it is found that the displacements of the double shear bolted connections assembled by the three different grades of stainless steel decrease as the load level increases. For instance, the displacement of the specimens at the load levels of 0.25Pu,N and 0.75Pu,N decreased from 18.1 to 4.5 mm, from 19.4 to 3.7 mm and from 17.2 to 3.9 mm, for specimens A-D-1-12, T-D-1-12 and L-D-1-12, respectively, as shown in Table 7.

5. Comparison of test strengths with nominal strengths

(b) Series T-D-1-12

The nominal (unfactored) strengths (PASCE and PEC) of the stainless steel double shear bolted connections were obtained from the current design specifications [5,7,25]. These design specifications are only applicable at room temperature, and elevated temperatures are not covered. The nominal strengths of the connections were calculated using the reduced yield stress and ultimate strength obtained from the coupon tests [24] at elevated temperatures as well as using the measured specimen dimensions. The design rules of stainless steel double shear

(c) Series L-D-1-12 Fig. 5. Test curves of double shear bolted connections at elevated temperatures: (a) Series A-D-1-12; (b) Series T-D-1-12; (c) Series L-D-1-12.

Fig. 6. Definition of the critical temperature or the critical displacement for double shear bolted connection.

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

bolted connections in the ASCE Specification [5] are identical to those in the AS/NZS Standard [6]. Hence, the connection strengths calculated from these two specifications are identical. The connection strengths and failure modes obtained from the tests were compared with the nominal strengths calculated from the ASCE Specification [5] and Eurocodes [7,25] as shown in Tables 2–6. It is shown that the nominal strengths obtained from both ASCE Specification [5] and Eurocodes [7,25] are conservative, except for the Eurocodes predictions for specimens A-D-2Pe-8 at nominal temperatures of 22 and 350 °C, T-D-2Pe-8 at nominal temperatures of 22 and 950 °C, L-D-2Pe-8

(a)

Specimens A-D-1-12, T-D-1-12 and L-D-1-12 at 0.25Pu,N

at nominal temperature of 350 °C, A-D-3-8 at nominal temperatures of 22 and 350 °C, and T-D-4-6 at nominal temperature of 350 °C. It is shown that the nominal strengths obtained from ASCE Specification (PASCE) are more conservative than the Eurocodes predictions (PEC) for the stainless steel double shear bolted connections at elevated temperatures. Moreover, the bolted connections assembled by lean duplex stainless steel type L are generally more conservative compared with the bolted connections assembled by austenitic stainless steel types A and T calculated from the Eurocodes [7,25]. For Series A-D-1-12 type A

(a)

Specimens A-D-1-12, T-D-1-12 and L-D-1-12 at 0.25Pu,N

(b) Specimens A-D-1-12, T-D-1-12 and L-D-1-12 at 0.50Pu,N

(b) Specimens A-D-1-12, T-D-1-12 and L-D-1-12 at 0.50Pu,N

(c)

(c) Specimens A-D-1-12, T-D-1-12 and L-D-1-12 at 0.75Pu,N

Specimens A-D-1-12, T-D-1-12 and L-D-1-12 at 0.75Pu,N

Fig. 7. Load–temperature curves of double shear bolted connection tests: (a) Specimens AD-1-12, T-D-1-12 and L-D-1-12 at 0.25Pu,N; (b) Specimens A-D-1-12, T-D-1-12 and L-D-112 at 0.50Pu,N; (c) Specimens A-D-1-12, T-D-1-12 and L-D-1-12 at 0.75Pu,N.

57

Fig. 8. Load–displacement curves of double shear bolted connection tests: (a) Specimens A-D-1-12, T-D-1-12 and L-D-1-12 at 0.25Pu,N; (b) Specimens A-D-1-12, T-D-1-12 and LD-1-12 at 0.50Pu,N; (c) Specimens A-D-1-12, T-D-1-12 and L-D-1-12 at 0.75Pu,N.

58

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

Table 7 Transient state test results for specimens of one-bolted connections.

Table 9 Transient state test results for specimens of two-perpendicular bolted connections.

Specimen

Pu,N (kN)

Load level

Actual load (kN)

Critical temperature (°C)

Displacement (mm)

Failure mode

Specimen

Pu,N (kN)

Load level

Actual load (kN)

Critical temperature (°C)

Displacement (mm)

Failure mode

A-D-1-12

38.5

B B B B B B B B B B B B B B B B B B

30.7

47.1

16.2 18.1 8.9 7.1 4.5 4.7 17.6 19.4 10.6 13.1 4.1 3.7 16.6 17.2 8.0 8.5 3.9 6.4

T-D-2Pe-8

L-D-1-12

807 809 668 652 328 295 882 889 734 746 408 417 770 768 667 653 304 563

32.5

35.3

9.63 9.63 19.25 19.25 28.88 28.88 8.83 8.83 17.65 17.65 26.48 26.48 11.78 11.78 23.55 23.55 35.33 35.33

A-D-2Pe-8

T-D-1-12

0.25 0.25 0.50 0.50 0.75 0.75 0.25 0.25 0.50 0.50 0.75 0.75 0.25 0.25 0.50 0.50 0.75 0.75

L-D-2Pe-8

40.9

0.25 0.25 0.50 0.50 0.75 0.75 0.25 0.25 0.50 0.50 0.75 0.75 0.25 0.50 0.50 0.75

8.13 8.13 16.25 16.25 24.38 24.38 7.68 7.68 15.35 15.35 23.03 23.03 10.23 20.45 20.45 30.68

822 827 708 689 352 459 890 894 774 777 621 609 781 677 688 495

10.8 11.4 6.0 5.7 2.6 3.3 11.6 10.6 6.7 6.6 3.2 3.8 9.9 5.0 5.5 3.5

B B B B B B B B B B B B B B B B

+ + + + + + + + + + + + + + + + + +

NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS

+ + + + + + + + + + + + + + + +

NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS

Note: B = Bearing failure; NS = Net section tension failure.

Note: B = Bearing failure; NS = Net section tension failure.

one-bolted connections as shown in Table 2, the mean values of Pu,T/PASCE and Pu,T/PEC are 1.35 and 1.21, with the corresponding coefficients of variation (COV) of 0.114 and 0.051, respectively, whereas the mean values of Pu,T/PASCE and Pu,T/PEC are 1.36 and 1.23, with the corresponding COV of 0.115 and 0.101, respectively, for Series T-D-1-12. The mean values of Pu,T/PASCE and Pu,T/PEC are 1.28 and 1.25, with the corresponding COV of 0.048 and 0.057, respectively, for Series L-D-1-12, as shown in Table 2. The comparison of the test strengths with the nominal strengths for two-parallel bolted, two-perpendicular bolted, three-bolted and fourbolted connections is shown in Tables 3–6, respectively. The deterioration of the nominal strengths of the stainless steel double shear bolted connections has a similar deterioration as those of the material properties (f0.2,T/f0.2,N and fu,T/fu,N) at elevated temperatures. This is due to the calculation of the connection strengths that used the same coefficients for both room and elevated temperatures. Fig. 10(a)–(c) shows the deterioration of the test strengths of the double shear bolted connections as the

Table 8 Transient state test results for specimens of two-parallel bolted connections. Pu,N (kN)

Load level

Actual load (kN)

Critical temperature (°C)

Displacement (mm)

Failure mode

A-D-2Pa-8

42.1

0.25 0.25 0.50 0.50 0.75 0.75 0.75 0.25 0.25 0.50 0.50 0.75 0.75 0.75 0.25 0.25 0.50 0.50 0.75 0.75 0.75

10.53 10.53 21.05 21.05 31.58 31.58 31.58 9.80 9.80 19.60 19.60 29.40 29.40 29.40 13.05 13.05 26.10 26.10 39.15 39.15 39.15

811 813 680 680 270 337 306 890 885 757 758 409 521 447 767 780 662 664 414 549 545

13.9 14.5 6.3 6.7 3.9 4.0 3.8 15.7 15.6 8.0 8.5 3.0 3.5 3.0 10.9 13.2 5.9 6.0 3.1 4.0 4.1

B B B B B B B B B B B B B B B B B B B B B

L-D-2Pa-8

39.2

52.2

Note: B = Bearing failure; NS = Net section tension failure.

Specimen

Pu,N (kN)

Load level

Actual load (kN)

Critical temperature (°C)

Displacement (mm)

Failure mode

A-D-3-8

33.4

T-D-3-8

31.8

L-D-3-8

42.8

0.25 0.25 0.50 0.50 0.75 0.25 0.25 0.50 0.50 0.75 0.75 0.25 0.25 0.50 0.50 0.75 0.75

8.35 8.35 16.70 16.70 25.05 7.95 7.95 15.90 15.90 23.85 23.85 10.7 10.7 21.40 21.40 32.10 32.10

821 829 704 703 455 903 889 775 786 633 621 781 784 677 684 577 540

9.2 9.6 5.2 5.1 2.6 9.1 9.3 5.4 6.2 3.3 3.3 7.9 8.1 4.5 4.5 3.4 3.4

B B B B B B B B B B B B B B B B B

+ + + + + + + + + + + + + + + + +

NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS

Note: B = Bearing failure; NS = Net section tension failure.

Specimen

T-D-2Pa-8

Table 10 Transient state test results for specimens of three-bolted connections.

+ + + + + + + + + + + + + + + + + + + + +

NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS

Table 11 Transient state test results for specimens of four-bolted connections. Specimen

Pu,N (kN)

Load level

Actual load (kN)

Critical temperature (°C)

Displacement (mm)

Failure mode

A-D-4-6

39.5

T-D-4-6

35.3

L-D-4-8

44.3

0.25 0.50 0.50 0.75 0.75 0.25 0.25 0.50 0.50 0.75 0.75 0.25 0.25 0.50 0.50 0.75

9.88 19.75 19.75 29.63 29.63 8.83 8.83 17.65 17.65 26.48 26.48 11.08 11.08 22.15 22.15 33.23

816 673 672 264 252 886 883 762 760 608 607 777 781 687 679 564

9.3 4.4 4.9 2.2 2.4 9.6 9.6 5.7 5.6 2.9 3.1 7.5 8.0 4.8 4.1 3.4

B B B B B B B B B B B B B B B B

Note: B = Bearing failure; NS = Net section tension failure.

+ + + + + + + + + + + + + + + +

NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

(a) One-bolted connections

(b) Two-parallel bolted connections

(c) Two-perpendicular bolted connections

(d) Three-bolted connections

59

(e) Four-bolted connections Fig. 9. Comparison of transient state test results and steady state test results for double shear bolted connections: (a) One-bolted connections; (b) Two-parallel bolted connections; (c) Two-perpendicular bolted connections; (d) Three-bolted connections; (e) Four-bolted connections.

temperature increases for cold-formed stainless steel types A, T and L, respectively. The horizontal axis plots the specimen temperature, while the vertical axis represents the normalized f0.2,T/f0.2,N, fu,T/fu,N and Pu,T/Pu,N. It is shown that the reduction factors of fu,T/fu,N generally agree well with those of the connection strengths. 6. Comparison of steady state with transient state test results The comparison of the steady state test results with the transient state test results for the three different grades of stainless steel double shear bolted connections is shown in Fig. 9(a)–(e). The specimens having the same number of bolt and bolt configuration are compared in each figure. It is shown that the critical temperatures obtained from

the transient state tests are generally slightly higher than those deduced from the steady state test results at the load level of 0.50Pu,N. For instance, the critical temperatures at the load level of 0.50Pu,N for specimens A-D-1-12, T-D-1-12 and L-D-1-12 were approximately 660, 740 and 660 °C, whereas the critical temperatures based on the steady state tests were approximately 630, 725 and 605 °C, respectively. It should be noted that in some cases the critical temperatures under the load level of 0.75Pu,N are scattered, for example, 304 and 563 °C at the load level of 0.75Pu,N for specimen L-D-1-12, as shown in Table 7. This may be due to the connection strengths that are similar for the temperature ranged from 200 to 500 °C in the steady state test results, as shown in Fig. 9(a). The critical temperature under the load level of 0.75Pu,N in the transient state tests generally occurs in the temperature

60

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

performed by the steady state tests for the cold-formed stainless steel double shear bolted connections. Generally, the transient state test results of the stainless steel double shear bolted connections are revealed to provide higher critical temperatures when compared with the coupon test results of fu,T/fu,N for the stainless steel grades A and T at the same load levels, while the stainless steel grade L generally provides higher critical temperatures when compared with the coupon test results of f0.2,T/f0.2,N at the same load levels. 7. Failure modes at elevated temperatures

(a) Series A

(b) Series T

(c) Series L Fig. 10. Comparison of transient state test results and steady state test results for different grades of stainless steel: (a) Series A; (b) Series T; (c) Series L.

Tables 2–6 show the observed failure modes of each stainless steel double shear bolted connection test specimen in steady state tests. Two failure modes were observed, namely the bearing failure (B) and the net section tension failure (NS). Interaction of these two failure modes (B + NS) was also observed. It should be noted that bearing failure mode occurred in all specimens, where the material in front of bolt hole was piled up and the bolt hole was elongated. The failure mode of net section tension involved the necking of cross section in the plate and cracks near the bolt holes. The bolt shear failure and tear out failure (end pull out failure) were not observed in the tests. These failure modes were deliberately avoided in the design of test specimens, as this study focused on bearing failure of stainless steel double shear bolted connections. The one-bolted and two-parallel bolted connection specimens failed in bearing only for the three different types A, T and L of stainless steel at elevated temperatures as shown in Tables 2–3. The net section tension failure was observed in addition to the bearing failure for twoperpendicular bolted, three-bolted and four-bolted connection specimens in the temperature ranged from 22 to 500 °C. The net section tension failure mode was disappeared as the temperature increases. Only bearing failure mode was found for these specimens in the temperature ranged from 650 to 950 °C. Therefore, the failure mode changed from interaction of bearing and net section tension to bearing only for these specimens as shown in Tables 4–6. The bearing failure mode of the specimens in Series T-D-1-12 at different temperatures using steady state test method is shown in Fig. 11. The failure modes of each cold-formed stainless steel double shear bolted connection specimen by transient state test method are listed in Tables 7–11. The cold-formed stainless steel double shear bolted connection specimens failed by the combination of bearing and net section tension failure modes based on the observation from the tests. However, it is found that the bearing failure mode is more obvious for specimens at the load level of 0.25Pu,N compared with those specimens at the load levels of 0.50Pu,N and 0.75Pu,N, i.e. the elongation of bolt hole as illustrated in Fig. 12. The tear out failure (end pull out failure) and bolt shear failure were not observed in any specimen in the transient state tests, and these failure modes were also not observed in the steady state tests. The combination of bearing and net section tension failure modes of specimens L-D-1-12 tested at three different load levels are shown in Fig. 12. 8. Conclusions

ranges 200 to 500 °C, as shown in Fig. 9(a)–(e). Furthermore, the heating rate of the furnace below 520 °C is quite fast that may cause the critical temperatures scattered. The steady state test results and transient state test results for double shear bolted connections with different numbers and diameters of bolts which are assembled by the same grade of stainless steel are plotted in Fig. 10(a)–(c). The deteriorations of the 0.2% proof stress (f0.2,T/f0.2,N) and tensile strength (fu,T/fu,N) obtained from the material coupon tests [24] using steady state test method are also plotted in Fig. 10(a)–(c). It was found that the tendency of reduction of the connection strengths conducted by the transient state tests is similar to those specimens

A test program on the structural behavior of cold-formed stainless steel double shear bolted connections at elevated temperatures conducted by steady state and transient state test methods has been presented. In the steady state tests, the stainless steel bolted connections were investigated in the temperature ranged from 200 to 950 °C, while in the transient state tests, the connections were tested under 3 different load levels, namely 0.25, 0.50 and 0.75 of the failure load at room temperature. The double shear bolted connection specimens were fabricated by three different grades of stainless steel with plate thickness of 1.5 mm. The three types of stainless steel are austenitic stainless steel EN 1.4301 (AISI 304) and EN 1.4571 (AISI 316Ti having small amount of titanium) as well as lean duplex stainless steel

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

(a) 22 ºC

(b) 206 ºC

(c) 359 ºC

(d) 508 ºC

(e) 652 ºC

(f) 806 ºC

61

(g) 950 ºC Fig. 11. Failure mode of double shear bolted connections (Series T-D-1-12) in the steady state tests: (a) 22 °C; (b) 206 °C; (c) 359 °C; (d) 508 °C; (e) 652 °C; (f) 806 °C; (g) 950 °C.

EN 1.4162 (AISI S32101). The connection specimens were tested by varying bolt diameters, number of bolts and arrangement of the bolts. The connection strengths and failure modes of the cold-formed stainless steel double shear bolted connections were examined. The connection strengths for different types of cold-formed stainless steel double shear bolted connections at elevated temperatures were compared. It was found that the austenitic stainless steel type EN

1.4571 generally has better resistance than the stainless steel types EN 1.4301 and EN 1.4162 at elevated temperatures in steady state tests and transient state tests. Furthermore, the test strengths of the double shear bolted connections were compared with the nominal strengths calculated from the design rules in the ASCE Specification, AS/NZS Standard and Eurocodes for stainless steel structures. The reduced material properties due to elevated temperatures were used in the calculation.

62

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63

(a) Load level of 0.25Pu,N

Nomenclature d nominal diameter of stainless steel bolt; nominal diameter of bolt hole; d0 elastic modulus at room (ambient) temperature; EN the elastic modulus at elevated temperatures; ET longitudinal 0.2% tensile proof stress at room temperature; f0.2,N longitudinal 0.2% tensile proof stress at elevated temperatures; f0.2,T longitudinal tensile strength at room temperature; fu,N longitudinal tensile strength at elevated temperatures; fu,T ultimate strain at room temperature; εu,N ultimate strain at elevated temperatures; εu,T nominal strength of bolted connection calculated based on Pu,ASCE ASCE Specification; Pu,AS/NZS nominal strength of bolted connection calculated based on AS/NZS Standard; nominal strength of bolted connection calculated based on Pu,EC Eurocodes; ultimate load of bolted connection test at room temperature; Pu,N ultimate load of bolted connection test at elevated Pu,T temperatures; Ppre,load pre-load for transient state test; θ gas temperature in fire compartment; t time in minute.

Acknowledgments

(b) Load level of 0.50Pu,N

The authors are grateful to STALA Tube Finland for supplying the test specimens. The research work described in this paper was supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project No. HKU718612E). References

(c) Load level of 0.75Pu,N Fig. 12. Failure modes of double shear bolted connections (Series L-D-1-12) in the transient state tests: (a) Load level of 0.25Pu,N; (b) Load level of 0.50Pu,N; (c) Load level of 0.75Pu,N.

It was found that the current design formulas in these three specifications by substituting the reduced material properties at elevated temperatures generally underestimate the connection strengths of the cold-formed stainless steel double shear bolted connections. A similar strength reduction tendency was found for the stainless steel double shear bolted connections in both the transient state tests and the steady state tests. The strength reduction tendency is also similar to the material coupon test results. The bearing failure and combination of bearing and net section tension failure modes of cold-formed stainless steel double shear bolted connection specimens were observed in the steady state tests, while only the combination of bearing and net section tension failure modes was observed in the transient state tests.

[1] Ashraf M, Gardner L, Nethercot DA. Structural stainless steel design: resistance based on deformation capacity. J Struct Eng ASCE 2008;134(3):402–11. [2] Gardner L. The use of stainless steel in structures. Prog Struct Eng Mater 2005;7: 45–55. [3] Gardner L, Baddoo NR. Fire testing and design of stainless steel structures. J Constr Steel Res 2006;62(6):532–43. [4] Gardner L, Ng KT. Temperature development in structural stainless steel sections exposed to fire. Fire Saf J 2006;41(3):185–203. [5] American Society of Civil Engineers (ASCE). Specification for the design of coldformed stainless steel structural members. ASCE Standard, SEI/ASCE-8-02, Reston, Virginia; 2002. [6] Australian/New Zealand Standard (AS/NZS). Cold-formed stainless steel structures. AS/NZS 4673:2001, Standards Australia, Sydney, Australia; 2001. [7] EC3-1.4, Eurocode 3. Design of steel structures — part 1. General rules — supplementary rules for stainless steels. European Committee for Standardization, BS EN 1993-14:2006, Brussels; 2006. [8] Zadanfarrokh F. Analysis and design of bolted connections in cold formed steel members. [PhD Thesis] The University of Salford; 1991. [9] Rogers CA, Hancock GJ. Bolted connection tests of thin G550 and G300 sheet steels. J Struct Eng ASCE 1998;124(7):798–808. [10] Rogers CA, Hancock GJ. Bolted connection design for sheet steels less than 1.0 mm thick. J Constr Steel Res 1999;51(2):123–46. [11] Rogers CA, Hancock GJ. Failure modes of bolted-sheet-steel connections loaded in shear. J Struct Eng ASCE 2000;126(3):288–96. [12] Bouchaïr A, Averseng J, Abidelah A. Analysis of the behaviour of stainless steel bolted connections. J Constr Steel Res 2008;64(11):1264–74. [13] Cai Y, Young B. Structural behavior of cold-formed stainless steel bolted connections. Thin-Walled Struct 2014;83:147–56. [14] Sakumoto Y, Nakazato T, Matsuzaki A. High-temperature properties of stainless steel for building structures. J Struct Eng ASCE 1996;122(4):399–406. [15] Makelainen P, Outinen J. Mechanical properties of an austenitic stainless steel at elevated temperatures. J Constr Steel Res 1998;46(1–3):455. [16] Chen J, Young B. Stress–strain curves for stainless steel at elevated temperatures. Eng Struct 2006;28(2):229–39. [17] Brnic J, Turkalj G, Canadija M, Lanc D. AISI 316Ti (1.4571) steel—mechanical, creep and fracture properties versus temperature. J Constr Steel Res 2011;67(12): 1948–52. [18] Brnic J, Niu JT, Turkalj G, Canadija M, Lanc D. Experimental determination of mechanical properties and short-time creep of AISI 304 steel at elevated temperatures. Int J Miner Metall Mater 2010;17(1):19–45. [19] Ng KT, Gardner L. Buckling of stainless steel columns and beams in fire. Eng Struct 2007;29(5):717–30.

Y. Cai, B. Young / Journal of Constructional Steel Research 104 (2015) 49–63 [20] To E, Young B. Performance of cold-formed stainless steel tubular columns at elevated temperatures. Eng Struct 2008;30(7):2012–21. [21] Yan S, Young B. Tests of single shear bolted connections of thin sheet steels at elevated temperatures — part I: steady state tests. Thin-Walled Struct 2011;49(10):1320–33. [22] Yan S, Young B. Tests of single shear bolted connections of thin sheet steels at elevated temperatures — part II: transient state tests. Thin-Walled Struct 2011;49(10): 1334–40. [23] Yan S, Young B. Effects of elevated temperatures on double shear bolted connections of thin sheet steels. J Struct Eng ASCE 2013;139(5):757–71.

63

[24] Cai Y, Young B. Behavior of cold-formed stainless steel single shear bolted connections at elevated temperatures. Thin-Walled Struct 2014;75:63–75. [25] EC3-1.8. Eurocode 3: design of steel structures—part 1.8: design of joints. European Committee for Standardization, BS EN 1993-1-8, Brussels; 2005. [26] BS EN ISO 3506-1. Mechanical properties of corrosion-resistant stainless steel fasteners—part 1: bolts, screws and studs. BS EN ISO 3506-1; 1998. [27] EC1-1.2. Eurocode 1 – actions on structures – part 1-2: general actions — actions on structures exposed to fire. European Committee for Standardization, EN 1991-12:2002, Brussels; 2002.