experimental behaviour of end-plate beam-to-column …luisss/eccs/ficheiros/a89-ir08-eccsv5.pdf ·...

DEPARTMENT OF CIVIL ENGINEERING

UNIVERSITY OF COIMBRA

ECCS TECHNICAL COMMITTEE 10 “CONNECTIONS”

TWG 10.2

EXPERIMENTAL BEHAVIOUR OF END-PLATE

BEAM-TO-COLUMN JOINTS UNDER BENDING AND

AXIAL FORCE

Database reporting and discussion of results

by

L. SIMÕES DA SILVA, L. LIMA, P. VELLASCO and S. ANDRADE

Meeting in Ljubljana

April 11-12, 2001

UC/DEC/GCOM/2002-008/RR/JOINT and ECCS TC10-02-WG2/0xx (Version 1 – 05/04/2002)

Experimental behaviour of endplate beam-to-column joints under bending and axial force

UC/DEC/GCOM/2002-008/RR/JOINT,and ECCS TC10-02-WG2/0xx 09-04-2002

2

EXPERIMENTAL BEHAVIOUR OF ENDPLATE BEAM-TO-COLUMN JOINTS UNDER BENDING AND AXIAL FORCE

Luís Simões da Silva1, Luciano R. O. de Lima2, Pedro C. G. da S. Vellasco3 e Sebastião A. L. de Andrade4

ABSTRACT

The objective of the present paper is to present the results of an experimental research project on endplate beam-to-column bolted steel joints subjected to ending and axial force currently being currently carried out at the University of Coimbra.

Because of the ongoing nature of this work, this paper is an updated version of a previous Document presented in Timisoara (Document TC10-01-WG2/011), 15 tests being now reported from flush and extended end-plate configurations. In addition, special emphasis is given to database reporting, to allow other research centres to use this experimental data for calibration of theoretical models.

1. INTRODUCTION

A great part of the beam-to-column connections and beams-splices are subject to axial forces, bending and shear. The interaction of these efforts modifies the behaviour of the joint with respect to its initial stiffness, bending resistance and rotation capacity.

In the Eurocode 3, [1], it is possible to evaluate the resistant capacity of a joint subject to bending moment and axial force whenever the latter is less than a maximum limit, given by the equation (1),

1.0≤plN

N (1)

where N is the beam axial force and Npl is the beam plastic resistance subjected to compression. Under these conditions, the axial force may be disregarded in the design of the joint. It is important to point out that there is no background to justify this empirical limit of 10%. For connections that exceed the limit range of this equation, Eurocode 3 does not make any specific recommendation. However, the general principles of the component method contemplate this case, since any component is characterized independently of the type of applied loading to the connection. It is up to the designer to define a coherent component 1 Department of Civil Engineering, University of Coimbra, Portugal 2 Department of Civil Engineering, PUC-Rio – Rio de Janeiro – Brazil (PhD Student - PDEE – CAPES – Brazil) 3 Department of Structures and Foundations, UERJ – Rio de Janeiro – Brazil 4 Department of Civil Engineering, PUC-Rio – Rio de Janeiro – Brazil and Department of Structures and

Foundations, UERJ – Rio de Janeiro – Brazil



3

model that identifies the relevant interactions between the various components. Recent attempts to establish specific procedures for these types of connections were performed for Jaspart et al, [2], Cerfontaine, [3], Silva and Coelho, [4] and Silva et al, [5]. To provide as sound basis for further theoretical developments, a set of 15 tests was carried out at the University of Coimbra and are briefly described in the next section [6, 7]

2. DESCRIPTION OF THE EXPERIMENTAL TESTS

A series of 8 (eight) experimental tests of beam-to-column steel connections with flush endplate, typically illustrated in Figure 1, were already carried out at the University of Coimbra, within a research program on behaviour of beam-to-column joints under bending and axial force. In the first test, FE1, only bending moment was applied through a hydraulic actuator, located a meter away from the face of the column flange, Figure 1(a). For the following tests - FE3, FE4, FE5, FE6, FE7, FE8 and FE9 - constants axial forces of, respectively, -4%, -8%, -20%, -27%, -20%, +10% and 20% of the beam plastic resistance were applied to the beam.

A further series of 7 (seven) tests with extended endplate joints were also performed, Figure 1(b). As in the first test series, only bending moment was applied for test EE1. For the subsequent tests – EE2, EE3, EE4, EE5, EE6 and EE7 - constant axial forces of, respectively, -10%, -20%, -27%, -15%, +10% and +20% of the beam plastic resistance were applied to the beam.

In all tests, the columns were simply-supported at both ends and consist of a HEB240, the beams consist of an IPE240 and the endplate is 15 mm thick, all manufactured from a steel S275. The bolts are M20, class 10.9.

The compressive force application system, Figure 2(a), was composed of a hydraulic jack that applies a tension force to four cables of prestressing with diameter φ = 15,2 mm. The transfer of this force for the connection was performed through a central load cell with capacity of 500kN (NOVATECH), Figure 2 (c). These cables pass by a deviator constituted by a profile HEM100 to guarantee that the axial force is always parallel to the beam axis. A load cell TML with capacity of 200 kN in each cable was also used, with the purpose of to obtain the force in each cable. The transmission of the applied axial force in the connection to the reaction wall was made through a reinforced concrete footing that was connected to the reaction wall by means of a profile HEB 200, and prestressed to the reaction slab through bars DYWIDAG.

The tensile axial force application system is shown in Figure 3. Four hydraulic jacks placed in one of the extremities of a circular profile transmit the tension axial force. These circular profiles are simply supported in the other extremity for allow the rotation of these profiles and to guarantee that the axial force is applied always parallel to the beam axis.



4

(a) (b)

Figure 1 –Endplate joints layout

(a)

(b)

(c)

Figure 2 – (a) Frame of application loading, (b) hydraulic jack and (c) central load cell

Figure 3 – Experimental test layout for tests subjected to bending and tension axial force

desviator



5

All tests were instrumented as shown in Figure 4 and Figure 5, with single strain gauges (FLK 6-11-TML), rosettes to 45º (FRA 5-11-TML), bolts axial strain gauges (BTM 6-C-TML), and displacements transducers (LVDT's), in order to evaluate the main characteristics of the connection, such as, bending moment resistance, initial stiffness and rotation capacity. The registration of measurements was done with a data acquisition system TDS602-TMlL.

IPE240

HEB240

34

56

11

18/19

20/21

2223

/24

2829/30

31/32

78

9

1213

14

252627

10

18 19

20 21

2223 24

25, 26 & 27

28

29 30

31 32

B35 B36

B37 B38

28 32 32 28

25

3778

7837

267

20

IPE240

HEB

240

CH39 CH40

CH41 CH42

CH44

CH43CH45

CH47

CH46

CH48

CH49

1000500

16080

6060

230.

2

Figure 4 – Layout of Single strain gauges, rosettes to 45º, bolts axial strain gauges layout and

displacements transducers

(a)

(b)

Figure 5 – (a) Displacements transducers and (b) data acquisition system

For all tests, a constant axial force was applied first, maintained constant throughout

the test, with subsequent application of a bending moment incremented to failure. Two unloadings were performed, the first for a bending moment of 25 kNm (down to 5 KNm, to eliminate possible slack in the joint) and the second for a rotation of 20 mrad. Force control was used in the first part of each test, subsequently changed to displacement control.

Tensile tests on coupons extracted from the beams were carried out, aiming at characterizing the actual properties of the material. Then, it was possible to calculate the beam plastic resistance and to determine which the true level of applied axial force to the beam for the other tests. These tensile tests were accomplished according to the following specifications, EN10002 [6], EN10020 [7] and EN10025 [8], yielding the results of Table 1.



6

Table 1 – Steel mechanical properties (beam of the first test - FE1)

Specimen fy (MPa) fu (MPa) Flange3A 312.1 439.2 Flange3B 319.7 459.2

Flange average 315.9 449.2 Web2B 345.1 451.4 Web3B 332.3 443.4 Wen3A 350.0 457.9

Web average 342.5 450.9

3. ANALYSIS OF THE EXPERIMENTAL RESULTS FOR THE FLUSH ENDPLATE JOINTS

3.1. Moment vs. Rotation Curves and Application of the Axial Force

The moment vs. rotation curves of the seven tests are presented below where it may be observed that even for a level of equivalent axial force of 20% of the beam plastic resistance, the bending moment is still higher than the Eurocode 3. This is due to the fact that the components in tension for two bolt row are alleviated by the compression axial force and the components in compression, even increased, don't reach the strength, even in presence of the axial force, Figure 6. The results of test FE9 (applied axial force of + 20% of the beam plastic resistance) will be added to the results shortly.

0

20

40

60

80

100

120

0 10 20 30 40 50 60 70 80 90 100Rotation (mrad)

Ben

ding

Mom

ent (

kN.m

)

FE1 (only M)FE3 (N = - 4% Npl)FE4 (N = - 8% Npl)FE5 (N = - 20% Npl)FE6 (N = - 27% Npl)FE7 (N = - 20% Npl)FE8 (N = + 10% Npl)

Figure 6 – Moment vs. rotation curves

With the purpose of calibrating the axial force application system, a initial test was

performed (FE2) in elastic regime. It was verified that the applied axial load with the cables



7

was transmitted by the central load cell to the connection as shown in Figure 7. In this graph, it was verified that the axial force applied to the connection, measured through the strain gages located in the web and flange beam and the measured by the central load cell is similar.

-300

-250

-200

-150

-100

-50

0-25 -20 -15 -10 -5 0

Displacement (mm)

Axia

l For

ce (k

N)

adds of the individual cells (1, 2, 3 e 4)strain gagesscentral load cell (5)

18 19

20 21

29 30

31 32

Figure 7 –– Applied axial force verification

For higher levels of bending moment, it was verified that the rotation of the beam

provoked a reduction of the force in the lower cables and an increment of load of the top cables. Consequently, hydraulic jacks were placed in the lower cables to make the correction of the axial force as the test is being performed. In Figure 8, the variation of the axial force is exemplified in the four individual cables with the moment applied for each test, as well as a detail of the hydraulic jacks. The correction can be observed in item (c).

The Table 2 presents the values obtained for the bending moment resistance and for the initial stiffness of the tests connections. The values of the initial stiffness were obtained from slope of the unloading part of the moment vs. rotation curves. The theoretical values calculated according to Eurocode 3 were respectively, 68.9 kN.m and 5915.7 kN.m/rad being disregarded the presence of the axial force.

Table 2 - Experimental values of bending moment resistance and initial stiffness

Test N (kN) MRd (kN.m) Si (kN.m/rad)

FE1 (only M) - 67.8 5785 FE3 (- 4% Npl) 52.7 72.0 5947 FE4 (- 8% Npl) 105.6 74.0 6250 FE5 (- 20% Npl) 265.0 75.5 6163 FE6 (- 27% Npl) 345.0 72.4 7869 FE7 (- 20% Npl) 265.0 76.4 5914 FE8 (+ 10% Npl) 130.6 58.9 5488

1111 2222

3333 4444 5555



8

y = 0.0005x - 52.679-60

-50

-40

-30

-20

-10

00 20 40 60 80 100

Bending Moment (kN.m)Ax

ial F

orce

(kN

)

(3) e (4)

(5)

(1) e (2)

(a) Test FE3

y = 0.0077x - 105.62

-140

-120

-100

-80

-60

-40

-20

00 20 40 60 80 100

Bending Moment (kN.m)

Axia

l For

ce (k

N)

(3) e (4)

(5)

(1) e (2)

(b) Test FE4

y = 0.0079x - 344.58

-400

-350

-300

-250

-200

-150

-100

-50

00 10 20 30 40 50 60 70 80 90 100

Bending Moment (kN)

Axi

al F

orce

(kN

)

(1), (2), (3) and (4)

(5)

(c) Test FE6

(d) Hydraulic jacks Figure 8 – Bending moment vs. axial force curves

3.2. Analysis of Individual Components

The Table 3 presents the theoretical values of the bending moment resistance and initial stiffness for all components of the connection in study, calculated in agreement with Eurocode 3.

Table 3 - Theoretical values of the resistance and stiffness of the components

Component Resistance (kN) Stiffness (mm) Column web in tension (3) 458,7 7,03 Column flange in bending (4) 397,2 38,22 Endplate in bending(5) 321,7 13,35 Beam web in tension (8) 455,2 ∞

Tension

Bolts in tension (10) 441,0 7,76 Column web in shear (1) 552,7 7,52 Column web in compression (2) 598,2 10,40 Compression Beam flange in compression (7) 503,6 ∞∞∞∞

In Figure 9, the moment vs. rotation curves of the four tests can be observed with the

experimental identification of the yielding sequence of the several components. It is clearly noticed that application of the compression axial force benefits the critical component of the tension zone (endplate in bending) and it decreases the capacity of the critical component of the compression zone (beam flange in compression).



9

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90 100Rotation (mrad)

Bend

ing

Mom

ent (

kN.m

)

Experimental - FE1Endplate in bending (5)Beam flange in compression (7)Bolts in tension (10)Column flange in bending (4)Beam web in tension (8)Column web in shear (1)

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90 100Rotation (mrad)

Bend

ing

Mom

ent (

kN.m

)

Experimental - FE3Endplate in bending (5)Beam flange in compression (7)Bolts in tension (10)Column flange in bending (4)Beam web in tension (8)Column web in shear (1)

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90 100Rotation (mrad)

Bend

ing

Mom

ent (

kN.m

)

Experimental - FE4Beam flange in compression (7)Endplate in bending (5)Bolts in tension (10)Column flange in bending (4)Beam web in tension (8)Column web in shear (1)

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90 100Rotation (mrad)

Bend

ing

Mom

ent (

kN.m

)

Experimental - FE5Beam flange in compression (7)Endplate in bending (5)Column flange in bending (4)Beam web in tension (8)Bolts in tension (10)Column web in shear (1)

Figure 9 – Moment vs. rotation curves with yielding sequence of the components

In the graph presented in the Figure 10 it can be observed that, for all the tests, the

column flange in bending presented deformations according to mode 1, that is, complete yielding of the flange. The measured displacements for this component are presented in Figure 11 where it is noticed that the behaviour of the component is similar for all the tests, independently of the applied axial force.

0

20

40

60

80

100

120

-1000 0 1000 2000 3000 4000 5000

Strain (µεµεµεµε)

Ben

ding

Mom

ent (

kN.m

)

FE1 (only M)

FE3 (N = - 4% Npl)

FE4 (N = - 8% Npl)

FE5 (N = - 20% Npl)

FE6 (N = - 27% Npl)

FE7 (N = - 20% Npl)

FE8 (N = + 10% Npl)

Figure 10 – Moment vs. strain curves for column flange in bending component

FE1 FE3

FE4 FE5



10

0

20

40

60

80

100

120

-6 -4 -2 0 2 4 6 8 10 12Displacement (mm)

Ben

ding

Mom

ent (

kN.m

)

FE1 (only M)

FE3 (N = - 4% Npl)

FE4 (N = - 8% Npl)

FE5 (N = - 20% Npl)

FE6 (N = - 27% Npl)

FE7 (N = - 20% Npl)

FE8 (N = + 10% Npl)

Figure 11 – Moment vs. displacements curves for column flange in bending component

In agreement with the design rules of Eurocode 3, the resistance of the endplate in

bending is equal to 321,7 kN. In the graph of the Figure 12, this component, in test FE1 reaches the yielding strain for a bending moment of, approximately, 45 kN.m. For this bending moment level, the tension load in the first bolt row, evaluated in agreement with the strain gages located in the beam flange, it is equal to 333,4 kN. However, for the test FE5, due to the contribution of the applied axial force, the endplate reached the yield strain for a higher level of applied moment. In Figure 13 the moment vs. displacements curves for this component are presented.

0

20

40

60

80

100

120

-80000 -70000 -60000 -50000 -40000 -30000 -20000 -10000 0 10000


Ben

ding

Mom

ent (

kN.m

)

FE1 (only M)

FE3 (N = - 4% Npl)

FE4 (N = - 8% Npl)

FE5 (N = - 20% Npl)

FE6 (N = - 27% Npl)

FE7 (N = - 20% Npl)

FE8 (N = + 10% Npl)

Figure 12 - Moment vs. strain curves for endplate in bending component



11

0

20

40

60

80

100

120

-25 -20 -15 -10 -5 0 5Displacement (mm)

Ben

ding

Mom

ent (

kN.m

)

FE1 (only M)

FE3 (N = - 4% Npl)

FE4 (N = - 8% Npl)

FE5 (N = - 20% Npl)

FE6 (N = - 27% Npl)

FE7 (N = - 20% Npl)

FE8 (N = + 10% Npl)

Figure 13 - Moment vs. displacements curves for endplate in bending component

Analysing the curves presented in Figure 14, it is clearly noticed that the beam flange

also reaches yielding. According to Eurocode 3, the beam flange in compression resistance is 529,3 kN. For this bending moment level, the average of the measured strains in the inferior beam flange was 2300µε, that it, is equal to a force of 570,0 kN, higher than the 529,3 kN presented above, fact that can be explained through the hardening of the steel. However, it is worthwhile to point out that for the first test, where the compression axial force was not applied to the beam, beam flange yielding occurred for a larger value of bending moment than in the other tests.

0

20

40

60

80

100

120

-70000 -60000 -50000 -40000 -30000 -20000 -10000 0 10000


Ben

ding

Mom

ent (

kN.m

)

FE1 (only M)

FE3 (N = - 4% Npl)

FE4 (N = - 8% Npl)

FE5 (N = - 20% Npl)

FE6 (N = - 27% Npl)

FE7 (N = - 20% Npl)

FE8 (N = + 10% Npl)

Figure 14 - Moment vs. strain curves for beam flange in compression component



12

As it can be seen in Figure 15, the column web in compression component did not reach yielding for any of the tests.

0

20

40

60

80

100

120

-1500 -1300 -1100 -900 -700 -500 -300 -100 100


Ben

ding

Mom

ent (

kN.m

)

FE1 (only M)

FE3 (N = - 4% Npl)

FE4 (N = - 8% Npl)

FE5 (N = - 20% Npl)

FE6 (N = - 27% Npl)

FE7 (N = - 20% Npl)

FE8 (N = + 10% Npl)

Figure 15 - Moment vs. displacements curves for column web in compression component

The column web in shear component reached yielding for all tests, Figure 16. This

graph was obtained through a rosette positioned in the centre of the column web panel where the used component was the ε45º.

0

20

40

60

80

100

120

-2000 0 2000 4000 6000 8000 10000 12000 14000 16000 18000


Ben

ding

Mom

ent (

kN.m

)

FE1 (only M)

FE3 (N = - 4% Npl)

FE4 (N = - 8% Npl)

FE5 (N = - 20% Npl)

FE6 (N = - 27% Npl)

FE7 (N = - 20% Npl)

FE8 (N = + 10% Npl)

Figure 16 - Moment vs. strain curves for column web in shear component

ε45º



13

4. ANALYSIS OF EXPERIMENTAL RESULTS FOR EXTENDED ENDPLATE JOINTS

4.1. Moment vs. Rotation Curves and Application of the Axial Force

The moment vs. rotation curves of the seven tests are presented below where it may be observed that even for a level of equivalent axial force of -10% of the beam plastic resistance, the bending moment is still higher than the Eurocode 3. This is due to the fact that the components in tension for the two bolt rows are alleviated by the compression axial force and the components in compression, even increased, don't reach the strength, even in presence of the axial force, Figure 17. The results of test EE6 (applied axial force of + 10% of the beam plastic resistance) show that the bending moment was reduced by 10%. In Figure 18, some deformations that occurred in these tests may be observed.

0

20

40

60

80

100

120

140

160

0 10 20 30 40 50 60 70 80 90 100Rotation (mrad)

Ben

ding

Mom

ent (

kN.m

)

EE1 (only M)

EE2 (N = - 10% Npl)

EE5 (N = - 15% Npl)

EE3 (N = - 20% Npl)

EE4 (N = - 27 % Npl)

EE6 (N = +10% Npl)

EE7 ( N = +20% Npl)

Figure 17 – Moment vs. rotation curves

Figure 18 – Column flange, beam flange and endplate deformations



14

5. DATABASE REPORT

The data for all flush endplate tests will be available on the website of the Mechanical Structural Laboratory of the University of Coimbra, Figure 19.

y = 0.167x + 72.307y = 4.763x + 5.7203

0

10

20

30

40

50

60

70

80

90

100

0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00Rotation (mrad)

Bend

ing

Mom

ent (

kN.m

)

FE1 (only M)

Figure 19 – DATABASE information

6. CONCLUDING REMARKS

The experimental results obtained to the moment for beam-to-column connections with flush endplate allow the elaboration of a moment vs. axial force interaction curve corresponding to the elastic resistance of the connection, just as it illustrates the Figure 20.

Database on website - http://www.dec.uc.pt/~luisss/eccs/eccspreview.htm

http://www.dec.uc.pt/LME



15

50

55

60

65

70

75

80

-30 -20 -10 0 10 20% Npl

Bend

ing

Mom

ent (

kN.m

)

Figure 20 – M-N interaction diagram for flush endplate joints

From the analysis of the Table 3 that illustrates the resistance and elastic stiffness of

the all components of the connection in study, calculated in agreement with Eurocode 3, it is possible to conclude that, for this connection, the tension zone presents a resistance of the critical component that is equal at 63% of the component of the compression zone. Thus, the compression axial force results in an increase of the bending moment resistance of the connection, as it is verified. The results evidence the need to prepare specific clauses for the design of connections subjected to axial forces and bending moment.

7. REFERENCES

[1] Eurocode 3 (1998), ENV - 1993-1-1:1992/A2, Annex J, Design of Steel Structures – Joints in Building Frames. CEN, European Committee for Standardisation, Document CEN/TC 250/SC 3, Brussels.

[2] Jaspart, J. P., Braham, M. and Cefontaine, F. (1999), “Strength of Joints Subjected to Combined Action of Bending Moments and Axial Forces”. In: Studnicka J, Wald F, Machacek J. Eurosteel’99, Proceedings of the 2nd European Conference on Steel Structures, Czech Technical University in Prague, Prague, p. 465-468.

[3] Cerfontaine, F. (2000), “Etude analytique de l’interaction entre moment de flexion et effort normal dans les assemblages boulonnés” (in french), Travail de fin d’etudes du DEA en Sciences Appliquées, Département MSM, University of Liège, Belgium.

[4] Silva, L.S. and Coelho, A.G. (2001), “A analytical evaluation of the response of steel joints under bending and axial force”, Computers & Structures, vol. 79, 873-881

[5] Luís Simões da Silva, Luciano R. O. de Lima, Pedro C. G. da S. Vellasco e Sebastião A. L. de Andrade, (2001). “Experimental and Numerical Assessment of Beam-to-Column Joints Under Bending and Axial Force”. In Chang-Koon Choi and Hyo-Gyoung Kwak (eds), Techno Press, Proceedings of the First International Conference on Steel & Composite Structures, 715-722, Pusan, Korea.

[6] IPQ – EN 10002 (1990) – “Materiais Metálicos – Ensaios de Tracção” (in Portuguese). [7] IPQ – EN 10020 (1989) – “Definição e Classificação dos Aços” (in Portuguese). [8] IPQ – EN 10025 (1994) – “Produtos Laminados a Quente de Aços de Construção Não-

Ligados” (in Portuguese).

DEPARTMENT OF CIVIL ENGINEERINGEXPERIMENTAL BEHAVIOUR OF END-PLATE BEAM-TO-COLUMN JOINTS UNDER BENDING AND AXIAL FORCEDatabase reporting and discussion of results

L. SIMÕES DA SILVA, L. LIMA, P. VELLASCO and S. ANDRADEMeeting in Ljubljana

experimental behaviour of end-plate beam-to-column …luisss/eccs/ficheiros/a89-ir08-eccsv5.pdf ·...

Documents