0% found this document useful (0 votes)
7 views12 pages

2023 the Cyclic Performance of Column Base Plate Connections Using Different Types of Stiffeners (1)

This study investigates the cyclic performance of column base plate connections with various stiffeners under cyclic loading. Four types of stiffeners were analyzed, revealing that all methods provided sufficient strength and stiffness, with built-up stiffeners exhibiting the highest moment capacity. The findings emphasize the importance of using stiffened connections to enhance seismic performance and reduce potential damage during earthquakes.

Uploaded by

scopedoo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
7 views12 pages

2023 the Cyclic Performance of Column Base Plate Connections Using Different Types of Stiffeners (1)

This study investigates the cyclic performance of column base plate connections with various stiffeners under cyclic loading. Four types of stiffeners were analyzed, revealing that all methods provided sufficient strength and stiffness, with built-up stiffeners exhibiting the highest moment capacity. The findings emphasize the importance of using stiffened connections to enhance seismic performance and reduce potential damage during earthquakes.

Uploaded by

scopedoo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 12

Article

The cyclic performance of column base plate connections


using different types of stiffeners
Mohammed Al-Sharmooteea , Seyed Hossein Hosseini Lavassanib , Mir Hamid Hosseinia , and Mahmoud Ali c

a
Department of Civil Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran; b Department of Civil
Engineering, Kharazmi University, Tehran, Iran; c School of Civil Engineering, College of Engineering, University of Tehran, Tehran,
Iran
Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23

Corresponding author: Mahmoud Ali (email: [email protected])

Abstract
Column base connections represent the support system of a structure, and they should be strong enough to sustain the
transmitted demands between the columns and foundation. This study describes numerically the performance of strengthened
column base plate connections under cyclic loading. Four types of stiffeners were considered in the analysis: bracket plates,
angles, channels with vertical stiffeners, and built-up stiffeners. The efficiency of used stiffeners, the cyclic performance of
the strengthened connections, and along with the failure modes were investigated. The numerical results showed that all the
proposed strengthening methods have sufficient flexural strength and stiffness higher than the column. A pinched hysteretic
response was observed for all strengthened connections under the applied cyclic loading. The built-up stiffener presented the
For personal use only.

maximum moment capacity and rotational stiffness among the other configurations. Using strengthened connections with
more stiffeners made the anchor rods the fuse elements in the connection.
Key words: column base connection, base plate, stiffeners, numerical analysis, strengthening

1. Introduction
sistance with the possibility of unintentional nonlinear be-
Steel buildings are relatively widespread all over the world. havior within the embedded portion of the ECB connection.
Column base plate connections correspond to one of the most They proposed a method to provide improved estimates of
sensitive parts of steel structures as they transfer the loads the anticipated flexural demands of nondissipative ECB con-
from the superstructure to the foundations. These connec- nections. Trautner and Hutchinson (2018) presented a rapid
tions are frequently used in low-rise steel buildings, and a parametric numerical modeling approach for analyzing ex-
typical connection comprises a steel column, base plate, an- posed column base plate connections of many configurations
chor rods, grout, and a foundation (Lim et al. 2017). subjected to combined axial and lateral loads. Additionally,
The 1994 Northridge, California and the 1995 Kobe, Japan a numerical modeling approach was proposed by Singh and
earthquakes showed unsatisfactory performance of column Woods (2022) to capture the nonlinear cyclic behavior of col-
bases in both moment-resisting frames and braced frames umn base connections, including their stiffness, strength,
(Fahmy 2000). The brittle fracture of such connections was and energy dissipation capacity. Abdolahzadeh et al. (2019)
unanticipated, emphasizing the need for more investiga- also provided a hybrid modeling for describing the hysteresis
tions to improve the existing ones and develop new con- behavior of column base plate connections under cyclic load-
nections that will give the expected seismic performance ing. For block-out connections, a simple method was devel-
(Grauvilardell et al. 2005). Many studies have investigated the oped by Richards et al. (2018) to predict the flexural stiffness
behavior of column base connections under different load of such connections using a model that combines the theory
situations. Torres-Rodas et al. (2021a) indicated that grav- of beams on elastic foundations with a base rotational spring.
ity framing systems profoundly affect the behavior of steel Current design methods, such as those listed in the AISC
moment-resisting frames (SMFs) by reducing their probabil- Steel Design Guide 1 (2006), provide design procedures to de-
ity of collapse. In such a case, the seismic demands for the termine the size of the base plate and the anchor rods. When
design of column-base connections can be reduced safely, in- the connection is subjected to lateral loads, the column base
corporating their high deformation capacity as part of the connection resists the applied flexural moment by develop-
energy dissipative mechanisms. Inamasu et al. (2021) indi- ing compressive force on the foundation’s concrete and ten-
cated that the current design approaches for embedded col- sile force in some or all anchor rods. In moment-resisting
umn base (ECB) connections are unconservative, underesti- steel frames, the column base connections should be stiff
mating the seismic demands and overestimating the ECB re- and strong enough to force yielding in the column (AISC-

Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156 1


Canadian Science Publishing

Fig. 1. Geometrical properties of the experimental specimen Test #2 (mm) (Shaheen et al. 2017).
Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23

Fig. 2. Finite element model. Fig. 3. Boundary conditions and applied load.
For personal use only.

341 2022). For stiff column base connections, the variation


of beam rotation demand is reduced, which is desirable, as
it allows for demand redistribution to more frame elements
and connections and increases the degree of redundancy in
the lateral force-resisting system (Aviram et al. 2010). The re-
duction in base stiffness alters the collapse mechanism of the
structure, such that structures with more flexible bases ex-
hibit mechanisms that concentrate deformations in only a
few stories such that that structures with greater base fix-
ity are significantly more resistant to collapse (Aviram et al.
2010; Zareian and Kanvinde 2013).
Many studies proved that the behavior of column base con- to the flexibility of the base plate, flexibility of anchor bolts,
nections is closer to the semi-rigid state under cyclic load- and gaps between the bearing surface and plate caused by
ing (Astaneh-Asl et al. 1992; Burda 1999; Lee and Goel 2001; concrete crushing during reversed cyclic (Aviram et al. 2010).
Kiani and Izadinia 2015). Such partial fixity condition is due Modeling the bases as pinned results in high error and is not

2 Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156


Canadian Science Publishing

Fig. 4. Cyclic loading protocol: (a) SAC loading protocol be used for strengthening the column base connections such
(Gomez et al. 2010), and (b) applied loading history. Copyright that the plastic hinge could be developed in the column’s bot-
© American Institute of Steel Construction. Reprinted with tom. In addition, grade beams also were referred to be used
permission. All rights reserved. at the bottom of the column such that the hinging will occur
in the grade beam. In such a case, the grade beam should be
designed in conformance with ACI-318 (2016).
Falborski et al. (2020b) have shown that acceptable per-
formance for SMFs could be achieved by designing their
base connections for overstrength seismic loads rather than
based on the column capacity, offering significant cost sav-
ings. Torres-Rodas et al. (2021b, 2022) proposed a strength-
ening method for column base connections by using vertical
and horizontal stiffeners such that the yielding of the con-
nection occurred in the extended region of the anchor rods.
The other components of the connection were designed to
sustain the strength of the anchor rods. Numerical results
Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23

showed a stable hysteretic response of the connection under


the applied cyclic loading with excellent deformation capac-
ity. The seismic energy can be dissipated through the yielding
of ductile anchor rods, improving the overall performance of
the connection and reducing the damage to it (Hassan et al.
2022).
To repair and maintain aged steel columns, Lin et al. (2017)
proposed several methods for strengthening the bottom of
these columns, such as glass-fiber-reinforced polymer plates,
rapid-hardening concrete, rubber-latex mortar, and reinforc-
For personal use only.

ing bars. Static loading tests and numerical analysis were per-
formed on steel columns with and without strengthening.
Both the experimental and numerical results indicated the
ability of the proposed strengthening methods to enhance
the stiffness and the load-carrying capacities of the column
recommended as these connections are susceptible to flex- base connections, resulting in a significant improvement of
ural yielding and failure, carrying moment that can be esti- seismic performance during earthquakes.
mated only by accurate characterization of base fixity (Torres- The failure mechanisms of stiffened and nonstiffened base
Rodas et al. 2018a, 2018b; Falborski et al. 2020a). Considering plates were evaluated numerically by Diaz et al. (2020). The
the semi-rigidity of column base connections can increase the effects of base plate thickness, location of anchor rods, and
accuracy of the analysis results and can lead to a decrease in location of stiffeners were considered in the analysis. The re-
structural weight and deformation (Dumas et al. 2006). For sults showed that the anchor rods work as fuse elements in
column base connections with two or four anchor rods, the the stiffened base plates. Furthermore, the models with an-
rigidity ratio (G) can be considered as GLY = 0.5 for weak-axis chor rods located outside the column flanges showed lower
buckling and GLX = 1.5 for strong-axis buckling, if there is no flexural strength and rotational stiffness than those with an-
relative displacement of column ends (Picard et al. 1987). The chor rods located between column flanges. Eventually, a new
compression force in the column can significantly increase method was proposed to design the base plates with the pro-
the flexural stiffness of base connections, offering a sufficient posed stiffened configurations.
rotational restraint to be considered in the design (Picard and Shafieifar and Khonsari (2018) carried out a numerical
Beaulieu 1985). study on six different types of base plates with different con-
Recent studies indicated that designing column base con- figurations of stiffeners. The models were subjected to a con-
nections according to current provisions will lead to a thick stant axial load; next, a monotonic lateral loading was ap-
base plate that is not commercially available and can cause plied. The results revealed that using stiffener plates perpen-
concrete crushing, resulting in unexpected connection be- dicular to the column flange and the base plate gave the best
havior (Kanvinde et al. 2013; Díaz et al. 2020). Althoughus- rigidity to the column base connection, and the column could
ing stiffened base plates is very common in detailing and reach its capacity accompanied by large deformation in the
constructing steel structures and can reduce the thickness of flanges.
base plates (Blodgett 1966; Ravari et al. 2011), just a few stud- Regarding the abovementioned studies, it is evident that
ies have addressed the behavior of column base connections using stiffened base plates can be considered a sufficient sub-
with stiffeners. stitute to avoid thick base plates and provide strong con-
The seismic provisions of AISC (2022) addressed the ability nections. Therefore, this research aims to evaluate the cyclic
to employ stiffeners in column base connections. It is men- behavior of stiffened column base connections using differ-
tioned that thick base plates, haunches or cover plates can ent types of stiffeners, assessing the effectiveness of these

Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156 3


Canadian Science Publishing

Table 1. Mechanical properties of steel material (Gomez et al. 2010).


Steel part Young’s modulus (GPa) Yield strength (MPa) Poisson’s ratio
Base plate 200 285 0.3
Anchor rods 203 760 0.3
Washers and nuts 200 – 0.3
Note: Copyright © American Institute of Steel Construction. Reprinted with permission. All rights reserved.

Fig. 5. Compressive stress–strain relationship: (a) EC2 (2004), and (b) this study.
Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23
For personal use only.

Fig. 6. Comparison of cyclic behavior curves. 2.1. Description of the finite element model
First, the specimen (Test #2), as tested by Gomez et al.
(2010), was used as a validation test bed for the simulation
methodology. Figure 1 demonstrates the parts and dimen-
sions of the experimental specimen. The diameter of the an-
chor rods was 19 mm (the threaded part was not modeled).
Geometrical and material nonlinear analysis was conducted,
and the numerical results were compared with the experi-
mental ones.

2.2. Element types


The FE model was composed of 8-node linear brick el-
ements with reduced integration (C3D8R). These elements
can provide accurate results for complex nonlinear analy-
sis involving contacts, constraints, complex boundary condi-
tions, large deformations, and nonlinear material behavior.
Reduced integration was chosen to reduce the analysis run-
ning time, where a lower-order integration will be used to
form the element stiffness (Abaqus 2019). The specimen was
fully modeled in the finite element analysis. Figure 2 shows
a representative FE model constructed for the column base
connection. To provide more accurate results, a refined mesh
strengthening methods and investigating the failure modes
was used on the parts with high-stress concentrations (i.e.,
and mechanical properties of such connections.
base plate, grout, and anchor rods). A mesh size of 25 mm
was considered for all parts, except for the foundation that
was considered as 50 mm.
2. Finite element modeling
A numerical study was carried out using the FE software 2.3. Contact and boundary conditions
package Abaqus/Standard (2019). A three-dimensional non- Regarding the experimental observations in Gomez et al.
linear FE model was developed to validate the experimental (2010), the column and base plate were merged since there
work conducted by Gomez et al. (2010). The FE simulation is was no fracture in the weld between them. Surface-to-surface
discussed in the subsequent sections. contact properties were defined between the interface of the

4 Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156


Canadian Science Publishing

Fig. 7. Deformed shape of base plate connection at peak load, (a) numerical (this study), and (b) experimental (Gomez et al.
2010). Copyright © American Institute of Steel Construction. Reprinted with permission. All rights reserved.
Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23

Fig. 8. Grout damage at 6% drift: (left) experimental (Gomez et al. 2010), and (right) numerical (this study). Copyright © Amer-
ican Institute of Steel Construction. Reprinted with permission. All rights reserved.
For personal use only.

base plate and grout, the base plate and anchor rods, grout ing protocol presents the relation between the drift ratio
and anchor rods, pedestal and anchor rods, and foundation and the number of cycles. However, Abaqus requires the
and anchor rods. The tangential behavior of the contact inter- displacement–time curve to be implemented as a loading his-
action was defined as isotropic friction following the penalty tory. Therefore, as demonstrated in Fig. 4b, the drift ratio in
method (Abaqus 2019). The friction coefficient used for steel- Fig. 4a was multiplied by the height of the column (2350 mm)
to-concrete interaction is 0.45, as suggested by Gomez et al. to calculate the required lateral displacement to be applied.
(2010). Still, the steel-to-steel interaction friction coefficient Then, a constant time of 1 s was considered for each half-
is considered 0.8, according to Grigoriev et al. (1997). The cycle. This time is cumulated, and it reached 80 s at the end
normal contact behavior was defined as “hard” contact with of loading.
allowable separation after contact for both contact interac-
tions.
As shown in Fig. 3, a lateral cyclic load was applied at the 2.4. Material modeling
top of the column and fixed boundary conditions were con- In this study, a nonlinear material obeying the von Mises
sidered at the foundation’s base to verify the cyclic behavior yield criterion was used to model the steel parts of the con-
of the experimental specimen. nection. The behavior of steel material is considered to be
The lateral cyclic load was applied according to the dis- elastic-perfectly-plastic with no hardening (as the impact of
placement history of SAC loading protocol proposed by hardening was negligible in the experiment). The material
Krawinkler et al. (2000). As shown in Fig. 4a, SAC load- properties of each steel part are listed in Table 1.

Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156 5


Canadian Science Publishing

Fig. 9. The 3D models of stiffeners.


Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23
For personal use only.

According to the experimental observations, the concrete is the strain, εc1 is the strain at peak stress, εcu1 is the nomi-
of the pedestal and the foundation remained elastic. There- nal ultimate strain, and fctm is the concrete tensile strength.
fore, the concrete was defined as an elastic material with a Figure 5 defines the parameters used in the equations above
Poisson’s ratio of 0.15 and Young’s modulus of 30 GPa calcu- along with the compressive stress–strain curve of the grout
lated according to eq. 1. On the other hand, the crushing of material.
grout was observed in the experimental test; therefore, the As shown in Fig. 5b, the response of the grout was ter-
grout was modeled employing the concrete damage plastic- minated at a high post-peak compression capacity of about
ity approach (CDP). The CDP parameters were considered de- 41 MPa, which will be fixed by Abaqus until the end of loading
fault values, and the viscosity parameter (μ) was regarded as (Abaqus assumes a horizontal plateau at the end of the user-
0.001 (as proposed by Shaheen et al. 2017). The definition of inputted data). However, the crushing of grout was noticed
concrete material in Abaqus requires the definition of stress– at large drift ratios in the experimental study and the plas-
strain curves (compression and tension) that were obtained tic mechanism occurred in Test#2 due to the simultaneous
according to the EC2 (2004) by using the following equations: yielding of both the base plate and the anchor rods. There-
fore, considering a high post-peak compression capacity of
(1) Ecm = 22[( fcm ) /10]0.3 the grout will not remarkably influence the results for such
a case.
σc Kη − η2
(2) =
fcm 1 + (K − 2)η
(3) K = 1.05Ecm × εc1 / fcm 2.5. Verification of FE model
The experimental specimen Test #2 conducted by Gomez et
(4) η = εc /εc1 al. (2010) was used as a validation test bed. Validation of the
cyclic behavior curve may be a suitable indicator of the sim-
(5) εc1 ( ) = 0.7 fcm 0.31 ≤ 2.8 ulation efficiency. The comparison of the moment–rotation
behavior between the finite element model and the exper-
(6) εcu1 ( ) = 2.8 + 27[(98 − fcm ) /100]4 imental test data are presented in Fig. 6. As shown in the
   figure, the shape of the cyclic curve along with the values
fcm of strength and stiffness of the finite element model had a
(7) fctm = 2.12ln 1 +
10 good and acceptable agreement with the corresponding ex-
perimental curve with a negligible difference.
where Ecm is the modulus of elasticity, fcm is the mean Since the finite element simulation provides an opportu-
concrete compressive strength (for grout: fcm = 55 MPa, for nity to view the stress distribution of the connection, Fig. 7 il-
pedestal and foundation: fcm = 27 MPa), σ c is the stress, εc lustrates the deformation shape of the connection along with

6 Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156


Canadian Science Publishing

Fig. 10. FE models for each specimen configuration. sor in the material is given by the scalar damage equation:
(8) σ = (1 − D) σ̄
where σ̄ is the effective (or undamaged) stress tensor that
would exist in the material in the absence of damage com-
puted in the current increment. As the material undergoes
loading, the scalar damage variable (D) increases, indicating
that the material is becoming more damaged and losing stiff-
ness. It ranges from 0 to 1, where 0 represents an undam-
aged material and 1 represents a completely damaged mate-
rial. Figure 8 shows the grout’s damage zone compared to the
connection’s experimental compression part. The SDEG val-
ues were close to one, which indicates zero residual stiffness
or fully failed material (Abaqus 2019).
Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23

3. Strengthening of column base


connection
After validating the finite element model, the developed
model was adopted for a more comprehensive study to inves-
tigate the cyclic behavior of stiffened base connections with
different configurations.
It is worth mentioning that applying stiffeners to the cur-
rent FE model makes the anchor rods the weakest component
of the connection (Díaz et al. 2020), inducing the concentra-
tion of yielding in the anchor rods. Therefore, to investigate
For personal use only.

the efficiency of the proposed strengthening methods, the an-


chor rods are redesigned according to the capacity base de-
sign method to withstand the tension load corresponding to
the column plastic moment (Mp ).
By calculating the plastic moment of the column according
to AISC-341 (2022) (eq. 9), the tension load of the anchor rods
can be obtained by dividing the column plastic moment (Mp )
by the distance between the centers of the column’s flanges,
as illustrated in eq. 10.

(9) Mp = Fy Zp

Mp
(10) Tu =
dc − tf

where Zp is the column plastic section modulus, Fy is the col-


the distribution of the von Mises stresses over the connection umn yielding stress, Tu is the tension load, dc is the column
region. section height, and tf is the column flange thickness.
As shown in the figure, the deformed shape and the yield- By using three anchor rods on each side of the base plate
ing over the connection region obtained from the numerical (parallel to the column flanges) with a yield strength of
analysis are similar to the observed results in the experimen- 760 MPa (not changed), the diameter of anchor rods is ob-
tal study. Furthermore, the von Mises stress values reached tained to be 30 mm such that the anchor rods could with-
the yielding strength of the anchor rods, indicating the com- stand the tension load Tu . Accordingly, the column base con-
bined plastic mechanism between the base plate and the an- nection is stiffened considering the new layout of anchor
chor rods, similar to the experimental results. rods.
The spalling of the grout was observed in the experimen- The stiffeners investigated in this study are shown in Fig. 9,
tal test at the extreme compression edge of the column base which are the most common configurations used for column
connection. To investigate the simulation of grout, the scalar base connections. They are bracket plates (Stiff-1), angle (Stiff-
stiffness degradation variable (SDEG) was used to compare 2), a channel with vertical stiffener plates (Stiff-3), and built-
the damage of the grout with the experimental observations. up stiffener (Stiff-4).
Abaqus/Standard assumes that the degradation of the elastic In practice, the Stiff-1 configuration (Fig. 9a) consists of
stiffness can be modeled using the scalar damage variable, D. many bracket plates that are attached to the face of the col-
At any given loading cycle during the analysis, the stress ten- umn flange using a fillet weld. The groove weld can be used

Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156 7


Canadian Science Publishing

Fig. 11. The cyclic behavior of stiffened base connections with plastic column.
Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23
For personal use only.

to fix the stiffeners to the base plate. Similarly, a fillet weld type contact was assigned, which prevents any relative move-
could be used to attach the vertical leg of the angle stiff- ment. Although the fracturing of welds is probable in such
ener (Stiff-2 in Fig. 9b) to the face of the column flange, but connections, it was not considered in this study (investigat-
the horizontal leg will be fastened to the base plate through ing the welds is out of the scope of this study).
the anchor rods. In Stiff-3 (Fig. 9c), another simple detail was The FE models in Fig. 10 were analyzed by applying the
shown whereby a wide-flanged channel serves as a stiffener. SAC loading protocol (Krawinkler et al. 2000) shown in Fig. 4.
The channel is set up just clear of the bearing on the base The hysteresis behaviors of the column base connections
plate such that the end of the column will take the compres- strengthened by the aforementioned stiffeners are presented
sive load without overloading the channel, and the channel in Fig. 11. The obtained curves were compared to the column
will not be designed for any bearing. In addition, the load of plastic moment (Mp ), as the models were designed to concen-
the anchor rods is supported by the top flange of the channel, trate the damage in the column rather than the connection
reinforced by the vertical stiffener plates welded between the (weak column/strong connection).
channel flanges such that the channel will be designed just As shown in Fig. 11, the connection was stronger than
for the tension from the anchor rods. A fillet weld could be the column and the connection flexural strength reached
used to fix the channel with additional stiffeners to the col- the plastic moment of the column for all models. The hys-
umn flange. In Stiff-4 (Fig. 9d), vertical and horizontal stiff- teretic behavior of all models was stable up to 8% drift with
ener plates are combined to strength the column base con- a slight pinching effect. It is worth mentioning that the ex-
nection. periment of Gomez et al. (2010) was designed to concentrate
Figure 10 shows the finite element models of the column damage on the connection, and the column was designed
base connection strengthened by the aforementioned differ- to remain elastic. Therefore, the hysteresis curve of the ex-
ent types of stiffeners. Stiff-1 and Stiff-2 are 10 mm thick (57% perimental specimen (Test#2) was not presented in Fig. 11
tf ), whereas Stiff-3 and Stiff-4 have a thickness of 25 mm to as it represents a different concept (strong column/weak
sustain the maximum tensile capacity of the anchor rods. The connection).
same material used for the column was considered for all the Figure 12 depicts the stress distribution of the deformed
stiffeners. The brick element (C3D8R) was used to generate FE models and the elements that are yielded after the cyclic
the mesh of the stiffeners. To simulate the welded joint be- loading, where the red regions indicate the onset of yield-
tween the column, the base plate and the stiffeners, a “Tie” ing in the steel material according to the AC yield criteria

8 Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156


Canadian Science Publishing

Fig. 12. (Left): the von Mises stress, MPa, and (right) the AC that the damage could be concentrated in the connection
yield criteria of the FE models. rather than the column (strong column/weak connection).
The hysteretic response of the four configurations of
strengthened base plates with elastic column is illustrated
in Fig. 13. As shown in the figure, the pinched hysteretic re-
sponse was similar for all configurations. However, the pinch-
ing was less prevalent for the models with the configurations
Stiff-1 and Stiff-2. This variance may be attributed to the elon-
gation rate of the anchor rods in these models, which is less
than the others. All the configurations showed a high flexural
strength (higher than Mp ) with a reasonable ductility. How-
ever, the model with Stiff-2 presented a more stable hysteretic
response than the others.
For comparison purposes, the FE models in Fig. 10 were an-
alyzed monotonically considering strong (or elastic) column.
Figure 14 shows the obtained monotonic curves, where the
Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23

monotonic curve of the strengthened connection with plas-


tic column was also plotted in the same figure for comparison
purposes.
The figure illustrates that designing the column base con-
nection with each of the studied strengthening methods
could be considered a conservative design since the strength
of the connection is higher than the column. The maximum
increase in the connection strength (56%) was obtained uti-
lizing a built-up stiffener compared to the plastic column
strength. Strengthening column base connection using an
For personal use only.

angle represents the weakest strengthening method since it


presents the minimum increase in the connection strength;
however, it increased the strength by approximately 38%.
Furthermore, using bracket plates or channels with verti-
cal plates can give the same increased ratio of connection
strength (47%). Using Stiff-1 and Stiff-4 led to a slight increase
in the connection’s initial stiffness.
Besides the peak moment capacity of the strengthened con-
nections, investigating the failure mode of each configura-
tion is considerable attention. Figure 15 demonstrates the
yielded parts in the strengthened connections (the red re-
gions), considering only the elastic behavior of the column.
As shown in the figure, using Stiff-1 and Stiff-2 caused a global
yielding distribution in all of the connection’s components
on both sides, including the base plate, the anchor rods,
and the stiffeners. The cantilever performance of Stiff-2 has
caused a degradation in the peak moment capacity compared
with Stiff-1.
Additionally, Utilizing Stiff-3 caused a yielding concentra-
(which are used to determine whether a material has reached tion in the anchor rods. Stiff-3 plays no role in the compres-
its yield point and undergone plastic deformation). sion side (a gap left between the channel and the base plate),
As shown in the figure, severe damage occurred in the col- so it remained elastic. On the other hand, Stiff-4 has a remark-
umn’s flanges above the stiffened part of the connection. The able contribution on the compression side and is designed to
AC yield contour demonstrates that yielding is concentrated keep elastic on the tension side (the yielding is concentrated
just in the column, and the other connection’s components just in the anchor rods), representing the best achievable con-
remained elastic. tribution of the connection’s component against the applied
Regarding the previous results, it is obvious that the pro- load.
posed configurations were able to make the connection
stronger than the column. However, the strengthened con-
nections should be investigated and compared such that the
4. Conclusion
most efficient strengthening method could be selected. For In this study, the cyclic behavior of stiffened column base
that aim, the models shown in Fig. 10 were reanalyzed cycli- plate connections was evaluated numerically using the FEM.
cally considering only the elastic behavior of the column so Four common strengthening methods were considered. The

Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156 9


Canadian Science Publishing

Fig. 13. The cyclic behavior of stiffened base connections with elastic column.
Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23
For personal use only.

Fig. 14. Comparison of the monotonic behavior of strengthened connections.

most general results and conclusions of this study are de- – Strengthening column base connections with either chan-
scribed next. nel with vertical plates or bracket plates gave the same
peak moment capacity.
– Using any of the applied strengthening methods could – Using stiffened connections with more stiffeners made the
make the connection more robust than the column. anchor rods the weaker components in the connections
– The cyclic behavior of all stiffened connections was ductile (fuse elements).
with pinched hysteresis loops. However, using angles as – Using bracket plates or angles as strengthening meth-
stiffeners could present a more stable hysteretic response ods seems practical and economical, as using plates with
with less pinching effects. a thickness of just 10 mm provided a flexural strength
– Utilizing a built-up stiffener provided the maximum mo- higher than the plastic moment of the column. However,
ment capacity and rotational stiffness of the connection. almost all the components of the connection will con-

10 Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156


Canadian Science Publishing

Fig. 15. Yielding concentration of strengthened connections with an elastic column.


Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23
For personal use only.

tribute to providing the flexural strength of the connec- Author contributions


tion. Formal analysis: MA
Methodology: MHH
Software: MA
This paper briefly explains the cyclic behavior of stiffened
Supervision: SHHL
column base plate connections and encourages full-scale ex-
Validation: MA
perimental works on them.
Visualization: SHHL
Writing – review & editing: MAS
Article information
Competing interests
History dates The authors declare that there are no competing interests.
Received: 3 April 2023
Accepted: 10 October 2023
Accepted manuscript online: 15 November 2023 Funding information
Version of record online: 7 December 2023 The authors declare no specific funding for this work.

Copyright
© 2023 The Author(s). Permission for reuse (free in most
References
cases) can be obtained from copyright.com. Abaqus. 2019. Dassault Systemes Simulia Corporation. Providence,
Rhode Island, USA.
Abdolahzadeh, G., Shabanian, S.M., and Tavakol, A. 2019. Experimen-
Data availability tal and numerical evaluation of rigid column to baseplate connec-
Data available within the article. tion under cyclic loading. Structural Design Tall Build, 28(6): e1596.
doi:10.1002/tal.1596.
ACI-318. 2016. Building code requirements for structural concrete and
Author information commentary. American Concrete Institute, USA.
AISC-341. 2022. Seismic provisions for structural steel buildings. Ameri-
can Institute of Steel Construction, USA.
Author ORCIDs AISC Steel Design Guide 1. 2006. Base plate and anchor rod design. Amer-
Mahmoud Ali https://orcid.org/0000-0003-3821-1199 ican Institute of Steel Construction, USA.

Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156 11


Canadian Science Publishing

Astaneh-Asl, A., Bergsma, G., and Shen, J. 1992. Behavior and design of Lim, W.Y., Lee, D., and You, Y.C. 2017. Exposed column-base plate strong-
base plates for gravity, wind and seismic loads. In Paper presented at axis connections for small-size steel construction. Journal of Con-
the Proceedings of the National Steel Construction Conference, Las structional Steel Research, 137: 286–296. doi:10.1016/j.jcsr.2017.06.
Vegas, USA, June. 018.
Aviram, A., Stojadinović, B., and Der Kiureghian, A. 2010. Performance Lin, W., Taniguchi, N., and Yoda, T. 2017. A preventive strengthening
and reliability of exposed column base plate connections for steel method for steel columns: experimental study and numerical anal-
moment-resisting frames. Pacific Earthquake Engineering Research yses. Journal of Constructional Steel Research, 138: 357–368. doi:10.
(PEER), University of California, Berkeley. 1016/j.jcsr.2017.07.019.
Blodgett, O.W. 1966. Design of welded structures. James F. Lincoln Arc Picard, A., and Beaulieu, D. 1985. Behaviour of a simple column base con-
Welding Foundation, USA. nection. Canadian Journal of Civil Engineering, 12(1). doi:10.1139/
Burda, J.J. 1999. Studies of seismic behavior of steel base plates. Univer- l85-013.126
sity of Nevada, USA. Picard, A., Beaulieu, D., and Perusse, B. 1987. Rotational restraint of a
Díaz, H., Nuñez, E., and Oyarzo-Vera, C. 2020. Monotonic response of simple column base connection. Canadian Journal of Civil Engineer-
exposed base plates of columns: numerical study and a new design ing, 14(1). doi:10.1139/l87-007.49
method. Metals, 10(3): 396. doi:10.3390/met10030396. Ravari, A.K., Othman, I.B., and Ibrahim, Z.B. 2011. Finite element analy-
Dumas, M., Beaulieu, D., and Picard, A. 2006. Characterization equations sis of bolted column base connection without and with stiffeners. In-
for steel column base connections. Canadian Journal of Civil Engi- ternational Journal of Physical Sciences, 6(1): 1–7. doi:10.5897/IJPS10.
neering, 33(4): 409–420. doi:10.1139/l05-054. 353.
EC2. 2004. 1-1: Eurocode 2: design of concrete structures- part 1-1: general Richards, P.W., Barnwell, N.V., Tryon, J.E., and Sadler, A.L. 2018. Flexu-
Can. J. Civ. Eng. Downloaded from cdnsciencepub.com by Mahmoud Ali on 12/07/23

rules and rules for buildings, European Committee for Standardiza- ral strength and stiffness of block-out connections for steel columns.
tion; CEN. Engineering structures, 173: 404–415. doi:10.1016/j.engstruct.2018.
Fahmy, M. 2000. Seismic behavior of moment-resisting steel column 06.055.
bases. University of Michigan, USA. Shafieifar, M., and Khonsari, V. 2018. A numerical investigation on behav-
Falborski, T., Hassan, A., and Kanvinde, A. 2020a. Column base fixity ior of column base plates with different configurations. Civil Journal,
in steel moment frames: observations from instrumented buildings. 4(6): 1223–1239. doi:10.28991/cej-0309169.
Journal of Constructional Steel Research, 168: 105993. doi:10.1016/j. Shaheen, M.A., Tsavdaridis, K.D., and Salem, E. 2017. Effect of grout prop-
jcsr.2020.105993. erties on shear strength of column base connections: FEA and analyt-
Falborski, T., Torres-Rodas, P., Zareian, F., and Kanvinde, A. 2020b. Effect ical approach. Engineering Structures, 152: 307–319. doi:10.1016/j.
of base-connection strength and ductility on the seismic performance engstruct.2017.08.065.
of steel moment-resisting frames. Journal of Structural Engineering, Singh, G., and Woods, J. 2022. 2D and 3D numerical modelling of exposed
146(5): 04020054. doi:10.1061/(ASCE)ST.1943-541X.0002544. steel base plate connections under cyclic loading. Canadian Journal
Gomez, I., Deierlein, G., and Kanvinde, A. 2010. Exposed column base of Civil Engineering, 49(8). doi:10.1139/cjce-2021-0463.1402
connections subjected to axial compression and flexure. University Torres-Rodas, P., Flores, F., and Zareian, F. 2018a. Seismic response of
For personal use only.

of California, Davis & Stanford University, USA. steel moment frame considering gravity system and column base
Grauvilardell, J.E., Lee, D., Hajjar, J.F., and Dexter, R.J. 2005. Synthesis flexibility. In Paper presented at the Proceedings of the 11th Na-
of design, testing, and analysis research on steel column base plate tional Conference on Earthquake Engineering, Los Angeles, USA, Au-
connections in high-seismic zones. University of Minnesota, USA. gust. Available from http://dspace.ucuenca.edu.ec/handle/123456789
Grigoriev, I.S., Meı̆likhov, E.Z., and Radzig, A.A. 1997. Handbook of phys- /34965.
ical quantities. CRC. Torres-Rodas, P., Zareian, F., and Kanvinde, A. 2018b. Seismic demands in
Hassan, A.S., Song, B., Galasso, C., and Kanvinde, A. 2022. Seismic per- column base connections of steel moment frames. Earthquake Spec-
formance of exposed column–base plate connections with ductile tra, 34(3): 1383–1403. doi:10.1193/062317EQS127M.
anchor rods. Journal of Structural Engineering, 148(5): 04022028. Torres-Rodas, P., Flores, F., Pozo, S., and Astudillo, B.X. 2021a. Seismic per-
doi:10.1061/(ASCE)ST.1943-541X.0003298. formance of steel moment frames considering the effects of column-
Inamasu, H., Kanvinde, A.M., and Lignos, D.G. 2021. Seismic design base hysteretic behavior and gravity framing system. Soil Dynam-
of non-dissipative embedded column base connections. Journal of ics and Earthquake Engineering, 144: 106654. doi:10.1016/j.soildyn.
Constructional Steel Research, 177: 106417. doi:10.1016/j.jcsr.2020. 2021.106654.
106417. Torres-Rodas, P., Medalla, M., Lopez-Garcia, D., Campos, L., and Zareian,
Kanvinde, A., Jordan, S., and Cooke, R. 2013. Exposed column base plate F. 2021b. Strength method for exposed base plates with ex-
connections in moment frames-simulations and behavioral insights. tended anchor bolts. In Paper presented at the Proceedings of the
Journal of Constructional Steel Research, 84: 82–93. doi:10.1016/j. 17th World Conference on Earthquake Engineering, Sandai, Japan,
jcsr.2013.02.015. September.
Kiani, A., and Izadinia, M. 2015. An overview on effects of geometric pa- Torres-Rodas, P., Medalla, M., Zareian, F., and Lopez-Garcia, D. 2022.
rameters in column connection behavior via finite element method. Cyclic behavior and design methodology of exposed base plates with
Indian Journal of Science and Technology, 8(28): 1. doi:10.17485/ijst/ extended anchor bolts. Engineering Structures, 260: 114235. doi:10.
2015/v8i28/81825. 1016/j.engstruct.2022.114235.
Krawinkler, H., Gupta, A., Medina, R., and Luco, N. 2000. Loading histo- Trautner, C.A., and Hutchinson, T.C. 2018. Parametric finite-element
ries for seismic performance testing of SMRF components and assem- modeling for exposed steel moment frame column baseplate con-
blies. SAC Joint Venture, Report No. SAC/BD-00/10. Richmond, Califor- nections subjected to lateral loads. Journal of Structural Engineering,
nia, USA. 144(6): 04018049. doi:10.1061/(ASCE)ST.1943-541X.0001990.
Lee, D., and Goel, S. 2001. Seismic behavior of column-base plate connec- Zareian, F., and Kanvinde, A. 2013. Effect of column-base flexibility on the
tions bending about weak axis. Report No. UMCEE 01-09. University seismic response and safety of steel moment-resisting frames. Earth-
of Michigan, USA. quake Spectra, 29(4): 1537–1559. doi:10.1193/030512EQS062M.

12 Can. J. Civ. Eng. 00: 1–12 (2023) | dx.doi.org/10.1139/cjce-2023-0156

You might also like