Evaluation of a developed bypass viscous damper performance

Abstract

In this study, the dynamic behavior of a developed bypass viscous damper is evaluated. Bypass viscous damper has a flexible hose as an external orifice through which the inside fluid transfer from one side to the other side of the inner piston. Accordingly, the viscosity coefficient of the damper can be adjusted using geometrical dimensions of the hose. Moreover, the external orifice acts as a thermal compensator and alleviates viscous heating of the damper. According to experimental results, Computational Fluid Dynamic (CFD) model, a numerical formula and the simplified Maxwell model are found and assessed; therefore, the verification of numerical and computational models are evaluated for simulating. Also, a simplified procedure is proposed to design structures with bypass viscous dampers. The design procedure is applied to design an 8-story hospital structure with bypass viscous dampers, and it is compared with the same structure, which is designed with concentric braces and without dampers. Nonlinear time history analyses revealed that the hospital with viscous damper experiences less structural inelastic demands and fewer story accelerations which mean fewer demands on nonstructural elements. Moreover, seismic behaviors of nonstructural masonry claddings are also compared in the cases of hospital structure with and without dampers.

Keywords

developed viscous damper / external orifice / energy dissipation / seismic behavior / CFD model of viscous damper / a simplified model

Introduction

Application of energy dissipating devices has been recognized as one of the most promising techniques to reach to a seismic-resistant design. As a velocity-dependent device, fluid viscous dampers are among the most reliable devices with widespread applications in different industries. During the last decades, different viscous dampers have been designed for seismic applications and successfully used in different buildings and bridges throughout the world [¹–⁴]. Design and manufacture of viscous dampers are multidisciplinary tasks which call for high levels of technology and expertise. That is why most commercial viscous dampers have unpublicized details. Behaviors of viscous dampers and structures equipped with them have been comprehensively investigated by different researchers. By using shaking table tests on small scale 1-story and 3-story specimens, Constantinou and Symans [⁵] have examined the contribution of fluid viscous dampers. According to using the same specimens, Seleemah and Constantinou [⁶] have investigated the effect of linear and nonlinear viscous dampers and their placements on seismic behavior of the 1-story and 3-story specimens. Although the above mentioned tests have been carried out on small-scale specimens, full-scale tests are also available in the literature. During one of the most unique shaking table tests, contributions of different energy dissipating devices, including viscous dampers, to seismic behavior of a full-scale 5-story 3-D building have been evaluated [⁷]. Infanti et al. [⁸] who have been reported results of full-scale tests of nonlinear viscous dampers which used in the cable-stayed Rion-Antirion Bridge. Yamamoto et al. [⁹] who have been investigated the behavior of a new full-scale multi-unit oil damper which a number of smaller damper units can be combined to make a higher capacity oil damper.

Although viscous dampers are acknowledged by current seismic codes for both designs of new buildings [¹⁰] and rehabilitation of existing buildings [¹¹], the use of present codes is limited for the building structure. On the other hand, there is no completed design method to design viscous dampers and specify their places. According to this issue, some researchers have studied to propose different procedures for designing and locating the optimal placement of viscous dampers [¹²–¹⁸]. One of the first studies on this topic was carried out by Zhang and Soong [¹⁹]. They proposed a sequential procedure to find the optimum characteristics and placement of viscoelastic dampers. The procedure is also applicable to viscous dampers or any other velocity-dependent damper. More sophisticated procedures, such as control theories, are also used in order to design dampers. For example, adopting a linear quadratic optimal control theory, Gluck et al. [²⁰] proposed a procedure for the optimal design of viscous and viscoelastic dampers. Another approach for achieving an optimal placement is to minimize the transfer function of the building. This technique was also used by Takewaki [²¹], Mousavi and Ghorbani-Tanha [²²] for optimal placement of velocity-dependent dampers. The procedure which was proposed by Mousavi and Ghorbani-Tanha [²²], is a more advanced technique that accounts for higher mode effects and is applicable for both viscous and viscoelastic dampers.

The first best novelty of the present study is related to the use of a flexible hose instead an external or internal orifice in viscous damper and evaluate the performance of this viscous damper. The second best novelty is to simulate and present the Computational Fluid Dynamic (CFD) models and formulas for demonstrating performance on this developed viscous damper. The third best novelty of this study is to present the approach for designing and specifying placements of this viscous damper. Although the aforementioned design procedures are robust and lead to satisfactory results, they are too complex to use and design. Accordingly, one of the main scopes of the present study is to propose and assess a simplified design procedure for buildings equipped with linear bypass viscous dampers or generally any linear velocity-dependent damper with zero stiffness. Therefore, in the subsequent sections, first bypass viscous damper is introduced and then experimental results which obtained from a small-scale specimen, would be evaluated and discussed. After verification of currently available numerical simulations, a step-by-step simplified design procedure would be proposed for buildings with linear viscous dampers. The study is followed by investigating the seismic behavior of structural and nonstructural elements of a hospital building with and without viscous damper.

Developed bypass viscous damper

The developed bypass viscous damper is similar to other viscous dampers with an exception that it has an external high pressure but flexible hose as the orifice, as depicted in Fig. 1. It is well-understood that during dynamic excitations, the input energy would be dissipated into heat and the temperature of the damper oil, especially inside the orifice, could greatly be increased up to 400 ^0F [²³,²⁴]. Accordingly, thermal compensation is important in viscous dampers, and the external orifice of the bypass viscous damper would act as a thermal compensator to alleviate viscous heating of the damper. Moreover, the damping coefficient and damping exponent can be adjusted by length, diameter, and flexibility of the hose, among others.

According to Fig. 1, the main cylinder of the damper has inner and outer diameters of 60 and 80 mm, respectively. Shaft diameter is 35 mm, and an inner diameter of the external orifice is 6.35 mm. Dynamic viscosity of the used fluid is also 1.8 Pa•·s. Such details would result in a viscous damper with a damping coefficient of 0.094 kN•·s/mm and velocity exponent of 1 (linear damper).

Based on the performed tests, the above damper is a linear viscous damper which can be upgraded to a nonlinear damper. However, its simple formula that presents a linear behavior, is based on the Bernoulli relationship. The damping coefficient for this viscous damper is calculated as:

(1)C={\displaystyle \frac{\mathrm{8}\mathrm{\pi }{}^{\mathrm{2}}\mu L}{d^{\mathrm{4}}}},

where C, D_c, D_s, L, d, and m are defined damping coefficient, the diameter of the cylinder, the diameter of the damper shaft, the length of high pressure hose, the diameter of high pressure hose, and fluid viscosity, respectively. Table 1 shows the feature of the bypass viscous damper which is tested in this study.

The above relation will only be an estimate of the damping coefficient because it does not include the two important effects, i.e., compressibility, as well as local drops (caused by bends and the direction change of the flow and the existence of different throttles).

Research methodology

To investigate the performance of the proposed viscous damper, the CFD simulation by using ABAQUS software and experimental tests are utilized in the present study. Section 4 explains the experimental program and includes two subsections that describe test setup and material properties in laboratory tests. Two numerical methods, which involved CFD in ABAQUS and a simplified Maxwell model in SAP 2000, are used to evaluate the performance of this viscous damper in a hospital structure. On the other hand, fluid of the damper is simulated by using the CFD model in ABAQUS software, which this method is explained in Section 5. The simplified design Procedure is described in Section 6, which a step-by-step simplified design method is presented to design buildings with viscous dampers. Finally, a case study is selected to design with and without this viscous damper for evaluating the performance of this viscous damper in the specified structure; therefore, two same hospital structures are modeled with and without this viscous damper by using SAP2000 which detail of this method is described in subsection 7.2.2. After analyzing these structures, one of the nonstructural masonry claddings at the 8th story which is located in the same place in two structures, is considered to evaluate the performance of the viscous damper on nonstructural masonry claddings; therefore, these walls are modeled and analyzed based on exported load of SAP models in Standard section of ABAQUS software.

Experimental program

Test setup and instrumentation

The adopted set-up for the experimental phase of this study is shown in Fig. 2. This setup has been successfully used for dynamic testing of viscous dampers [²⁵,²⁶]. The damper and the actuator are connected through a middle steel element with a pinned connection at its base as shown in Fig. 2. In this way, out-of-plane displacements of the damper would be constrained, and the axial displacement of the actuator would be transferred to the damper via the rocking motion of the middle steel element. Figures 2(a) and 2(b) demonstrate graphical schema of test setup which the location of each part of this test setup is specified. Bypass viscous damper is connected to the column and fixed supported by two pins. An actuator is located other side of the column, and it is pinned to a solid wall and column. The actuator includes LVT and load cell which record the force and displacement during an experiment. Figures 2(c) and 2(d) show actual test setup in the present study.

Dynamic loadings of the specimen are selected based on the proposed protocols of ASCE 7-16 for velocity-dependent dampers with some conservative modifications. Instead of considering only one frequency, as suggested by ASCE 7-16, five frequencies of 0.25, 0.33, 0.50, 0.67, and 1.00 Hz are imposed on the damper. Ten continuous cycles with amplitude 0.33D_max, five continuous cycles with amplitude 0.67D_max and three continuous cycles with maximum D_max are applied to bypass viscous damper during test, respectively. These continuities amplitudes are applied separately. Moreover, the maximum amplitude (corresponding to a displacement of the damper under Maximum Considered Earthquake (MCE)) is imposed in 3 continuous cycles as suggested by ASCE 7-16. It should be noted that the maximum amplitude in each frequency is selected based on the maximum velocity of the actuator which is 175 mm/s.

Material properties

There is an actuator, developed viscous damper, solid wall and one rigid pinned column in the present study as shown in Fig. 2. The actuator can record the force and displacement during an experiment. According to the previous section and Fig. 1, a developed viscous damper includes a main cylinder shaft and external orifice and flexible hose. The main cylinder of the damper has inner and outer diameters of 60 and 80 mm, respectively. Shaft diameter is 35 mm, and the inner diameter of the external orifice is 6.35 mm. A flexible hose as an external orifice through which the inside fluid transfer from one side to the other side of the inner piston. Dynamic viscosity of the used fluid is also 1.8 Pa•·s. Such details would result in a viscous damper with a damping coefficient of 0.094 kN•s/mm and velocity exponent of 1 (linear damper).

CFD and Maxwell models

To simulate the behavior of the bypass viscous damper, two numerical models are adopted. One sophisticated CFD model in Abaqus/CFD (Dassault Systemes Simulia, 2015) and another simplified Maxwell model in SAP 2000 [²⁷,²⁸]. Details of the CFD model are shown in Fig. 3 which part (a) of this figure shows the dimension and size mesh of each part of the model. The fluid is considered to be incompressible with a density of 950 kg/m³ and viscosity of 1.8 Pa•·s as specified by the manufacturer.8-node linear fluid brick (FC3D8) elements with an approximate size of 3 and 1 mm are used to mesh the fluid at the main cylinder and the bypass orifice, respectively. Selected mesh sizes are obtained from a sensitivity study. It is found that finer meshes would not change CFD results while significantly increase computational efforts. Figure 3(b) demonstrates flow and pressure in cylinder and hose of this damper due to applying force. According to Fig. 3, only the fluid parts of the damper are modeled, and effect of the other solid parts (piston, cylinder, hose walls, etc.) on the turbulent fluid flow is considered by defining no slip boundary condition at the fluid-solid interface. Different input velocities are imposed on the model and from the obtained pressure difference at the front and the back of the piston, the corresponding damper force can be evaluated. In this way, the force-velocity curve of the damper can be estimated, as mentioned in the experimental results. CFD model is not able to directly account for friction between the piston and the cylinder. Accordingly, the forces obtained from the CFD model are increased by 2.5 kN.

Simplified design procedure

In this section, a step-by-step simplified design procedure is proposed for designing buildings with viscous dampers. According to Cimellaro and Retamales [²⁹], in order to attain a desired target damping ratio, the required total damping coefficient of linear viscous dampers can be estimated as follows:

(2){{\displaystyle C}}_{\mathrm{a}}={\displaystyle \frac{KT}{\mathrm{\pi }}}\left({\xi }^{*}-\xi \right),

where K is sum of the lateral stiffness of all stories along the considered direction, x* and x are the target and the inherent damping ratios, respectively, T is the fundamental period of the building along the considered direction, and the parameter C_a is the total damping ratio required which should be added to the building for achieving the target damping ratio. The designer should decide how to optimally distribute the obtained total damping into different stories. Viscous dampers are depended to velocity; therefore, selecting a damper is depended to a velocity of the story. As a result, the drift of the upper story requires to select a placement for a damper. On the other hand, the most suitable stories would be selected to place a viscous damper relate to the highest inter-story velocity. According to period elongation due to inelastic behaviors, the stories with the highest inter-story drifts are expected to also experience the highest inter-story velocities. According to the dependence of velocity and inter-story drift, more damping coefficient should be applied in placements which there is more inter-story drift. Based on the first mode of structure frame analysis, the pattern of inter-story drifts during a seismic event can be estimated. Accordingly, the total damping coefficient is calculated by Eq. (2) which should be distributed for each story based on the inter-story drift value of each story per summation of all inter-story drift. This procedure would be explained in more details in the subsequent section. Considering the abovementioned assumptions, the step-by-step design procedure of buildings with viscous dampers can be summarized as follow:

Step 1. Design the building without damper considering all gravity loads and 0.75 seismic loads (per ASCE 7-16). In this step, no check is required for inter-story drifts. For retrofit purposes, skip this step.

Step 2. Define the desired target damping ratio. In this step, the damping ratio should be adjusted such that inter-story drifts do not exceed a predefined value. This can be done by performing a spectral analysis considering the increased damping ratio of the building. Another simple technique is to reduce the designed spectrum by B-factors as proposed by ASCE 7-16. Considering viscous damper costs and available damper capacities in the market, damping ratios in excess of 40% that are not recommended.

Step 3. Obtain the total required damping coefficient from Eq. (2).

Step 4. Distribute damping coefficient along different stories based on their fundamental mode shape inter-story drifts.

Step 5. Select the final viscous damper for each story according to the previous step and the available viscous dampers in the market.

Step 6. Perform a final check using nonlinear time history procedure.

Figure 4 shows a schematic flowchart for the proposed design procedure.

Results and discussions

Experimental results

Obtained cyclic behaviors of the damper in different frequencies and its force-velocity curve are shown in Fig. 5. It can be seen that the damper revealed quite stable behaviors under different amplitudes and frequencies. During set-up adjustment, it is seen that a force of about 2.5 kN is required to impose a quasi-static deformation to the damper. This is the friction force between the piston and the cylinder. As a result, the force-velocity curve of the damper in Fig. 5(f) is offset from origin. The results show that the force of damper is increased by rising frequency and period. The maximum force which was applied in the present study, is 1.7 ton. According to Fig. 5(f), the force of damper is increased by rising velocity. According to ASCE 7-16, for accepting viscous damper, the change of minimum and maximum amounts of damper force in zero displacement should not be more than 15% of the average of force in all cycles. Bypass viscous damper satisfies this condition. Experimental tests are done in frequencies of 0.25, 0.33, 0.5, and 0.67 Hz, and the results of them are shown in Figs. 5(a), 5(b), 5(c), and 5(d), respectively.

This frictional behavior of the viscous damper is commonly negligible in a full-scale viscous damper, but it has a pronounced effect on the tested small-scale specimen. As a sample, displacement, velocity, and damper force time histories, as well as dissipated energy during the 0.50 Hz loading protocol, are illustrated in Figs. 6(a), 6(b), 6(c), and 6(d), respectively. According to test results, when displacement value is 20 mm during 10 cycles in the first 20 s, energy dissipation value is 5 kN•·m and then by increasing displacement from 20 to 25 mm during next 5 cycles during 25–35 s, the energy dissipation is increased from 5 to 10 kN••·m. Finally, by rising displacement from 25 to 45 mm during the last 5 cycles during 37–47 s, the energy dissipation is increased from 10 to 22.5 kN••·m.

Numerical and analytical results

Table 2 shows the results of the CFD model that are determined based on different speeds. Based on the results of the CFD model, the damping coefficient is determined 0.105 kN••·s/mm, which this value is larger than the numerical damping coefficient that is calculated based on Eq. (1). According to the previous subsection, the numerical damping coefficient is calculated 0.094 kN••·s/mm.

The results show that the reliability of the numerical approach is more than a computational method. Figure 7 shows pressure in the piston and flexible hose when fluid speed is 100 mm/s. This evaluation is carried out by ABAQUS software. CFD model shows maximum and minimum pressure which are occurred in this viscous damper, are 5.622e⁺⁶ and −1.41e⁻³, respectively.

According to the calculated values of the damping coefficient by Eq. (1), CFD method and experimental results, there are good agreements between the results of computational and numerical results and experimental results; therefore, the behavior of bypass viscous dampers can be simulated by currently available computational tools. Figure 8 demonstrates results of force-velocity of developed bypass viscous damper in experimental, computational and numerical results. Based on Fig. 8, numerical and computational results are verified and matched with experimental results.

In the simplified Maxwell model (SAP 2000), two parallel link elements are used which one of them is used for the viscous component, and another is utilized for a small frictional component of the damper. The results, which obtained from the numerical models, should be compared with the results of tests as shown in Fig. 9. The experimental test setup and computational model are shown in Fig. 9(a). The computational model is simulated like experimental test setup. The experimental and computational tests are carried out in 0.25, 0.33, 0.50, and 0.67 Hz which are demonstrated in Figs. 9(b), 9(c), 9(d), and 9(e), respectively. According to the results of Fig. 9, the computational results which are obtained from SAP2000, are verified by experimental results.

Case study

In recent years, many useful studies on the subject of computational methods for steel structure analysis, fracture and crack modeling are carried out [³⁰–⁴¹]. Commonly effect of viscous dampers has been shown by comparing the seismic behavior of a case study with viscous dampers and without damper. There are some architectural details and openings which might prevent to select the placement of viscous dampers on a moment resisting frame. In this study, comparison of frames with viscous dampers and frames without viscous dampers (with concentric braces) is carried out. For a fair comparison, viscous dampers and concentric braces are placed at the same bays in two separate same structures. On the other hand, all columns and beams of two structures are the same.

An 8-story steel hospital building is selected to evaluate the effect of using viscous damper. Dead loads at the floors and the roof are 8 and 6 kN/m², respectively. Floor live loads are applied 2 kN/m² on rooms and 4 kN/m² on corridors and passages. Roof live load is applied also 2 kN/m². Weight of building is considered 595 ton and 601 ton for using without and with viscous damper, respectively. Cladding walls impose 7 kN/m load on the perimeter beams. The minimum and expected yield strengths of the used steel are 240 and 265 MPa, respectively. This hospital is located in Tehran with the seismicity equivalent to seismic design category D per ASCE 7-16 (S_DS = 0.96g and S_D1 = 0.72g).

The building is designed in two different scenarios which all columns and beams of two structures are same but braces are used to design one of them, and another is designed with bypass viscous damper. In the first scenario, the building is designed per Iranian Seismic Code (Standard No. 2800, 2014) without viscous dampers. The adopted lateral load resisting system, in this case, is a dual system with special moment resisting frames and special concentrically braced frames, as shown in Fig. 10 [⁴²].

In the second scenario, the building is designed with viscous dampers according to the proposed design procedure in the previous section. Detail of each step is explained as follow:

Step 1. Preliminary design of the main structure: the building is analyzed and designed by a combination system of special moment resisting frames combination with viscous dampers, in this scenario. In this step, neglecting the viscous dampers, the bare frames would be designed for gravity loads as well as 0.75 seismic loads which would be a seismic coefficient of 0.074. Although no drift control is required in this step in order to limit the required damping ratio, the bare frames are designed such that in absence of viscous dampers inter-story drifts do not exceed 3%. The first three modes of the building are depicted in Fig. 11(a). Note that after placement of viscous dampers, natural periods and mode shapes of the building would not change as bypass viscous dampers virtually have zero stiffness.

Step 2. Target damping ratio: a 20% target damping ratio is selected in this scenario. As mentioned in the previous step, the maximum inter-story drift of the bare structure (without dampers) is estimated to be 3%. According to ASCE 7-16, damping reduction factor would be B = 1.5 for a damping ratio of 20%. As a result, it can be expected that for a target damping ratio of 20%, the maximum inter-story drift would not exceed 2%. The target damping ratio can be increased if less inter-story drifts are required. Note that as the building is a hospital, an importance factor of 1.4 is applied to all seismic loads per Standard No. 2800 (Iranian Seismic design and analysis code) [³¹]. As a result, it can be expected that less inter-story drifts would be observed during a Design Based Earthquake (DBE) with a return period of 475 years.

Step 3. Total required damping coefficient: from Eq. (2) the total required damping coefficient can be estimated as,

{{\displaystyle \mathrm{C}}}_{\mathrm{a}-X}=\mathrm{151}\mathrm{k}\mathrm{N}\cdot \mathrm{s}/\mathrm{m}\mathrm{m},

{{\displaystyle \mathrm{C}}}_{\mathrm{a}-Y}=\mathrm{157}\mathrm{k}\mathrm{N}\cdot \mathrm{s}/\mathrm{m}\mathrm{m}.

It should be noted that many software such as ETABS (Computers and Structures, 2017), automatically report lateral story stiffness in each direction. However, lateral story stiffness can also be estimated through dividing story shear by relative displacement of the story under a previously applied lateral load, for example with triangular pattern [³²].

Step 4. Placement of viscous dampers: total viscous dampers would be distributed to different stories based on the relative displacement of the stories in the fundamental mode shape. This issue is done in Table 3, and obtained results are shown in Fig. 11(b).

Step 5. Selecting final damper models: in this step, the final damper specifications would be selected based on the available viscous dampers on the Iranian market. The final selected viscous dampers are also presented in Table 3.

Step 6. Final check: finally the satisfactory seismic performance of the building which is designed by viscous dampers, should be verified through nonlinear time history analyses.

The buildings with and without viscous dampers are subjected to 10 pairs of scaled ground motions. All ground motions are scaled for 475 years target spectrum (DBE) and 2500 years target spectrum (MCE). So the performance of the building with and without viscous dampers are evaluated under both DBE and MCE hazard levels. P-M-M hinges are defined at both ends of the columns in all stories, flexural hinges are defined at both ends of the beams, and axial hinges with pinched cyclic behaviors are defined at the middle of the concentric braces. Plastic hinges are defined per ASCE 41-17 for columns, beams, beam-column connections, and braces. Fully restrained reduced beam section (RBS) connections are selected for all beam-to-column connections. Moreover, all bypass viscous dampers are modeled by using the pre-validated simplified Maxwell link elements.

Averages of maximum inter-story drifts, along with both directions and under both seismic hazards, are illustrated in Fig. 12. Obtained results of designed structure with viscous damper show that many maximum inter-story drifts of stories are smaller than the results of designed braced structure (without viscous dampers). Figure 12(a) shows inter-story drift in the X-direction with MCE hazard level. Inter-story drift in most stories of structure with damper is more than the inter-story drift of structure without damper although, based on results of MCE hazard level in Y-direction, inter-story drift values of structure with damper are more than inter-story drift values of the structure without damper in stories 5–7 that is shown in Fig. 12(b). According to Figs. 12(c) and 12(d), inter-story drift values of structure with damper are more than the values of this parameter in structure without damper which these results are obtained based on DBE hazard level. However, it should be clarified that the building with and without viscous dampers have two different lateral load resisting systems and focusing only on inter-story drifts might be misleading.

Figures 13 and 14 show performance of the building with and without viscous dampers under the ChiChi earthquake with MCE and DBE hazard levels, respectively. It is obvious that many structural elements in the building without this damper, especially braces, failed to satisfy the intended performance objective (Life Safety-LS under MCE and Immediate Occupancy-IO under DBE). In contrast, the building with viscous dampers satisfied the performance objectives.

Figures 13 (a) and 13(b) show performance objective of the structure without and with bypass viscous damper under MCE hazard level. According to Fig. 13(a), there are some elements of structure such as some braces and columns which do not satisfy the LS criteria, although all elements of structure with damper satisfy the LS criteria as shown in Fig. 13(b). The performance objective of the structure are evaluated under DBE hazard level by OI criteria as shown in Figs. 14(a) and 14(b) which they show an evaluation of IO criteria in structure without and with bypass viscous damper, respectively. Based on Fig. 14(a), some structural elements such as braces cannot satisfy IO criteria, although IO criteria is satisfied in structure with bypass viscous damper.

To evaluate the better effect of using bypass viscous damper, both scenarios are analyzed and assessed under 9 earthquake events. According to Table 4, the same results are obtained during the other 9 earthquakes. Accordingly, a building designed per the conventional force-based procedures will not necessarily satisfy performance objectives of the so called performance-based design. This is probably more pronounced for braced frames where they will surpass LS criteria in a rather small inter-story drift, while the current force-based procedure allows a story drift of about 2% for them.

Maximum base shear of the structure with and without viscous dampers are also presented in Table 4. It can be concluded that the building with viscous damper has experienced from 25% to 45% smaller base shear depending on the level of seismic hazard and direction. Obviously, in addition to the increased damping, flexibility of the building with viscous damper also contributed to this result. As a result of Table 4, base shear in both directions is decreased when the building is designed with viscous damper.

Another important response, especially for nonstructural elements and equipment, is the story acceleration. This parameter is also illustrated in different stories of the structure in Fig. 15. According to Fig. 15, story acceleration of structure with viscous damper is less than story acceleration of structure without viscous damper. On the other hand, average maximum story accelerations indicated that the structure with viscous dampers experienced up to 60% smaller story accelerations than structure without viscous damper. Again both flexibility and increased damping have contributed to such a pronounced reduction in story acceleration.

According to the main aim of the present study, the building with viscous damper was designed to achieve a target damping ratio of 20%. To investigate the level of accuracy of the proposed design procedure, viscous dampers are removed from the building, and its inherent damping ratio artificially is increased from 3% to 20%. The building is subjected to all 10 pairs of ground motions with MCE seismic hazard. Obtained results from the building with 20% damping ratio and the building with the designed viscous dampers are compared in Fig. 16. It can be found that unless inter-story drifts in upper stories, the response of both structures are in good agreement. The discrepancy in the upper stories might be due to the fact that higher mode effects are neglected during the design and placement of the viscous dampers.

This study highlighted that viscous dampers can significantly reduce story accelerations. Accordingly, it can be expected that non-structural elements in the hospital structure with viscous dampers are damaged less than those in the hospital structure without viscous dampers. In recent years, some useful studies on simulating FEM models have been published which evaluate solid material such as concrete [⁴³–⁴⁸]. To further investigate this claim, one of the masonry claddings at the 8th story is considered. Details of the considered wall are shown in Fig. 17. A 3-D finite element model of the masonry wall is created in ABAQUS software, and story acceleration from the Manjil earthquake (with DBE level) is imposed on the wall edges along its out-of-plane direction. According to ACI 530, mortar S with using masonry cement is considered in this study. Elastic modulus and ultimate shear stress of this material are considered as 0.26 and 0.4 MPa in this study, respectively. Contact and cohesive interactions are simultaneously defined at the bed-joint and head-joint of the masonry wall. It turned out that the masonry cladding would collapse in the hospital structure without damper due to the excessive story acceleration. The hospital structure with viscous dampers experiences significantly less story accelerations; therefore, the situation of masonry cladding is stable.

The collapse of the wall in the scenario of the structure without viscous dampers occurs which is shown in Fig. 18. This result verifies the contribution of viscous dampers to seismic performance of nonstructural elements. On the other hand, the vulnerability of non-structural elements in the structure which is design with damper, is less in comparing with the same structure which is design without viscous damper. According to Fig. 18(a), cracks of masonry wall are created and expanded, when t is reached to 11.95 s. Figure 18(b) shows the time that the masonry wall is collapsed. If t is reached to 12.60 s, the masonry wall is collapsed to outside. When t is reached to 13 s, the masonry wall collapsed to both of inside and outside as shown in Fig. 18(c).

Conclusions

A recently developed bypass viscous damper with an external orifice is introduced and experimentally evaluated in this study. Bypass viscous dampers have external orifices made from a high pressure flexible hose by which the damping coefficient can be adjusted. The external orifice also would act as a thermal compensator to alleviate viscous heating of the damper during dynamic excitations. Bypass viscous damper specimen revealed quite stable hysteretic behavior in different frequencies and amplitudes. Moreover, its behavior is simulated accurately by the currently available analytical and numerical tools from sophisticated CFD models to simplified Maxwell models.

A simplified step-by-step design procedure is proposed by which the damping ratio of a building under design can be increased to a predefined target value. Details of this procedure are explained for an 8-story hospital building. To make a faire comparison, the same building is also designed without viscous damper but with conventional special concentric braces. Nonlinear time history analyses indicated that the hospital structure without viscous damper would experience less inter-story drifts in many stories. This issue is due to the fact that in this scenario the design of the hospital building has concentric braced frames with high lateral stiffness. Contrary to the story drifts, the building with viscous damper performed quite satisfactory according to the so called performance-based design. The building with viscous dampers satisfies performance objectives of a hospital building, i.e., IO criteria in DBE and LS criteria in MCE seismic hazards. Although the hospital structure without viscous damper experiences less inter-story drifts, it is failed to satisfy the performance objectives. As a result, focusing only on story drifts might be misleading. Time history analyses have also shown that viscous dampers reduce significantly story accelerations. As a result, better seismic performance can be expected for non-structural elements and other equipments of the building with viscous dampers. This claim is verified by considering a 3-D finite element model of one of the masonry walls located at the 8th story of the hospital structure. Story accelerations of the building with and without viscous dampers are applied to the wall along its out-of-plane direction. It turned out that the considered nonstructural masonry wall in the building without viscous dampers collapse even under a DBE earthquake. But the same wall in the building with viscous damper remains stable under the same earthquake because of the reduced story accelerations.

References

[1] Miyamoto H, Scholl R E. Design of steel pyramid using seismic dampers. In: Building to Last. Portland, OR: Structures Congress XV, 1997

[2] Papanikolas P K. Deck superstructure and cable stays of the Rion-Antirion bridge. In: Proceedings of the 4th National Conference on Steel Structures. Patras: ASCE, 2002

[3] Zhang J, Makris N, Delis T. Structural characterization of modern highway overcrossings-case study. Journal of Structural Engineering, 2004, 130(6): 846–860

[4] Bacht T, Chase J G, MacRae G, Rodgers G W, Rabczuk T, Dhakal R P, Desombre J. HF2V dissipator effects on the performance of a 3 story moment frame. Journal of Constructional Steel Research, 2011, 67(12): 1843–1849

[5] Constantinou M C, Symans M D. Experimental and Analytical Investigation of Seismic Response of Structures with Supplemental Fluid Viscous Dampers. Technical Report NCEER-92-0032. Buffalo, NY: National Center for Earthquake Engineering Research, State University of New York at Buffalo, 1992

[6] Seleemah A A, Constantinou M C. Investigation of Seismic Response of Buildings with Linear and Nonlinear Fluid Viscous Dampers. Technical Report NCEER-97-0004. Buffalo, NY: National Center for Earthquake Engineering Research, State University of New York at Buffalo, 1997

[7] Kasai K, Matsuda K. Full-scale dynamic testing of response-controlled buildings and their components: Concepts, methods, and findings. Earthquake Engineering and Engineering Vibration, 2014, 13(S1): 167–181

[8] Infanti S, Papanikolas P, Catellano M G. Seismic protection of the Rion-Antirion bridge. In: The 8th World Seminar of Seismic Isolation, Energy Dissipation and Active Vibration Control of Structures. Yerevan, 2003

[9] Yamamoto M, Minewaki S, Nakahara M, Tsuyuki Y. Concept and performance testing of a high-capacity oil damper comprising multiple damper units. Earthquake Engineering & Structural Dynamics, 2016, 45(12): 1919–1933

[10] ASCE 7-16. Minimum Design Loads and Associated Criteria for Buildings and Other Structures. Reston, VA: American Society of Civil Engineers, 2016

[11] ASCE 41-17. Seismic Evaluation and Retrofit of Existing Buildings. Reston, VA: American Society of Civil Engineers, 2017

[12] Singh M P, Moreschi L M. Optimal placement of dampers for passive response control. Earthquake Engineering & Structural Dynamics, 2002, 31(4): 955–976

[13] Miguel L F, Miguel L F, Lopez R H. Simultaneous optimization of force and placement of friction dampers under seismic loading. Engineering Optimization, 2016, 48(4): 582–602

[14] Yousefzadeh A, Sebt M H, Tehranizadeh M. The optimal TADAS damper placement in moment resisting steel structures based on a cost-benefit analysis. International Journal of Civil Engineering, 2011, 9(1): 23–32

[15] Murakami Y, Noshi K, Fujita K, Tsuji M, Takewaki I. Simultaneous optimal damper placement using oil hysteretic and inertial mass dampers. Earthquakes and Structures, 2013, 5(3): 261–276

[16] Sonmez M, Aydin E, Karabork T. Using an artificial bee colony algorithm for the optimal placement of viscous dampers in planar building frames. Structural and Multidisciplinary Optimization, 2013, 48(2): 395–409

[17] Miguel L F F, Fadel Miguel L F, Lopez R H. A firefly algorithm for the design of force and placement of friction dampers for control of man-induced vibrations in footbridges. Optimization and Engineering, 2015, 16(3): 633–661

[18] Miguel L F, Miguel L F, Lopez R H. Robust design optimization of friction dampers for structural response control. Structural Control and Health Monitoring, 2014, 21(9): 1240–1251

[19] Zhang R H, Soong T T. Seismic design of vescoelastic dampers for structural applications. Journal of Structural Engineering, 1992, 118(5): 1375–1392

[20] Gluck N, Reinhorn A M, Gluck J, Levy R. Design of supplemental dampers for control of structures. Journal of Structural Engineering, 1996, 122(12): 1394–1399

[21] Takewaki I. Optimal damper placement for minimum transfer functions. Earthquake Engineering & Structural Dynamics, 1997, 26(11): 1113–1124

[22] Mousavi S A, Ghorbani-Tanha A K. Optimum placement and characteristics of velocity-dependent dampers under seismic excitations. Earthquake Engineering and Engineering Vibration, 2012, 11(3): 403–414

[23] Fahiminia M, Saeed M H. Experimental investigation and control of structures by using of semi-active viscous dampers. Thesis for the Master’s Degree. Iran: Islamic Azad University, 2017

[24] Black C, Makris N. Viscous Heating of Fluid Dampers under Wind and Seismic Loading: Experimental Studies, Mathematical Modeling and Design Formulae. Report No. EERC 2006-01. Berkeley, CA: The University of California, 2006

[25] Luo Z, Yan W, Xu W, Zheng Q, Wang B. Experimental research on the multilayer compartmental particle damper and its application methods on long-period bridge structures. Frontiers of Structural and Civil Engineering, 2019, 13(4): 751–766

[26] Dong B, Ricles J M, Sause R. Seismic performance of steel MRF building with nonlinear viscous dampers. Frontiers of Structural and Civil Engineering, 2016, 10(3): 254–271

[27] Abaqus. Analysis User’s Manual, Version 6.13 2013. Providence, RI: Dassault Systemes Simulia Corp, 2015

[28] SAP. Structural Analysis Program. Version 19.1.1. Edinburgh, CA:Computers and Structures Inc., 2000

[29] Cimellaro G P, Retamales R. Optimal softening and damping design for buildings. Structural Control and Health Monitoring, 2007,14(6): 831–857

[30] Ghorashi S S, Valizadeh N, Mohammadi S, Rabczuk T. T-spline based XIGA for fracture analysis of orthotropic media. Computers & Structures, 2015, 147: 138–146

[31] Amiri F, Anitescu C, Arroyo M, Bordas S P A, Rabczuk T. XLME interpolants, a seamless bridge between XFEM and enriched meshless methods. Computational Mechanics, 2014, 53(1): 45–57

[32] Areias P, Msekh M A, Rabczuk T. Damage and fracture algorithm using the screened Poisson equation and local remeshing. Engineering Fracture Mechanics, 2016, 158: 116–143

[33] Areias P, Rabczuk T. Steiner-point free edge cutting of tetrahedral meshes with applications in fracture. Finite Element Analysis, 2017, 132: 27–41

[34] Shishegaran A, Amiri A, Jafari M A. Seismic performance of box-plate, box-plate with UNP, box-plate with L-plate and ordinary rigid beam-to-column moment connections. Journal of Vibroengineering, 2018, 20(3):1470–1487

[35] Shishegaran A, Rahimi S, Darabi H. Introducing box-plate beam-to-column moment connections. Vibroengineering PROCEDIA, 2017, 11: 200–204

[36] Areias P, Reinoso J, Camanho P P, César de Sá J, Rabczuk T. Effective 2D and 3D crack propagation with local mesh refinement and the screened Poisson equation. Engineering Fracture Mechanics, 2018, 189: 339–360

[37] Ren H, Zhuang X, Rabczuk T. Dual-horizon peridynamics: A stable solution to varying horizons. Computer Methods in Applied Mechanics and Engineering, 2017, 318: 762–782

[38] Ren H, Zhuang X, Cai Y, Rabczuk T. Dual-horizon Peridynamics. International Journal for Numerical Methods in Engineering, 2016, 108(12): 1451–1476

[39] Desombre J, Rodgers G W, MacRae G A, Rabczuk T, Dhakal R P, Chase J G. Experimentally validated FEA models of HF2V damage free steel connections for use in full structural analyses. Structural Engineering & Mechanics, 2011, 37(4): 385–399

[40] Joshi S, Hildebrand J, Aloraier A S, Rabczuk T. Characterization of material properties and heat source parameters in welding simulation of two overlapping beads on a substrate plate. Computational Materials Science, 2013, 69: 559–565

[41] Bacht T, Chase J G, MacRae G, Rodgers G W, Rabczuk T, Dhakal R P, Desombre J. HF2V dissipator effects on the performance of a 3 story moment frame. Journal of Constructional Steel Research, 2011, 67(12): 1843–1849

[42] ETABS. Integrated Building Design Software. Version 16.2.0. CA: Computers and Structures, Inc., 2017

[43] Hamdia K M, Silani M, Zhuang X, He P, Rabczuk T. Stochastic analysis of the fracture toughness of polymeric nanoparticle composites using polynomial chaos expansions. International Journal of Fracture, 2017, 206(2): 215–227

[44] Ghasemi M R, Shishegaran A. Role of slanted reinforcement on bending capacity SS beams. Vibroengineering PROCEDIA, 2017, 11: 200–204

[45] Rabczuk T, Ren H, Zhuang X. A nonlocal operator method for partial differential equations with application to electromagnetic waveguide problem. Computers, Materials & Continua, 2019, 59(1): 31–55

[46] Anitescu C, Atroshchenko E, Alajlan N, Rabczuk T. Artificial neural network methods for the solution of second order boundary value problems. Computers, Materials & Continua, 2019, 59(1): 345–359

[47] Guo H, Zhuang X, Rabczuk T. A deep collocation method for the bending analysis of Kirchhoff Plate. Computers, Materials & Continua, 2019, 59(2): 433–456

[48] Shsihegaran A, Ghasemi M R, Varaee H. Performance of a novel bent-up bars system not interacting with concrete. Frontiers of Structural and Civil Engineering, 2019, 13(6): 1301–1315

RIGHTS & PERMISSIONS

Higher Education Press