Content uploaded by Nikola Naumovski
Author content
All content in this area was uploaded by Nikola Naumovski on Nov 22, 2022
Content may be subject to copyright.
10911st Croatian Conference on Earthquake Engineering - 1CroCEE
1st
Croatian Conference on Earthquake Engineering
1CroCEE 22-24 March 2021 Zagreb, Croatia
DOI: https://doi.org/10.5592/CO/1CroCEE.2021.132
Shake table tests for seismic assessment of non-structural
elements
L. Krstevska1, A. Bogdanovic2, R. Rimboeck3, A. Poposka4, F. Manojlovski4,
A. Shoklarovski5, N. Naumovski4, I. Markovski6, D. Filipovski7
1 Professor, University “Ss. Cyril and Methodius”, Skopje, North Macedonia, Institute of Earthquake
Engineering and Engineering Seismology, lidija@iziis.ukim.edu.mk
2
Associate Professor,, University “Ss. Cyril and Methodius”, Skopje, North Macedonia, Institute of Earthquake
Engineering and Engineering Seismology, saska@iziis.ukim.edu.mk
3
Dipl.ing., GiB GmbH, Arnstorf, Germany, Robert.Rimboeck@Gib-GmbH.com
4
Research Assistant,University “Ss. Cyril and Methodius”, Skopje, North Macedonia, Institute of Earthquake
Engineering and Engineering Seismology, angela@iziis.ukim.edu.mk; filipmanojlovski@iziis.ukim.edu.mk;
nikolan@iziis.ukim.edu.mk
5
Ext. Professional Associate, University “Ss. Cyril and Methodius”, Skopje, North Macedonia, Institute of
Earthquake Engineering and Engineering Seismology, antonio@iziis.ukim.edu.mk
6 Ext. Professional Associate, University “Ss. Cyril and Methodius”, Skopje, North Macedonia, Institute of
Earthquake Engineering and Engineering Seismology, igorm@iziis.ukim.edu.mk
7
Professional Associate, University “Ss. Cyril and Methodius”, Skopje, North Macedonia, Institute of
Earthquake Engineering and Engineering Seismology, dejan@iziis.ukim.edu.mk
Abstract
With the development of the new technologies and the modern residential and industrial
buildings typology the non-structural components and systems become major part of the total
value in building construction. This damage may result in loss of functionality, economic loss due
to damage and even life safety hazards. Since they are not amenable to traditional structural
analysis, full-scale experimental testing is crucial to understand their behavior under earthquake.
In this paper, results of the experimental testing of non-structural elements including raised
floors, ceilings and cleanroom systems, performed at the Dynamic Testing Laboratory in the
Institute of Earthquake Engineering and Engineering Seismology - IZIIS, Skopje, Republic of North
Macedonia will be presented. Seismic certification of this systems has been conducted according
to AC-156 - Acceptance criteria for seismic qualification by shake table testing for non-structural
components. For this reason, a steel cube structure was properly designed in order to simulate
the seismic effects at a generic building story where different types and configurations of non-
structural elements were implemented. Results have been obtained in terms of accelerations,
displacements and strains in characteristic points. The systems showed good performance and it
was confirmed that all acceptance criteria have been fulfilled during and after the seismic tests.
Based on the experimental research it was observed that most of the tested non-structural
elements successfully passed the seismic acceptance criteria for shake table testing of non-
structural components and systems according to ICC AC-156.
Key words: partition walls, shake table test, non-structural elements, AC-156
1092 SEISMIC PERFORMANCE OF STRUCTURES
1st Croatian Conference on Earthquake Engineering - 1CroCEE
1 General appearance
The use of non-structural elements in modern buildings like industrial and business
centers or in laboratories and hospitals with precious equipment is constantly gaining
attention due to the intensity of their implementation as well as their value. Even tho-
ugh they are considered as completely non-structural elements, in the events of ear-
thquakes, they are more vulnerable than the primary system. Their damage may re-
sult in loss of functionality, economic loss due to damage and even life safety hazards
[1,2]. Since there are many different types of products whose seismic response has
not been fully experimentally investigated and there is lack of strict design standards,
the investors require seismic certification by shake table testing. For the purpose of
seismic qualification of specific types of raised floors, ceilings and cleanroom systems
by shake table testing in the Institute of Earthquake Engineering and Engineering Sei-
smology in Skopje, Republic of N. Macedonia, a series of shake table seismic tests have
been performed according to the ICC AC-156 criteria [3]. Analysis of dynamic behavior
has been performed and results on earthquake response of the considered elements
have been obtained. Many different configurations of products were tested based on
the most frequent installation scenarios. Hence, there were 32 different raised floor
configurations, varying in height, constraints, etc., 6 ceiling configurations and 11 clean-
room system configurations. All of them were subjected to biaxial, artificially produced
earthquake excitation time history. Although most of the tested elements successfully
fulfilled the standard requirements, valuable data regarding the seismic response of the
specific nonstructural elements was also acquired. The testing procedure and selected
results are briefly presented in this paper.
2 Testing procedure
The testing program has been selected to comply with the seismic certification test
procedure described in chapter 6 of AC156 - Acceptance Criteria for Seismic Certifica-
tion by Shake Table Testing of Non-structural Components [3]. Specifically, the criteria
require simultaneous testing in both horizontal and vertical direction. Each specimen
was tested under two different types of excitations, resonant frequency search excita-
tions and time-history seismic excitations, for determination of the natural frequencies
of the specimens and the global dynamic response, respectively. All non-structural ele-
ments were tested in the Laboratory for Dynamic Testing in the Institute of Earthquake
Engineering and Engineering Seismology. The IZIIS’ shake table comprises a 5m by 5m
platform supported by two lateral and four vertical actuators, providing 5 degrees-of-
freedom (DOF). The table can carry up to 40 tons and a peak table acceleration of 3.0g
with peak displacements of ±125 mm in horizontal direction and acceleration of 1.5g
with peak displacement of ±60 mm in vertical direction. The shake table uses new, sta-
te-of-the-art digital control system produced by MTS and data acquisition system pro-
duced by National Instruments.
1093SEISMIC PERFORMANCE OF STRUCTURES
1st Croatian Conference on Earthquake Engineering - 1CroCEE
2.1 Resonant frequency test
For the resonant frequency search tests, random and sine sweep excitations have been
applied in each direction, horizontally and vertically independently, before testing of each
specimen (initial state), after certain biaxial tests and after all performed tests (final state).
The performed random and harmonic tests were in a frequency domain of 1.0-35.0(40.0)
Hz, with sweep rate of 2.0 octave/min and with peak excitation level of 0.01g-0.05g, in ho-
rizontal and vertical direction. Beside the standard resonant tests performed for each type
of non-structural elements, additional sine-sweep tests were carried out for determination
of the dynamic characteristics for different sizes and heights of raised floor systems. These
excitations were performed with 0.02g input acceleration in both X and Z direction, in frequ-
ency range of 1.0-40 Hz and 1.0-45.0 Hz for Y and Z direction, respectively.
2.2 Seismic tests
The main purpose of the testing procedure is to generate acceleration or test spectrum
time histories in accordance with the prescribed standardized spectra in vertical and
horizontal directions, respectively. Bi-axial time history tests, in accordance with the
6.5 Multi-frequency Seismic Simulation Tests of AC156 [3], were carried out by simul-
taneous, but independent inputs in the horizontal and vertical axes, each producing the
Required Response Spectrum (RRS) along the respective reference axis calculated with
1/12 octave of frequency bandwidth, 5 % damping and prescriptions reported in AC156.
The non-symmetric, non-structural elements were rotated for 90° in order each direc-
tion to be tested. The tests were conducted by gradually increasing the intensity of the
time history earthquake, starting with 0.23g and 0.17g and finishing with 0.92g and
0.64g in Y and Z direction, respectively. Fig. 1 shows the maximum input acceleration
time history plots in horizontal and vertical direction as well as the corresponding Test
Response Spectra (TRS) that match the Required Response Spectra given in AC 156.
Figure 1. Input time histories of the test with the highest excitation and TRS vs RRS
1094 SEISMIC PERFORMANCE OF STRUCTURES
1st Croatian Conference on Earthquake Engineering - 1CroCEE
3 Raised floors
3.1 Description of the tested specimen
The tested units consisted of vertical bearing support elements, steel pedestals with
varying heights of 0.5 m, 1.0 m and 1.5 m, fixed on the bottom and on the top, connec-
ted by a grid of perpendicular steel beams, each one of them intersecting the longitu-
dinal axis of the pedestals [4-6]. The longitudinal and transverse distance between the
pedestals was 0.6 m, i.e., the same as the sides of the floor panels. The only difference
between the setups were the 0.5 m high raised floors, where on the top of the pede-
stals, the floor panels were put directly on special shaped joints on the pedestals, wit-
hout the steel beam grid substructure. Additionally, the different configurations varied
in terms of bracings, i.e., they were without and with adjustable steel bracings inserted
in order to limit the horizontal motions of the raised floor sub-structure. Depending on
the type of the raised floor, different bracings were used.
In order to additionally explore the behavior of the increasing number of modules, a
varying number of panels were tested, i.e. one panel, three by three, five by five and
seven by seven panels, filling the space in the steel cube representing a real life scenario
where the floor is constrained by the elements of the primary structural system, resul-
ting in 32 different configurations.
The last variation considered in the investigation was the type and amount of additional
load added to the floor panels, simulating distributed and concentrated load.
In Fig. 2, a five by five panel configuration without the additional load from the 1.0m tall
pedestals is shown on the left and a view of a raised floor structure from the bottom is
shown on the right side.
Figure 2. Five by five panel configuration – left. Raised floor structure – right
1095SEISMIC PERFORMANCE OF STRUCTURES
1st Croatian Conference on Earthquake Engineering - 1CroCEE
3.2 Instrumentation
The specific and numerous experimentally tested raised floor configurations had a type
of modular instrumentation that was easily adjustable as the number of panels was
increasing. Four different types of transducers, namely, accelerometers (ACC), linear
potentiometers (LP), linear variable differential transformers (LVDT) and strain gauges
(SG) were installed to measure acceleration, absolute displacement and relative displa-
cement. Their precise placement on the raised floor structure and steel cube is shown
in Fig. 3.
Figure 3. Instrumentation set up for a raised floor system
3.3 Selected results
For the raised floor systems 1.0 m high, based on the presented results, the following
frequencies in horizontal direction were obtained: for the configuration with bracings
starting from 1x1 panel, 3x3 panels, 5x5 panels and 7x7 panels, frequencies of 30.02
Hz, 11.35 Hz, 7.32 Hz, 6.19 Hz respectively were measured (Fig. 4a). It was observed
that the increase of the number of panels led to a decrease of the value of the reso-
nant frequency. The same tendency is visible for the panels without bracings where,
in the same configurations, the measured respective frequencies of 6.36 Hz, 4.25 Hz,
3.81 Hz, 4.67 Hz are significantly lower compared to the previous ones (configurations
with bracings) due to the lower stiffness. If the results obtained for the frequencies are
compared, it is obvious that the increased number of panels and the use of bracings
significantly reduce the value of the resonant frequency.
1096 SEISMIC PERFORMANCE OF STRUCTURES
1st Croatian Conference on Earthquake Engineering - 1CroCEE
Figure 4. Change of dynamic characteristics of raised floors a) 1.0m high, b) 1.5m high
Analyzing the 1.5 m high raised floor system, the following frequencies in horizontal
direction were obtained: for the configuration with bracings starting with 1x1 panel,
2x2 panels, 3x3 panels and 5x5 panels, frequencies of 33.1 Hz, 20.1 Hz, 11.1 Hz, 5.3
Hz were measured, respectively (Fig. 4b). It can also be observed that the increase of
the number of panels leads to a decrease of the value of the resonant frequency. The
presence of bracings had a clear influence on the frequency value. The removal of the
bracings leads to stiffness degradation of the floor and lower frequency value. A simi-
lar tendency is also visible in the case of panels without bracings where, for the same
configurations, the measured frequencies of 4.8 Hz, 4.3 Hz, 3.5 Hz, 3.3 Hz, respectively,
are significantly lower compared to the configurations with bracings due to the lower
stiffness.
The plots in Fig. 5 and Fig. 6 show the input acceleration time history versus the accele-
ration time history at the level of floor panels. The maximum input acceleration is 0.85g
and the top acceleration with certain amplification ranges from 3.7g to 5.1g for the
specific measured points.
Figure 5. Input TH vs output TH of acceleration at point 5, a) braced, b) unbraced, 1.0m high
Figure 6. Input TH vs output TH of acceleration at point 5, a) braced, b) unbraced, 0.5m high
1097SEISMIC PERFORMANCE OF STRUCTURES
1st Croatian Conference on Earthquake Engineering - 1CroCEE
Related to the dynamic behavior during the most intensive seismic tests, there was
noticeable sliding and slight separation of the panels, but there was no damage of the
elements of the system and no change in the stability of the tested configurations.
4 Ceiling systems
Ceiling systems represent suspended nonstructural elements intended to architectu-
rally form space and simultaneously provide space for the necessary air ventilation
systems or other types of installations.
4.1 Description of the tested specimen
Testing of ceiling systems has been carried out by examining 10 different types of pa-
nels divided into two layouts [7]. Each layout consisted of 5 different types of panels,
supported by steel profiles set in a basic steel cube. Both layouts were tested biaxially
in both orthogonal directions (Y-Z and X-Z). For the ceilings of layout 1, some additional
components as seismic clips, screws and stoppers were used, while for the ceilings of
layout 2, stoppers and ropes were used. During the tests, additional load of different
weight was added to some of the panels simulating real-life exploitation scenarios.
4.2 Instrumentation
The testing protocol consisted of sine sweep tests and biaxial test, simultaneously in
horizontal and vertical direction, according to the general testing methodology. A total
number of 46 tests were performed for testing all 6 configurations. The response of
the models was monitored by data acquisition system and transducers consisting of 20
accelerometers (ACC), 12 linear variable differential transducers (LVDT), 1 linear poten-
tiometer (LP) and 5 strain gages (SG), providing information about accelerations, relative
and absolute displacement and deformation at different points. Graphical representati-
on as well as photo of the test setup and instrumentation are shown in Fig. 7.
1098 SEISMIC PERFORMANCE OF STRUCTURES
1st Croatian Conference on Earthquake Engineering - 1CroCEE
Figure 7. Ceiling systems - test setup and instrumentation
4.3 Selected results
The maximum relative horizontal displacement of the most flexible ceiling panel was
13.5 mm measured by LVDT 1 (Layout 1 – Configuration 1), while the maximum re-
lative vertical displacement between the steel profile and the panel was 14.9 mm as
measured by LVDT1 (Layout 1 – Configuration 2), Fig. 8. The maximum measured acce-
leration in horizontal direction was 4.25g (accelerometer ACC01) and the maximum
measured acceleration in vertical direction was 3.20g (accelerometer ACC02) (Layout
1 – Configuration 2). The selected time histories are shown in Fig. 9.
Figure 8. Time histories of maximum relative displacement, a) configuration 1, b) configuration 2
Figure 9. Time histories of maximum measured accelerations, a) ACC 01, b) ACC 02
1099SEISMIC PERFORMANCE OF STRUCTURES
1st Croatian Conference on Earthquake Engineering - 1CroCEE
During and after the seismic test, all units of the ceiling systems and their supporting
structures showed neither damage nor loss of function, except two different types of
panels, which fell during the strongest tests.
5 Cleanroom systems
Cleanroom systems are special type of closed rooms inside the main structural systems.
They have specific dust-proof, humidity-proof or some other special characteristics.
5.1 Description of the tested specimen
Cleanroom systems are generally more complicated non-structural systems than ra-
ised floors and ceilings because they contain more non-structural elements, ceilings,
partitions, raised floors, etc. The tested specimen for clean room systems were divided
into two different layouts [8]. The first layout consisted of one type of a ceiling system,
15 different partition walls, 2 different types of doors and one type of installations. The
second layout consisted of one type of a ceiling system, 20 different partition walls,
2 different types of doors and one type of installations. Each layout had an additional
or optional component as silicone wall or bracing system. Moreover, an imposed load
of 150 kg/m2 or a point load of 4x150 kg were used for testing of the systems. Since
the cleanroom systems were not symmetrical, they were rotated for 90° in order both
horizontal directions to be tested. Taking into consideration combinations of variables
including non-structural element, silicone wall, bracing systems, type of load and rota-
tion of specimens, eleven different configurations of clean-room systems were tested.
5.2 Instrumentation
Fig. 10 presents a photo of one of the tested configurations as well as the specific in-
strumentation. Each configuration was tested biaxially, simultaneously in horizontal
and vertical direction, according to the general testing methodology described in section
2. A total number of 62 tests were performed for testing all 11 configurations. To obtain
valuable data, 26 different transducers were used, out of which 11 accelerometers, 4
strain gauges and 11 linear variable differential transformers.
Figure 10. Photo of the tested clean room and instrumentation scheme
1100 SEISMIC PERFORMANCE OF STRUCTURES
1st Croatian Conference on Earthquake Engineering - 1CroCEE
5.3 Selected results
Maximum relative displacement of 19.6 mm and maximum acceleration of 6.4g were
measured during the intensive seismic tests. The corresponding time histories are pre-
sented in Fig. 11. After the test performed according to the test procedure, no visible
damage nor loss of function was noticed in 9 configurations. In the remaining 2 configu-
rations, slight bending of some panels and braces of the ceiling system was observed.
Figure 11. Time histories of maximum rel. displacement and acceleration during the seismic tests
6 Conclusions
During earthquakes, non-structural components can be seriously damaged, and their
failure can result in loss of human lives and extensive repair costs. Their stability and
seismic performance are necessary to be verified according to the acceptance criteria in
the prescribed standards.
The presented testing of different configurations and layouts of raised floors, ceilings
and cleanroom systems was performed at the Dynamic Testing Laboratory of the In-
stitute of Earthquake Engineering and Engineering Seismology, “Ss. Cyril and Methodi-
us” University in Skopje, Republic of N. Macedonia according to the AC156 acceptance
criteria for seismic qualification by shake table testing of non-structural components.
Qualification by biaxial testing in horizontal and vertical direction was simultaneously
performed with corresponding Test Response Spectra (TRS) that match the Required
Response Spectra (RRS) given in AC156.
The obtained results of this complex experimental research showed that almost all te-
sted systems successfully passed the seismic acceptance criteria for shake table te-
sting of non-structural components and systems according to the applied standard.
References
[1] Petrone, D., Calvi, P.M., Nascimbene, R. et al. (2019): Seismic Performance of Non-structural
Elements during the 2016 Central Italy Earthquake. Bulletin of Earthquake Engineering, 17 (10),
5655-5677.
[2] FEMA 74 (1994): Reducing the Risks of Nonstructural Earthquake Damage. A practical guide
issued by FEMA in furtherance of the Decade for Natural Disaster Reduction. Third edition.
[3] ICC-ES AC156: Acceptance Criteria for seismic Certification by Shake-Table Testing of Non-
structural Components: 2010 Edition
1101SEISMIC PERFORMANCE OF STRUCTURES
1st Croatian Conference on Earthquake Engineering - 1CroCEE
[4] Krstevska, L., Bogdanovic, A., Poposka, A., Manojlovski, F., Shoklarovski, A. (2019): Dynamic Testing
of Nortec Standard Floor (h = 1500mm). Technical Report IZIIS 2019-56.
[5] Krstevska, L., Bogdanovic, A., Poposka, A., Manojlovski, F., Shoklarovski, A. (2019): Seismic
Certification of Nortec Standard Floor 3kN/m2 According to AC156. Technical Report IZIIS 2018-
55.
[6] Krstevska, L., Bogdanovic, A., Poposka, A., Manojlovski, F., Shoklarovski, A. (2019): Seismic
Certification of Nortec Heavy Duty Floor 5kN/m2 +15kN/m2 Concentrated Load According to
AC156. Technical Report IZIIS 2018-53.
[7] Krstevska, L., Bogdanovic, A., Poposka, A., Manojlovski, F., Shoklarovski, A. (2019): Seismic
Certification of Ceiling Systems According to AC156. Technical Report IZIIS 2019-43.
[8] Krstevska, L., Bogdanovic, A., Poposka, A., Manojlovski, F., Shoklarovski A. (2019): Seismic
Certification of Cleanroom Systems According to AC156. Technical Report IZIIS 2019-56.
1102 1st Croatian Conference on Earthquake Engineering - 1CroCEE