International Association of Museum Facilities Administrators (IAMFA) The 2003 IAMFA Annual Conference in San Francisco, California September

21-24, 2003

BASE ISOLATION TECHNOLOGIES FOR SEISMIC PROTECTION OF MUSEUM ARTIFACTS

Bujar Myslimaj, Scott Gamble, Darron Chin-Quee and Anton Davies Rowan Williams Davies & Irvin Inc., Consulting Engineers 650 Woodlawn Road West, Guelph, Ontario, Canada, N1K 1B8 Brian Breukelman Motioneering Inc. 650 Woodlawn Road West, Guelph, Ontario, Canada, N1K 1B8

INTRODUCTION Base isolation technologies have been used traditionally to improve the seismic performance of buildings and other large structures such as bridges, etc.. In recent years the application of base isolation has been gradually extended to smaller structures - private housing, computer servers storing valuable data as well as in the seismic protection of museum artifacts. Installation of base isolation systems beneath showcases or sculptures displayed inside or outside museums provides effective protection of important and irreplaceable cultural properties and works of art (Fig. 1). Display cases or sculptures are often rigidly connected to the floor (Fig. 2) thus being prone to intensive shaking and damage to contents or internal structures during seismic events.

Fig. 1 Conceptual representation of a base isolation system installed beneath a pod.

Fig. 2 Current practice in museum displays. Current seismic design codes consider showcases, preservation racks and shelves as nonstructural elements or components. Their seismic design is covered by code provisions for non-structural elements, which focus mainly on the design of the connection of the nonstructural elements to the main structural system. Ensuring the seismic integrity of the connection between the building structure and shelves, showcases, etc. does not guarantee the safety of the showcase or shelf contents. Significant motion of artifacts supported on or housed within display cases can occur, leading to damage. To improve the seismic performance of non-structural components and avoid the permanent loss or breakage of irreplaceable or expensive assets (Fig. 3), application of effective technologies that can control the seismic response of non-structural components is needed.

Fig. 3 Avoiding the permanent loss or breakage of irreplaceable or expensive assets during a seismic event should be the top priority.

RECENT DEVELOPMENTS AND APPLICATIONS Recently developed compact base isolation systems for small-scale structures are based on sliding, rolling and rubber bearing techniques [1,2]. The rolling type design has proved to be very effective in improving the seismic performance of non-structural components. In Japan, a rolling type base isolation system called Tuned Configuration Rail (TCR) has been successfully applied during recent years in seismic base isolation of private housing, computer servers and more widely in museum showcases [3-6]. This system consists of eight wheels and eight tuned configuration rails installed between two parallel platforms. These platforms can move freely against each other in one orthogonal direction only (Fig. 4), which provides for movement in any direction in the horizontal plane. By adjusting the curvature of the rails, the system can be tuned so that its motion in the presence of a seismic event offsets the motion of the supporting structure. It is a simple and compact base isolation system that can be easily installed underneath existing (Fig. 5a) or new showcases (Fig. 5b).

Fig. 4 A TCR isolator designed for small size artifacts (courtesy of AS Inc., Japan).

Fig. 5 Base isolation systems installed under existing (a) or new (b) showcases (courtesy of AS Inc., Japan).

Since the force that brings the system back to its initial/original position and the damping force generated as a result of the friction between wheels and rails are both proportional to the weight, the system can be easily adjusted for a wide range of museum applications. It reduces the seismic response acceleration up to one tenth of the input excitation as shown in Fig. 6, where the input motion (i.e. motion of the non-isolated platform) and the seismic response of the base isolated platform in terms of acceleration are plotted for comparison. Results shown in Fig. 6 are taken from a recent 3-dimensional seismic performance shaking table test. The shaking table can simulate the earthquake ground motion. In the example, the input acceleration wave corresponds to the North-South direction of ground motion recorded during the Kobe earthquake of January 17, 1995 with peak acceleration of 818gal (1gal=1cm/s2=0.001g, where g is gravity acceleration). Peak response acceleration of the system is 72gal, or approximately 1/12 of the magnitude of the input motion.
INPUT ACCERALATION N/S 1000 800 600 400 200 0 -200 -400 -600 -800 -1000 0 5 10 15
TIME(sec)

ACC(gal

20

25

30

35

OUTPUT ACCERALATION N/S 1000 800 600 400 200 0 -200 -400 -600 -800 -1000 0 5 10 15
TIME(sec)

ACC(gal

20

25

30

35

Fig. 6 Shaking table test results for a TCR isolator. Table 1 illustrates the significance of the input earthquake ground motion levels and the output or the base-isolated motion levels shown in Fig. 6. Depending on the showcase or display location inside the museum, the TCR design can be easily adapted to meet the aesthetic and seismic performance requirements (Fig. 7). For the existing showcases enclosed directly against a wall, the base isolation system can be designed in the form of an integrated set of isolated platforms that can be installed within the showcases, offering thus a cost and time effective solution. Applications are also not limited to indoor locales as a TCR can be installed outdoor where valuable art works (sculptures, statues) are often displayed (Fig. 8). For these applications the TCRs can be readily designed to meet stringent aesthetic and performance requirements.

Table 1 Overview of earthquake ground motion levels in relation to human perception and damage potential [7]
Earthquake Intensity IMM I II III Description Detected with sensitive instrumentation Felt by few persons on upper levels; suspended objects may swing Felt noticeably indoors, but not always recognized as an earthquake; parked cars rock slightly Felt indoors by many, some people awaken; parked cars rock noticeably Felt by most people; cracked plaster in a few places; disturbances of trees, poles, and other tall objects sometimes noticed Felt by all; many are frightened; a few instances of fallen plaster; slight damage Everybody runs outdoor; damage to buildings varies, depending on the quality of the construction Panel walls thrown out of frames; walls, monuments, chimneys fall; drivers disturbed Buildings shifted off foundations, cracked, thrown off plumb, ground cracked; underground pipes broken Landslides; rails bent; most masonry and framed structures destroyed; ground cracked Bridges destroyed; broad fissures in ground; earth slumps and land slips in soft ground Total destruction Approximate peak ground horizontal acceleration (gal) <3

3-7 7-15 15-30 30-70 70-150 150-300 300-700 700-1500 1500-3000 3000-7000

IV V

VI San Francisco 1957 Taft, 1952 El Centro 1940 Northridge, 1994 Kobe, 1995 VII

VIII IX

X XI XII

Fig. 7 Works of art on base isolated display platforms (courtesy of AS Inc., Japan).

THE NATIONAL MUSEUM OF W ESTERN ART 2 SETS OF TCR SEISMIC ISOLATORS

Fig. 8 A base isolated statue in display outside the museum (courtesy of AS Inc., Japan).

OTHER IMPORTANT CONSIDERATIONS The evaluation of seismic performance of the base isolation systems for museum artifacts requires detailed information on the systems fundamental dynamic response properties (i.e., factors affecting its characteristic motion and response to a disturbing force). In addition, a reliable prediction of input motion characteristics (i.e., ground motion during an earthquake) is required at the site where the museum is located. This would normally lead to additional analyses to generate site-specific ground motions [8] or seismic design spectrum compatible input ground motions needed for performance evaluation [9]. Besides the use of analytical methods for seismic performance evaluation, shaking table testing has been also used to verify the seismic performance of the isolation systems. This approach has been used in addition to the analytical one, and has proven to be very important, particularly at the early stage of the design. CONCLUDING REMARKS Rolling type base isolation systems have been proven to be very effective in improving the seismic performance of operational and functional components attached to the main structural system. Recently, a rolling type base isolation system called Tuned Configuration Rail (TCR) has been successfully applied during the last few years in seismic base isolation of private housing, computer servers and more widely in museum showcases. It is a compact isolator that significantly reduces the acceleration response and can be easily installed underneath new or existing showcases, preservation racks, shelves and statues. REFERENCES 1. Iiba, M., Midorikawa, M., Yamanouchi, H. and Myslimaj, B. (1999), Three dimensional shaking table tests on seismic behavior of isolators for houses, Proceedings of the 30-th Joint Meeting of U.S.-Japan Panel on Wind and Seismic Effects, UJNR, May 1999, Tsukuba, Japan. 2. Myslimaj, B., Iiba, M. and Midorikawa, M. (1999), 3-dimensional shaking table tests on base-isolation systems for houses, International Workshop on Seismic Isolation, Energy Dissipation and Control of Structures, 6-8 May 1999, Guangzhou, China. 3. Yamada, C., Iiba, M., Myslimaj, B., Inoue, K., Seki, M., Hasegawa, O., Yatsuhashi, M., Yasui, Y. and Akimoto, M. (1999), Three dimensional shaking table tests on seismic behavior of isolators for houses - Part 2: Effect of bi-directional and vertical earthquake

4.

5.

6.

7. 8.

9.

motions on characteristics of isolators, Summaries of Technical Papers of Annual Meeting of Architectural Institute of Japan, Vol. B-2, pp. 743744 (in Japanese). Inoue, K., Iiba, M., Myslimaj, B., Yamada, C., Seki, M., Hasegawa, O., Yatsuhashi, M. and Yasui, Y. (1999), Three dimensional shaking table tests on seismic behavior of isolators for houses - Part 3: Effect of unbalanced weight on characteristics of isolators, Summaries of Technical Papers of Annual Meeting of Architectural Institute of Japan, Vol. B-2, pp. 745746 (in Japanese). Enomoto, T., Omori, Y., Iiba, M. and Myslimaj, B. (1999), Three dimensional shaking table tests on seismic behavior of isolators for houses - Part 6: Effect of base isolation on the response of superstructure, Summaries of Technical Papers of Annual Meeting of Architectural Institute of Japan, Vol. B-2, pp. 751752 (in Japanese). Egmond J.V. and Myslimaj, B. (2002), Seismic damage control technologies for protection of national assets and treasures, Presentation at Public Works and Government Services Canada, August 2002, Ottawa, Canada. Richter, C.R. (1958), Elementary Seismology, W.H. Freeman, San Francisco. Myslimaj, B. and Matsushima, Y. (1997), Stochastically based estimation of sitespecific ground motion parameters: - A design oriented approach -, Proceedings of the Seventh International Conference on Computing in Civil and Building Engineering (ICCCBE-VII), 19-21 August 1997, Seoul, Korea, VOLUME 2, pp. 12651270. Myslimaj, B. and Matsushima, Y. (1997), Inelastic earthquake response of structures accounting for local soil conditions, Journal of Structural and Construction Engineering, Transactions of AIJ, No. 497, pp. 4755.