Seismic Rehabilitation of Historic Brick Buildings by Inserting Stainless Pins
N. Araki, Associate Professor, Department of Architecture and Architectural Engineering, and Takiyama, Kyoto University, Dr. Y. Kitamura, Chief Designer, M. Maeda, Chief Engineer, Kozosoken Corporation, H. Nagae, Researcher, Hyogo Earthquake Engineering Research Center, National Research Institute for Earth Science and Disaster Prevention, Dr. N. Yoshida, N, Research Engineer, Katsura Int’tech Center, Kyoto University.
We report our research projects on seismic rehabilitation of historic brick buildings, wherein un reinforced walls are strengthened by inserting stainless pins. In strengthening historic buildings, preservation of their appearance, durability of strengthening materials, and minimization of parts to be damaged pose main difficulties. Inserting stainless pins into un reinforced walls is known as an effective method to overcome the difficulties, and has been applied to some buildings in practice in Japan. In the method, stainless pins are inserted from the mortar joints between bricks in the direction of 45 degree from the vertical, while preserving the appearance of the walls. No research efforts, however, have been devoted that evaluate the improvement of the seismic performance by the method, taking its unique way of inserting stainless pins into account. Our research aims to evaluate and improve, if necessary, the seismic performance of the reinforced walls through a set of experimental studies. As the first step toward this goal, we have conducted cyclic out-of-plane bending tests. The main findings from the tests can be summarized as follows: (1) Significant differences exit between the restoring force characteristics in one and the other loading directions. The reason of this is that the reinforced wall is subject to tension of stainless pins in one direction, while it is subject to dowel action of the stainless pins in the other direction. (2) Too much increase of the number and/or strength of stainless pins decrease the capacity of the reinforced wall. The increase makes the difference greater between the reinforced and un reinforced parts of mortar joints .The unbalance of the strengths causes stepping pattern fracture at the mortar joints where no stainless pins go through. Once stepping pattern fracture takes place, no tension element exists in one loading direction, which leads to loss of capacity.
Introduction
There are many historic un reinforced masonry (URM) buildings in Japan. Most of such buildings were constructed from the late 18th to the early 19th centuries, when westernization became widespread in Japan. In recent years, the number of retrofitting such buildings is increasing for cultural and commercial uses. Seismic performance of this kind of buildings is, however, usually poor because they have URM walls. With frequent occurrence of near-fault earthquakes of M7 in recent years and prediction of occurrence of very strong earthquakes of M8, such as Tokai and Tonankai earthquakes, with high probability, the demand is increasing for seismic rehabilitation of such historic URM buildings in Japan.
In seismic rehabilitation of historic buildings, it is desirable to preserve their cultural values, especially their appearance. Many kinds of techniques are available for this purpose as shown in Table 1. Placement of new reinforced concrete (RC) walls and/or steel braces is a widely used technique because of its cost-effectiveness. Nonetheless, this method has limitations in strengthening historic buildings since the deterioration of the appearance is unavoidable. Although seismic isolation is effective to preserve the upper part of the existing building, it is often problematic from the view point of the change of outside appearance due to clearance necessary for seismic isolation. Also construction cost for the isolation system is much higher than the other methods. As such, there are still many problems to be resolved in seismic rehabilitation of historic URM buildings. Preservation of the appearance of existing buildings, durability of strengthening materials, and minimization of the parts to be damaged create main difficulties.
In Japan, inserting stainless pins into URM walls is expected as an effective method to overcome the difficulties mentioned above, and already applied to several historic brick buildings in practice. In this paper, we present the progress of our research projects on seismic rehabilitation of historic brick buildings, where in un reinforced walls are strengthened by inserting stainless pins.
In seismic rehabilitation of historic buildings, it is desirable to preserve their cultural values, especially their appearance. Many kinds of techniques are available for this purpose as shown in Table 1. Placement of new reinforced concrete (RC) walls and/or steel braces is a widely used technique because of its cost-effectiveness. Nonetheless, this method has limitations in strengthening historic buildings since the deterioration of the appearance is unavoidable. Although seismic isolation is effective to preserve the upper part of the existing building, it is often problematic from the view point of the change of outside appearance due to clearance necessary for seismic isolation. Also construction cost for the isolation system is much higher than the other methods. As such, there are still many problems to be resolved in seismic rehabilitation of historic URM buildings. Preservation of the appearance of existing buildings, durability of strengthening materials, and minimization of the parts to be damaged create main difficulties.
In Japan, inserting stainless pins into URM walls is expected as an effective method to overcome the difficulties mentioned above, and already applied to several historic brick buildings in practice. In this paper, we present the progress of our research projects on seismic rehabilitation of historic brick buildings, where in un reinforced walls are strengthened by inserting stainless pins.
Construction Method and Rehabilitated Buildings
This section explains the construction method for inserting stainless pins, and provides practical examples of seismic rehabilitation of historic brick buildings using the method. In this method, stainless pins are inserted from the mortar joints of the inside walls of the building in the direction of 45 degree from the vertical. The reasons for applying this unique way of inserting stainless pins include:
This technique was applied to at least 6 retrofit projects up to now in Japan. In those projects, historic brick warehouses were converted into 3 museums, 2 restaurants, and 1 public bathhouse. Through one of the projects, we had an opportunity of conducting out-of-plane flexural test of brick walls reinforced by inserting stainless pins. That was a retrofit project of a historic brick warehouse built in 1902 at Maizuru, a port city located about 60km north of Kyoto (1,2). The warehouse before the retrofit is shown in Figure 2.
For the warehouse having a rectangular plan of 10.5m x 70.2m, structural design of the seismic rehabilitation project was planned as follows:
(a) preservation of the appearance of the existing walls,Figure 1 illustrates the construction procedures for inserting stainless pins. Each figure shows the following procedure, respectively.
(b) ease of construction for drilling and inserting epoxy resin as a grout, and
(c) preventing water infiltration into the wall, which may occur if stainless pins are inserted from the outside of the building. Stainless steel was selected as a strengthening material because of its high durability.
(a) Mark the position of the inside wall at which stainless pins are inserted.This method, however, makes deficit parts of existing construction materials. Caution should be taken, therefore, when all or most of the existing construction materials must be preserved.
(b) Drill the holes for inserting stainless pins using an apparatus for fixing the angle.
(c) Clean the holes using air pressure.
(d) Insert epoxy resin as a grout.
(e) Insert stainless pins.
(f) After inserting stainless pins. As shown in Figure 1, no clear change is found before and after inserting stainless pins. This is a strong point of the technique.
This technique was applied to at least 6 retrofit projects up to now in Japan. In those projects, historic brick warehouses were converted into 3 museums, 2 restaurants, and 1 public bathhouse. Through one of the projects, we had an opportunity of conducting out-of-plane flexural test of brick walls reinforced by inserting stainless pins. That was a retrofit project of a historic brick warehouse built in 1902 at Maizuru, a port city located about 60km north of Kyoto (1,2). The warehouse before the retrofit is shown in Figure 2.
For the warehouse having a rectangular plan of 10.5m x 70.2m, structural design of the seismic rehabilitation project was planned as follows:
a) Horizontal force in the direction of shorter sizes is sustained by core RC frames newly incorporated into the building.Since the building has very long sides in one-direction, collapse due to in-plane shear along this direction was not a serious concern. On the other hand, collapse due to out-of-plane bending caused by seismic forces was considered critical in the structural design of the seismic rehabilitation project. No literature was, however, available that evaluates the capacity of the walls reinforced by inserting stainless pins taking its unique way of inserting pins into consideration. The initial design of reinforcement, therefore, depended entirely on the design formula for vertical and horizontal steel bars for new masonry buildings (3,4). This led us to conducting monotonic out of-plane flexural testing of the reinforced walls to quantify the improvement of their capacity and to clarify the collapse mode. Through this test, we confirmed that the reinforcement provides much improved ductility under monotonic loading.
b) Horizontal force in the direction of longer sizes is supported by both the core RC frames and the existing brick walls.
c) To resist out-of-plane bending, brick walls are reinforced by inserting stainless pins.
Cyclic Out-of-plane Bending Tests Background
Some of fundamental characteristics of the brick walls reinforced by inserting stainless pins were made clear through the out-of-plane bending test mentioned in the last section. Nevertheless, since the loading condition of the test was limited to monotonic loading. In this section, we present the results of the experimental research projects conducted for studying the out-of-plane flexural characteristics of the reinforced walls under cyclic loading.
Test Specimens
We conduct experiments for 5 brick wall specimens. The differences among the specimens are only the number and/or diameter of the stainless pins. Table 2 shows the classification of the reinforcement type and the loading conditions of the test specimens.
Figures 3 to 5 depict the schematic illustrations of the specimens. The specimens are comprised of 22 brick layers. Figure 3 shows the layout of the bricks at the even and odd number of layers. The height, width, and thickness of the specimens are 1,600mm, 1,420mm, and 430mm, respectively. The size of each brick is 210x100x60mm. The thickness of mortar joins is 10mm.
The main physical characteristics out of material tests are as follows. The flexural tensile strength of the mortal was 0.46N/ mm2. The tensile strengths of the stainless pins were 13.4 kN for f6mm, and 25.9 kN for f8mm.
The black circles in the front elevation of Figs. 4 and 5 indicate the locations at which stainless pins are inserted. The solid line in the side elevation of Figure. 4 indicates that 4 stainless pins are inserted on the line. And the dotted line in the same Figure indicates that 3 pins are inserted on the line. This pattern of inserting stainless pins has been used in practice and called normal in Table 2. The solid line in Figure. 5 shows that 7 stainless pins are inserted at the line. This pattern of inserting stainless pins is called double in Table 2.
Figures 3 to 5 depict the schematic illustrations of the specimens. The specimens are comprised of 22 brick layers. Figure 3 shows the layout of the bricks at the even and odd number of layers. The height, width, and thickness of the specimens are 1,600mm, 1,420mm, and 430mm, respectively. The size of each brick is 210x100x60mm. The thickness of mortar joins is 10mm.
The main physical characteristics out of material tests are as follows. The flexural tensile strength of the mortal was 0.46N/ mm2. The tensile strengths of the stainless pins were 13.4 kN for f6mm, and 25.9 kN for f8mm.
The black circles in the front elevation of Figs. 4 and 5 indicate the locations at which stainless pins are inserted. The solid line in the side elevation of Figure. 4 indicates that 4 stainless pins are inserted on the line. And the dotted line in the same Figure indicates that 3 pins are inserted on the line. This pattern of inserting stainless pins has been used in practice and called normal in Table 2. The solid line in Figure. 5 shows that 7 stainless pins are inserted at the line. This pattern of inserting stainless pins is called double in Table 2.
Loading Conditions
Loading conditions are illustrated in Figure. 6. The base of the brickwall specimen is located on strong steel beams and fixed by the wide-flange shape and equal-legangle members. Two horizontal hydraulic jacks are connected to the top of the wall specimen. At the top, two wide-flange shape members are attached to the specimen using PC steel bars. Grout or hard rubber was attached between the brick specimen and the steel members. Monotonic loading was applied to un reinforced specimen No. 1 to examine its maximum strength.
Cyclic loading was applied to the reinforced specimen Nos. 2 to 5. The cyclic loading was conducted by increasing the amplitude of the rotation angle Ru/L. Here, u is the horizontal displacement of point A in Figure. 6, and L is the distance between point A and the base of the measurement frame as shown in Figure. 6. The amplitude of R was increased as follows: 0.0025, 0.005, 0.01, 0.0015, 0.02, 0.03, 0. 05 (radian).
Cyclic loading was applied to the reinforced specimen Nos. 2 to 5. The cyclic loading was conducted by increasing the amplitude of the rotation angle Ru/L. Here, u is the horizontal displacement of point A in Figure. 6, and L is the distance between point A and the base of the measurement frame as shown in Figure. 6. The amplitude of R was increased as follows: 0.0025, 0.005, 0.01, 0.0015, 0.02, 0.03, 0. 05 (radian).
Test Results
Figures 7 to 9 illustrate the test results. Specimen No. 1 recorded the maximum strength of 32.2kN. After reaching the maximum strength, brittle failure occurred at the mortar joint right above the pair of the steel members used to fix the base. The failure pattern of mortar joints of specimen No. 2 is depicted in Figure 7(a). Figure 7(b) shows the photograph of the fracture line of mortar joints from the side elevation, where yellow dotted line highlights the mortar fracture. The restoring force characteristics of specimen No. 2 are shown in Figure 7(c). Figure 8 shows the same illustrations and photographs for specimen No. 4 as those in Figure 7. Figure 9 summarizes the force-deformation hysteretic characteristics of all the specimens. The bar graph indicates the residual strengths at the load reversals for specimen Nos. 2 to 5, and the dotted line indicates the maximum strength of specimen No.1.
Observations from the test results can be summarized in what follows.
Observations from the test results can be summarized in what follows.
- As indicated in Figures 7(c), 8(c), and 9, all the reinforced specimens have the larger
- Residual strength on the deteriorating skeleton than the un reinforced specimen.
- The skeletons for one and the other loading directions are significantly different.
- Small energy dissipation is observed at the negative direction loading of specimen No. 4. Similar results were obtained for specimen Nos. 3 and 5, although the hystereses of the specimens are not shown in the Figures.
Discussions
Major findings from the tests and discussions on them are summarized as follows:
a) Significant differences exit between the restoring force characteristics in the positive and negative loading directions. As shown in Fig. 10(a), the joints of the brick wall are efficiently constrained by tension of stainless pins in the duration of positive loading. However, in the duration of negative loading, the stainless pins constrain brick walls not by tension but by dowel action. Thus, the constraint of stainless pins does not work well in the direction of the negative loading.
b) Increasing the number and/or strength of stainless pins can decrease the capacity of the reinforced wall. That is to say, unbalanced reinforcement makes the difference greater between the strengths of the reinforced and un reinforced parts of mortar joints. As shown in Figure 10(b), this causes stepping pattern fracture in the negative loading direction through the mortar joints where no stainless pins are inserted. The stepping pattern fracture can be observed in Figure 8(b). Once stepping pattern fracture takes place, no tension element exists in the joints in the duration of negative loading. Thus, the reinforced wall loses its strength capacity.
a) Significant differences exit between the restoring force characteristics in the positive and negative loading directions. As shown in Fig. 10(a), the joints of the brick wall are efficiently constrained by tension of stainless pins in the duration of positive loading. However, in the duration of negative loading, the stainless pins constrain brick walls not by tension but by dowel action. Thus, the constraint of stainless pins does not work well in the direction of the negative loading.
b) Increasing the number and/or strength of stainless pins can decrease the capacity of the reinforced wall. That is to say, unbalanced reinforcement makes the difference greater between the strengths of the reinforced and un reinforced parts of mortar joints. As shown in Figure 10(b), this causes stepping pattern fracture in the negative loading direction through the mortar joints where no stainless pins are inserted. The stepping pattern fracture can be observed in Figure 8(b). Once stepping pattern fracture takes place, no tension element exists in the joints in the duration of negative loading. Thus, the reinforced wall loses its strength capacity.
Conclusion
We have reported the progress of our research project on seismic rehabilitation of historic brick buildings, which adopts the method of inserting stainless pins into the un reinforced walls. Future topics of the research project include:
a) Study of the locations, strengths, and directions of inserting stainless pins that can prevent the stepping pattern fracture of mortar joints.
b) Examination of in-plane shear characteristics, which are also essential in evaluating the seismic performance of the brick-wall system.
c) Minimization of the damaged area of existing members by the use of new type of stainless steel with very high strength and ductility, which has been developed recently in Japan.
Acknowledgment
This research was founded by the Ministry of Land, Infrastructure and Transport of Japan. We gratefully acknowledge the support.
References
- Y. Araki and N. Yoshida, “Flexural tests of historical brick walls reinforced by stainless-pin insertion,” Proceedings of the 6th International Symposium on Architectural Interchanges in Asia, Vol. II, pp. 1081-1084, Daegu, Korea, October, 2006.
- Y. Araki and Y. Nobutoshi, “Out-of- Plane Flexural Strength of Historic Brick Walls under Monotonic Loading Reinforced by Inserting Stainless Pins,” AIJ J. Technol. Des., No. 25, pp. 147-152, 2007.6 (in Japanese).
- R.G. Drysdale, A.A. Hamid, and L.R.Baker, Masonry Structures, Behavior and Design, Prentice Hall, NJ, 1993.
- J.E. Amrhein, Reinforced Masonry Engineering Handbook: Clay and Concrete Masonry, CRC Press, FL, 1998.
NBM&CW May 2008