A Study on Seismic Performance and Seismic Diagnosis, Seismic Retrofit of Japanese Temple

13th World Conference on Earthquake Engineering Vancouver, B.C., Canada August 1-6, 2004 Paper No. 853

A STUDY ON SEISMIC PERFORMANCE AND SEISMIC DIAGNOSIS, SEISMIC RETROFIT OF JAPANESE TEMPLE

Shunsuke MIYAMOTO1, Kenji MIYAZAWA2, Yasutaka IRIE3, Juwei WU4, Yoshinori NOMATA5, Osamu GOTO6

SUMMARY

Structural style of Japanese temple building is peculiar and different from a wooden dwelling house. This research aims to clarify seismic performance of Japanese temple building by field investigation, seismic diagnosis and seismic retrofit. The actual seismic retrofit construction of the existing temple is investigated for this aim, and the effect of seismic retrofit are confirmed from the vibration measurement after seismic retrofit. This result, roof is large and occupies the great portion of weight. The measured natural period of Japanese temple building was about 2[Hz]. It is long period as compared with a wooden construction house, because it is said that the natural period of wooden dwelling houses is 5-6[Hz]. In order to confirm the effect of seismic retrofit, we carried out the microtremor measurements before and after the seismic retrofit. The main results are following; the natural frequency of this temple has increased from 2.0[Hz] to 3.6[Hz]. The damping constant has decreased from 3.5[%] to 2.7[%]. Putting in diagonal steel braces in the horizontal ceiling structure, the effect in reduction of the torsional deformation was confirmed.

INTRODUCTION

Japanese temple building is timber structure. It has long history. Structural style of Japanese temple building is peculiar. And it has experienced many earthquakes. This research clarifies seismic performance of Japanese temple building which exists really by field investigation, seismic diagnosis and seismic retrofit. The investigated temple building is Rensho-ji which is really existing at Fujieda-city in Shizuoka- prefecture. And the vibration characteristic of the temple building comprehends by microtremor measurement. Moreover, it shows the actual example of seismic retrofit construction, and we propose effective seismic retrofit in earthquake load.

1 Graduate School of Eng., Kogakuin Univ., Ms. Eng. 2 Prof., Department of Architecture, Kogakuin Univ., Dr. Eng. 3 Assoc. Prof., Department of Information and Control Systems Science, Graduate School of Eng., Utsunomiya Univ., Dr. Eng. 4 Fujitsu FIP Corporation, Ms. Eng. 5 Technical Official, Department of Eng., Utsunomiya Univ. 6 Assoc. Prof., Department of Design in Architecture, Kogakuin Univ., Dr. Eng. STRUCTURAL OF JAPANESE TENPLE BUILDING

Structural style Rensho-ji of target temple building is in the southern part of Fujieda-city in Shizuoka- prefecture. And it is the Jodo-Shinshu-Otani group's temple building. The Rensho-ji Hondo as of now will be built in 1913. A panorama of temple building is shown in Photo.1 and a ground plan is shown in Fig. 1. Japanese temples building such as Rensho-ji are architectural structure with the very large weight of a roof truss. Therefore, roof weight occupies most gross weight, and it is influenced seismic performance greatly. Moreover, since Japanese temple building has little bearing wall to earthquake load, it is thought that seismic performance is low. However, Japanese temple building has seismic elements, such as frame and rail, wall upper than picture rail.

1970 2970 1000

1805 2707. 5 5415 1805 Adytum 2707. 5 1805 1540 3080

20375 1540

20375[mm] 1910 Main room 5730 1910

1910

3180 19730[mm] 1470 1470 1950 1950 1940 2170 1940 1950 1950 1470 1470 3900 3900 2940 13850 2940 19730

Photo.1 Panorama of Rensho-ji Hondo Fig.1 Ground plan of Rensho-ji Hondo

Weight The gross weight of Japanese temple building is greatly influenced by the kind of roof. Some typical roof classification1) of Japanese temple building and roof weight of the temple building which exists really is shown in Table.1. Weight of roof tile roofing is heavy in roof classification, and Hongawara roofing is as the heaviest as 250 [kgf/m2] also in some roof classification. Subsequently, Pantaile roofing is as heavy as 120 [kgf/m2]. Although the roof tile of the temple building is heavy, clay roofing which superimpose on a roof tile also occupies 30 - 40 percent of roof weight. Fig.2 shows the roof area and unit roof weight of roofing. Although clay roofing of Rensho-ji was also heavy, clay roofing was carried away by seismic retrofit construction. As a result, roof weight of before seismic retrofit was shown 1.7 times of roof weight of after seismic retrofit. Table.1 Roof weight of the temple building 0.30 Unit roof weight and Roof Area of Temple № Name Construction data The kind of roof Roof Area Roof Weight Unit roof weight ② ① ③ ④ [year] [m2][ｋN] [kN/m2] 0.25 ] ① Jizoubu-ji Hondo ～1513 Hongawara roofing 192 48 0.250 2 Rensho-ji ② Joutok-ji Entuden 1401 Hongawara roofing 151 38 0.250 (Befor Seismic Retrofit) ⑯ ③ Kyutomyo-ji Hondo 1393～1453 Hongawara roofing 519 130 0.250 0.20 ④ Kansin-ji Kondo 1346～1370 Hongawara roofing 790 197 0.250 Rensho-ji ⑤ Joshin-ji Hondo 1243～1247 Hongawara roofing 452 43 0.094 (After Seismic Retrofit) Hongawara roofing 0.15 ⑥ Chikurin-ji Hondo 1511 Wood shingle roofing 274 5 0.019 Wood shingle roofing Joshin-ji ⑦ Shokai-ji Hondo 1648～1652 Wood shingle roofing 230 8 0.033 need roofing (After Seismic Retrofit) ⑯ ⑧ Kozo-ji Hondo 1591 Wood shingle roofing 158 3 0.019 0.10 ⑪ roofing of cypress bark shingles ⑨ Horin-ji Hondo 1667 Wood shingle roofing 570 14 0.024 ⑤ ⑬ Pantile roofing ⑩ Kiyonizu-dera Hondo 1390～1394 Wood shingle roofing 914 23 0.025 ⑫ ⑪ ～ Unit roof weight [kN/m 0.05 ⑦ ⑭ Senpuku-ji Yakushido 1513 1573 need roofing 137 12 0.085 ⑨ ⑫ Tenjuin 1651 need roofing 140 10 0.072 ⑮ ⑧ ⑥ ⑩ ⑬ Shoren-ji Amidado 1542 need roofing 372 25 0.067 0.00 ⑭ Kuwanomi-dera Hondo 1393～1453 roofing of cypress bark shingles 472 18 0.039 ⑮ Kongosho-ji Hondo 1610 roofing of cypress bark shingles 661 22 0.033 0 200 400 600 800 1000 1913 600 124 0.207 2 ⑯ Renshoji Hondo Pantile roofing Roof Area [m ] 2003 600 72 0.120 Fig.2 Roof area and unit roof weight

Vibration characteristic Also structural and building construction Japanese temple building differ from a wooden house. Therefore, vibration characteristic of temple building, traditional wooden house and wooden house is considered by microtremor measurement. The result of microtremor measurement has draw anamnestic research2-29). Fig.3 shows natural frequency of ridge direction and span direction. Fig.4 shows damping factor of ridge direction and span direction. Natural frequency of Japanese temple building is 1.5[Hz] to 2.7[Hz]. Natural frequency is small in order of temple building- traditional wooden house -wooden house. Damping factor of a temple building is concentrated on 4[%] from 3[%]. And damping factor of a wooden house is concentrated on 2[%] from 1.5[%]. Therefore, deformation performance of temple building can say that it is high. Fig5 shows natural frequency and damping factor of each building. In ridge direction of the temple building, if natural frequency becomes large, damping factor will become small. In span direction of temple building, if natural frequency becomes large, damping factor will become large. However, vibration characteristic of the traditional wooden house showed the characteristic contrary to temple building. Moreover, natural frequency and damping factor of wooden house has an inverse proportion relation of the temple building.

10 Natural frequency 5 Damping factor

20% 10% 0% 10% 20% 10% 0% 10% 8 20% 4 20%

6 3

4 2 Damping factor

Natural frequency Temple building of span direction [%]

of span directionof span [Hz] Temple building 2 Traditional wooden house 1 Wooden house Traditinal wooden house Wooden house 0 0 0246810 012345 Natural frequency of ridge direction [Hz] Damping factor of ridge direction [%] Fig.3 Natural frequency Fig.4 Damping factor

5 Natural frequency - Damping factor (Wooden house) 6 Natural frequency - Damping factor (Temple building) 10 Natural frequency - Damping factor (Traditional wooden house) Span direction(Single Moode) 2 4 R = 0.0972 8 Ridge direction(Single Moode) 2 Span direction(Single Moode) R = 0.0168 4 RENSHOU-ji-temple 3 2 Ridge direction(Single Moode) (Befor S eis mic Retro fit) R = 0.0317 6

2 2 4 R = 0.0297 RENSHOU-ji-temple

2 (After Seismic Retrofit) [%] factor Damping Damping factor [%] factor Damping Damping factor [%] factor Damping 1 2 2 R = 0.4478 Span direction(Single Moode) 2 R = 0.0197 Ridge direction(Single Moode) 0 0 0 0246810 02460246810 Natural frequency [Hz] Natural frequency [Hz] Natural frequency [Hz]

Japanese temple building Traditional wooden house Wooden dwelling house

Fig.5 Natural frequency and damping factor of each building

SEISMIC DIAGNOSIS

Outline of seismic diagnosis The method of seismic performance evaluation of the wooden building generally used is dependent on wooden brace or bearing wall. This method is laid emphasis on strength. Therefore, it is thought that this method of seismic performance evaluation cannot evaluate appropriately seismic performance of temple building which hardly has wooden brace or bearing wall. However, temple building also has many earthquake resisting elements which resist shear force, such as mud wall, compressive strain inclined to the grain by beam-column(Nuki) joint, tenacity of bracket complex(Kumimono) and restoring force to column rocking produced by a thick column and roof weight. Therefore, in order to evaluate rationally a temple building with the earthquake resisting element which cannot be evaluated only by wall length ratio, it is necessary to decide the deformation domain of a wooden building and to evaluate the response in case of an earthquake quantitatively. This diagnostic method evaluates the limit state where damage as the story drift 1/120[rad.] of a building, and it evaluates shear force of building. Moreover, it reduces strength in order to take eccentricity of a building into consideration. Evaluation of eccentricity not only takes balance of alignment of a wall into consideration, but also can evaluate that it is an irregular shape in three dimensions.

Earthquake resisting element The evaluation method of mud wall has referred QW:Shear Force Hysteresis characteristics of mud wall 30), 31), 32) by anamnestic research anamnestic research . Fig.6 shows structural model t0･t･l of mud wall. The evaluation method of column with 0.83t 0･t･l Kokabe frame has adopted the one where the shear Load 0.75t 0･t･l strength of story drift 120[rad.] and bending strength of t0:Shearstrength column is smaller. The evaluation method of mud wall is of mud wall Mud Wall h 0.5t 0･t･l the same as that of the evaluation method shown in Fig.6. t:Thickness of wall The column was calculated by assumption of cantilever l:Length of wall beam. This outline is shown in Fig.7. Moreover, the h : Higth of mud wall l evaluation method of beam-column(Nuki) joint, bracket

dW:Displacement complex(Kumimono) and restoring force to column 0 rocking has draw reference33), 34), 35), 36). The each h/250 h/120h/90 h/60 h/15 earthquake resisting elements is shown Fig.8, Fig.9, and Fig.6 Structural model of mud wall Fig.10.

QSW:Shear Force Q:Shear Force Hysteresis characteristics model Hysteresis characteristics Hysteresis characteristics model of mud wall portion model of column of mud wall portion QWy 0.83t 0･t･l Hysteresis characteristics model of Kokabe QCy Hysteresis characteristics model of column QSWy QCy ＝ ( 、 ) 0.75t 0･t･l QSWy Mi n QWy QCy δSW＝δW＋δCa

l δSW: Displacement of Kokabe t0: Shear strength δW: Displacement of mud wall of mud wall portion l 0.5t 0･t･l Mud wall δCa: Displacement of column t:Thickness of Kokabe portion QSWa:Shear strength of QCa l:Length of Kokabe story drift 1/120[rad.] h : Higth of Kokabe QSWa h h ＋＝ l : Length of Kokabe Column h1 h : Higth of Kokabe h1

h1: Door height

δ δ δSW δ W Ca :Displacement δ δSW:Displacement 0 0 SWa

h/250 h/120DCy h/90 h/15 h/120 δSWy h/15

Fig.7 Structural model of column with Kokabe

Equation (1) shows the calculation method of Kokabe. 2 δ ＋δ ＝δ G: Shear modulus [N/mm ] W C E: Young modulus of a column [N/mm2] P ･ h P ･ h 3 (1) ＋ 1 ＝δ l : Length of wall to pay [mm] G ･ l･ t 3･ E ･ I fb: Allowable bending stress [mm] t : Thickness of mud wall [mm] Equation (2) shows shear force of a column determined by deformation. 3･E･I･G･l･t P＝ (2) ･･･ 3＋･･･ G l t h1 3 E I h Equation (3) shows shear force determined by bending strength of a column ･＝ fb Z Pcr (3) h1

CQa: Allowable shear force of a column CQa=Min (P, Pcr)

The calculation method of beam-column(Nuki) joint shows below.

Mjn:Moment Rotational rigidity: Kθ[N･cm/rad.]

Hysteresis characteristics model 2 x 4Z 1 Kθ＝x ･y･E { (1＋－ )} (4) of beam-cloumn(Nuki) joint by anamnestic research C Z 3x 3 Mjny X X Beam Column Beam Allowable moment: Mjna[N･cm] Mjna Zy1 Beam Mjna ＝･ M jna Kθ (5) h Compressive strain 120 Nuki joint Column inclined to the grain XX Column Elevation surface Planar surface θjn:Rotational angle 0 K? θ h/120 jny h/15 Fig.8 Structural model of beam-column(Nuki) joint

The calculation method of bracket complex(Kumimono) and restoring force to column rocking shows below.

H:Horizontal force Load K1:The bottom of Daito and To bearing stiffness※ K2:0(the slip with friction)※ Hysteresis characteristics model of K3:Dabo of Daito and To bearing stiffness ※ ※Refarence:34) restoring force to column rocking(Kumimono)

H0 by anamnestic research K2 μN K3 H0＝P･B/h 0.8H0

0.75H0 K1 P δ Displaceme H0 0 1/120[rad]

0.5H0 h θ clearance of member B δ :Displacement 0.02B 0.06B0.1B 0.2B Fig.9 The restoring force to column rocking Fig.10 The bracket complex(Kumimono)

Result of seismic diagnosis We calculated the strength of span and ridge direction at the time of story drift (1/120[rad.]) by the evaluation method of each earthquake resisting elements. The story shear force of span direction is 305[kN], and ridge direction is 93[kN]. However, when eccentricity of the building by arrangement of a wall or a column was taken into consideration, the story shear force of only the span direction showed 222[kN]. Table.2 shows the seismic diagnosis result before seismic retrofit. Table.3 shows calculation of earthquake load before seismic retrofit. The gross weight before seismic retrofit is 1198[kN], and the earthquake load is 287[kN]. Since the strength of each direction has not arrived at safety margin to earthquake load, it is evaluated that there is danger of collapse. And this building needs to reconsider seismic performance. Fig.4 shows calculation of necessary strength. Table.2 Seismic diagnosis result (Before seismic retrofit) Strength in Horizontal Force Reduction Earthquake Eccentricity consideration of Retrofit direction ratio by load Strength / Earthquake Adjudication burden of all column burden of all walls Sum total ratio Eccentricity Eccentricity [kN] [kN] [kN] [kN] [kN] Span(X) direction 160 145 305 0.26 0.73 222 287 0.77 NG Ridge(Y) direction 93 0 93 －－－ 287 0.32 NG

Table.3 Calculation of earthquake load (Before seismic retrofit)

Calculation of Unit weight Roof Area Weight Shera coefficient（※1） Earthquake load Kind of load earthquake load [kN/㎡][㎡][kN]Ci[kN] RoofBy results of an investigation 706 169

Fixed load Roof truss（※2） 0.83 306 253 61 0.24 fittings（※2） 0.78 306 239 57 Sum total 1198 287 (※1)The area coefficient which considered the possibility of the Tokai earthquake by specification of Shizuoka Prefecture=1.2, Ci＝Co×Z＝0.2×1.2＝0.24 (※2)References37) is referred to. Table.4 Calculation of necessary strength (Before seismic retrofit) Earthquake load Strength Necessary Strength Retorfit wall length Target Strength Retrofit direction [kN] [kN] [kN] [m] [kN/m] Span(X) direction 287.47 222.46 65.01 16.50 915 Ridge(Y) direction 287.47 93.10 194.37 9.40 3031

SEISMIC DIAGNOSIS AFTER WEIGHT SAVING AND SEISMIC RETROFIT PLANNING

Fig.5 shows result of seismic diagnosis after weight saving of roof, and Fig.6 shows calculation of earthquake load after weight saving of roof. Before seismic retrofit, shortage of bearing wall was conspicuous in span(X) direction and ridge(Y) direction owing to the weight of roof load. However, the gross weight decreased to 878[kN] by weight saving of roof, and earthquake load also became small with 211[kN]. Shortage of bearing wall was conspicuous only in the ridge(Y) direction with weight saving of a roof. Since the necessary strength of before seismic retrofit is 65[kN], it was thought that it was necessary to establish more 16.5[m] bearing wall of 4 times the magnification. However, the wall of span(X) direction brought a result with which safety is filled by weight saving of roof. The seismic retrofit method, which raises seismic performance from result of seismic diagnosis, is proposed. 1) Weight saving of roof, and extension of a ceiling brace. The intendment fortify of horizontal plane of structure. And abatement of weight decreases the imposition of horizontal load. 2) Extension and earthquake retrofit of the bearing wall The strength of a building is made to increase. 3) Fixation of a column base By shear capacity of a wall is strengthened, the tensile force of a column base becomes strong. Table.5 Seismic diagnosis result (Weight saving of roof) Strength in Horizontal Force Reduction Earthquake Eccentricity consideration of Retrofit direction ratio by load Strength / Earthquake Adjudication burden of all column burden of all walls Sum total ratio Eccentricity Eccentricity [kN] [kN] [kN] [kN] [kN] Span(X) direction 160 145 305 0.26 0.73 222 212 1.05 OK Ridge(Y) direction 93 0 93 －－－ 212 0.44 NG

Table.6 Calculation of earthquake load (Weight saving of roof)

Calculation of Unit weight Roof Area Weight Shera coefficient（※1） Earthquake load Kind of load earthquake load [kN/㎡][㎡][kN]Ci[kN] RoofBy results of an investigation 386 93

Fixed load Roof truss（※2） 0.83 306 253 61 0.24 fittings（※2） 0.78 306 239 57 Sum total 878 211 (※1)The area coefficient which considered the possibility of the Tokai earthquake by specification of Shizuoka Prefecture=1.2, Ci＝Co×Z＝0.2×1.2＝0.24 (※2)References37) is referred to.

Table.7 Calculation of necessary strength (Weight saving of roof) Earthquake load Strength Necessary Strength Retorfit wall length Target Strength Retrofit direction [kN] [kN] [kN] [m] [kN/m] Span(X) direction 212 222 Strength＞Earthquake load Ridge(Y) direction 212 93 119 9.4 21.47

ACTUAL EXAMPLE OF SEISMIC RETROFIT

Repair work of existing mud wall and existing column with Kokabe Fig.11 and Fig.12 show earthquake retrofit position of existing mud wall and existing column with Kokabe. Photo.2 shows condition of earthquake retrofit. Since the defect junction with a boundary frame was seen on the whole by deterioration of existing mud wall and existing column with Kokabe used as earthquake resisting element, we thought that sufficient seismic performance was not demonstrated. Therefore, a mud wall is removed, and it is made replacement the structural plywood of thickness: 12[mm].

19,730[mm] 19,730[mm] 1,470 1,470 1,9509,950 1,950 1,470 1,470 1,470 1,470 1,9509,950 1,950 1,470 1,470

L L

wX3wX3' wX3' wX4 wX3' wX3' wX3 1,970

WX5 WX5' WX5'WX6' WX6' WX6 1,970 K K wY5 1,000 Seismic retrofit wall J 1,000 J Number Method of seismic retrofit 2,707.5

2,707.5 I I Kokabe frame H' wX1 H' (plywood-single sides) H Kokabe frame H wY4 FL+240 wX1' FL+240 (plywood-single sides)

G wX2 Kokabe frame G [mm] (plywood-single sides) [mm]

F wY3 xW3 Kokabe frame

F Seismic retrofit wall 20,375 (plywood-single sides) 9,607.5 WX7 WX8 20,375 9,607.5 E Kokabe frame E Number Method of seismic retrofit wX3' (plywood-single sides)

wY2 wY4 FL FL Hanging wall + Frame Kokabe frame D WX5 (plywood-single sides) D wX4 (plywood-single sides)

Hanging wall + Frame wY2 wY2 wY3 wY4 wY4 wY5 Kokabe frame WX5' wY1 C (plywood-single sides) C (plywood-single sides) Hanging wall + Frame Kokabe frame wY1

WX6 wY1 wY2 (plywood-single sides) wX1wX1' wX1'wX2 wX1' wX1' wX1 1,910 wY2 (plywood-single sides) 1,910 B B Hanging wall + Frame wY3 Kokabe frame WX6' FL-180 (plywood-single sides) FL-180 (plywood-single sides)

Hanging wall + Frame 3,180 Kokabe frame 3,180 WX7 wY4 (plywood-single sides) (plywood-single sides) A A Hanging wall + Frame Kokabe frame WX8 (plywood-single sides) wY5 (plywood-single sides)

132 4 576 8 9 10 1112 132 4 5 6 7108 9 11 12 Fig.11 Earthquake retrofit alignment Fig.12 Earthquake retrofit alignment of existing mud wall of existing column with Kokabe

Expansion of bearing wall In order to compensate the absence of shear capacity to earthquake load, six bearing walls were extended in north and south direction and, five hanging walls were extended in east and west direction. The extended bearing wall constructed a stud and a bracket of plywood in lattice-shaped, and jointed the structural plywood of thickness: 12[mm] to both sides at the interval of 8-9 [cm] with nail(CN50). Moreover, the extension bearing wall prepared the wooden sill of sectional area: 72×72[mm] newly, and joined it on RC underground beam with the anchor bolt. And a metallic material (Hold-down) is attached in the column base, in order to forfend pull-out of the column base. Fig.13 shows alignment of extended bearing wall. Fig.14 shows detail drawing of extended bearing wall.

Existing beam:140× 240 19,730[mm] Joint metallic material of seismic retrofit 1,470 1,470 1,9509,950 1,950 1,470 1,470 number Name of joint metallic material Corbel:33 × 72 X1 , X1' HD-B10 Stud:72 × 72 X2 , X2' HD-B20 Y1 L Y1' X3 , X3' HD-B20 X1' WY1

X1 WY6 K 1,970 X4 , X4' HD-B20 and HD-B20 Y2 Y2' X5,Y6 , X5',Y6' HD-B20 and HD-B20 1,000 [mm] Lintel:60 × 180

J Y3 Y3' X6 , X6' HD-B20 WY3 WY5 Y1 , Y1' HD-B20 and HD-B20 2,707.5 I 4,665 Y2 , Y2' HD-B20 and HD-B20 Corbel:33× 72

H' Y4 Y4' H Y3 , Y3' HD-B20 and HD-B10 FL+240 Y4 , Y4' HD-B15 G Y5 , Y5' HD-B20 and HD-B20 HD [mm]

F Seismic retrofit wall

WX1 WX2 20,375 9,607.5 Sleeper:120 × 190 E Number Method of seismic retrofit X2 X3 X4 X4' X3' X2' FL WX1 Bearing wall(plywood-both sides) New sill:72 × 72 Upper bed of cornerstone D WX2 Bearing wall(plywood-both sides) Y5 Y5' C WX3 Bearing wall(plywood-both sides) 1,980 [mm] WX4 Bearing wall(plywood-both sides) WY4 1,910 WY2 WX3 WX4 B Y6 Y6' X5 X6 FL-180 X6' X5' WY1 Bearing wall(plywood-both sides) WY2 Bearing wall(plywood-both sides) Common element 3,180 A WY3 Bearing wall(plywood-both sides) The structual plywood is stretched to both sides using a Douglas fir.(Thickness of 12mm) WY4 Bearing wall(plywood-both sides) The structual plywood is joined with nail of CN50.(@8-9cm) WY5 Bearing wall(plywood-both sides) Anchor of a lower hole down fixed to RC base. WY6 Bearing wall(plywood-both sides) Corbel is binded with a coach bolt. 13211124 5 6 7108 9 The junction part of components, such as brace, a column, and Kamoi, use Z mark ironmongery goods. (or equivalent of Z mark ironmongery goods)

Fig.13 Alignment of extended bearing wall Fig.14 Detail drawing of extended bearing wall

Repair work of footing The column base of the existing mud wall of east and west direction at northern end of the temple building and bearing wall newly extended in a circumference of the temple building is on upheaped earth retaining with the stone. It has the danger of crash by an earthquake load. In order to utilize bearing wall effectively, seismic retrofit of a stone is needed. The method of seismic retrofit undertakes construction reverse of L-type retaining wall of RC at circumference of upheaped stone. Moreover, the footing and a column base are made to connect. Fig.15 shows alignment of repair work of footing. Photo.3 shows condition of repair work of footing.

Seismic retrofit of horizontal plane of structure in roof truss The setting of horizontal brace utilize existing mud wall and extended bearing wall effectively as opposed to an earthquake, and it makes horizontal rigidity high. We installed a steel joint plate(PL-6) which carried out a bolt(M16) fasten in a capital of roof truss, and it connected columns and neighboring beams of 30 places by plain bar(M16). Fig.16 shows alignment of horizontal brace. Photo.4 shows condition of setting.

Photo.2 Column with Kokabe Photo.3 Repair work of footing Photo.4 Setting of horizontal 19,730[mm] 19,730[mm] 1,470 1,470 1,950 9,950 1,950 1,470 1,470 1,470 1,470 1,9509,950 1,950 1,470 1,470

L L Level brace 12φ (with a turn buckle) K 1,970 K 1,970 1,000 J 1,000 J

number Size of a column 2,707.5 2,707.5 I I × H' H' C9C8 C8 C9 C2 205mm 205mm H H C7 360mm in diameter FL+240 C8 260mm in diameter

C9 C8 C8 C9 C9 230mm in diameter G 5 G [mm]

G [mm] Ｆ F mark The construction method F 5

G 20,375 20,375 Number Explanatory notes

Ｆ 9,607.5

9,607.5 Beam joint E E C2C7 C7 C2 Cornerstone + Column Column joint FL Column joint D Cornerstone + Column D The level brace of the mark carries out construction which hangs the Cornerstone + floor post intersection of a brace from a hut C C beam by iron wire φ 3.2. 1,910 1,910 Base The construction method B B A level brace is set to 16φ with a turn buckle. FL-180 Basic beam and ground stop range A turn buckle position is set to less than 500mm from a column and a beam joint. 3,180 3,180 Outer wall extension range Intensity of brace steel materials and a coach bolt A A is the 400[N/mm2] class. 132 4 5 6 7108 9 11 12 132 4 5 6 7108 9 11 12 Fig.15 Alignment of repair work of footing Fig.16 Alignment of horizontal brace

Repair work of thatching The ratio of the roof weight occupied in earthquake load is large. For the purpose of mitigation of roof weight, this seismic retrofit also carried out conversion of the thatching of the whole roof. The situation is shown in Photo.5. The roof was deconstructed from the ridge. Work was removed in order of plain roof tile, cedar bark roofing, and sheathing roof board. And the new sheathing roof board was affixed densely simultaneously. Roofing, pantile, plain roof tile and the ridge were constructed. By this construction, roof weight was able to be decreased compared with seismic retrofit before. It shows below. 1) The clay roofing of the whole roof was removed and the tile was transposed to the light tile. 2) The ridge part was made smaller than seismic retrofit before, and was made light. 3) Onigawara was made smaller than C before, and was made light. Photo.6 shows panorama of the temple building(Rensho-ji) after seismic retrofit.

Photo.5 Repair work of thatching Photo.6 Rensho-ji after seismic retrofit

CONFIRMATION OF THE EFFECT OF SEISMIC RETROFIT BASED ON MICROTREMOR MEASUREMENT

In order to confirm the effect of seismic retrofit mentioned above, we carried out the microtremor measurements before and after this seismic retrofit.

Outline of microtremor measurement The sensing instruments for microtremors are one EW direction

component moving-coil velocity type seismometer 6.7m 5.8m 6.7m N 5.95m 7.5m having a natural period of 1.0[sec]., damping constant 3.4m of 0.7 sensitivity of 3[volts/kine]. All of the data are 4m recorded by digital form at sampling frequency of NS direction 2nd floor

100[Hz]. Two types of seismometer locations are Vertical comp. 8.8m shown in Fig.17. Horizontal comp. 1st floor

Fig.17 Two types of seismometer locations Natural frequency As an example, microtremors and its Fourier spectra in micron micron 5 5 NS component and EW component are shown in Fig.18, Center(before Seismic Retrofit ） Center(before Seismic Retrofit ） 0 0 Fig.19 and Fig.20, respectively. From Fig.19, we can 0 2 4 6 8 sec.10 0 2 4 6 8 sec.10 -55 -55 recognize a natural frequencies around 2.0[Hz] in both West end(before Seismic Retrofit ）South end(before Seismic Retrofit ） 0 0 sides of component and a torsional frequency in 0 2 4 6 8 sec.10 0 2 4 6 8 sec.10 -55 -55 Center(after Seismic Retrofit ） Center(after Seismic Retrofit ） 2.9[Hz]. Comparing the Fig.3 and Fig.4, we can 0 0 0 2 4 6 8 sec.10 0 2 4 6 8 sec.10 recognize an increase of natural frequency from 2.0[Hz] -55 -55 West end(after Seismic Retrofit ） South end(after Seismic Retrofit ） to 3.5[Hz] in NS component, from 2.0[Hz] to 3.2[Hz] in 0 0 0 2 4 6 8 sec.10 0 2 4 6 8 sec.10 EW component, and from 2.9[Hz] to 4.9[Hz] in -5 NS Comp. -5 EW Comp. torsional component. Moreover, the effect of reduction of the torsional deformation can be seen from the Fig.18 Microtremors decrease of the spectral value around the 4.9[Hz].

µ・sec µ・sec µ・sec µ・sec 15 10 NS comp. EW comp. NS comp. Average EW comp. Average Average 3.5Hz Average 10 (Center) (Center) (Center) (Center) 7.11 1.9Hz6.04 2.0Hz5.05 5 3.2Hz 3.43 5 2.9Hz2.82 150 100 NS comp. Average EW comp. Average NS comp. Average EW comp. Average 10 (West end) (South end) (West end) (South end) 1.8Hz9.30 3.6Hz 4.61 1.9Hz6.33 5 3.2Hz 3.59 5 2.9Hz2.52 2.9Hz1.70 4.9Hz 0.94 4.9Hz 0.43 150 100 NS comp. Average EW com. Average NS comp. Average EW com. Average (East end) (North end) (East end) (North end) 10 2.0Hz6.94 5 3.6Hz 3.96 2.0Hz 3.2Hz 3.37 5 2.9Hz3.22 3.38 2.9Hz3.63 4.9Hz1.02 4.9Hz 0.80 0 0 0 2 4 6 (Hz)0 2 4 6 (Hz)8 0 2 4 6 (Hz) 0 2 4 6 (Hz) 8 before Seismic Retrofit after Seismic Retrofit Fig.19 Fourier spectra (before retrofit) Fig.20 Fourier spectra (after retrofit)

Damping constant Free vibration waves produced by man power are shown in Fig.21. The damping constant has micron micron 80 Cal. time 80 decreased owing to the seismic retrofit from 3.5[%] before Seismic Retrofit before Seismic Retrofit 0 0 to 2.7[%] in NS component, and from 3.9[%] to 0 2 4 6 sec.8 0 2 4 6 sec.8 -8080 -8080 2.3[%] in EW component. The reason of this after Seismic Retrofit after Seismic Retrofit 0 0 reduction is caused by the tightening of the 0 2 4 6 sec.8 0 2 4 6 sec.8 -80 -80 connection of members. NS comp. EW comp.

Fig.21 Free vibration waves Deformation of torsional mode and soil-building interaction mode To investigate the mode shape around the natural frequency, smoothing of waves are carried out. The Fourier amplitude spectra (F.A.S.) around the natural frequency are cut out by a band pass filter. A band width of the filter is 0.5[Hz] and height of it is 1.0[Hz], and both side of this rectangular filter having the cosine taper of 0.5[Hz]. The smoothed waves are obtained by transforming these cut out F.A.S. to the waves of time domain. Deformation of a translational motion and a torsional motion of the 2nd floor level of the temple are shown in Fig.22. The translational motion is larger in NS component than in EW component, and this motion becomes a bow shape after the seismic retrofit. The torsional motion included the translational one in EW component is recognized before the seismic retrofit, and this translational motion almost vanishes after the seismic retrofit. The reason of this phenomenon is explained by the tightening of horizontal structure caused by putting in diagonal horizontal steel braces. 50 5 50 5 The time histories of swaying and rocking NS direction EW direction 40 （before Seismic Retrofit） 4 40 （before Seismic Retrofit） 4 ratios (Rs and Rr) are shown in Fig.23. The Rs 30 3 30 3 and Rr are defined as Rs=Xb/Xt and Rr=Xr /Xt 20 2 20 2 10 1 10 1 where Xb is the base displacement including 500 50 500 50 NS direction EW direction ground motion and the horizontal deformation 40 （after Seismic Retrofit） 4 40 （after Seismic Retrofit） 4 Swaying ratio(%) Swaying 30 3 ratio(%) Rocking 30 3 ratio(%) Rocking of the soil, Xt is the absolute displacement of a ratio(%) Swaying ● ○ ● ○20 2 20 2 temple at the 2nd floor level and Xr is the 10 1 10 1 displacement caused by the rotational 0 0 0 0 deformation of the soil at the 2nd floor level. 0 2 4 6 8 sec.10 0 2 4 6 8 sec.10 From this figure, Rs is about 20[%] and Rr is Fig.23 Swaying and rocking ratios about 1[%] and these values don’t transit after the seismic retrofit. Translational Mode

Torsional Mode

Translational Mode before Seismic Retrofit

Torsional Mode

after Seismic Retrofit Fig.22 Deformation of a translational and torsional motion CONCLUSIONS

1) Structural style of Japanese temple building is peculiar, and the weight of a roof is very large. As a heavy reason, the roof tile of the temple building is heavy. And clay roofing which superimpose on a roof tile also occupies 30 - 40 percent of roof weight. Moreover, a coiumn exists mostly. Howevre, it has the characteristic that wall length ratio is little. 2) Natural frequency of Japanese temple building is 1.5[Hz] to 2.7[Hz]. Natural frequency is small in order of temple building- traditional wooden house -wooden house. Damping factor of a temple building is concentrated on 4[%] from 3[%]. And damping factor of a wooden house is concentrated on 2[%] from 1.5[%]. Therefore, deformation performance of temple building can say that it is high. 3) Seismic retrofit of Japanese temple building needs consideration of design. Therefore, it is difficult to secure shear capacity of a temple building only by bearing wall. Consequently, in order to reduce earthquake load, it is effective to make weight light. And weight saving of roof carried out as the seismic retrofit method. 4) As a confirmation of the seismic retrofit effect, natural frequency of translation first mode and natural frequency of torsional mode became large 1.7 times. However, rigidity of temple building was high by seismic retrofit. Therefore, since deformation of a joint etc. was controlled, damping factor became small with 0.7 times.

REFERENCES

1. “Guideline For Seismic Capacity Evaluation of Important Cultural Assets (Building Structure) Background Information” Agency for Cultural Affairs Cultural Properties Protection Department Architecture Division, 2000 2. Ishitomi,K., Ohba.S., Inoue,K. “Dynamic Characteristics of Three-story at Yakushi-ji Temple” Transactions of the architectural institute of Japan Kinki network, 1998, p225-p228 3. Kinoshita,A., Ohba,S. “Dynamic Properties of Five-storied Prediction Pagoda and Three-storied Pagoda at Kohuku-ji Temple” Transactions of the architectural institute of Japan Kinki network, 2000, p401-p404 4. Kinoshita,A., Ohba,S., Murao,M. “Dynamic Properties of Three-storied Wooden Pagoda at Kinzan-ji Temple” Transactions of the architectural institute of Japan Kinki network, 2001, p65-p68 5. Kinoshita,A., Ohba,S. “Dynamic Properties of Traditional Wooden Buildings at Kyouougokokuji Temple” Transactions of the architectural institute of Japan Kinki network, 2002, p357-p360 6. Araki,Y., Kinoshita,A., Ohba,S. “Changes in Dynamic Properties of Three-storied Wooden Pagoda during restoration” Transactions of the architectural institute of Japan Kinki network, 2002, p361-p364 7. Nagase,T., Saburi,K., Imanishi,Y., Kaneko,T. “Micro Tremor Observation at Toshodai-ji Kondo” Transactions of the architectural institute of Japan Kinki network, 2000, p101-p104 8. Fujita,K., Hanazato,T., Sakamoto,I. “Vibration Characteristics Of Traditional Timber Five Storied Pagodas, (Part3) Earthquake Monitoring of Tsukannon Pagoda” Summaries of technical papers of annual meeting architectural institute of Japan, 2003.9, C-1 fascicle p465-p466 9. Maekawa,H. “Micro Tremore Measurement on Wooden Buildings, Which are Important Cultural Properties” Summaries of technical papers of annual meeting architectural institute of Japan, 1997.9, C-1 fascicle p223-p224 10. Kawai,N., Uchida,A., Maekawa,H. “Dynamic Characteristics of Traditional Wooden Building, Part1:Micro Tremor Measurements on Rebuilt Buildings of Heijo Palace Site” Summaries of technical papers of annual meeting architectural institute of Japan, 1995.8, C-1 fascicle p55-p56 11. Uchida,A., Kawai,N., Maekawa,H. “Dynamic Characteristics of Traditional Wooden Building, Part2: Micro Tremor Measurements on Horyuji Pagada” Summaries of technical papers of annual meeting architectural institute of Japan, 1996.9, C-1 fascicle p171-p172 12. Kawai,N., Uchida,A., Maekawa,H. “Dynamic Characteristics of Traditional Wooden Building, Part3: Micro Tremor Measurements on Horyuji Kondo and Chumon” Summaries of technical papers of annual meeting architectural institute of Japan, 1996.9, C-1 fascicle p173-p174 13. Maekawa,H., Uchida,A., Kawai,N. “Dynamic Characteristics of Traditional Wooden Building, Part4: Micro Tremor Measurements on Horyuji Dai-kodo” Summaries of technical papers of annual meeting architectural institute of Japan, 1996.9, C-1 fascicle p175-p176 14. Uchida,A., Kawai,N., Maekawa,H. “Dynamic Characteristics of Traditional Wooden Building, Part5: Micro Tremor Measurements on Yakushiji Pagadas” Summaries of technical papers of annual meeting architectural institute of Japan, 1997.9, C-1 fascicle p219-p220 15. Kawai,N., Uchida,A., Maekawa,H. “Dynamic Characteristics of Traditional Wooden Building, Part6: Micro Tremor Measurements on Yakushiji Toindo and Todaiji Tegaimon” Summaries of technical papers of annual meeting architectural institute of Japan, 1997.9, C-1 fascicle p221-p222 16. Maekawa,H., Uchida,A., Kawai,N. “Dynamic Characteristics of Traditional Wooden Building, Part7: Micro Tremor Measurements on Kiyomizudera Nio-mon and Nanzenji San-mon” Summaries of technical papers of annual meeting architectural institute of Japan, 1998.9, C-1 fascicle p259-p260 17. Maekawa,H., Uchida,A., Kawai,N. “Dynamic Characteristic of Traditional Wooden Building, Part8:Estimation of Load-displacement Relationships and Natural Frequencies” Summaries of technical papers of annual meeting architectural institute of Japan, 1999.9, C-1 fascicle p157-p158 18. Maekawa,H., Kawai,N., Uchida,A. “Dynamic Characteristic of Traditional Wooden Buildings, Part9: Estimation of Load-displacement Relationships and Natural Frequencies on Houses” Summaries of technical papers of annual meeting architectural institute of Japan, 2000.9, C-1 fascicle p145-p146 19. Maekawa,H., Kawai,N., Uchida,A. “Dynamic Characteristic of Traditional Wooden Buildings, Part10: Micro Tremor Measurements on Gankoji Hondo and Zenshitsu” Summaries of technical papers of annual meeting architectural institute of Japan, 2001.9, C-1 fascicle p173-p174 20. Tosaka,T., Matsuda,S., Maekawa,H., Shibuya,I., Uchida,A., Kawai,N., Minowa,C. “Dynamic Characteristic of Traditional Wooden Buildings, Part11:Vibration Test on Kaneiji Pagoda” Summaries of technical papers of annual meeting architectural institute of Japan, 2003.9, C-1 fascicle p467-p468 21. Maekawa,H., Kawai,N. “Dynamic Characteristic of Traditional Wooden Houses, Part8:Micro Tremor Measurement on Tyugoku and Shikoku District” Summaries of technical papers of annual meeting architectural institute of Japan, 2003.9, C-1 fascicle p449-p450 22. Saito,T., Ooki,Y., Sakata,H. “Dynamic Characteristics of Traditional Wooden House from Microtremor Measurment” Summaries of technical papers of annual meeting architectural institute of Japan, 2003.9, C-1 fascicle p447-p448 23. Tabuti,A., Nisizawa,H. “Seismic Performance of Traditional Wooden Structure (Microtremor Measurments of The Noya House in Tuyama city)” Summaries of technical papers of annual meeting architectural institute of Japan, 2000.9, C-1 fascicle p133-p134 24. Matsuno,H. “The role of Microtremor Observation and Future of The Usage” The Kenchiku Gijutsu 2003.11, p156-p159 25. Moda,S., Miyazawa,K. “The resistance to earthquake Noh play of the existent tree quality residences and damage estimation” Summaries of technical papers of annual meeting architectural institute of Japan, 1999.9, C-1 fascicle p137-p138 26. Hosaka,T., Moda,S., Kohara,K., Miyawzawa,K. “A Study on Seismic Retrofit System of Wooden Dwelling Houses, Part4” Summaries of technical papers of annual meeting architectural institute of Japan, 2000.9, C-1 fascicle p303-p304 27. Moda,S., Hosaka,T., Kohara,K., Miyawzawa,K. “A Study on Seismic Retrofit System of Wooden Dwelling Houses, Part5” Summaries of technical papers of annual meeting architectural institute of Japan, 2000.9, C-1 fascicle p305-p306 28. Okamoto,A., Moda,S and Miyazawa.K; “A study on Prediction Method of Seismic Performance for Existing Wooden Houses, Part1” Summaries of technical papers of annual meeting architectural institute of Japan, 2001.9 C-1 fascicle p267-p270 29. Moda,S., Miyazawa,K. “A study on Prediction Method of Seismic Performance for Existing Wooden Houses, Part2” Summaries of technical papers of annual meeting architectural institute of Japan, 2001.9, C-1 fascicle p267-p270 30. Sugiyama and Others, “Structural Mechanics Research of Tradition Wooden Structure” Clinical Study Report of Preservation Science and Humanity-Science About Old Cultural Assets, 1984 31. Maekawa,H., Uchida,A., Kawai,N. “Dynamic Characteristic of Traditional Wooden Buildings, Part8: Estimation of Load-displacement Relationships and Natural Frequencies” Summaries of technical papers of annual meeting architectural institute of Japan, 1999.9, C-1 fascicle p157-p158 32. Maekawa.H, Kawai.N and Uchida.A; “Dynamic Characteristic of Traditional Wooden Buildings, Part9: Estimation of Load-displacement Relationships and Natural Frequencies on Houses” Summaries of technical papers of annual meeting architectural institute of Japan, 2000.9 C-1 fascicle p145-p146 33. “Allowable Stress Design of Wooden Frame Structural”: Japan Housing and Wood Technology Center 34. Fujita,K., Kimura,M., Ohashi.Y., Sakamoto,I. “Hysteresis Model and Stiffness Evaluation of Bracket Complexes Uses in Traditional Timber Structures Based on Static Lateral Loading Tests” J.Struct.Constr.Eng.,AIJ,No.543,121-127,May,2001 35. Tsuwa,I., Suzuki,S., Asano,K. “Modeling of Restoring-Force Characteristics of a Traditional Wooden-Frame Structure and its Earthquake Response Analysis” Transactions of the architectural institute of Japan Kinki network, 2003, p25-p28 36. Nakahara,K., Nagase,T. “Earthquake Response of Rocking column in Japanese Traditional Wooden Structure” Summaries of technical papers of annual meeting architectural institute of Japan, 2000.9, C-1 fascicle p141-p142 37. Kataoka., Nomura. “Improvement of Structural Performance of Traditional Wooden Frame Structures.” Summaries of technical papers of annual meeting architectural institute of Japan, 1996.9, C-1 fascicle p181-p182