Bamboo Structure Constructed with Laminated Bamboo Cotter and Epoxy Resin Connection at the Yokohama Sesquicentennial Festival

YOSHIDA Keito1), MATUI Masasumi2) and TANAKA Shigeto3)

1) Assistant Professor, Tokyo Institute, Polytechnic University, 2-32-1Ogawa Nishi-Machi, Kodaira-shi, Tokyo, e-mail kyoshida@tokyo-pc.ac.jp

2) Representative director, Team Zoo Atolier ToDo, 2-7-6 Tukiji-Chuo-ku, Tokyo e-mail todo.matsui@nifty.ne.jp

3) Yokohama City Office, 1-1 Minato-cho, Naka-ku, Yokohama

The purpose of this paper is to investigate possiblity of using bamboo as an architectural material. To meet the objective, we assessed basic mechanical characterustics of moso bamboo and proposed an architetural building which has truss system. The chord of the system has bamboo laminated lumber as a cotter and a joint using epoxy resin which is devised by us. The strength of the joint is also examined through force application test. The building was constructed at a memorial event site celebrating the 150th anniversary of the opening of Yokohama Port. Finally, the strength of the bamboo structure was assesed by force application tests. Key words: bamboo, bamboo laminated lumber, cotter, epoxy resin

1. INTRODUCTION As a theme of the memorial event celebrating the sesquicentennial of the opening of the Yokohama City Port, the hillside stage is targeted for the use of bamboo as an architectural material. To meet that objective, the authors assessed basic mechanical characteristics of mousou bamboo, which is available in the Yokohama vicinity, and proposed an architectural building using bamboo as a structural material. This building has a truss structure in its construction that necessitates the joining of bamboo. The authors then devised a joint using bamboo laminated lumber as a cotter and a joint using epoxy resin for connecting chord members of the truss and performed experiments before the structure was constructed. This report describes strength experiments of the bamboo joints performed to examine the possibility of utilization of bamboo materials for architecture and strength experiments of a building constructed for the opening of the Yokohama City Port

2. OUTLINE OF EXPERIMENTS 2.1 Experimental items For the experiments, the respective strengths of four items were confirmed: orthogonal joints and parallel joints for confirmation of mounting strength of chord members and lattice members of the truss; the bamboo laminated lumber used for cotter components; and the beam butt joint.

2.2 Orthogonal joining 1) Profile of test specimens As test specimens, vertical and horizontal materials were taken from the same bamboo. The vertical material and horizontal material were, respectively, 690 mm and 570 mm long. Application of a load to the test specimen is presented in Fig. 1 and Photograph 1. Three test specimens were prepared. Measurements of the bamboo diameter and thickness are shown in Table I

and Table II. Thickness of the bamboo was measured at four locations on the cross-section and at two locations on the diameter.

A

40

150

150

150

490

PL-12ｘ100ｘ490

150

150

370

φ18

φ18A B

B

570

B-B　SECTION

340

690

150

150

50

SPECIMEN

SPECIMEN

SPECIMEN

100 100

Fig. 1 (left) and Photograph 1 (right) Profile of orthogonal test specimen.

Table I Measurement results (Thickness) THICNESS (mm)

UPPER BOTTOM AVE.

1 DB1(U) 7.9 7.9 7.6 8.2 8.0 7.6 8.0 8.0 7.9DB1(D) 8.0 8.6 7.9 8.7 8.2 8.5 8.9 8.9 8.4

2 DB2(U) 9.7 8.9 9.5 9.5 10.8 10.9 9.7 10.0 9.9DB2(D) 8.5 8.5 8.4 9.7 8.4 9.0 8.2 8.7 8.7

3 DB3(U) 10.3 10.0 10.8 10.2 10.7 10.5 11.5 8.6 10.3DB3(D) 9.8 8.4 8.9 8.9 8.8 9.4 9.9 9.9 9.2

(U): Vertical Specimen (D): Horizontal Specimen

Table II Measurement results (Diameter)

(U): Vertical Specimen

Diameter (mm) UPPER BOTTOM AVE.

1 DB1(U) 690 103.0 96.8 98.7 92.0 97.6DB1(D) 570 105.4 99.8 103.3 97.8 101.6

2 DB2(U) 690 105.1 111.7 111.7 114.5 110.7DB2(D) 570 109.0 102.5 107.0 100.3 104.7

3 DB3(U) 690 101.1 98.8 96.1 96.5 98.1DB3(D) 570 88.2 91.1 95.0 96.0 92.6

(D): Horizontal Specimen

923

Transactions of the Materials Research Society of Japan 35[4]923-928(2010)

2) Results of tests Failure properties of test specimen DB1, 2, 3 showed nearly identical trends to those shown below. With increased load, the compressive strain caused by the bearing occurred at the hole part of the cotter for mounting of the vertical member. Then a fracture along with the axis occurred to the horizontal member. No fracture was caused to the cotter. Table III presents results of tests of the test specimen and Photograph 2 shows test specimens after tests.

Table III Results of orthogonal joining test SPECIMEN Max Load (ｋN)

1 DB1(U) 690 2.95DB1(D) 5702 DB2(U) 690 3.34DB2(D) 5703 DB3(U) 690 4.33DB3(D) 570

Photograph 2 Profile of an orthogonal test specimen after failure (DB1).

2.3 Parallel joining 1) Test specimen profile As the test specimen, two vertical members were taken from the same bamboo. The member located at the upper part is designated as (U), and that located at the lower part is designated as (B). Three test specimens were used. Table IV and Table V respectively show the bamboo diameter and thickness. Continuous thread bolts (φ16) were used for fixing of the bamboo and jig. A washer was inserted in each part and fixed with a nut. The nut was fastened to such an extent that turning by hand was not possible. Cotters (φ18) made from bamboo laminated lumber were used for connection of upper and lower bamboo components, and a washer was also inserted in the mounting part of the bamboo and jig and fixed with a nut. A wedge was used at the end of the cotter to prevent its extraction. Diameters of all holes for mounting of the specimen were 18 mm. Fig. 2 and Photograph 3 show the test specimen.

2) Results of tests Results of the test are presented in Table VI. With the PB1 test specimen, a deformation caused by the bearing occurred at the contact portion with the cotter, the wedge at the outside of the cotter was then pulled inside and sank into the inside. The bearing portion of the bamboo fell off eventually (Photograph 4). With the PB2 test specimen, nodes of upper and lower bamboo components made contact at the point of time of preparation of the test specimen, resulting in the creation of a clearance on the surface of both. Consequently, a bending moment was generated at the cotter, which caused breakage (Photograph 4). Breakage of the PB3

test specimen resulted from shearing of the mounting cotter. This test specimen showed higher strength than other test specimens because it had higher machining accuracy (Photograph 4).

Table IV Profile of test specimen parallel to the axis (Diameter)

(U): Upper Specimen, (B): Bottom Specimen

Table V Profile of test specimen parallel to the axis (Thickness)

(U): Upper Specimen, (B): Bottom Specimen 150

150

50

340

40

150

150

490

15040

150

150

90

150

150

690

1,280

PL-12ｘ100ｘ490

φ18

φ18

200

390

390

SPECIMEN

BAMBOO

BAMBOO

690

150

Specimen Max. Load（ｋN）1 PB1 4.22 2 PB2 5.88 3 PB3 7.34

Diameter (mm)UPPER BOTTOM AVE.

1 PB1 (U) 92.7 97.1 91.6 86.5 92.0 PB1 (B) 101.7 99.2 97.2 95.8 98.5

2 PB2 (U) 101.4 94.0 104.0 97.0 99.1PB2 (B) 96.6 89.4 101.6 93.3 95.2

3 PB3 (U) 96.2 87.5 102.5 94.3 95.1PB3 (B) 102.4 92.1 109.3 98.0 100.5

Thickness (mm) UPPER BOTTOM AVE.

1 PB1 (U) 6.5 9.3 7.8 8.2 8.2 8.1 8.8 8.3 8.1PB1 (B) 8.3 8.2 8.8 8.6 7.6 7.3 8.8 8.1 8.2

2 PB2 (U) 9.5 9.5 9.5 9.9 10.5 10.8 10.9 11.1 10.2PB2 (B) 9.3 9.8 10.0 10.1 10.0 10.3 10.2 10.7 10.1

3 PB3 (U) 7.8 8.3 8.1 8.0 8.2 8.8 9.2 9.3 8.5PB3 (B) 8.1 8.6 9.0 8.7 9.4 9.7 8.6 8.8 8.9

Fig. 2 (left) and Photograph 3 (right) Profile of Parallel test specimen.

Table VI Experimental results of test specimen

Photograph 4 Failure profile of a test specimen parallel to the axis (PB1, PB2, PB3).

924 Bamboo Structure Constructed with Laminated Bamboo Cotter and Epoxy Resin Connection at the Yokohama Sesquicentennial Festival

2.4 Material strength test of bamboo laminated lumber (cotter)

Table VIIIb Results of shear test for bamboo laminated lumber

1) Test specimen profile Bamboo laminated lumber was employed as the cotter to be used for joining of bamboo truss. To grasp material characteristics, the bending test and shear test were conducted according to the following procedures. As presented in Photograph 5a, the span of the bending test specimen was 250 mm. For support, one end was joined by a pin; the other end was supported by a roller, and a concentrated load was applied to the center. Shear testing was performed as shown in Photograph 5b. Three test specimens were used for each test. Measurements of the bending test specimen diameter were taken at three locations: at the center and 100 mm to the right and left. Measurements are shown in Table VIIa. The diameter of the test specimen for shear test was measured in two directions. Results obtained are shown in Table VIIb. As depicted in Fig. 3, the bamboo laminated lumber is a lamination of bamboo pieces with 5 mm thickness. Because the bamboo laminated lumber is designed to bear shear capacity, influences by direction of lamination should be known. To determine them, experiments were performed for SH test specimen placed horizontally with respect to the force application direction and SV test specimen placed vertically. However, the direction of lamination for the bending test was in the vertical direction only.

Table VIIa Profile of bamboo laminated lumber (Diameter, for bending test)

Diameter (mm) Sec. Aremm2

End-1 Middle End-2 Ave.X Y X Y X YB1 18.0 18.0 18.0 18.0 18.0 18.0 18.0 254.3B2 18.0 18.0 18.0 18.0 18.0 18.1 18.0 254.3B3 18.0 18.0 18.0 18.0 17.9 18.1 18.0 254.3

Table VIIb Profile of bamboo laminated lumber (Diameter, for shear testing)

Fig. 3 Section of bamboo laminated lumber

2) Results of tests (Bending and shearing) Results of tests are shown in Table VIIIa and Table VIIIb.

Table VIIIa Results of bamboo laminated lumber bending test

( )( )max(*1) : (*2) : 4 3Q A Q Aτ τ= =

(*3) : Percentage of Water Content

3) Status of failure Method of force application and status of failure after the bending test and shear test are shown in Photographs 5c and 5d.

2.5 Strength test for epoxy resin joint-1 1) Profile of test specimen To construct a chord member of the truss structure with a large span, the bamboo materials themselves should be butt-joined. For the connection to be used in this joining, joining experiments to use adherence strength of epoxy resin were performed. Figure 4 shows that epoxy resin was injected into upper and lower bamboos to join them by adhesion force. The cotter was inserted to ensure safety of the bamboo laminated lumber. The joints’ external appearance is portrayed in Photograph 6.

Diameter (mm)X Y Ave.

SH1 18.0 18.0 18.0 SV1 18.0 18.1 18.1 SH2 17.9 18.0 17.9 SV2 18.0 18.0 18.0 SH3 18.0 18.0 18.0 SV3 18.0 18.0 18.0

Sec. Area

(mm2)

Span (mm)

Young’sModukus (N/mm2)

Bending Stress

(N/mm2)B1 254.5 250.0 6844.3 98.8B2 254.5 250.0 8566.1 141.2B3 253.8 250.0 7480.8 109.0

Sec.Area Max.Load Disp. τ 1 (*1) τ 2(*2) (*3)(mm2) (kN) (mm) (kN/mm2) (kN/mm2) ％

SH1 254.5 5.77 5.4 2.27E-02 3.02E-02 7.6SV1 255.9 4.69 5.5 1.83E-02 2.44E-02SH2 251.7 5.58 11.7 2.22E-02 2.96E-02 7.2SV2 253.8 5.29 9.9 2.08E-02 2.78E-02SH3 253.8 6.03 11.9 2.38E-02 3.17E-02 7.1SV3 253.8 6.14 12.1 2.42E-02 3.23E-02

Photograph 5a Bending test.specimen.

Photograph 5b Shear test specimen.

Photograph 5d Results of shear test.

Photograph 5c Results of bending

Load(H）

18m

m

Loa

d（V）

360

196

196

520

80

80

146

14650

EPOXY

50

Photograph 6 Profile of bamboo material

joint-1 test specimen.

Fig. 4 Profile of bamboo material

joint-1 test specimen.

925K. Yoshida et al Transactions of the Materials Research Society of Japan 35[4]923-928(2010)

2) Test specimens 2.6 Strength test for epoxy resin joint-2 Three test specimens were used. Dimensions of these specimens are depicted in Table IX.

1) Profile of test specimen Experiments revealed that adhesion of epoxy resin and bamboo are not expected without modification. Therefore, we devised a new method of joining that is free from adhesion. A hole (φ6) was made in two locations for bamboo materials to be joined. Epoxy resin was injected into these holes to form a protrusion by epoxy resin, and these upper and lower protrusions were integrated by epoxy resin. This is a so-called brace from bamboo inside by epoxy. Because epoxy resin might not fill all voids in the bamboo material, a bamboo component that was one size smaller was inserted so that the thickness of the epoxy resin connecting the protrusions became about 5 mm.

Table IX Profiles of bamboo joint-1 test specimen

Table XII Results of tests of bamboo joint-2

Specimen Max.Load(kN)

J4 16.2J5 12.3J6 11.3

(U): Upper Specimen, (B): Bottom Specimen

J-1 J-2 J-3Dim(U) （mm）

Upper 121.0 114.6 104.0Bottom 126.8 125.0 114.0

Dim(D) （mm）

Upper 118.1 127.9 117.3Bottom 124.9 117.4 109.7

Thickness (U)

（mm）

1 9.3 10.2 10.62 9.3 9.9 9.63 9.9 9.5 9.34 10.7 9.3 10.3

Thickness (D)

（mm）

1 9.3 10.1 10.22 9.3 9.2 10.03 9.1 10.8 10.54 9.1 10.1 10.5

Length of Cotter（mm） 100 100 100Length of Specimen (mm） 519 518 518

It was considered that this joining method shown in Fig. 5 and Photograph 8 prevents rattling of both and provides strength required for joining of bamboo materials. Three test specimens were used in the experiments. Their profiles are presented in Table XI.

3) Results of tests Table X presents test results. Although we expected adhesion with the bamboo by adoption of epoxy, the end of the bamboo fell off because of the bearing of the cotter. Because experiment results and values by shear resistance of the bamboo are nearly identical (Eq. 1), the result showed that adhesion force between the bamboo and epoxy is disappointing because the adhesion force was disturbed by the membrane inside the bamboo.

BAMB

OO

EPOXY　

RESI

N

PROT

RUSI

ON　

Meanwhile, the shear strength of the cotter is 5.58 kN on average. It is 11.16 kN at the most, even if both ends are examined. It is considered that for this load, the cotter undergoes shear failure earlier. The reason why this did not occur is considered to be attributable to the difference of force application portion. In the material experiment, the force application device was linear. The contact portion then acted as a concentrated load, and stress concentration occurred, resulting in lower values. In this experiment, the load was regarded as applied uniformly to the cotter in semi-circular fashion. Stress concentration did not occur at the contact portion, thereby showing higher values. The failure status is shown in Photograph 7.

Photograph 8 Bamboo joint-2 test specimen.

Fig. 5 Profile of bamboo joint-2 test specimen.

Table XI Profiles of bamboo joint-2 test specimen

Table X Results of test of bamboo joint-1

(*): Percentage of Water Content

Max.Load (kN)

Thickness (mm)

Shear (kN/cm)

(*) （％）

J1 18.1 9.2 19.6 22.7J2 21.7 10.1 21.5 19.7J3 22.0 10.0 21.3 32.5 (U): Upper Specimen, (B): Bottom Specimen

J-4 J-5 J-6Dim(U) （mm）

Upper 96.0 86.0 116.0Bottom 93.0 84.0 110.0

Dim(D) （mm）

Upper 90.0 91.0 111.0Bottom 90.0 87.0 106.0

Thickness (U)

（mm）

1 7.0 6.9 8.52 7.0 6.6 9.33 6.8 6.5 8.34 7.1 7.4 9.3

Thickness (D)

（mm）

1 6.5 7.1 9.02 6.8 7.0 8.73 6.7 7.8 9.34 6.45 7.2 8.8

Distance of Cotter（mm） 146 149 143

(1) 2 2.13N mm t kN= ⋅max (41 ) 4 13 /Q mm t= × × ×2) Results of tests Test results are shown in Table XII. Breakage occurred at the center portion of joining of the epoxy resin and caused brittle failure, although strength comparable to the shear strength of the cotter was obtained (Photograph 9).

Photograph 7 Status of breakage of bamboo joint-1.specimen

926 Bamboo Structure Constructed with Laminated Bamboo Cotter and Epoxy Resin Connection at the Yokohama Sesquicentennial Festival

3. CONSTRUCTION OF ARCHITECTURE 3.1 Architectural work Based on the experimental results described above, a bamboo building 4 m high, with maximum span 10 m, and total floor area of 150 m2 was constructed. Cotters (φ18) were used for mounting of lattice members and joints using epoxy resin were used for connection of chord members. Photographs 10 and 11 were taken upon completion of the bamboo buildings. These buildings were constructed at a memorial event site celebrating the 150th anniversary of the opening of Yokohama Port.

3.2 Force application test To assess the truss structure strength, a force application test was performed upon completion of the event. Two frames were used: one with 10 m span (designated as FR10) and the other with angle brace and 6 m effective inner size span (designated as FR6). A concentrated load was applied to the center for both frames (Photographs 12, 13).

3.3 Results of experiments When the movable load to FR10 frame reached 1.76 kN, 2.16 kN, shear failure was exerted on the cotters with the lattice member, respectively. When the load

reached 2.25 kN later, deformation of the applied force part reached 107 mm and upper chord materials caused breakage and the load is reduced. The experiment was terminated at this point. Meanwhile, with the FR6 frame, the cotter of lattice member at the center of beam caused shear failure when the movable load reached 3.92 kN. When it reached 4.41 kN, the upper chord material joint was broken. An outline of results of experiments is shown in Table XIII. Results of deformations under the load are depicted in Fig. 6. Breakage of FR10 and FR6 is shown, respectively, in Photograph 14 and Photograph 15.

Photograph 9 Profiles of failure of bamboo joint-2 test specimen.

Load (kN)

Deformation (mm) Breakage

FR63.92 56 (1/179) Cotter4.41 74 (1/135） Upper Chord

FR101.76 34 (1/294) Cotter2.16 68 (1/147） Cotter 2.25 107 (1/93) Upper Chord

FR10 FR6

Photographs 12 Test Frames FR6 and FR10

Photographs 13 Test Frames FR6 and FR10 Photographs 10 Constracted Bamboo building-1

Table XIII Outline of results of experiments

Photographs 11 Constracted Bamboo building-2

00.5

11.5

22.5

33.5

44.5

5

0 20 40 60 80 100 120

Load  (kN)

Load

(kN

)

Fig. 6 Relation of Load-Deformation Deformation (mm)

FR10 FR6

927K. Yoshida et al Transactions of the Materials Research Society of Japan 35[4]923-928(2010)

Next, a comparison was made between cotter strength obtained using the material experiment and results of the calculation. Assumptions used for stress calculation are such that the own weight of the bamboo is 50 N/m, the building truss beam length is 750 mm, and the beam is supported immovably at both ends. Back calculation of the load concentrated at the center when the cotter of the building was broken revealed that shear stresses acting to the cotter of FR6 and FR10 are, respectively, 9.5 kN and 10.7 kN. These values are greater than 5.6 kN, which is an average of the shear strength of the cotter because, as stated previously, stress concentration did not occur.

[3] International Code Council Evaluation Service (ICC-ES), “Mechanical Properties of Bamboo,” April 12, 2002

[4] Institute for Structural Design Dr.-Ing. Evelin Rottke Reiffmuseum Schinkelst.1 52062 Aachen

[5] K. Kitahara, "Physics of wood," Morikita Shuppan, Nov. 30, 1972

[6] “Bamboo Structural Design,” International Standard Design ISO/TC 165/N313, N314, N315

[7] ICBO Evaluation Service, Inc., “Acceptance Criteria for Structural Bamboo,” AC162, March 2000

[8] "Report on basic research experiments of material characteristics of mousou bamboos for utilization to architectural structure," Employment and Human Resources Development Organization of Japan, Tokyo Institute, Polytechnic University,, Funded research, Oct. 31, 2007

[9] "Memorial event celebrating 150th Anniversary of opening of the Yokohama Port," "Report on verification experiments of field of bamboo," Ando First-class Architect Office, Employment and Human Resources Development Organization of Japan,Tokyo Institute, Polytechnic University, Feb. 29, 2008

Photograph 14b (FR6)Breakage of upper chord . Photograph 14a (FR6)

Breakage of cotter.

Photograph 15b (FR10)Breakage of joint.

Photograph 15a (FR10) Breakage of cotter.

4. CONCLUSIONS Results obtained through this series of experiments are enumerated below.

1. The type of failure of the bamboo material at orthogonal joints depends on the fracture at the mounting part and the shear force of the cotter. The average fracture stress intensity is assumed to be 1.15 N/mm2.2. The type of failure of the parallel-joined bamboo material depends on the cotter shear strength or edge clearance. The shear strength of the cotter used this time was 5.6 kN. 3. Adhesion of epoxy and the bamboo interior are not promising.4. The connection method using protrusions of epoxy filled in the holes drilled into bamboo materials to be connected can prevent rattling. Moreover, its joining strength is promising. 5. Cotters show higher values than values obtained from material experiments.

References [1] "Report on an experimental forest, Agricultural

Faculty, The University of Tokyo," No. 36, December 1948, pp. 134-186

[2] Homepage of Wood Structure Laboratory, Welfare and Environmental Engineering Course, Oita University

928 Bamboo Structure Constructed with Laminated Bamboo Cotter and Epoxy Resin Connection at the Yokohama Sesquicentennial Festival

(Received April 20, 2010; Accepted November 4, 2010)