The 12th International Symposium on Fiber Reinforced Polymers for Reinforced Concrete Structures (FRPRCS-12) & The 5th Asia-Pacific Conference on Fiber Reinforced Polymers in Structures (APFIS-2015)

Joint Conference, 14-16 December 2015, Nanjing, China

REPAIR OF FATIGUE CRACKS INITIATED AT OUT-OF-PLANE WELDED GUSSET JOINTS USING PRE-TENSIONED CFRP STRIPS

Hitoshi Nakamura1, Hiroya Itoh2, Daisuke Mukaida3 and Fan Lin4

1 Department of Civil and Environmental Engineering, Tokyo Metropolitan University 1-1 Minami-Osawa, Hachioji-shi, Tokyo, 192-0397, Japan

Email: hnaka@tmu.ac.jp

2 Department of Civil and Environmental Engineering, Tokyo Metropolitan University 1-1 Minami-Osawa, Hachioji-shi, Tokyo, 192-0397, Japan

3 Izumi Civil Engineering Office, City of Yokohama

4623 Izumi-Cho, Izumi-Ku, Yokohama-shi Kanagawa, 245-0016, Japan

4 School of Urban Construction and Management, Yunnan University No.2 Cuihu North Rd, Kunming City, China

Email: lin-fan@ynu.edu.cn

Keywords: CFRP strip, Pre-tensioning technique, Repair of fatigue crack, Welded gusset joints

ABSTRACT

This paper deals with the repair of typical fatigue cracks initiated at out-of-plane welded gusset joints in steel bridges using externally-bonded pre-tensioned CFRP strips. The pre-tensioning technique for CFRP strips has been developed in the previous study. Test specimens (L1,350×W400× t9 mm) with real size out-of-plane welded gusset plates (L500 W300×t9×mm) were fabricated and initial fatigue cracks (2a=50mm) occurred at welded toes by fatigue tests. In order to evaluate the effect of repair, fatigue tests of three kinds of methods have been conducted as follows: the specimen without repair, with repair by not pre-tensioned CFRP strips and with repair by pre-tensioned CFRP strips. The compressive force due to the release of the pre-tension is approx. 65 kN by four pre-tensioned CFRP strips. As a result, under the nominal stress range of 100 MPa, fatigue lives were improved drastically, and the effect of 7.7 times in pre-tensioned CFRP strips and 1.6 times in not pre-tensioned CFRP strips was obtained as compared without repair. The prediction of fatigue lives based on LEFM was evaluated conservatively. The repair method using compressive pre-stress results in a reduction of the nominal stress range in consideration of the crack closure condition. 1 INTRODUCTION

In urban highway steel bridges, fatigue damage has occurred with an increase in traffic and the passage of overloaded vehicles. As a simple and easy method for repairing fatigue cracks, the repair method using externally-bonded CFRP strips attracts attention [1], and we have developed a repair method using externally-bonded multi-layered CFRP strips [2], [3]. The more effective repair method by using externally-bonded pre-tensioned CFRP strips has been developed and investigated [4], [5], [6], [7], [8].

In this paper, the repair effect for typical fatigue cracks initiated at welded joints in steel bridges has been investigated experimentally by using externally-bonded pre-tensioned CFRP strips. The pre-tensioning technique for CFRP strips have been developed in the previous study [8].

Test specimens with real size out-of-plane welded gusset plates were fabricated and initial fatigue cracks occurred at welded toes by fatigue tests. In order to evaluate the repair effect, fatigue tests of three kinds of methods have been conducted. The prediction of fatigue lives based on linear elastic fracture mechanics (LEFM) has been investigated.

Hitoshi Nakamura, Hiroya Itoh, Daisuke Mukaida and Fan Lin

2 EXPERIMENTAL AND ANALITICAL PROCEDURE

2.1 Test specimen with real size out-of-plane welded gusset joints

Figure 1 shows the test specimen with real size out-of-plane welded gusset joints. Both sides of the steel plate (L1350×W400×t9 mm) were attached to out-of-plane gusset plates (L500×W300×t9 mm) by fillet welding. The gusset plate was fabricated as a real size in typical plate girder bridges. To initiate fatigue cracks from only one side, the condition of welded joint in the other side was to install a half of hole (R50) and to be full penetration weld as shown in Figure 1. Table 1 shows material properties.

9

400

150

45 75 75

150

5050

75 75 45

5015

015

050

CFRP strip

5451350

Cracked section

500

300

9

R50

Strain gage

Side A

Side B

2525

Initial crack

(a) Side view (b) Plan view

Figure 1 Test specimen with real size out-of-plane welded gusset joints and repair method

Table 1 Material properties

Steel plate JIS SM400

CFRP strip Epoxy resin adhesive

(Konishi E250) Elastic modulus (GPa) 202 150 1.5 Yield strength (MPa) 292 – – Tensile strength (MPa) 420 2808 30 Elongation (%) 37 1.9 –

2.2 Repair method and condition of fatigue test

All fatigue tests have been conducted using a servo-hydraulic test machine (dynamic load capacity: 750 kN) under load control mode. By fatigue test, the initial crack, whose length is approx. 25 mm, has been propagated from the center of gusset plate to both ends, and fatigue cracks were repaired using externally-bonded CFRP strips.

Three kinds of methods have been examined as follows: (a) not repaired for a reference specimen (PWGN), (b) repaired by externally-bonded four CFRP strips (L450×W50×t1.2 mm) without pre-tension (PWGC) and (c) repaired by externally-bonded four CFRP strips (L450×W50×t1.2 mm) with pre-tension (PWGP).

The pre-tension was installed to CFRP strips using the pre-tensioning device [8]. Figure 2 shows the developed pre-tensioning device. Pre-tensioned CFRP strips were bonded. After hardening, the pre-tension was released and the compressive stress was installed into the specimen. In the specimen of PWGP, the tensile strain of 2,500×10-6, namely, the tension of 22.5 kN was installed into the CFRP strip. Table 2 shows the condition of fatigue test.

12

150Pre-tension Length

2Lp100 5010050

450470

Developement Length LnDevelopement Length Ln

70

Steel Wedge Steel Anchor P1. CFRP StripBonding Area

Unit:mm Strain Gauge Movable Part Fixed Part

Hydraulic Nut

Long Bolt for Adjustment

450 (a) Steel anchor plates with CFRP strip (b) Pre-tensioning device assembled by bolts

Figure 2 The configuration of the developed pre-tensioning device [8]

The 12th International Symposium on Fiber Reinforced Polymers for Reinforced Concrete Structures (FRPRCS-12) & The 5th Asia-Pacific Conference on Fiber Reinforced Polymers in Structures (APFIS-2015)

Joint Conference, 14-16 December 2015, Nanjing, China

Table 2 Condition of fatigue test

Repair method Specimen No. Stress range (MPa)

Maximum stressmax (MPa)

Minimum stressmin (MPa)

Stress ratio R

Loading speedf (Hz)

Not repaired PWGN_01 100 111.1 11.1 0.1 3 PWGN_02 80 88.9 8.9 0.1 3 PWGN_03 60 66.7 6.7 0.1 3

Repaired by CFRP strips with pre-stress

PWGP_04 100 111.1 11.1 0.1 3 PWGP_05 100 111.1 11.1 0.1 3 PWGP_07 80 88.9 8.9 0.1 3

Repaired by CFRP strips without pre-stress

PWGC_06 100 111.1 11.1 0.1 3 PWGC_08 80 88.9 8.9 0.1 3 PWGC_09 60 66.7 6.7 0.1 3

(a) Fatigue test system (b) Repairing with pre-stressing device (c) PWGP specimen after repair

Figure 3 Set up of fatigue test and repaired test specimen

2.3 Analysis method analytical model

The 3D finite element analysis (FEA) has been conducted using MSC Marc 2013. The minimum element size of crack tips was 0.04 mm square. According to the symmetry of the test specimen, a quarter of the specimen was modelled using solid element. As the equivalent initial stress, the compressive pre-stress of 375 MPa was set into the CFRP strip longitudinally. The load and analysis condition was longitudinal and uniform distributed stress of 100 MPa and the linear elastic analysis.

The crack was modelled using double nodes definition. The energy release rate was calculated using virtual crack closure technique (VCCT) and the stress intensity factor was evaluated. The crack propagation analysis has been conducted based on LEFM and the evaluation of fatigue life has been investigated. The relationships between the stress intensity factor range KI and the fatigue crack growth rate based on Paris' law is expressed by Eq. (1)

mKCdN

da (1)

Where C and m are material coefficients. In this study, recommendation values [9] in fatigue design for welded joints is adopted and m=2.75, C=1.5×10-11, respectively.

3 EXPERIMENTAL RESULTS AND DISSCUSSIONS

3.1 Evaluation of installed compressive stress into test specimens

Figure 4 shows the stress distribution due to release of pre-tension in the cracked cross section of the specimen (Figure 1). This specimen is PWGP_05. The average value of four strain gages placed longitudinally on CFRP strips was 2,450×10-6 after hardening of adhesive and then it became

Hitoshi Nakamura, Hiroya Itoh, Daisuke Mukaida and Fan Lin

2,250×10-6 after release of pre-tension. Accordingly, The strain difference before and after release of pre-tension is the compressive pre-stress installed into the steel plate.

The experimental result shows the uniform compressive pre-stress of –18.1 MPa or the compressive pre-stress on CFRP strips of –32.1 MPa is installed into the cracked cross section. On the other hand, the analytical result indicates the former is –23.2 MPa or the latter is –37.2 MPa. Although there is the stress deference between side A and B in the specimen, the compressive pre-stress is installed properly into the specimen. The compressive force due to the release of the pre-tension is approx. 65 kN by four pre-tensioned CFRP strips. Moreover, there is the large deference of the installed compressive pre-stress near the out-of-plane gusset plate (X = 0 mm). This is attributed to release of the residual stress installed into a weld toe.

-18.1

-32.1 -23.2

-37.2

-70

-60

-50

-40

-30

-20

-10

0

10-200 -150 -100 -50 0 50 100 150 200

Lon

gitu

dina

l str

ess

(MPa

)

Distance from the center of specimen X (mm)

Side A (Exp.)

Side B (Exp.)

Ave. Section (Exp.)

Ave. CFRP (Exp.)

Center thicness (FEA)

Ave. Section (FEA)

Ave. CFRP (FEA)

Bonding rangeby CFRP strip

Bonding rangeby CFRP strip

Figure 4 Stress distribution due to release of pre-tension in the cracked cross section of the specimen

3.2 Relationships between stress intensity factor and crack length

As a result of FEA, Figure 5 shows the relationships between the stress intensity factor range and crack length under the applied stress of 100 MPa. The result of PWGN without repair indicates that the values of stress intensity factor increase with increasing crack lengths. The result of PWGC and PWGP with repair by CFRP strips indicate that increasing the stress intensity factor controls with increasing crack lengths and the effect is especially remarkable in the range bonded by CFRP strips.

0

10

20

30

40

50

60

70

80

0 10 20 30 40 50 60 70 80 90 100

SIF

rang

e K

I(M

Pa√m

)

Crack length a (mm)

PWGN

PWGC

PWGP

Bonded range by CFRP

Figure 5 Stress intensity factor range vs. crack length

The 12th International Symposium on Fiber Reinforced Polymers for Reinforced Concrete Structures (FRPRCS-12) & The 5th Asia-Pacific Conference on Fiber Reinforced Polymers in Structures (APFIS-2015)

Joint Conference, 14-16 December 2015, Nanjing, China

3.3 Relationships between crack length and number of cycles

Figure 6 shows the relationships between crack lengths and the number of cycles under the stress range of 100 MPa. The relationships between crack lengths and the number of cycles were obtained by the beach mark method experimentally. The experimental result indicates the prolongation of fatigue life is in case of PWGP, 7.7 times and in case of PWGC, 4.8 times as compared without repair, respectively. The results of fatigue life prediction indicate that in case without repair, the experimental data can be simulated with accuracy, however in case with repair, the predicted fatigue lives are less than experimental fatigue lives and are evaluated conservatively.

0

50

100

150

200

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Cra

ck le

ngth

a (

mm

)

Number of the cycles Np (×106 )

PWGN (Exp.)PWGC (Exp.)PWGP (Exp.)PWGN (LEFM)PWGC (LEFM)PWGP (LEFM)Failure

Analysis conditionsn=100 (MPa)C=1.5×10-11

m=2.75

Bonding range by CFRP strip

Figure 6 Relationships between the crack length and the number of cycles

3.4 Results of fatigue test and prediction of fatigue life

Figure 7 shows the relationships between the stress range and the number of cycles when the crack length a is from 25 mm to 150 mm. The result of regression lines of fatigue tests (dotted lines) and the fatigue life prediction (solid lines) are also showed in this figure. Although there is little test data, the scatter of experimental fatigue lives is small and in case with repair, predicted fatigue lives are less than experimental ones as mentioned above. In the case of PWGP with pre-stress, fatigue lives can be estimated roughly by crack propagation analysis.

105

106

107

50

60

70

80

90

100

110110

120

sn

(N/m

m2 )

Number of cycles N

PWGNPWGCPWGPPWGNPWGCPWGPPWGNPWGCPWGP

Test results

Regression line of test results

Prediction by crack propagation analysis

Figure 7 Relationships between the stress range and the number of cycles

Hitoshi Nakamura, Hiroya Itoh, Daisuke Mukaida and Fan Lin

4 CONCLUSIONS

Therefore, these results indicate that the compressive pre-stress can be installed into the cracked steel plate with out-of-plane welded gusset plates experimentally and analytically and the stress intensity factor and the fatigue crack growth rate can be reduced by release of pre-tension. Moreover, fatigue lives can be estimated roughly by crack propagation analysis and predicted fatigue lives are evaluated conservatively.

REFERENCES

[1] X.L. Zhao, FRP-Strengthened Metallic Structures, CRC Press, 2013. [2] H. Nakamura, W. Jiang, H. Suzuki, K. Maeda and T. Irube, Experimental study on repair of

fatigue cracks at welded web gusset joint using CFRP strips, Thin-Walled Structures, 47(10), 1059–1068, 2009.

[3] F. Lin, J. G. Sun, H. Nakamura and K. Maeda, Fatigue crack repair using drilled holes and externally bonded CFRP strips, Proceedings of the 5th International Conference on Bridge Maintenance, Safety and Management, IABMAS 2012, 1308–1315, 2012.

[4] I. Stöcklin and U. Meier, Strengthening of concrete structures with prestressed and gradually anchored CFRP strips, Proceedings of the Seventh International Symposium on Fiber-Reinforced Polymer Reinforcement for Concrete Structures (FRPRCS-5), 291–296, 2001.

[5] P. Colombi, A. Bassetti and A. Nussbaumer, Analysis of cracked steel members reinforced by pre-stressed composite patch, Fatigue and Fracture of Engineering Materials and Structures, 26(1), 59–66, 2003.

[6] B. Täljsten, C.S. Hansen and J.W. Schmidt, Strengthening of old metallic structures in fatigue with prestressed and non-prestressed CFRP laminates, Construction and Building Materials, 23(4), 1665–1677, 2009.

[7] Y. Huawen, C. König, T. Ummenhofer, Q. Shizhong and R. Plum, Fatigue performance of tension steel plates strengthened with prestressed CFRP laminates, Journal of Composites for Construction, 14(5), 609–615, 2010.

[8] H. Nakamura, Y. Yamamura, H. Itou, F. Lin and K. Maeda, Development of pre-tensioning device for CFRP strips and applicability to repair of cracked steel members, Advances in Structural Engineering, 17(12), 1705–1717, 2014.

[9] Japan Society of Steel Construction (JSSC), Fatigue design recommendations for steel structures and commentary, Gihodo Publishing, 1993. [in Japanese]