aci structural journal technical paper behavior of cfrp for prestressing and shear ... ·  ·...

10
ACI STRUCTURAL JOURNAL TECHNICAL PAPER Title no. 94-S9 Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway Bridges by Amir Z. Fam, Sami H. Rizkalla, and G. Tadros This paper describes the behavior of five I-girders, 9.3 m (30.5 ft) each, exclusively reinforced for shear and prestressing, by carbon fiber reinforced plastic reinforcements (CFRP), and one beam prestressed by conventional steel strands and reinforced by steel stirrups. The test beams are 1:3.6 scale models of bridge girders to be built in Manitoba, Canada, using two types of CFRP reinforcements for shear and prestressing. To simulate the composite action of the bridge deck, the stirrups were projectedfrom the girder into the slab, which was cast after a minimum age of 7 days of the prestressed beams. Various web reinforcement ratios were used for each type of the CFRP rein- forcements. Test results were compared to the ACI building code and the modified compression field theory. Effect of CFRP stirrups configuration and size on the shear behavior and their peiformance in providing the dowel action between the girder and top slab are discussed. Draping effect of the prestressing CFRP tendons is also presented. Keywords: concrete; strain; cracks; carbon; fiber; reinforcements; shear; prestressing; FRP. INTRODUCTION One of the major problems that reduces the lifetime service- ability of concrete structures is related to corrosion of the steel reinforcements. With the development of fiber reinforced plastic, FRP reinforcements, and their outstanding character- istics of being noncorrosive material with high strength-to- weight ratio ranging from 3 to 5 times higher than the pre- stressing steel in addition to their excellent fatigue properties, FRP reinforcements have been used for the last few years as reinforcements and prestressing for concrete structures. 1 The use of FRP for shear reinforcement is not yet fully utilized due to the unidirectional characteristics of the reinforcements and presence of shear cracks at an angle with the fibers. 2 Due to the severe environmental conditions and use of salt for deicing, the province of Manitoba, Canada, has decided to use the advanced carbon fiber reinforced plastic (CFRP rein- forcements) for the shear and prestressing reinforcement of four large concrete highway bridge girders in a five span bridge, 32.5 m (106.6 ft), each, to be built in Headingley, Manitoba. The two types of CFRP reinforcements are the car- bon fiber composite cables (CFCC) produced by Tokyo Rope, Japan, and Leadline produced by Mitsubishi Kasei, Japan. This paper describes the experimental program undertaken to determine the effect of the type, percentage of shear reinforce- ments, stirrups configuration, and draping of the tendons on the flexural and shear behavior of the girders, including the mode of failure. RESEARCH SIGNIFICANCE The paper provides unique data, using large scale span gird- ers, to describe the performance of CFRP stirrups and their ef- fect on the behavior of diagonal shear cracks and their capability to provide the dowel action between the girder and the top slab. The study also examines the feasibility and effi- ciency of draping the prestressing CFRP tendons. Test results also provide an assessment of the validity of the current ana- lytical and design approaches for shear using FRP stirrups. The program is crucial for the design of the bridge girders due to the lack of codes and standards in this field. The findings should significantly affect and contribute to the development of future design guidelines and codes. BRIDGE OUTLINE The bridge consists of five spans, 32.5 m (106.6 ft) each, covering a total length of 165.1 m (541.7 ft). The bridge is lo- cated over the Assiniboine River in the Parish of Headingley, Winnipeg, Manitoba, Canada. The deck slab of the bridge is 200 mm (7.9 in.) thick and supported by a total of 40 precast pretensioned simply supported girders. The girders have an 1- shape cross section of the AASHTO type and are transversely spaced at 1.8 m (5.906 ft). The original design of the bridge consists of girders each prestressed by forty pretensioned conventional steel strands of 13 mm el2 in.) diameter. Sixteen strands out of the 40 strands were draped with angles ranging from 3 to 5 deg at distances of 12.7 m (41.7 ft) from both ends. A typical pretensioned concrete girder of the bridge is shown in Fig. 1. ACI Structural Journal, V. 94, No.1, January-February 1997. Received Oct. 31, 1995, and reviewed under Institute publication poliCies. Copy- right © 1997, American Concrete Institute. All rights reserved, including the making of copies unless permission is obtained from the copyright proprietors. Pertinent dis- cussion will be published in the November-December 1997 ACI Structural Journal if received by July I, 1997.

Upload: vuongdang

Post on 20-May-2018

230 views

Category:

Documents


3 download

TRANSCRIPT

Page 1: ACI STRUCTURAL JOURNAL TECHNICAL PAPER Behavior of CFRP for Prestressing and Shear ... ·  · 2016-08-12Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway

ACI STRUCTURAL JOURNAL TECHNICAL PAPER Title no. 94-S9

Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway Bridges

by Amir Z. Fam, Sami H. Rizkalla, and G. Tadros

This paper describes the behavior of five I-girders, 9.3 m (30.5 ft) each, exclusively reinforced for shear and prestressing, by carbon fiber reinforced plastic reinforcements (CFRP), and one beam prestressed by conventional steel strands and reinforced by steel stirrups. The test beams are 1:3.6 scale models of bridge girders to be built in Manitoba, Canada, using two types of CFRP reinforcements for shear and prestressing. To simulate the composite action of the bridge deck, the stirrups were projectedfrom the girder into the slab, which was cast after a minimum age of 7 days of the prestressed beams. Various web reinforcement ratios were used for each type of the CFRP rein­forcements. Test results were compared to the ACI building code and the modified compression field theory. Effect of CFRP stirrups configuration and size on the shear behavior and their peiformance in providing the dowel action between the girder and top slab are discussed. Draping effect of the prestressing CFRP tendons is also presented.

Keywords: concrete; strain; cracks; carbon; fiber; reinforcements; shear; prestressing; FRP.

INTRODUCTION One of the major problems that reduces the lifetime service­

ability of concrete structures is related to corrosion of the steel reinforcements. With the development of fiber reinforced plastic, FRP reinforcements, and their outstanding character­istics of being noncorrosive material with high strength-to­weight ratio ranging from 3 to 5 times higher than the pre­stressing steel in addition to their excellent fatigue properties, FRP reinforcements have been used for the last few years as reinforcements and prestressing for concrete structures. 1 The use of FRP for shear reinforcement is not yet fully utilized due to the unidirectional characteristics of the reinforcements and presence of shear cracks at an angle with the fibers.2

Due to the severe environmental conditions and use of salt for deicing, the province of Manitoba, Canada, has decided to use the advanced carbon fiber reinforced plastic (CFRP rein­forcements) for the shear and prestressing reinforcement of four large concrete highway bridge girders in a five span bridge, 32.5 m (106.6 ft), each, to be built in Headingley, Manitoba. The two types of CFRP reinforcements are the car­bon fiber composite cables (CFCC) produced by Tokyo Rope, Japan, and Leadline produced by Mitsubishi Kasei, Japan. This paper describes the experimental program undertaken to

determine the effect of the type, percentage of shear reinforce­ments, stirrups configuration, and draping of the tendons on the flexural and shear behavior of the girders, including the mode of failure.

RESEARCH SIGNIFICANCE The paper provides unique data, using large scale span gird­

ers, to describe the performance of CFRP stirrups and their ef­fect on the behavior of diagonal shear cracks and their capability to provide the dowel action between the girder and the top slab. The study also examines the feasibility and effi­ciency of draping the prestressing CFRP tendons. Test results also provide an assessment of the validity of the current ana­lytical and design approaches for shear using FRP stirrups. The program is crucial for the design of the bridge girders due to the lack of codes and standards in this field. The findings should significantly affect and contribute to the development of future design guidelines and codes.

BRIDGE OUTLINE The bridge consists of five spans, 32.5 m (106.6 ft) each,

covering a total length of 165.1 m (541.7 ft). The bridge is lo­cated over the Assiniboine River in the Parish of Headingley, Winnipeg, Manitoba, Canada. The deck slab of the bridge is 200 mm (7.9 in.) thick and supported by a total of 40 precast pretensioned simply supported girders. The girders have an 1-shape cross section of the AASHTO type and are transversely spaced at 1.8 m (5.906 ft).

The original design of the bridge consists of girders each prestressed by forty pretensioned conventional steel strands of 13 mm el2 in.) diameter. Sixteen strands out of the 40 strands were draped with angles ranging from 3 to 5 deg at distances of 12.7 m (41.7 ft) from both ends. A typical pretensioned concrete girder of the bridge is shown in Fig. 1.

ACI Structural Journal, V. 94, No.1, January-February 1997. Received Oct. 31, 1995, and reviewed under Institute publication poliCies. Copy­

right © 1997, American Concrete Institute. All rights reserved, including the making of copies unless permission is obtained from the copyright proprietors. Pertinent dis­cussion will be published in the November-December 1997 ACI Structural Journal if received by July I, 1997.

Page 2: ACI STRUCTURAL JOURNAL TECHNICAL PAPER Behavior of CFRP for Prestressing and Shear ... ·  · 2016-08-12Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway

Amir Z. Fam is a graduate student at the University of Manitoba, Winnipeg, Mani­toba, Canada, in the Department of Civil and Geological Engineering. He received his BScfromAlexandria University, Alexandria, Egypt, in 1991 and his MScfrom the University of Manitoba in 1996. His research interests include applications of fiber reinforced plastics in concrete structures.

Sami H. Rizkalfil is an ACI Fellow, Professor of Civil Engineering, and President of the Canadian Network of Centres of Excellence on Intelligent Sensing for Innovative Structures. He is a Fellow of ASCE, CSCE, and EIC, as well as a member of ACI Committee 440, FRP Reinforcements, 550 Precast Concrete. Rizkalla also serves as chief editor of the "FRP International" newsletter.

Gamil Tadros is a structural engineering consultant involved primarily with bridge design and construction. He graduated from Cairo University with a BSc in 1962, and he received his Ph.D from the University of Calgary in 1970. He has won numerous awards and is a member of ACI, CSCE, PCI, ASCE, and IABSE.

EXPERIMENTAL WORK The test beams were designed using the same span-to-depth

ratio of 17.8 and the same prestressed level which induced a compressive stress of 24 MPa (3481 psi) and a tensile stress of 6 MPa (870 psi) at the bottom and top extreme fibres of the section respectively, similar to the bridge girders. Due to lack of information in the literature on the performance of FRP as

32500 .. ··· .... • ......................... ·· .. , •••••••••• T ~

Half elevation 4-: a

c.l.

if »-" :1 Sec.( a-a) - t!> 16 mm

Fig. l-Generallayout of the bridge girder

20500 60 100 36x 110-3960

Sec. (b-b)

500

185 Sec. (a-a)

Fig. 2-Typical configuration of test beams using CFCC strands

Fig. 3-D raped tendons and the hold-down system

shear reinforcement, various stirrups sizes and shapes were used to study their effect on shear and flexural behavior.

All test beams were 9.3 meters (30.5 ft) long, having an I-shape cross section with an overall depth of 500 mm (19.69 in.), as shown in Fig. 2. Similar to the prototype gird­ers, 40 percent of the prestressing tendons were draped at distances of 40 percent of the span from both ends at an av­erage angle of 4 degrees. The hold-down system consisted of stainless steel pins of 33 mm (1.3 in) diameter and sleeves free to rotate, supported by the two sides of the steel form. Therefore, it is recommended that rotating type sleeves be used for the hold-down system in the bridge girders. Fig. 3 shows the hold-down system and the draped tendons. A top slab of 500 mm (19.69 in) wide and 50 mm (1.97 in) in depth was cast after a minimum age of seven days of the pre-ten­sioned beam to simulate the composite action of the deck and the girder. All stirrups were projected from the girders into the slabs to provide the dowel action needed to simulate the composite behavior. Fig. 4 shows a girder before and after casting of the top slab.

Table 1 provides a description of the six test beams, includ­ing the flexural and web reinforcements, p and Pw respective­ly, as tested in this program. The stirrups of the six beams were uniformly spaced at 110 mm (4.33 in) using the same scale factor of the prototype girders. Within the end blocks, similar to the bridge girders, rectangular stirrups were used with a re­duced spacing of 50 mm (1.97 in) for shear and confinement of the concrete within transfer zone. Configurations and di­mensions of the CFRP stirrups are shown in Fig. 2 and 5. Me­chanical characteristics of the two types of CFRP stirrups and steel stirrups are given in Table 2.

Since the scale factor is not applied to the unit weight of the concrete, the resulting stresses at the top surface of the

Fig. 4-Test beam before and after casting the top slab

Page 3: ACI STRUCTURAL JOURNAL TECHNICAL PAPER Behavior of CFRP for Prestressing and Shear ... ·  · 2016-08-12Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway

test beams, due to the effect of prestressing and self weight, exceeded the allowable tensile strength;3 therefore, tempo­rary external post tensioning was used to provide additional stresses on the cross section as shown in Fig. 2 and 4. The ex­ternal post tensioned strands were released before the test, and after an application of a small load of 12 kN (2.7 kips).

Construction details Test beams were fabricated by Con-Force Structures

Company Ltd. in Winnipeg, Manitoba, Canada, using the setup shown in Fig. 6. Steel couplers were used to coupl~ the CFCC and Leadline prestressing reinforcement to conven­tional steel strands to minimize the cost of CFRP and facili­tate using the existing jacking system of the precast fabricators.4 To prestress the draped reinforcement, a special Fig. 5-Configuration of CFRP stirrups

Table 1-Description of test beams

Flexural P =AjAc, Shear Pw =AvlAw, Beam prestressing percent reinforcement percent

TR-I-7.5/7 7.5 mm cP 7-wire stirrups, 0.789 7.5/7 Identical flexural reinforcement, 5 mm cP single wire stirrups, TR-2-5/l five 15.2 mm CP, CFCC strands, 1.03 0.395

TR 5/1

TR-3-5/7 5 mm cP 7-wire stirrups, 0.262 5/7

Double-legged stirrups with rect-

LL-4-2B angular cross section of same 1.0 Identical flexural reinforcement, ten area as 7 mm cP rods,

8 mm cP Leadline rods, 0.858 2B LL Same as in Beam LL-4-2B but

LL-5-1B single-legged stirrups, 0.5 IB

ST-6-C Five 13 mm cP steel strands, ST 0.898 6 mm cP epoxy-coated deformed 0.737 steel stirrups

Table 2-Material properties of concrete Concrete properties

Girder Slab

Beam f~, MPa f,.,MPa f~,MPa

TR-I-7.5/7 50.5 5.83 61.1

TR-2-5/1 60 6.15 58.5

TR-3-5/7 61.1 6.25 60.6

LL-4-2B 60.6 6.39 65.5

LL-5-1B / 65.5 6.91 61.1

ST-6-C 51 6.39 50.5

Prestressing tendons properties

(Jacking! Ultimate Guaranteed Jacking guaranteed) Elastic

Diameter, Area, strength, strength, strength, strength, modulus, Type mm mm2 MPa MPa MPa percent GPa

CFCC' 15.2 113.6 2150 1750 1012 58 137

Leadline . 8 47.3 2950 1970 1216 61 147

Steel 13 98.7 1888 1860 1165 62 205.5

S' tlrrups properties

Area of one Yield Ultimate Elastic Diameter, branch, strength, strength, modulus,

Type mm mm2 MPa MPa GPa

7.5 (7 wires) 30.4 1880 137 CFCC' 5.0 (single wire) 15.2 - 1840 137

5.0 (7 wires) 10.1 1780 137

Leadline • 7.0 (equivalent) 38.5 - 1886 140

Steel ~ 6.0 (deformed) 28.8 600 650 205

'Linear elastic material; given values specified by manufacturers.

Page 4: ACI STRUCTURAL JOURNAL TECHNICAL PAPER Behavior of CFRP for Prestressing and Shear ... ·  · 2016-08-12Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway

setup was used as shown in Fig. 6. The hold-down system,

shown in Fig. 3, was supported by the steel sides of the form,

which was braced to the floor. Fig. 6 also shows the hold-up

system used to provide the change in the direction of the

draped reinforcement to the horizontal position for jacking

purposes.

Typical fabrication of the test beams started by jacking the

bottom straight tendons followed by assembling the stirrups.

The steel sides of the form were assembled to support the

hold-down pins required for jacking draped tendons. After

the concrete reached the specified strength, the beams were

supported downwards by vertical posts at the pin locations in

order to remove the two sides of the form supporting the

hold-down pins. The draped prestressing reinforcement was

released and the vertical posts holding the beam down were

removed. External post tensioning was applied to control the

camber using two steel strands located beneath the top flange and anchored to the end blocks, as shown in Fig. 2 and 4, fol­

lowed by releasing the bottom straight prestressed tendons.

The top slab was cast after a minimum age of seven days

from releasing the prestressed tendons.

c.l.

s~~s :::::~:::.:::::: t{:.:!ns pins supported by sides of the form

Sec. (a·a)

Fig. 6-Generallayout of the jacking system

Fig. 7-Test setup

Test setup and instrumentations The testing program was conducted at the Structural

Engineering and Construction R&D Facility, University of Manitoba, using the setup shown in Fig. 7. Spreader beams were used to apply four concentrated loads to simu­late an equivalent AASHTO HSS 25 truck loading condi­tion. Lateral supports were provided at four locations along the span. A ±5338 kN (±1.2 million pounds) MTS testing machine was used to apply the load using stroke control mode. Deflection at midspan was measured from both sides using two linear motion transducers (LMTs). Dial gauges were used to monitor slip of the tendons and the relative slip between the girder and top slab. Demec point stations of the "Rosette" type were used at the high shear stress locations to measure the strain on the concrete surface in three directions. Other demec point stations were located at the mid-span zone to measure the strain distribution along the section. Electrical strain gauges were also attached to the prestressing tendons to monitor jacking strains, prestress losses, and the strains during testing.

FLEXURAL BEHAVIOR Fig. 8 shows the load-deflection curves of the six beams

tested in this program up to failure. Comparing the post­cracking stiffness of beams prestressed by CFCC strands, the behavior indicates slight differences in the stiffness due to the changing of shear reinforcement ratio. The same phe­nomenon was also observed in comparing beams prestressed by Leadline bars, LL-4-2B and LL-5-lB, using double and single legged stirrups of the same size. Results indicate that reducing the web reinforcement ratio by 50 percent does not influence the overall deformation.

All beams exhibited almost similar crack patterns at the maximum moment zone as shown in Fig. 9. Eleven cracks were developed within the 1200 mm (47.2 in) constant mo­ment zone, approximately at the location of the stirrups with an average spacing of 110 mm. Therefore, it appears that the number and spacing of cracks was mainly controlled by the location of the stirrups rather than the type of the prestressing tendons, in spite of their different bond characteristics. Fig. 9(a), (b), and (c) show the flexural crack patterns of beams prestressed by CFCC strands, Leadline bars, and steel strands respectively.

400

350

300

Z 250 e. ~

200 0

150 ...J

100

50

00 50

I 5

6

1. TR-1·7.5/7 2. TR·2·5/1 3. TR·3-5/7 4. LL·4-2B

5. LL·5-1B

6. ST·6-C

100 150 200 Deflection (mm)

250 300

Fig. 8-Mid-span load-deflection diagrams of test beams

Page 5: ACI STRUCTURAL JOURNAL TECHNICAL PAPER Behavior of CFRP for Prestressing and Shear ... ·  · 2016-08-12Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway

MODES OF FAILURE Test beams prestressed by CFCC strands and reinforced by

7.5 mm 7-wire, and 5 mm solid stirrups, TR-I-7.5/7 and TR-2-5/1 respectively, failed by rupture of the bottom draped strand at the location of the steel pin, located 400 mm (15.7 in.) outside the constant moment zone, as shown in Fig.l O( a). The initial rupture was followed by rupture of the straight bottom strands, and finally by rupture of the upper draped strand. Failure of the bottom draped strand, located above the bottom straight strands, was due to the higher jacking force for this particular strand in comparison to the straight bottom ones as recorded by the strain gage readings.

Failure of the third beam prestressed by CFCC strands and reinforced by 5 mm 7-wire stirrups, TR-3-5/7, confirms the above as shown in Fig. 10(b). Failure occurred by rupture of two straight strands within the constant moment zone, about 100 mm (3.9 in.) from the loading point. The third strand failed outside the constant moment zone.

Test Beam LL-4-2B, prestressed by Leadline bars and re­inforced by double legged Leadline stirrups, failed at higher load level compared to the others due to the high tensile strength of the Leadline bars in comparison to CFCC strands. Failure occurred at the maximum shear location, 2.6 m (8.53 ft) from the support. Before failure, spalling of concrete cover was observed at the bend of the stirrup between the web and the bottom flange, which suggests straightening of the stir­rup at this location. Increasing the applied load caused fur­ther straightening of the stirrups, until they became unable to resist the applied shear as demonstrated by the crushing of the web and the dramatic,failure, shown in Fig. 1O(c). This behavior is clearly related to the stirrup configuration (shown as the second shape in Fig. 5), originally planned for the bridge, and certainly is not related to the characteristics of the CFRP reinforcements.

Beam LL-5-lB, prestressed by Leadline bars and rein­forced by single leg Leadline stirrups, failed by rupture of the bottom straight bars within the constant moment zone at a load level higher than the beam reinforced by double legged stirrups, LL-4-2B, as shown in Fig. 1O(d). Although the web reinforcement ratio in this beam is 50 percent less than in Beam LL-4-2B, it did not exhibit the same mode of failure due to the proper anchorage and shape of the single legged stirrups as shown in Fig. 5. This supports the above conclusion that failure of Beam LL-4-2B is mainly due to the shape of the stirrups rather than the capability of CFRP stir­rups to resist the applied shear.

Beam ST -6-C, prestressed by steel strands and reinforced by 6 mm (0.24 in) steel stirrups, failed. in flexure by yielding of the bottom steel strands followed by crushing of the con­crete at top surface, as shown in Fig. 1O( e), at a much lower load level compared to beams prestressed by CFCC strands and Leadline bars.

SHEAR BEHAVIOR Similar to the prototype bridge girders, the design of the

test beams was controlled by flexural strength requirements and sufficient shear reinforcement to avoid shear failure. All tested beams exhibited a considerable number of diagonal cracks within the maximum shtar span before failure. The diagonal crack patterns were almost similar in number and

spacing and covered about 50 perc~nt of the maximum shear span before failure of the beams in flexural mode. For beams prestressed and reinforced by Leadline stirrups, failures oc­curred at significantly higher load levels compared to the other beams due to the higher ultimate strength of Leadline rods in comparison to CFCC and steel strands. As a result ad­ditional web diagonal cracks developed near the supports. Similarity of the crack patterns in different beams is due to the location of all types of stirrups at the same spacing and prestressing reinforcement layout, consequently the same geometry of the truss mechanism.8 Although the diagonal crack patterns were similar, the crack width and the strain level in the concrete at the location and direction of the stir­rups varied according to the different types of the stirrups used. Fig. II(a), (b), and (c) shows the diagonal crack pattern of beams reinforced by two different sizes of CFCC stirrups and a beam reinforced by steel stirrups, respectively.

Five demec stations of the "Rosette" type were used at the lo­cation of the stirrups where the maximum shear stresses were ex­peeted, as shown in Fig. 12. The demec stations were used to measure the concrete strain of the web in the horizontal, vertical, and diagonal directions using a 200 min (7.87 in) gage length. The readings of strains, combined with the measured angle of the cracks, were used to calculate the average diagonal crack width. The vertical measurements in the direction of the stir­rups were considered equivalent to the nominal strain in the stirrups at the same location.

(a)

(b)

(c)

Fig. 9-Flexural crack patterns

Page 6: ACI STRUCTURAL JOURNAL TECHNICAL PAPER Behavior of CFRP for Prestressing and Shear ... ·  · 2016-08-12Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway

Effect of web reinforcement ratio The effect of the web reinforcement ratio was considered

for the beams reinforced by the CFCC stirrups, Group (1),

and the beams reinforced by the Leadline stirrups, Group (2). In each group, the beams were identical in terms of the pre­stressing reinforcement, layout, and stirrups spacing. The main variable within the two groups was the web reinforce­ment ratio, which was controlled by the stirrups size in the first group and the stirrups configuration in the second group. Fig. Il(a) and (b) shows the diagonal shear crack

(a)

(b) I

Fig. lO-Different modes offailure of test beams

patterns oftwo beams, TR-1-7.SI7 and TR-2-S/1 respective­ly, of the first group using CFCC reinforcements.

The average of the measured strains in direction of stirrups at the maximum shear zone location using the five demec stations was used to compare the beams within each group, as shown in Fig. 13(a) and (b) for CFCC and Leadline re­spectively. The strains for Beams TR-3-SI7 and LL-S-IB are incomplete as they were tested after the brittle failure ob­served for Beam LL-4-2B due to the inappropriate configu­ration of the stirrups as discussed earlier. Comparison of the

(c)

(d)

(e)

Page 7: ACI STRUCTURAL JOURNAL TECHNICAL PAPER Behavior of CFRP for Prestressing and Shear ... ·  · 2016-08-12Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway

three beams reinforced by the three different sizes of CFCC stirrups, shown in Fig. 13(a), and the two beams reinforced by two different configurations of Leadline stirrups, shown in Fig. 13(b), indicates that reducing the web reinforcement ratio will result in higher strain level in the stirrups. Howev­er, the increase in strain is not directly proportional to the web reinforcement ratio as seen when comparing Beams TR-I-7.S/7 and TR-2-S/1.

At different shear levels, the total crack width within each de­mec station was calculated using the demec readings in three di­rections, as shown in Fig. 14. The crack width w was determined as the movement perpendicular to the crack, while the movement parallel to the diagonal crack plane s was considered to be repre­sentative of the slide along the crack. Both quantities were deter­mined based on the measured crack angle and the deformation in two different directions.5 The average of the total crack width is used to compare the cracking behavior of the different beams. The calculated slides along the cracks were found to be very small in comparison to crack widths. Fig. IS(a) and (b) shows the vari­ation of the average of the total crack width for the beams rein-

(a)

(b)

(c),

Fig. II-Diagonal crack patterns

forced by CFCC and Leadline respectively. Crack widths of Beams TR-3-S/7 and LL-S-lB were not included due to the lack of readings after cracking. Comparing Beams TR-I-7.5n and TR-2-S/1 in Fig. IS(a) indicates that the diagonal crack width in­creased by reducing the stirrups size. However, the effect is not directly proportional to the web reinforcement ratio.

Effect of elastic modulus In order to evaluate the effect of the elastic modulus, the strain

in the direction of the stirrups of the steel reinforced beam, ST -6-C, is compared to Beam TR-I-7.5n reinforced by CFCC in Fig. 16. The two beams have almost identical area of stirrups. In this figure, the behavior was considered only after diagonal cracking [V-Vcr] to exclude the effect of variation of the concrete strength in tension. The figure suggests that the slight increase in the strain of the beam reinforced by CFCC in comparison to the steel is not proportional to the value of the elastic modulus.

I Section # 5 I I

I Section # 1 ~ ""'\i ~ I I

I ~

"' I ~

fl"l

~ Demec stations # 1 2 3 4 5 I ... 2.15 m 0.44m

Fig. 12-Demec stations of the "Rosette" type

200

180

Z 160 e. "C 140

~ 120

i 100 --.c Av=20.2 mm2 <II 80 TR-1-7.SI7 "C -~ 60 a. TR-2-5/1

~ 40 ..... 20

TR-3-SI7

~0.5 0 0.5 1.5 2 2.5 3 3.5 4 4.5 Strain in direction of stirrups * 1000

(a)

200 failure load

180

Z 160 e. "C 140

~ 120 "-as 100 (]) .c <II 80 "C ---(]) 60 LL-4-2B '5. Q. 40 -e-< LL-S-1B

20

0 -0.5 0 0.5 1.5 2 2.5 3 3.5 4 4.5

Strain in direction of stirrups * 1000 (b)

Fig. I3-Effect of web reinforcement ratio on the stirrups strain

Page 8: ACI STRUCTURAL JOURNAL TECHNICAL PAPER Behavior of CFRP for Prestressing and Shear ... ·  · 2016-08-12Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway

Fig. 17 shows also the total diagonal crack width of the two beams. The behavior also does not show definite differences due to the lower elastic modulus of the stirrups. However, it should be noted that these beams were designed to fail in flexural mode.

DISCUSSION OF TEST RESULTS Behavior of the beams in shear was analyzed using two analyt­

ical models, the modified compression field theory, and the ACI code, using the material properties given in Table 2.6,7 A brief de­scription of the applicability of each model follows:

Modified compression field theory The analysis was performed using the computer program

"RESPONSE" version 1.0,9 developed to determine the load-deformation response of a prestressed concrete section sub-

V typical cracks

gauge lengtb - 200 mm

Fig. I4-Diagonal cracks within the "Rosette" demec stations

200.-----------------------------,

180

Z 160 e. "C 140

~ 120 iii 100..L~-=--

.J::. Ul 80

~ 60 a. ~ 40

failure load

--TR-1-7.5/7

20 TR-2-5/1

00 . 0.2 0.4 0.6 0.8 1.2 1.4 1.6 Total crack width (mm)

/

(a)

200.-------------------------~--~·

180

Z160 e. "C 140

~ 120

iii 100 .J::. Ul "C Q)

'5. ~

80

60

40

20

failure load

--LL-4-2B

00 0.2 0.4 0.6 0.8 1.2 1.4 1.6 Total crack width (mm)

(b) ~

Fig. I5-Effect of web reinforcement ratio on the diagonal crack width

jected to a combination of moment, shear, and axial load. For each beam, the two critical sections considered were located at Demec Stations #1 and #5, shown in Fig. 12. De­tailed information to account for the material properties of CFCC and Leadline reinforcements in the program are giv­en in Reference 10. The average of the measured concrete strain in the vertical direction at the five demec stations zone is compared to the predicted response at Stations 1 and 5 for the tested beams in Fig. 18. The comparison certainly reflects an excellent agreement of the measured strains to the predicted values using the modified compression field tl}eory.

ACI approach The ACI code assumes that the shear resistance consists of two

components, concrete contribution, Vc,and web reinforcement contribution, VS. It was found that for this particular design of beams, the concrete contribution is controlled by the flexural shear cracking resistance Vcb given by Eq. (1), due to the high contribution of moment and shear at the critical section locations. Since Vci' given by ACI, is independent of the type and amount of reinforcement and affected only by the concrete properties, ge­ometry of the section and prestressing losses, the calculated val­ues at Sections #1 and #5 for a typical beam were found to be 93.3 and 81.8 kN, respectively, using concrete strength and prestress­ing losses of 58 MPa, and 23 percent, respectively.

(1)

To predict the full response of Sections #1 and #5 in a similar fashion to the program "RESPONSE," the code equation was re-

Z 70

e '::'

60 0 > 50 2:-m 40

.J::. Ul

30 C) c

:iii: 20

~ .,!. tJl 10 0 Q.

failure

E=137GPa

--TA-1-7.5/7

ST-6-C

0.5 2.5 3 3.5 4 4.5 Strain in direction of stirrups * 1000

Fig. I6-Effect of elastic modulus on the stirrups strain

Z 70

failure

~ '::'

60

~ 50 2:. .... 40 III

Q) .J::. tJl

30 C) c :i2

20 ~ --TR-1-7.S/7 0

~ 10 Q.

0.2 0.4 0.6 0.8 1.2 . 1.4 1.6 Total crack width (mm)

Fig. 17-Effect of elastic modulus on the diagonal crack width

Page 9: ACI STRUCTURAL JOURNAL TECHNICAL PAPER Behavior of CFRP for Prestressing and Shear ... ·  · 2016-08-12Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway

arranged to predict the applied shear V in terms of the strain in the stirrups £ as follows

(AvdE ) V = V.+---£

Cl s (2)

The previous relation represents a straight line with an intersect to the load axis at a value of Vcb which is equivalent to the shear cracking load. The slope of the line is related to effective depth d, spacing s, the cross section area of the stirrups Av and elastic mod­ulus E. Equation 2 was used to predict the load-strain response at Sections # 1 and #5 for each beam. Fig. 19 shows the average of the measured vertical strains at five demec stations compared to the predicted values at Stations 1 and 5. The comparison shows good agreement in predicting the diagonal cracking load; howev­er, it underestimates the stress level in stirrups after cracking. The behavior suggests that after diagonal cracking occurs, the con­crete contribution Vci is not fully maintained and reduces gradual­ly due to the excessive opening of the diagonal cracks. Fig. 20 proposes a reduction of the Vci contribution as the applied load in­creases to provide a better prediction of the test results. Due to the limited data at this stage, further work is needed to qualify the pro­posed reduction.

CONCLUSIONS Based on an experimental program including testing large

scale girders totally reinforced with CFRP for prestressing and shear reinforcements, the following conclusions could be drawn:

1. The web reinforcement ratio certainly affected the induced stress level in stirrups and the diagonal crack width; however, the effect was not directly proportional to the web reinforcement ra­tio.

2. Due to the relatively high elastic modulus of CFRP in com­parison to other FRP reinforcements, the effect of the elastic mod-

Beam TR·1·7.5/7 200r-~---------------~

180 160

Z 140 C. 120 i .Q 100

j 80

(J) 60 40

/

20

0 ·1 0 2 3 4 5 6 7

stirrups strain * 1000

Beam LL·4·2B 200

180 160

Z 140 C. 120 i .Q 100 (ij 80 ! 60 (J)

40

ulus on the induced strain in the stirrups and the diagonal crack width was insignificant and was not directly proportional to the elastic modulus ratio.

3. The flexural behavior of beams prestressed by CFRP ten­dons showed similar stiffness to the beam prestressed by steel strands after flexural cracking up to yielding of the steel.

4. The ACI code predicted the shear cracking load well; how­ever, it underestimated the stirrups strain after diagonal cracking. This suggests that the concrete contribution Vci is gradually re­duced after cracking. The modified compression field theory pre­dicted the entire response well.

5. Draping of CFRP tendons is practical and does not influence the flexural capacity; however, flexural failure could occur at the bent point locations.

6. No slip of the prestressing reinforcements was observed, nor between the top slab and the girder. This suggests that the strength of concrete at the interface between the girder and the slab was adequate to transfer the horizontal shear stresses.

7. For beams controlled by the flexural capacity, variation of the web reinforcement ratio does not significantly affect the flex­ural behavior of the beams.

ACKNOWLEDGMENTS The authors gratefully acknowledge the support provided by the Manito­

ba Department of Highways and Transportation, Con-Force Structures Ltd., and Natural Sciences and Engineering Research Council of Canada. Special thanks are extended to Messrs. M. McVey, E. Lemke, S. Sparrow, and H. Louka, for their assistance during fabrication and testing of the specimens.

NOTATION Av = area of shear reinforcement bw = web width of the beam d = distance from extreme compression fibers to centroid of

longitudinal reinforcement E = elastic modulus of shear reinforcement

Beam TR·2·5/1 200 180 160

Z 140 C. 120 i .Q 100 ...

80 ~ (J) 60

40 20

0 ·1 0 2 3 4 5 6 7

Stirrups strain * 1000

Beam ST-6-C 200

180 160

Z 140 C. 120 i .Q 100 !ii 80 Q) J: 60 (J)

40 20 20 o O~~-T--~~-T--~~~ -1 0 2 3 4 5 6 7 ·1 0 2 3 4 5 6 7 I Stirrups strain * 1000 Stirrups strain * 1000

Fig. IS-Measured versus predicted response based on the modified compression field theory

Page 10: ACI STRUCTURAL JOURNAL TECHNICAL PAPER Behavior of CFRP for Prestressing and Shear ... ·  · 2016-08-12Behavior of CFRP for Prestressing and Shear Reinforcements of Concrete Highway

Beam TR-1-7.5/7 Beam TR-2-5/1 200 200 180 180 160

SI#5 160 Z 140 Z 140 ~ 120 ~ 120 11 ~ ..Q 100 100 a 80 a 80 .! 60 .! 60 (J) (J)

40 40 20 20 0 -1 0 2 3 4 5 6 7

0 -1 0 2 3 4 5 6 7

Stirrups strain * 1000 Stirrups strain * 1000

Beam LL-4-2B Beam ST-e.C 200 200

180 SI#l SI#5 180

160 160

Z 140 Z 140 ~a .. ~ 120 ~ 120 ,/i

~ 100 ~ 100

J 80 if 80

60 .r:; 60 (J) (J)

40 40

20 20

0 -1 0 2 3 4 5 6 7

0 -1 0 2 3 4 5 6 7

Stirrups strain * 1000 Stirrups strain * 1000

Fig. 19-Measured versus predicted response based on the ACI code

Typical lest results

~

) i ... ""'

.;r 1~ .~ 1 ~I 01 iii

I~ / >2

'6 .eo

Strain level In stirrups

Fig. 2V-Proposed reduction in the concrete contribution in shear, Vej, of the ACI code

Ie = compressive strength of concrete cylinders fr = modulus of rupture of concrete Mer = flexural cracking moment Mmax = maximum moment at section due to externally applied loads s = stirrups spacing V = applied shear Ve = nominal shear strength provided by concrete Vci = nominal shear strength provided by concrete when diagonal

cracking results from combined shear and flexure Ver = measured cracking shear Vd = shear force due to unfactored dead load Yj = factored shear force at section due to externally applied loads

occurring simultaneously with Mmax Vs = nominal shear strength provi;ed by shear reinforcement E = stirrups strain

p = flexural reinforcement ratio Pw = web reinforcement ratio

CONV'1~~~t! fe~TORS 304.8 mm = 1 ft 4.448 kN = 1 kip

6.895 MPa = 1 psi

REFERENCES 1. Abdelrahman, A. A.; Tadros, G.; and Rizkalla, S. H., "Test Model for

the First Canadian Smart Highway Bridge," ACI Structural Journal, V. 92, No.4, July-Aug. 1995, pp. 451-458.

2. Maruyama, T.; Honma, M.; and Okamura, H., "Experimental Study on the Di&gonal Tensile Characteristics of Various Fiber Reinforced Plastic Rods," Transactions, Japan Concrete Institute, Vol. ll, 1989, pp. 193-198.

3. Zia, P.; White, R. N.; and Vanhorn, D. A., "Principles of Model Analysis," Models for Concrete Structures, ACI Special Publication No. 24, 1970, pp. 19-39.

4. Grant, L.; Tadros, G.; and Rizkalla, S., ''Toward Development of Bridges in the Next Century," Proceedings, Second International RILEM Symposium (FRPCS-2), Aug. 1995, pp. 654-662.

5. Pincheira, J. A, "Welded WIre Fabric as Shear Reinforcement in Concrete T-Beams Subjected to Cyclic Loading," MSc thesis, University of Manitoba, Wmnipeg, July 1988, 172 pp.

6. Tokyo Rope Mfg. Co., Ltd., ''Technical Data on CFCC," Japan, Oct. 1993. 7. Mitsubishi Kasei, "Leadline Carbon Fiber Rod Technical Data," Japan,

Dec. 1992,50 pp. 8. Collins, M: P., and Mitchell, D., Prestressed Concrete Structures, Prentice

Hall, Englewood Cliffs, 1991,766 pp. 9. Collins, M. P.; Mitchel, D.; Felber, A. J.; and Kuchma, D. A.,

"RESPONSE Version 1.0," 1990, For Use with Prestressed Concrete Structures.

10. Fam, A Z., "Carbon Fiber Reinforced Plastics Prestressing and Shear Reinforcements for Concrete Highway Bridges," MSc thesis, University of Manitoba, Wmnipeg, Nov. 1995,210 pp.

11. ACI Committee 318, "Building Code Requirements for Reinforced Con­crete (ACI 318-89) and Commentary (ACI 318R-89," American Concrete Insti­tute, Detroit, 1989,353 pp.

Authorized reprint from: January/February 1997 issue of ACI Structural Journal ;:',