CC BY-NC-ND
0Available online at www.sciencedirect.com
SciVerse ScienceDirect Procedía
Engineering
Procedía Engineering 54 (2013) 752 - 763 =
www. el sevi er. com/1 ocate/procedi a
The 2nd International Conference on Rehabilitation and Maintenance in Civil Engineering
Torsional Strength of Ferrocement "U" Wrapped Normal Strength Beams with only Transverse Reinforcement Gopal Charan Beheraa*, Rao T.D.Gunneswarb and Rao C.B.Kb
aBhadrak Institute of Engineering and Technology, India bNational Institute of Technology, Warangal, India
Abstract
Structural repair and rehabilitation of concrete structures is necessary for all deteriorated or damaged structures to restore and enhance the load bearing capacity as well as to increase the life span of the structure In the recent past 3Rs of construction Technology (Retrofitting, repair and Rehabilitation) became popular due to the natural calamities or the updates of the codal provisions. Use of FRPs have been gaining world wide acceptance as retrofitting material for their high strength, light weight and good fatigue life. However, from cost benefit point of view ferrocement can be used as wrapping for rehabilitation of deteriorated or damaged reinforced concrete beams. Ferrocement, a thin structural composite material, exhibits better crack arresting capacity, higher tensile strength to weight ratio, ductility and impact resistance. Hence it can be used as an alternative for FRPs in the field of repair and rehabilitation. Very little information is available from literature on the repair and rehabilitation of reinforced concrete beams with ferrocement wrapping especially in resisting pure torsional loads. Repair and rehabilitation of the distressed concrete structures is well addressed by researchers for the basic structural actions such as axial loads, flexure and shear in past few years, but torsional repair and rehabilitation has not attained much importance in research due to its complex nature and its occurrence with other basic structural forces. Circulatory torsion induces shear stresses on all four sides and would be well resisted by closed form reinforcement. But due to monolithic construction of beam and slabs, U wrap retrofitting is the most practical solution. Experimental results and analytical models of previous researchers revealed that wrapping on three sides of the beam also enhance the ultimate torsional strength. An experimental investigation is conducted to address the torsional capacity and twist of reinforced concrete beams with reinforcement only on transverse direction and strengthened with ferrocement "U" wraps (leaving the top side free from wrapping) as this situation is more common in the rehabilitation of concrete structures. The results revealed that single type of reinforcement (only on transverse
* Corresponding author.
E-mail address: beheragb@yahoo.co.in
ELSEVIER
1877-7058 © 2013 The Authors. Published by Elsevier Ltd.
Selection and peer-review under responsibility of Department of Civil Engineering, Sebelas Maret University doi: 10. 1016/j .proeng .2013.03.069
direction) is not an effective way of enhancing the torsional strength while increase in toughness is found to be marginal.
© 2013 The Authors.Published by Elsevier Ltd.
Selectirn and peer-review under responsibility of Depsrtment of Civil Engineering, Sebelas Maret University
Keywords: wrapping; torsional rigidity; reinforced concrete; torsional stiffness; torsional toughness; ultimate torque.
1. Introduction
Torsion is a basic structural action to be considered in the design. But due to its complex nature and occurrence with other basic forces, it is ignored by the designers. Increased service loads, aging of structures, Manmade havocs, natural calamities and updates in the codes have necessitated many of the structures to be retrofitted.
Due to its high tensile strength and ductility, FRPs have capacity to alter the failure mode of structural members from brittle to ductile and hence dominated the field of retrofitting material (Deifalla and Ghobarah 2005; Ghobarah 2001; Ghobarah et.al. 2002). Ferrocement has a better strain distribution across the cross section and crack arresting capacity ACI-549-1979 (Ghobarah et al. 1979) and ACI 549-R -1997 (Ghobarah et al. 1997), hence can be an alternative for FRPs' in the field of retrofitting material from the cost benefit point of view. Strengthening of the distressed concrete structures is well addressed by researchers for the basic structural actions such as axial loads, flexure and shear in past few years, but torsional strengthening has not attained much importance in research due to complex nature of the problem and its occurrence with other basic structural forces (Adrian and Riadh 2007; Behera et al. 2007). Circulatory torsion induces shear stresses on all four sides and would be well resisted by closed form reinforcement. But due to monolithic construction of beam slabs, U wrap retrofitting is the most practical solution (Saravanan and AbdeldeliBelarbi 2002; Rao et al. 2007). Experimental results and analytical models have well proven the enhancement of torsional capacity with U wrap. Hence, it is necessary primarily to investigate whether the six possible reinforcement configurations can enhance the torsional capacity and torsional toughness. Out of six possible reinforcement configurations i.e. beams with only longitudinal reinforcement, beams with only Transverse reinforcement, under reinforced Beams, longitudinally over reinforced and transversely under reinforced, transversely over reinforced and longitudinally under reinforced and completely over reinforced, in this study an experimental investigation has been taken up for only longitudinal reinforcement to identify efficacy of reinforcement configuration.
Improving the torsional behaviour both in terms of strength and ductility deserves better understanding of different possible schemes of reinforcement configuration and their relative efficiency. This paper highlights the torsional response of reinforced concrete beams with only transverse steel and with ferrocement "U" wraps.
2. Experimental Program
The experimental program consists of casting and testing of beams with only transverse reinforcement along with a ferrocement "U" wrap on three faces. Three reinforced beams of size (125 X 250) mm and 2000 mm long were cast with 25 mm thick ferrocement shell on outer perimeter for mentioned configurations. In order to force the failure in the central portion of the beam, end zones of 250mm length were heavily reinforced. To allow the beam to form two helices a length of 1500 mm was required without end stiffening portions. Hence the total length of the beam was fixed 2000 mm to allow at least two helical cracks in the central portion. Three control beams of same size without transverse reinforcement were cast to compare the results. The details of the beams are shown in Table 1.
Table 1. Details of tested Beams and control beams
SN Designation of Sectional dimensions Compressive strength (MPa) mesh
the beam Breadth Depth mortar Concrete Layers
1. BQ3N 125 250 40 35 3
2. BQ4N 125 250 40 35 4
3. BQ5N 125 250 40 35 5
4. T3N 125 250 40 35 3
5. T4N 125 250 40 35 4
6. T4N 125 250 40 35 5
2.1 Reinforcement
All the three beams are reinforced with only transverse reinforcement 8mm dia. 100 mm c/c having yield strength of 465 MPa. There was no longitudinal reinforcement in the testing zone. The control beams were cast without any transverse reinforcement.
2.2 Ferrocement Material Properties
Three control beams and three test specimens were wrapped with 3, 4 and 5 numbers of layers of galvanized square grid wire mesh of diameter 0.72mm, grid spacing 6.35mm with yield stress of 250 MPa and modulus of elasticity of 180 GPa. The required strips were taken from a roll of width 1200 mm. The mortar mix was prepared with 1:1.5 ratio (cement: sand) with water cement ratio 0.55 which gives a flow value more than 120 for easy flow of mortar into wire mesh. The local available sand is taken passing through 1.18mm sieve. The target mean strength of mortar (100 mm cube) was found to be 40 MPa.
2.3 Specimen Preparation
The mix proportion for the core concrete of test beams and control beams was 1:1.5:3 with w/c ratio 0.55. The cement was 53 grade of cement confirming to IS: 12269-1987 and the fine aggregate confirms to zone II of IS: 383-1978. Crushed hard granite passing through 20mm sieve and retaining on 4.75mm sieve was used. The cube strength of concrete was found to be 35MPa. For "U" wrap , mesh layers were cut and fabricated to required U shape for U wrapping. Ferrocement wrap was cast first and
core concrete was cast later. The process of casting was continuous and there were no construction joints. Fig. 1 shows making of ferrocement U wrap while Fig.2 to Fig.3 shows the total casting procedure. After casting of ferrocement wrap, core concrete was cast without any reinforcement in core for control beams while transverse reinforcement was kept in core portion for test specimens The end portions of all beams were heavily reinforced with cages to avoid local crushing of concrete.
Figure 1. Bending of wire mesh
Figure 2. Mould ready for placing mortar
Figure 3. Mould ready with end anchorage for placing concrete
2.4 Designation of Beams
For control beams, keeping the core concrete strength (N=35 MPa), aspect ratio (B=2.0) and grade of ferrocement matrix (Q=40 MPa) constant, number of layers (3, 4 and 5) was varied. Three control beams designated as BQ3N, BQ4N and BQ5N were cast without any transverse reinforcement in core concrete. The three test beams with aspect ratio 2,concrete strength 35 MPa, ferrocement matrix strength 40 MPa and numbers of layers as three, four and five were cast along with 8mm dia.100mm c/c and designated as T3N, T4N and T5N respectively.
3. Test Setup
Beams were cured for 28 days in a curing tank .The specimens were white washed to observe crack growth and crack pattern. The beams were tested using torsion test set up shown in Fig.4.
Figure 4. Torsion Test rig mesh
It consists of a frame to apply load and another frame to act as reaction frame. The load was applied through load frame. Rollers at reaction end were provided for allowing the beam to elongate along longitudinal direction. The loading end was supported on a roller to allow the beam to rotate under pure torsion. The twist was measured with the help of the two twist meter frames placed 500 mm apart. Proper care was taken to keep the loading lever arm perpendicular to longitudinal axis of beam to avoid bending. Neoprene pads were provided between beam sides and loading frame to avoid local crushing of concrete.
4. Results and Discussions
The results of the tested beams viz., cracking torque, ultimate torque, twist at ultimate torque, torsional stiffness and torsional toughness are presented in Table 2. The plain "U" wrapped beams showed no cracking when tested under pure torsion at cracking torque of un-wrapped plain beams. All the beams failed with a single potential crack developed in the middle of the unwrapped concrete face i.e. top face unlike the plain beams where the crack is initiated on longer face (Hsu 1984). Similar behaviour of formation of single potential crack is observed while testing plain concrete beams and plain fibrous beams by earlier researchers (Hsu 1984) and (Gunneswara 2006).
A single crack was found on the un-wrapped face having inclination of approximately 450 with the longitudinal axis of the beam. For controlled beams the top face is without reinforcement and the wrapped faces as provided with equal volume of reinforcement, crack angle should be 450 (Hsu 1984). The tested beams have same reinforcement but the on top face only transverse reinforcement was provided without longitudinal steel. The crack angle analytically is found to be 450 while the crack angle for tested and control specimen experimentally is found to be nearly equal to the expected analytical results as reported in Tab.-2.The deviation may be due to experimental errors in measurement.
On further attempts of loading beyond the ultimate, de-bonding of ferrocement layer was noticed at the interface of concrete and snapping of wires of ferrocement was also observed. The torque-twist diagram of ferrocement "U" wrapped beam is linear. This linearity ends once the torque reaches to elastic torque. Torque beyond this point of inflection is almost coincides with the onset of cracking on the specimen. The physical observation when correlated with the torque-twist behaviour gives an understanding that the stiffness has reduced after initiation of this micro-cracking. Visible crack is noticed beyond certain stage of the end of the linearity in the torque-twist diagram. That means, in between the stage from change of linearity to formation of visible crack, there could have been formation of few micro-cracks and stiffness might have been reduced. So, the micro-cracking stage is initiated from change of linearity and ends with formation of first macro crack.
The first crack was noticed on the un-wrapped face of each beam. In the beams BQ3N, BQ4N and BQ5N, the cracks were initiated in un-wrapped faces as tensile stress induced there was more than the tensile strength of un-wrapped face while stress in wrapped face was below the tensile strength the ferrocement material (Behera et al 2008). Up to ultimate loading, there was formation of new cracks on the wrapped face while at the ultimate load the cracks on the un-wrapped face were widened and got connected with cracks of ferrocement wrap face. The tested specimens T3N, T4N and T5N also behaved in the same manner.
Beyond cracking, the beams provided with only transverse reinforcement could not resist additional torque to a substantial amount but the twist at ultimate was improved. Concrete member when subjected to torsion, longitudinal reinforcement, transverse reinforcement and the concrete present in the diagonal strut resist the load. For a single type of reinforcement, as one of the load resisting elements is absent, the load carrying capacity is limited to plain beams only. Thus, the beams with single type of reinforcement (transverse only) with ferrocement "U" wrap can be analyzed as plain ferrocement "U" wrapped beams.The torque of control beams were found be nearly equal to the test beams as reported in Tab.-2.
Table 2. Experimental Results of Transversely reinforced "U" Wrapped normal Strength Beams with
control Specimen Data
Beam BQ3N BQ4N BQ5N T3N T4N T5N
Experimental torsional stiffness (KN-M2) Initial 1337 1458 1403 1434 1466 1484
Cracking 993 1021 1027 1006 1044 1006
Ultimate 993 1021 1027 839 852 814
Expt. Torque (KNm) Cracking 5.415 5.415 5.491 5.53 5.53 5.53
Ultimate 5.415 5.415 5.491 5.45 5.45 5.45
Expt. Toughness(KNm/m) 0.017 0.017 0.0177 0.0228 0.0232 3 0.026
Expt. Twist X 10-3 (rad/m) cracking 5.45 5.3 5.23 5.5 5.3 5.5
Ultimate 5.45 5.3 5.23 6.3 6.4 6.7
Crack angle (on ferrocement face) 47.58 46.54 48.78 47.23 45.95 48.67
Crack angle (on concrete face) 42.34 45.08 45.6 44 44.63 46.28
4.1 Ultimate Torque, Crack Angle
Experimental torque of control beams and tested specimens are found be sufficiently greater than that of theoretical values of plain concrete beams. This shows that experimental values are higher and hence it is evident that the plain concrete beam even it is wrapped in three sides, there is a substantial increase in torque. The reason could be attributed that the stiff ferrocement "U" wrap subjected to tensile action restrains by concrete. Thus, concrete is experiencing a prestressing effect (Saravanan 2002; Behera et al 2008). Such prestressing can improve the torque carrying capacity of plain concrete beams. The theoretical crack angle for concrete and ferrocement face should be 450 due to same volume fraction of reinforcement in jacketed faces of plain beams as square mesh was provided. The experimental crack angles for concrete un-wrapped face and ferrocement longer face were are found to be very close to 450. The crack pattern of T4N is shown in Fig.5.
Figure 5. Crack pattern of T4N
The torques and twists of beams T3N, T4N and T5N are found be nearly equal upto cracking with their control beams BQ3N, BQ4N and BQ5N respectively and beyond cracking , twist is more for transversely reinforced beams than the control beams. Torque twist response of transversely reinforced beams cannot be theoretically evaluated by Hsu's softened truss model (Hsu 1988) as it lacks one shear resisting
component i.e. longitudinal reinforcement, but can be evaluated as plain "U" wrapped ferrocement beams using Behera (2008), Mansur (2002) and Rasid (2002).When analyzed by Behera et.al.(2008) it was also found that when the stress produced in the shorter unwrapped face is equal to its tensile stress ,the tensile stress produced on the longer face is less than that of the tensile stress of the composite material of ferrocement matrix on the longer face. That is the reason why the crack is initiated on the longer
4.2 Torque
The cracking torque of all these beams T3N, T4N and T5N were found to be 5.53 kNm while cracking torque of their control beams BQ3N, BQ4N and BQ5N were 5.415 kNm, 5.415 kNm and 5.49 kNm respectively. The torque increased by 2.21%, 2.21% and 0.72% for beams T3N, T4N and T5N over their plain "U" wrapped beams BQ3N, BQ4N and BQ5N respectively. This shows that the improvement is very marginal. The beams provided with only longitudinal reinforcement shows some improvement on torsional strength over the controlled beams as reported by Behera et al. (2010).
/ g Torque Vs Twist for3 Layers
0 0.002 0.004 I' 006 0.00?
. Twist(rad/m)
Figure 6. Torque Vs Twist of BQ3N and T3N
Figure 7. Torque Vs Twist of BQ4N and T4N
Figure 8: Torque Vs Twist of BQ5N and T5N
Figure 9: Torque Vs Twist of T3N, T4N and T5N
This also conforms provision of longitudinal reinforcement enhances the torsional capacity as stated (Hsu 1984). The torque-twist response of these beams T3N, T4N and T5N were presented in the Fig- 6 to Fig- 9.
4.3 Effect of Number of Layers
In ferrocement wrapped concrete beams without any reinforcement in core (control specimens BQ3N, BQ4N and BQ5N), the most important parameters influencing the torque-twist response are number of mesh layers, strength of ferrocement mortar matrix and core concrete. When it is analyzed with layers from 3, 4 and 5, the ultimate torques are found to be 5.54 kNm for all beams without any variation. This is due to the fact that the crack is initiated on un-wrapped face for 3 layers also. Increasing the number of layers beyond three layers only increases the tensile strength of ferrocement, but unable to change the failure plane. The failure is initiated on shorter unwrapped concrete face. The difference in beams T3N, T4N and T5N to that of plain ferrocement "U" wrapped beams BQ3N, BQ4N and BQ5N were that the latter were provided with 8 mm diameter bars with 100 mm c/c.
Table 3. Analytical Results of control Beams
Beam Analytical torsional stiffness (KNm2) Analytical torque Analytical toughness (KNm/m) Analytical twist (rad/m)
Initial Cracking
BQ3N 1458 1030 5.54 0.01742 0.00537
BQ4N 1458 1041 5.54 0.0171 0.00531
BQ5N 1458 1050 5.54 0.01697 0.00525
4.4 Twist
The ultimate twist for these beams T3N, T4N and T5N were found to be 0.0063 rad/m, 0.0064 rad/m and 0.0067 rad/m respectively while the twist of BQ3N, BQ4N and BQ5N were found to be 0.00545 rad/m, 0.0053 rad/m and 0.00523 rad/m respectively. Provision of only transverse reinforcement in "U" wrapped beams cannot enhance the ultimate torque, but capable of providing better toughness due to increase in twist. Twist of tested specimens T3N, T4N and T5N were found to be increased by 15.60 %, 20.74 % and 28.10 % over their control specimens BQ3N, BQ4N and BQ5N respectively. The torque twist diagram of the beams shows that the control beams fails just after cracking while there exists a drooping portion for tested specimens which conforms their better ductility over control beams.
4.5 Stiffness
The initial stiffness of the beams T3N, T4N and T5N were found to be 1434 kNm2, 1466 kNm2 and 1484 kNm2 respectively against their predicted values 1458 kNm2 which was equal for all beams. The secant stiffness at ultimate torque for the above beams was found to be 839 kNm2, 852 kNm2 and 814 kNm2 while the secant stiffness at ultimate torque for the beams BQ3N, BQ4N and BQ5N were found to be 993 kNm2, 1021 kNm2 and 1027 kNm2 respectively.
5. Conclusion
Based on the test results and the discussions thereof following conclusions can be drawn from the present study:
a) A significant increase in torsional strength is observed with ferrocement "U" wrapped normal strength concrete beams over their plain concrete beams.
b) Ultimate torque is dependent upon the core concrete, mortar strength, mesh layers and aspect ratio combinedly.
c) The torque twist response of ferrocement "U" wrapped reinforced concrete beams was similar to the reinforced concrete beam.
d) The increase in torsional strength over the number of layers for longitudinal reinforcement is very less.
e) Single type of reinforcement i.e. only with transverse reinforcement is not capable of increasing the torsional capacity but capable of enhancing the toughness to some
extent.
Acknowledgement
The authors are grateful to the Director of National Institute of Technology, Warangal for providing financial support to this research project. The authors are also thankful to the persons those who are associated with this work directly or indirectly.
References
Adrian KYHii and Riadh Al-Mahaidi 2007.Torsional Capacity of CFRP strengthened Reinforced Concrete Beams. Journal of composites for construction ASCE, January/February , pp 71-80.
Behera GC, TDG Rao, Rao CBK 2007.Retrofitting- A State of Art. National Conference on 'Materials and Structures December 14-15,2007, NIT Warangal, A.P. INDIA, pp 280--287.
Behera G.C, TDG Rao. and Rao C.B.K.,2008. Torsional Capacity of High Strength Concrete Beams jacketed with ferrocement U wraps. Asian Journal of Civil Engineering, Vol. 9,No. 4,August 2008, pp. 411-422.
Behera GC, Rao TD Gunneswara Rao and Rao C.B.K.,2010. Torsional Strength of Ferrocement "U" wrapped beams with only Single Type of Reinforcement. Proceedings of Seventh Structural Engineering Convention,8-10 December-2010,Annamalai University, Tamil Nadu,India,Vol-II,pp-872-882.
Deifalla A and Ghobarah A 2005. Simplified Analysis For torsionally Strengthened RC Beams Using FRP. Proceedings of the Int. Symposium on bond behavior of FRP in Structures (BBFS2005) Chen and Teng (eds) 2005, International Institute for FRP in Construction, pp 377-378.
Ghobarah A 2001. FRP composites in civil engineering. proceedings FRP composites in civil Engineering 2001, CICE, Hong Kong, Balkema, London, pp. 701-712.
Ghobarah A, Ghorbel MN and Chidiac SE 2002 Upgrading Torsional Resistance of Reinf. Concrete Beams Using Fiber-Reinforced Polymer. J. of composites for Const, ASCE, Vol- 6, No.4, Nov, pp 257-263.
_''Ferrocement-Materials and Applications", ACI Committee 549, ACI Symposium
Proceedings,SP-61, American Concrete Institute, Farmington Hills, Michigan, 1979, pp195 .
_"State of art report on ferrocement" , ACI-549 committee report R-97, ACI, Detroit,1997
Gunneswar Rao TD, D RamaSeshu 2006. Torsional Response of fibrous reinforced concrete beams; Effect of single type reinforcement. Construction and Building materials, 20 (2006), pp. 187-192.
Hsu TTC 1984. Torsion of reinforced concrete. Von Nostrand Reinhold Co., New York.
Hsu TTC 1988. Softened truss model theory for shear and torsion . ACI Structural Journal, Nov-Dec 1988, pp 624-635.
Mansur MA and MM Islam 2002. Interpretation of Concrete Strength for Nonstandard Specimens. ASCE
journal of materials in civil engineering, March-April 2002, pp 151-155.
Rao CBK, TDG Rao, and Behera G 2007. Torsional Behavior plain concrete beams skinned with U wrap. North Cyprus, 14-17 November 2007, pp 303-310.
Rashid MA, Mansur MA and Paramsivam P 2002. Correlations between mechanical properties of high strength concrete. Journal of Materials in Civil Engineering, May-June , pp 230-238.
Saravanan Panchacharam and Abdeldelil Belarbi 2002. Torsional Behavior of Reinforced Concrete Beams Strengthened with FRP Composites. First FIB Congress October 13-19, 2002, Osaka, Japan, pp 1-11.
