Tensile Capacity of U-bar Loop Connections with Precast Fiber Reinforced Dowels

Transcription

1 Downloaded from orbit.dtu.dk on: Oct 15, 018 Tenile Capacity of U-bar Loop Connection with Precat Fiber Reinforced Dowel Sørenen, Jeper arrild; oang, Linh Cao; Oleen, John Forbe; Ficher, Gregor Publihed in: fib Sympoium 016: Performance-baed approache for concrete tructure Publication date: 016 Document Verion Peer reviewed verion Link back to DTU Orbit Citation (APA): Sørenen, J.., oang, L. C., Oleen, J. F., & Ficher, G. (016). Tenile Capacity of U-bar Loop Connection with Precat Fiber Reinforced Dowel. In fib Sympoium 016: Performance-baed approache for concrete tructure: Proceeding General right Copyright and moral right for the publication made acceible in the public portal are retained by the author and/or other copyright owner and it i a condition of acceing publication that uer recognie and abide by the legal requirement aociated with thee right. Uer may download and print one copy of any publication from the public portal for the purpoe of private tudy or reearch. You may not further ditribute the material or ue it for any profit-making activity or commercial gain You may freely ditribute the URL identifying the publication in the public portal If you believe that thi document breache copyright pleae contact u providing detail, and we will remove acce to the work immediately and invetigate your claim.

2 TENSILE CAPACITY OF U-BAR LOOP CONNECTIONS WIT PRECAST FIBER REINFORCED DOWELS Jeper. Sørenen, Linh C. oang, John F. Oleen and Gregor Ficher Department of Civil Engineering, Technical Univerity of Denmark, Kg. Lyngby, Denmark ABSTRACT Thi paper decribe an invetigation of the tenile capacity of in-itu cat U-bar loop connection between precat concrete element. The baic idea i to introduce a mall precat cylindrical dowel of fiber reinforced mortar that fit into the bend diameter of the overlapping U-bar. The remaining part of the connection i cat in-itu with a regular mortar, which then encapulate the precat dowel. Different dowel configuration have been invetigated, including the ue of teel or ynthetic fiber with or without lacer bar placed within the precat dowel. The experimental reult how that ue of a precat fiber reinforced dowel perform at a lightly lower load level, a compared to a connection grouted olely with regular mortar and reinforced with the ame amount of tranvere reinforcement. owever, the load-diplacement repone of pecimen with a fiber reinforced dowel i cloer to ideal ductile behavior than that of the pecimen grouted with regular mortar. The experimental reult of the tenile tet are compared with calculation baed on an upper bound platicity model and atifactory agreement ha been obtained. Keyword: Loop Connection, Tenile Capacity, Fiber Reinforced Concrete, Upper Bound Method 1. Introduction Contruction with precat concrete component i known to be fat, economic, and labor efficient. The challenge of contructing with precat component i in the aembly detail where tructural continuity and integrity have to be enured, typically in a narrow contruction zone. The connection deign may be contituted by lap plice, U-bar loop, rebar welding or imilar reinforcing detail grouted by in-itu cat concrete or mortar. Common for all olution i the demand for eay aembly, minimization of labor work, and tructurally well performing olution. In ome application, the U-bar loop connection i preferred, a it require the leat work within the mallet contruction zone and the performance of the connection i increaed due to the uperior anchorage propertie of the loop bar (Mattock 1994). Thi connection type can be ued to tranfer tenile force, bending moment or hear force (FIB 008) and the deign i widely applicable from connection in bridge deck at intermediate pier (Jørgenen & oang 013) to connection of vertical hear wall in building. In bridge deck application over intermediate pier, the demand for ufficient tenile capacity i obviou a the connection appear in the tenile zone of the compoite bridge girder. owever, for hear wall ytem, where the joint interface often i indented, the tenile capacity of the connection i required to activate the full load carrying capacity of the indentation and to enure connectivity of the joint during hear deformation. The loop bar connection conit of bent U-bar protruding from the precat element and overlapping the loop of the adjacent element. To obtain a trong connection, the overlapping loop area i reinforced and the joint area i grouted on-ite with a concrete or a mortar. The connection i preferably deigned for reinforcement yielding and not concrete failure. A relatively limited amount of work ha been publihed on the behavior of loop connection in tenion. Leonhardt et al. (1973) wa the firt to publih tet reult, invetigating the required overlapping area of the U-bar in order to obtain yielding of the U-bar and not concrete failure of the grout material. The tet were deigned a a -to- connection where the overlapping loop were placed cloely together. The deign i an example of a ymmetric unit (Gordon 006) where the main reinforcement i placed ymmetrically about the line of loading. Gordon teted both ymmetric 3-to-4 connection and non-ymmetric connection and reported that non-ymmetric pecimen

3 are unuitable for pure tenion due to in-plane rotation of the pecimen. The remaining experimental work on ymmetric connection include the work of Ong et al. (006) who teted 1-to- deign and compared to a theoretical trut-and-tie model and Jørgenen & oang (013) who teted -to-3 pecimen and provided an upper bound platic model. In the literature it i generally recognized that the overlapping area of the U-bar loop, the pacing of the loop, the trength of the joint material and, finally, the tranvere reinforcement in the overlapping area influence the tenile capacity. Fig. 1. Geometry of tet pecimen with a fiber reinforced dowel placed in the overlapping loop area The tranvere reinforcement i of pecial interet a the reinforcement area, the yield trength and the anchoring condition influence the tenile capacity. Gordon (006) teted a ingle connection with teel fiber reinforced concrete a replacement for rebar concluding that the fiber alone did not enure a ductile behavior a the capacity decreaed once the concrete material failed. Thi paper decribe the tenile capacity of a ymmetric -to- loop connection, ee Fig. 1. The deign invetigate the ue of a precat fiber reinforced dowel a reinforcement of the circular overlapping loop area while the remaining connection i grouted with regular low trength mortar. In addition, the paper preent a model for aement of the ultimate capacity of the loop connection regarding concrete failure. The model i baed on the upper bound theorem of rigid-platicity accounting for the geometry of the connection.. Experimental invetigation The experimental program conited of 7 pecimen invetigating the tenile capacity and the loaddiplacement relation of the reinforced loop connection. The overlapping loop area wa reinforced with a precat fiber reinforced dowel with and without lacer reinforcement in hape of a double T-headed bar. A erie without fiber reinforcement erved a a reference. The program wa divided into three erie, a given in Table 1. Table 1. Overview of experimental program Serie No. of pecimen Precat Dowel Fiber type Shape Length [mm] Diameter [mm] R 9 No P 8 Ye Polypropylen Random Random - S 10 Ye Steel Straight Reinforcement configuration and material propertie Table contain an overview of the teted pecimen with their material propertie. The fiber reinforced concrete mix wa deigned with mall aggregate ize due to the limited ize of the precat dowel. For the

4 polypropylene fiber mix an aggregate ize of maximum mm wa ued and for the teel fiber mix and the regular mortar mix maximum 4 mm aggregate were ued. The compreive trength,, given in the table i related to the trength of the joint mortar for Serie R and related to the trength of the dowel mortar for the remaining erie. The compreive trength of the concrete in the prefabricated element were in the order of 60 MPa and the grouting mortar in Serie P and S were kept lower than the trength of the dowel a it wa ued a filling material outide the overlapping loop area; in thi cae,, wa around 16 MPa. Table. Specimen detail and tet reult Specimen Dowel Lacer reinforcement Loop reinforcement D [mm] f c [MPa] L [mm] f yl [MPa] [mm] f y [MPa] P u,tet [kn] R1 a,b,c / 77.6 / 78.4 R14 a,b,c / 79. / 88.0 R16 a,b,c / 97. / P a,b,c 56± / 33.0 / 36.4 P1 a,b,c 56± / 53.0 / 53.8 P16 a,b 56± / 75.9 S a,b,c 56± / 7.9 / 31.1 S1 a,b,c 56± / 58.0 / 63.7 S14 a,b 56± / 70.1 S16 a,b 56± / Geometry and fabrication Each pecimen conited of two precat concrete element linked by a loop connection, ee Fig. 1. In each precat element a reinforcement bar wa placed centrally in order to achieve precie centric loading condition. The precat dowel were produced in advance. The mold ued for the precat dowel conited of polytyrene plate with mechanically drilled hole. The procedure of drilling frayed the edge and made the urface of the dowel rough, ee Fig. (a). (a) Precat dowel (b) Poitioning of dowel (c) Without precat dowel Fig.. Example of (a) a precat dowel after demolding from polytyrene mold, (b) a precat dowel placed in the overlapping loop area and (c) the poitioning of the lacer reinforcement in Serie R

5 The urface improved the anchorage of the dowel to the remaining joint mortar. The dowel wa produced with a diameter of 56± mm leaving a tolerance of 4± mm to the internal bend diameter of the loop bar. The overlapping area of the U-bar loop and the connection width were kept contant. For the pecimen cat with regular mortar, Serie R, the lacer bar wa attached to the outer loop bar; ee Fig. (c)..3 Setup and teting procedure The tet were performed in a 500 kn ervo hydraulic teting machine applying uniaxial tenion to the centered bar in each of the two precat element. The tet were performed a a diplacement controlled quai-tatic tet with a rate of 0.5 mm/min followed by 4 mm/min after entering the pot peak region. The pecimen were poitioned vertically, and in the lower end the rib were removed from the reinforcement bar in order to enure ufficient anchorage between the hydraulic grip and the bar. At the top of the pecimen a connecting device wa applied to accommodate any unintentional eccentricitie and to eae the intallation of the pecimen in the teting machine, ee Fig. 3. The load wa applied through a pherically haped nut which automatically centered the load. (a) Device accounting for eccentricitie (b) Element mounted Fig. 3. Tet device including (a) geometry and (b) tet device with pecimen prepared for teting The relative diplacement of the loop connection were monitored on the urface of the pecimen uing diplacement tranducer placed centrally on each ide of the pecimen. On the remaining face high reolution digital camera were ued to record photo for digital image correlation (DIC). owever, a a complex tre ditribution developed in the loop connection, no accurate prediction of the cracking behavior could be meaured on the urface. Conequently, the reult of the digital image correlation were primarily ued for verification of the diplacement tranducer meaurement. 3. Experimental reult The experimental reult are reported a load-diplacement curve, where the axial diplacement i taken a an average of the meaurement from the diplacement tranducer. The diplacement meaurement from the tranducer were confirmed by the digital image correlation. The general behavior of all pecimen wa a relatively tiff elatic repone up to a certain point where cracking became viible on the urface of the connection. The reult for the pecimen reinforced with precat fiber reinforced concrete without lacer reinforcement identified thi load level a the cracking load of the concrete material, ee Fig. 4. The behavior wa common to both pecimen with precat dowel and with regular mortar, and thereby it contituted the point where the lacer reinforcement wa activated. The ultimate load level reached after the initial cracking wa dependent on the amount of lacer reinforcement. In general it wa een that an increaing reinforcement ratio led to an increaed ultimate load. Figure pecimen with a precat dowel the average of the tet reult are alo given a bold curve. The average

6 of the reult from the three pecimen without lacer reinforcement i indicated a dahed line. Beide the identification of the cracking load, it can be een in the figure that the peak load of the pecimen with a precat dowel were not a high a for the connection cat entirely with mortar, however, the reidual capacity after peak wa comparable a the fiber reinforced pecimen experienced a far le pronounced drop in load-deflection repone at peak. (a) nt (b) Fig. 4. Tet reult for pecimen with lacer reinforcement

7 From the tet it wa alo found, that the two fiber reinforced configuration, Serie P and S, performed imilarly. It wa found that the pecimen without lacer reinforcement experienced a reduction in load level after the cracking load, wherea the lacer reinforced fiber dowel maintained a relatively contant load level. (a) With precat dowel (b) Without precat dowel Fig. 5. Failure pattern for (a) a pecimen with a precat dowel reinforced with polypropylene fiber and (b) a loop connection fully grouted with mortar and tranverely reinforced with teel Figure 5 contain picture of the general failure pattern oberved at a late tage of teting. For pecimen reinforced with a precat dowel, the ultimate failure wa governed by concrete failure in the dowel in term of crack extending from the contact point of the outermot loop of an element to the contact point of the outermot loop protruding from the adjacent precat element; ee Fig. 5(a). Before thi failure, the urrounding mortar had already palled off. For the pecimen with a connection olely grouted with mortar, the failure pattern wa imilar; however the geometry wa much more complex a the mortar outide the overlapping area of the loop influenced the crack pattern. Figure 5(b) demontrate the crack appearance between the loop while alo revealing the correlation to the cracking of the remaining mortar. 4. Failure mechanim From the oberved failure, ee Fig. 5, and the correponding failure load, it i indicated that the capacity of the joint material regarding concrete failure i le than the tenile capacity of the reinforcement loop. Furthermore it i indicated that the concrete yield line between the loop bar dictate the capacity of the connection. A the overlapping loop area determine the total area of the diagonal yield line, the capacity of the connection i alo governed by thi area, a uggeted by Jørgenen & oang (013). In the following, an upper bound platicity model for concrete failure of the examined connection will be preented. The concrete and the reinforcement loop are aumed to be rigid-perfect platic material obeying the aociated flow rule. Furthermore the reinforcement i aumed to carry only axial load, the concrete i aumed to have no tenile trength and the failure urface are a a implification aumed a plane, repreented by a linear yield line in a D repreentation. In Fig. 6(a) the implified two-dimenional failure mechanim i hown and in Fig. 6(b) the relative diplacement for the yield line are indicated. Figure 6(c) repreent the characteritic area adopted in the calculation.

8 (a) Failure mechanim (b) Relative diplacement (c) Concrete core area Fig. 5. Simplified failure mechanim (a) for tenile action on a loop connection, (b) a repreentation of the relative diplacement in each yield line and (c) the characteritic concrete area The relative diplacement are related in the following way: u u co and u in l u (1) t inclination of the yield line related to vertical: arctan () 4.1 Upper bound olution The load carrying capacity i determined from the work equation, where the rate of external work and the rate of internal work performed in the failure mechanim are equated. The rate of external work performed by the axial load i given by: WE Nu l (3) The rate of internal work i contituted by contribution from diipation in the concrete yield line and in the lacer reinforcement croing the yield line. The contribution from the lacer reinforcement i given a: W A S I L f yl u t The diipation in a concrete yield line for a plane train problem i given by (Nielen & oang 011): C 1 WI fc(1 in) Ai u (5) Where A i i the area of the yield urface, for concrete. It i neceary to introduce the effectivene factor a concrete i not a perfect platic material. The factor ha not been etablihed for the pecific problem at hand, however in Jørgenen & oang (013) the factor applied for beam hear problem wa adopted. When dowel action i ignored and (4)

9 the axial length of the overlapping area i choen a the characteritic hear length,, the effectivene factor appear a: , f c in MPa and in m. (6) f c A the concrete i aumed to have no tenile trength, the contribution from the concrete yield line outide the overlapping area i zero a the angle of the relative diplacement vector i 90, ee Fig. 5(c) and Eq. (5). The capacity will thereby olely depend on the two diagonal yield line between the adjacent loop bar protruding from each precat element. A the hape of the overlapping area i circular, the plane area of the yield line i given a: Ac A (7) co 4 co Equating Eq. (3) with (4) and (5) the following upper bound olution i found: N 1 in L A f tan c c co co Where the mechanical reinforcement ratio of the lacer reinforcement i introduced: AL f yl L Ac fc The minimum of the upper bound olution i found for the following diplacement angle: L 1 arcin 1 From the optimal angle, it can be found, that the tenile capacity of the loop connection will be zero when no lacer reinforcement i preent. Thi i of coure due to the fact that the concrete i aumed to have no tenile trength. From the geometry of the connection it can alo be found, that an additional retriction i impoed on the diplacement angle. The diplacement angle cannot be maller than the inclination of the yield line, a uch a ituation would caue area II to move inward, ee Fig. 5(b). Conidering thee retriction, the capacity can be ummarized a: (8) (9) (10)

10 N A f c c L 1 1 L tan co 1 tan L tan A the capacity of the connection i governed either by the concrete failure or tenile yielding of the loop bar, the capacity for the preented loop deign will be given by: N U N Concrete failure a given in Eq. (11) (a) min (1) 4A f y Yielding of U-bar in tenion (b) A it can be een, the olution i dependent on the internal angle of friction for the cementitiou material ued to fabricate the dowel or to grout the joint. If an internal angle of friction i choen a the value for regular concrete, tan() =3/4, the olution appear equal to the finding of Jørgenen & oang (013). owever, a the material ued for prefabrication of fiber reinforced dowel contain maller aggregate ize than regular concrete, the olution mut be able to account for the actual value of the internal angle of friction of the grouting material. 5. Comparion of theory with tet reult Figure 6 contain a comparion between the peak load oberved in the experimental tet and the model preented. In the calculation a mean compreive trength of 39.5 MPa ha been ued and the ame characteritic length of =0.076 m ha been ued for all pecimen in the determination of the effectivene factor. The effectivene factor wa calculated to The internal angle of friction wa, Serie P and S, 0 for the pecimen without precat dowel, Serie R. For the pecimen with fiber reinforced dowel, the urrounding mortar did crack and break off before the ultimate load wa reached; hence it i aumed that the mortar did not contribute to the capacity at ultimate. A the precat dowel itelf had a maller diameter than the circular overlapping area, the inclination of the yield line,, would be larger than preented in Eq. (). For a more accurate trength determination, the characteritic height,, wa replaced by the diameter of the dowel, D, changing the concrete area A c and thereby the mechanical reinforcement degree. Thi wa conidered reaonable a the diameter of the dowel wa cloe to the diameter of the overlapping area of the U-bar. owever, Fig. 6(a) how that the calculation underetimate the capacity, which i regarded the fact that the improved tenile capacity of the fiber reinforced concrete wa not included in the model. Furthermore, the effectivene factor for the fiber reinforced material might attain a larger value than what ha been utilized in thee calculation. (11)

11 (a) Serie P and S (b) Serie R Fig. 6. Comparion of tenile capacity for (a) pecimen reinforced with precat fiber reinforced dowel and (b) pecimen with joint grout of regular mortar The internal angle of friction i an important parameter in the theoretical calculation. For the fiber to fit the tendency oberved in the experimental tet, however a the aggregate ize are mall in the fiber reinforced concrete the internal angle of friction might in fact be maller. Further tudie are needed to verify thi value. For the pecimen grouted with regular mortar with a maximum aggregate ize of 4 mm, the internal angle of friction i believed to be reduced compared to 37 o, which i the value normally adopted for regular concrete with larger aggregate ize. The choen value of 30 capture the development oberved in the tet, however a the number of tet are limited, further reearch on thi parameter i needed to verify current initial reult. Furthermore it i well known that a mortar material behave in a more brittle manner compared to regular concrete a the effect of aggregate interlocking i reduced. Thi could be accounted for by reducing the effectivene factor; however the current initial tet do not warrant uch a need. 6. Concluion The experimental program of the invetigation reported on in thi paper conited of tenion tet on loop connection reinforced with precat fiber reinforced dowel. The tet revealed a pronounced ductile behavior of pecimen reinforced with precat fiber reinforced dowel compared to a connection grouted with regular mortar. The ultimate load level wa lower than that in reference tet completely grouted with mortar; however the load diplacement development wa clearly ductile. Two different fiber material were examined and no ditinctive difference wa oberved between the two in term of load-diplacement repone. Tet on pecimen with a fiber reinforced precat dowel without additional lacer reinforcement, revealed the tranition point where additional teel reinforcement wa activated. In general, the preented upper bound olution capture the tendency in the tet regarding the concrete failure of the concrete core for increaing tranvere reinforcement area. The tet on pecimen with precat fiber reinforced dowel reveal a maller ultimate capacity compared to a grouting without precat dowel. Thi i accounted for in the model by adjuting the area of the yield urface to the cro ectional area of the precat dowel. Thi predict a conervative capacity a the tenile trength of the fiber reinforced compoite i diregarded. Thi adjutment hould only be ued for precat dowel with a diameter in the ame order of magnitude a the diameter of the overlapping loop area. For the pecimen without precat dowel, the model capture the tet reult well, when chooing an internal angle of friction for the mortar of 30.

12 Acknowledgement The experimental reearch preented in thi paper wa financially upported by the COWI Foundation. The tet element were produced at CR Concrete and the experimental work wa conducted with upport from M.Sc. J. Svejgaard, M.Sc. L. Øvrelid, B.Sc. L. Anderen and B.Sc. M. aan. The author gratefully acknowledge thee valuable contribution. Reference FIB, bulletin 43, 008. Structural connection for precat concrete building Structural connection for precat concrete building, Gordon, S.R., 006. Joint for Precat Deck in Steel Concrete Compoite Bridge. eriot Watt Univerity. Jørgenen,.B. & oang, L.C., 013. Tet and limit analyi of loop connection between precat concrete element loaded in tenion. Engineering Structure, 5, pp Leonhardt, F., Walther, R. & Dieterle,., Veruche zur Ermittlung der Tragfähigkeit von Zugchlaufentößen, Bulletin no. 6, Deutcher Auchu für Stahlbeton. Mattock, A.., Effectivene of Loop Anchorage for Reinforcement in Precat Concrete Member. PCI Journal, 39(6), pp Nielen, M.P. & oang, L.C., 011. Limit Analyi and Concrete Platicity 3rd ed., CRC Pre, Taylor & Franci Group. Ong, K.C.G., ao, J.B. & Paramaivam, P., 006. A trut-and-tie model for ultimate load of precat concrete joint with loop connection in tenion. Contruction and Building Material, 0(3), pp