interface shear transfer of lightweight-aggregate ... journal/2016/march-april/interface...

March–Apr i l 2016 | PCI Journal38

The shear friction design provisions presented in the American Concrete Institute’s (ACI’s) Building Code Requirements for Structural Concrete (ACI

318-14) and Commentary (ACI 318R-14)1 and the PCI Design Handbook: Precast and Prestressed Concrete2 are based on physical test data, though most published test data pertain to concrete that is normalweight. Lightweight-aggregate concretes are often considered for use when it is desirable to reduce member weight, such as in precast concrete construction to reduce transporta-tion costs, and enhance fire resistance. Precast concrete elements commonly incorporate connections that are designed based on the shear friction concept to transfer forces across an interface. Previous studies on shear fric-tion have shown that interface condition, reinforcement ratio, concrete strength, and concrete type (normalweight, sand-lightweight, or all-lightweight) influence the shear transfer strength.3–10 However, few studies have investi-gated the direct shear transfer of structural lightweight-aggregate concretes, especially for conditions in which concretes are cast at different times, that is, cold joint conditions. Lightweight aggregates commonly used in the production of structural lightweight concretes include expanded shale, expanded slate, and expanded clay. These aggregates can have different unit weights and mechani-cal properties, depending on the aggregate source and production process. The overall goal of this experimental

■ This paper presents the results from the second phase of an ongoing investigation of the direct shear transfer across an interface of lightweight-aggregate concretes.

■ The lightweight concretes were made with different lightweight-aggregate materials (expanded shale, expanded clay, or expanded slate) and tested with the following variables: con-crete type (normalweight, sand-lightweight, or all-lightweight), lightweight-aggregate material, surface preparation of the shear interface, reinforcement ratio, and crack interface condition.

■ Shear strengths computed by ACI 318-14 Eq. (22.9.4.2) and PCI Design Handbook Eq. (5-32a) using the coefficient of friction µ approach were conservative for the sand-lightweight and all-lightweight monolithic and cold joint specimens in this study.

Interface shear transfer of lightweight-aggregate concretes with different lightweight aggregates

Lesley H. Sneed, Kristian Krc, Samantha Wermager, and Donald Meinheit

39PCI Journal | March–Apr i l 2016

taken as 0.85 for sand-lightweight concrete.1,2 Table 1 also summarizes upper limits on the values of nominal shear strength for the ACI 318-14 and PCI Design Handbook shear-friction design provisions.

The nominal shear strength given by Eq. (22.9.4.2) and (5-32a) can also be expressed in terms of nominal shear stress vn (Eq. [1]), where the reinforcement ratio ρ equals Avf /Ac, and Ac is the area of concrete shear interface (Ac is denoted as Acr in the PCI Design Handbook):

vn = µρfy (1)

An alternative approach to designing the shear friction reinforcement is presented in the PCI Design Handbook, in which µ in Eq. (5-32a) is replaced with an effective co-efficient of friction µe. However, this alternative approach is not applicable for certain crack interface conditions, namely cases 3 and 4 in Table 1, or when load reversals occur.

This paper presents the results from the second phase of an ongoing investigation of the direct shear transfer across an interface of lightweight-aggregate concretes. Results from the first phase of the study were presented by Shaw and Sneed,10 who studied the shear transfer strength of lightweight-aggregate concretes made with expanded shale lightweight aggregate and a cold joint interface condition. In the second phase of the study, the lightweight-aggregate concretes were made with different lightweight-aggregate materials (expanded shale, expanded clay, or expanded slate) and different interface conditions (monolithic or cold joint). In this paper, the applicability of Eq. (22.9.4.2) and (5-32a) to lightweight-aggregate concrete with dif-ferent lightweight aggregates is examined. The use of the effective coefficient of friction µe approach in the PCI Design Handbook will be discussed in a future paper by the authors.

study was to extend previous research and examine the applicability of the shear friction design model for differ-ent types of lightweight-aggregate concretes with various crack interfaces.

Current ACI 318-14 and PCI Design Handbook shear-friction design provisions present a similar approach based on the simplified model originally proposed by Birkeland and Birkeland,11 in which the nominal shear strength Vn is computed as a function of the coefficient of friction µ, the area of shear reinforcement across the shear plane Avf, and the yield stress of reinforcement fy. In ACI 318-14, the nominal shear strength Vn across the assumed shear plane is given by Eq. (22.9.4.2):

Vn = µAvf fy (22.9.4.2)

PCI Design Handbook Eq. (5-32a) is similar, though written in terms of Avf required to resist the factored shear force Vu. However, the equation can be rearranged into the same equation by substituting Vn with Vu/ϕ where ϕ is the strength reduction factor.

A

Vvf

u=fyμϕ

(5-32a)

In Eq. (22.9.4.2) and (5-32a), the coefficient of friction µ is intended to globally account for friction between the surfaces of the crack interface, shearing of protrusions, and dowel action of the reinforcement. The value of µ is taken to be a function of the crack interface condition and the concrete type (Table 1). The modification factor λ for con-crete type is intended to account for reduced values of the mechanical properties of lightweight-aggregate concrete relative to normalweight concrete of the same compressive strength. The value of λ is taken as 1.0 for normalweight concrete and 0.75 for all-lightweight concrete and may be

Table 1. Coefficient of friction and maximum values of nominal shear strength for different interface conditions

Case Crack interface condition μ ACI 318-14 maximum Vn, lb-in.PCI Design Handbook maximum Vu /ϕ,

lb-in.

1 Concrete to concrete, cast monolithically 1.4λ For normalweight concrete 0.2 fc

'Ac ≤ (480 + 0.08 fc' )Ac ≤ 1600Ac

For all other cases 0.2 fc

'Ac ≤ 800Ac

0.30λ Acr ≤ 1000λAcr

2 Concrete to hardened concrete with roughened surface

1.0λ0.25λfc

'Acr ≤ 1000λAcr

3 Concrete placed against hardened concrete not intentionally roughened

0.6λ

0.2 fc'Ac ≤ 800Ac

0.20λfc'Acr ≤ 800λAcr

4 Concrete to steel 0.7λ 0.20λfc'Acr ≤ 800λAcr

Note: Ac = area of concrete shear interface (ACI 318-14); Acr = area of concrete shear interface (PCI Design Handbook); fc' = 28-day concrete com-

pressive strength; Vn = nominal shear strength; Vu = ultimate shear force; λ = modification factor reflecting the reduced mechanical properties of lightweight concrete relative to normalweight concrete of the same compressive strength; µ = coefficient of friction; ϕ = strength reduction factor. 1 in. = 25.4 mm; 1 lb = 4.448 kN.


Concrete properties

Seven different concrete mixtures were designed in this study: one normalweight concrete; three sand-lightweight concretes with expanded shale, expanded slate, or ex-panded clay coarse aggregates and normalweight fine aggregates; and three all-lightweight concretes with either expanded shale, expanded slate, or expanded clay coarse and fine aggregates.

Normalweight coarse aggregate was crushed dolomite from the Jefferson City formation readily available in Mis-souri. Missouri River sand was used as the fine aggregate in the normalweight and the sand-lightweight concretes. Table 2 lists the sources of the different lightweight ag-gregates. It also lists values of loose bulk density (ASTM C2912) and specific gravity (ASTM C12713/12814) for the aggregates used in this study. Wetted surface-dry condi-tions for lightweight aggregates, analogous to saturated surface-dry conditions for normalweight aggregates, were established using the provisional surface (metal pan) and paper towel tests in ASTM C128.

The normalweight coarse aggregate gradation used was 100% passing the 1 in. (25.4 mm) sieve and less than 5% passing the no. 8 (2.36 mm) sieve. For the sand-lightweight and all-lightweight concretes, the goal of the mixture proportioning was to obtain a consistent gradation of the combined lightweight aggregates for direct comparison of the results. However, the relative proportions of coarse and fine aggregates might not be typical, and the mixtures designed in this study may have slightly different densities than those that would have been achieved otherwise.

The expanded shale coarse aggregate gradation used in the production of the sand-lightweight concrete was 3⁄8 in. (9.5 mm) × no. 8 (2.36 mm) sieve. The all-lightweight concrete with expanded shale was made with a premixed

Experimental program

Test variables

The experimental program included 52 push-off specimens used to investigate the direct shear transfer of different types of concretes with different interface conditions. The main objective in selecting the variables in the test matrix was to extend the research conducted in the first phase of the project10 and fill in gaps in the literature with respect to sand-lightweight concrete, all-lightweight concrete, and the condition of the interface of concretes cast at different times—that is, a cold joint condition.

The selected test variables included concrete type, light-weight-aggregate material, shear interface surface condi-tion, and reinforcement ratio. In this paper, the term con-crete type refers to either normalweight, sand-lightweight, or all-lightweight concrete, where each type is designated by its aggregate composition. Interface conditions included monolithic uncracked concrete, monolithic precracked con-crete, cold joint roughened (1⁄4 in. [6 mm] amplitude), or cold joint smooth. In the PCI Design Handbook, the mono-lithic uncracked and precracked interface is represented as a case 1 condition, the cold joint roughened interface represents a case 2 condition, and the cold joint smooth interface represents the lower-bound condition of case 3. Table 1 summarizes these crack interface conditions.

Monolithic specimens were made of only normalweight concrete, shale sand-lightweight concrete, or shale all-lightweight concrete to extend the first phase of the study10 to the monolithic interface condition. Cold joint specimens were made with shale or clay lightweight aggregates to extend the first phase of the study10 to different lightweight-aggregate materials and reinforcement ratios. Additional discussion on the test matrix is included in the test speci-men description section.

Table 2. Description of aggregates

Aggregate Origin Density,* lb/ft3 Specific gravity† Size

Normalweight coarse Rolla, Mo. 99 2.63 3⁄4 in. to no. 8

Normalweight fine Jefferson City, Mo. 110 2.55 No. 8 to 0

Shale, coarse New Market, Mo. 44 1.35 3⁄8 in. to no. 8

Shale, coarse/fine premix New Market, Mo. 54 1.69 3⁄8 in. to 0

Clay, coarse Livingston, Ala. 33 1.30 3⁄8 in. to no. 8

Clay, fine Livingston, Ala. 40 1.42 No. 8 to 0

Slate, coarse Gold Hill, N.C. 52 1.60 3⁄8 in. to no. 16

Slate, fine Gold Hill, N.C. 60 1.75 No. 4 to 0

* Loose bulk density, ASTM C29† Bulk specific gravity saturated surface-dry, ASTM C127/C128Note: no. 4 = 4.75 mm; no. 8 = 2.36 mm; no. 16 = 1.18 mm; 1 in. = 25.4 mm; 1 ft = 0.305 m; 1 lb = 0.454 kg.


Test specimen description

The specimen geometry was designed to be similar to that used in the Shaw and Sneed10 study so that a direct com-parison of results could be made with previous tests. This specimen geometry was also similar to that used by Hof-beck et al.,4 Mattock and Hawkins,5 Mattock,6 Mattock et al.,7 and Kahn and Mitchell.9 The shear plane was 11.0 in. (279 mm) long and 4.5 in. (110 mm) wide with an area of concrete shear interface Ac of 49.5 in.2 (31,900 mm2). Shear reinforcement consisting of no. 3 (10M) closed-tie, deformed reinforcing bar stirrups was provided normal to the shear plane for all specimens. The reinforcement ratios ρ were 0.009, 0.013, 0.017, and 0.022. The stirrups were ASTM A61521 Grade 60 (420 MPa) with a measured yield strength fy of 72.2 ksi (498 MPa). Figure 1 shows test specimen dimensions and reinforcement.

Figure 2 shows custom-built formwork that was used to construct the specimens. All concrete specimens in the same series were cast on the same day. The monolithic specimens were cast on their side faces in one lift. The cold joint specimens were cast in two stages. For the two-stage casting, after casting the first half of the specimen, the shear interface was troweled smooth. Approximately four hours after casting, the interface roughening was completed on specimens that were designated to have a roughened interface.

Roughening was accomplished by scoring 1⁄4 in. (6 mm) deep indentations into the surface of the shear interface in the direction perpendicular to the direction of loading. The surface roughness was measured using a digital caliper at several random locations on the shear interface. After roughening was completed, the interface was cleaned with compressed air. The second half of the specimen was cast

3⁄8 in. × 0 gradation. The expanded slate coarse and fine aggregates were provided in gradations of 3⁄8 in. × no. 16 (1.18 mm) and no. 4 (4.75 mm) × 0, respectively. For the all-lightweight concrete, a mixture of 30% coarse aggre-gate and 70% fine aggregate by weight was used because it produced a gradation similar to that of the expanded shale gradation. The expanded clay coarse aggregate provided in a 3⁄8 in. × no. 8 gradation was used in the production of the sand-lightweight concrete. The fine aggregate was provided in a no. 8 × 0 gradation. For the all-lightweight concrete, a mixture consisting of 55% coarse aggregate and 45% fine aggregate by weight was used because it was similar in gradation to the expanded shale gradation. Additional information about the lightweight-aggregate properties and gradings is summarized in Krc15 and Wermager.16

The target concrete compressive strength of each mixture was 5000 psi (34 MPa). Type I/II portland cement was used, with a water-cement ratio ranging from 0.43 to 0.59. Table 3 summarizes the concrete mixture proportions. In producing the sand-lightweight and all-lightweight concretes, the lightweight aggregates were saturated for a minimum of 48 hours prior to casting the specimens. Tables 4 and 5 present fresh and hardened properties of each concrete, respectively. Values of concrete density reported in Table 4 correspond to the unit weights mea-sured on the fresh concrete according to the procedure outlined in ASTM C138.17 Table 5 presents hardened concrete properties, including the 28-day compres-sive strength fc

' (ASTM C123118), splitting tensile strength fct (ASTM C49619), and modulus of elasticity Ec (ASTM C46920). It is interesting to note that the measured values of Ec for the all-lightweight concrete with shale and slate aggregates are similar to that of the sand-lightweight with clay aggregate, and Ec appears to be dependent on concrete density rather than concrete type.

Table 3. Concrete mixture proportions

Concrete typeLightweight

aggregate material

Mixture design quantities, lb/yd3

Coarse aggregate Fine aggregate Water Type I/II cement w/c

Normalweight* n/a 1728 1302 305 517 0.59

Sand-lightweight†

Shale 834 1498 281 535 0.53

Slate 975 1125 265 530 0.50

Clay 692 1251 263 612 0.43

All-lightweight‡

Shale§ 1885 260 610 0.43

Slate 528 1233 378 801 0.47

Clay 692 556 263 796 0.46

* Normalweight concrete coarse and fine aggregate satisfied ASTM C33† Sand-lightweight concrete coarse aggregate was ASTM C330, fine aggregate was ASTM C33‡ All-lightweight concrete coarse and fine aggregate satisfied ASTM C330§ All-lightweight expanded shale aggregate was premixed by the manufacturerNote: n/a = not applicable; w/c = water-cement ratio. 1 yd = 0.914 m; 1 lb = 0.454 kg.


Table 4. Fresh concrete properties

Concrete type Lightweight-aggregate material Density,* lb/ft3 Air,† % Slump,‡ in.

Normalweight n/a 148 2.5 5.5

Sand-lightweight

Shale 117 2.0 4.25

Slate (S-SL-CJ-XX-09 series) 117 1.5 2.0




Clay (S-CL-CJ-XX-09 series) 105 2.5 1.25




All-lightweight

Shale 108 3.0 2.5

Slate 106 2.8 6.5

Clay 88 2.5 0.5

* Density of freshly mixed concrete, ASTM C138† Gravimetric method used for normalweight concrete, ASTM C138; volumetric method used for lightweight concrete, ASTM C173‡ ASTM C143Note: n/a = not applicable. 1 in. = 25.4 mm; 1 ft = 0.305 m; 1 lb = 0.454 kg.

Table 5. Hardened concrete properties

Specimen series Target fc', psi fc

' ,* psi fc' at test day,* psi fct,† psi fct,† fc

' Ec,‡ psi

N-MO-XX-13 5000 4840 4840 420 6.0 3,900,000

S-SH-MO-XX-13 5000 4770 4770 460 6.7 3,300,000

A-SH-MO-XX-13 5000 4700 4700 515 7.5 2,650,000

S-SL-CJ-XX-09 5000 5380 5380 595 8.1 3,300,000

S-SL-CJ-XX-13 5000 5570 5570 570 7.6 3,500,000

S-SL-CJ-XX-17 5000 4950 4950 670 9.5 3,050,000

S-SL-CJ-XX-22 5000 5000 5000 445 6.3 3,450,000

A-SL-CJ-XX-13 5000 4380 4380 420 6.3 2,450,000

S-CL-CJ-XX-09 5000 4770 4770 340 4.9 2,500,000

S-CL-CJ-XX-13 5000 4640 4640 360 5.3 2,650,000

S-CL-CJ-XX-17 5000 4550 4550 410 6.1 2,600,000

S-CL-CJ-XX-22 5000 4790 4790 485 7.0 2,700,000

A-CL-CJ-XX-13 5000 4460 4460 405 6.1 1,700,000

* Values reported are the average of three cylinder tests, ASTM C1231† ASTM C496‡ ASTM C469Note: Ec = modulus of elasticity of concrete; fc

' = concrete compressive strength; fct = tensile strength of concrete. 1 psi = 6.895 kPa.


a minimum of eight hours after casting the first half of the specimen. This amount of time was selected to provide enough time to allow formation of the cold joint but to minimize differences in concrete compressive strengths at test date and was consistent with the procedure used in the first phase of the study.10

Table 6 lists the test specimens. Two replicate specimens were tested with each combination of test variables. Speci-men designation indicates concrete type (N = normal-weight, S = sand-lightweight, or A = all-lightweight); lightweight-aggregate material for the sand-lightweight and all-lightweight specimens (SH = expanded shale, SL = ex-panded slate, or CL = expanded clay); interface type (MO = monolithic or CJ = cold joint); interface preparation (U = uncracked, P = precracked, R = roughened, or S = smooth); shear friction reinforcement ratio ρ (in units of × 10-3); and specimen number of replicate specimens (1 or 2).

Test setup, procedure, and instrumentation

The specimens were tested 28 days after casting the con-crete. For the monolithic specimens that were designated to be precracked, the shear plane was first painted with white paint on both sides of the specimen. The specimen was placed on its side, and a line load was applied to opposite sides of the specimen using a steel bar with a V-shaped cross section placed in and along the chamfer on each side of the shear plane. The specimen was loaded gradually until a significant drop in load occurred. At this point, the load was paused and the specimen was examined visually for hairline cracks along the shear plane.

Figure 1. Test specimen (dimensions shown in the figure are measured to the nearest 0.25 in.). Note: Details shown are for a specimen with shear friction reinforce-ment ratio ρ of 0.013. For specimens with reinforcement ratios of 0.009, 0.017, and 0.022, the number of no. 3 closed-tie stirrups crossing the shear plane is 2, 4, and 5, respectively, and they are distributed uniformly across the 11 in. dimension of the shear plane. No. 3 = 10M; 1 in. = 25.4 mm.

Flange

12.00 in. 12.00 in.

17.50 in.11.00 in.

5.50 in.

5.50 in.2.00 in.4.50 in.

0.50 in.

No. 3 closed tie stirrups Shear plane(4.5 in. x 11 in.)

Figure 2. Formwork for test specimens.

Monolithic interface

Cold joint interface


Table 6. Summary of test results

Specimen identification ρ fc' at test day, psi Vu, lb vu, psi vu,ave, psi Failure mode

N-MO-U-13-1 0.013 4840 63,410 1281 1269

ShearN-MO-U-13-2 0.013 4840 62,203 1257 ShearN-MO-P-13-1 0.013 4840 61,071 1234

1192Shear

N-MO-P-13-2 0.013 4840 56,973 1151 ShearS-SH-MO-U-13-1 0.013 4770 55,434 1120

1132Shear

S-SH-MO-U-13-2 0.013 4770 56,588 1143 ShearS-SH-MO-P-13-1 0.013 4770 50,593 1022

1035Shear

S-SH-MO-P-13-2 0.013 4770 51,884 1048 ShearA-SH-MO-U-13-1 0.013 4700 52,032 1051

1056Shear

A-SH-MO-U-13-2 0.013 4700 52,549 1062 ShearA-SH-MO-P-13-1 0.013 4700 46,120 932

998Shear

A-SH-MO-P-13-2 0.013 4700 52,692 1064 ShearS-SL-CJ-R-9-1 0.009 5380 49,340 1000

1010Shear

S-SL-CJ-R-9-2 0.009 5380 50,475 1020 ShearS-SL-CJ-S-9-1 0.009 5380 26,945 540

600Shear

S-SL-CJ-S-9-2 0.009 5380 32,500 660 Shear S-SL-CJ-R-13-1 0.013 5570 63,167 1276

1238Shear


891Shear

S-SL-CJ-S-13-2 0.013 5570 48,767 985 ShearS-SL-CJ-R-17-1 0.017 4950 62,385 1260

1290Shear


955Shear

S-SL-CJ-S-17-2 0.017 4950 47,120 950 ShearS-SL-CJ-R-22-1 0.022 5000 64,455 1300

1230Shear


1075Shear

S-SL-CJ-S-22-2 0.022 5000 56,535 1140 ShearA-SL-CJ-R-13-1 0.013 4380 46,525 940

944Shear

A-SL-CJ-R-13-2 0.013 4380 46,925 948 ShearA-SL-CJ-S-13-1 0.013 4380 37,842 764

774Shear

A-SL-CJ-S-13-2 0.013 4380 38,751 783 ShearS-CL-CJ-R-9-1 0.009 4770 37,060 750

810Shear

S-CL-CJ-R-9-2 0.009 4770 42,910 870 ShearS-CL-CJ-S-9-1 0.009 4770 31,920 650

710Shear

S-CL-CJ-S-9-2 0.009 4770 37,960 770 ShearS-CL-CJ-R-13-1 0.013 4640 50,785 1026

986Shear

S-CL-CJ-R-13-2 0.013 4640 46,885 947 ShearS-CL-CJ-S-13-1 0.013 4640 41,006 828

822Shear


1095Splitting

S-CL-CJ-R-17-2 0.017 4550 56,660 1150 SplittingS-CL-CJ-S-17-1 0.017 4550 43,140 870

930Shear


1111Splitting

S-CL-CJ-R-22-2 0.022 4790 53,225 1075 SplittingS-CL-CJ-S-22-1 0.022 4790 52,405 1059

1061Splitting

S-CL-CJ-S-22-2 0.022 4790 52,590 1062 SplittingA-CL-CJ-R-13-1 0.013 4460 41,858 846

865Shear

A-CL-CJ-R-13-2 0.013 4460 43,816 885 ShearA-CL-CJ-S-13-1 0.013 4460 36,966 747

750Shear

A-CL-CJ-S-13-2 0.013 4460 37,324 754 Shear

Note: fc' = concrete compressive strength; vu = ultimate shear stress; vu,ave = average value of vu for the two replicate specimens; Vu = ultimate shear

force; ρ = shear friction reinforcement ratio. 1 lb = 0.454 kg; 1 psi = 6.895 kPa.


Applied shear force V-slip relations Figures 5, 6, and 7 show applied shear force-slip (V-slip) relations for the normalweight, sand-lightweight, and all-lightweight series specimens with a reinforcement ratio ρ of 0.013, respectively. Figure 5 shows that for normalweight con-crete specimens with a monolithic interface, the slip cor-responding to the peak shear force was slightly larger for uncracked specimens than for precracked specimens. After the peak shear force was achieved, the applied shear force reduced with increasing slip until a nearly constant value of applied shear force was reached for all specimens in a given series. The uncracked specimens behaved in a more quasi-brittle manner than the corresponding precracked specimens; that is, after the peak shear force was achieved, the shear force decreased rapidly with increasing slip.

Figure 6 shows that the shale sand-lightweight concrete monolithic interface specimens exhibited similar initial stiffness. The peak shear force was larger for the uncracked specimens than the precracked specimens. Quasi-brittle post-peak behavior was observed with uncracked speci-mens.

For the slate and clay sand-lightweight concrete cold joint interface specimens, Fig. 6 shows that the initial stiffness of the smooth and roughened interface specimens was similar. Specimens with a roughened interface (dashed lines in the figure) behaved in a more quasi-brittle manner

Specimens were loaded in compression concentric to the shear plane with a hemispherical head at the top of the specimen to allow rotational freedom and a fixed plate at the base of the specimen. Testing was conducted by con-trolling the displacement of the testing machine at a rate of 0.015 in. (0.38 mm) per minute. The specimens were tested until the peak load was reached and then one of the follow-ing conditions occurred:

• a target slip of 0.3 in. (8 mm) was reached

• the applied load dropped to 60% of the peak load

An external confinement system was added to the top and bottom flanges of the specimens to mitigate premature failure in the flanges that occurred in the Shaw and Sneed10 tests.

Two direct-current linear variable displacement transducers (DC-LVDTs) were installed on each face of the specimen to monitor dilation (separation) of the shear plane. One ad-ditional DC-LVDT was installed on each face of the speci-men to monitor slip of the interface in the direction of the load. Uniaxial electrical resistance gauges were attached to three stirrup legs crossing the shear plane. Figure 3 shows the test setup and instrumentation.

Results and discussion

This section summarizes the results of this study. Addition-al information is provided in Krc15 and Wermager.16

Observed behavior

Failure mode and cracking Table 6 lists the failure mode of each test specimen. Failure of most specimens was associated with shear failure of the shear interface. Specimens S-CL-CJ-R-17-1, S-CL-CJ-R-17-2, S-CL-CJ-R-22-1, S-CL-CJ-R-22-2, S-CL-CJ-S-22-1, and S-CL-CJ-S-22-2, which all had expanded clay aggregates and ρ of 0.017 or 0.022, failed due to concrete splitting22 prior to shear failure. The sand-lightweight concrete with expanded clay aggregates had the lowest splitting tensile strength and elastic modulus of all sand-lightweight concretes in this study (Table 5). Specimens S-SL-CJ-R-22-1, S-SL-CJ-R-22-2, S-SL-CJ-S-22-1, and S-SL-CJ-S-22-2, which had expanded slate aggregates and ρ of 0.022, failed in shear but also exhibited several splitting cracks. Figure 4 shows examples of the type of cracking observed at peak load in specimens with the different cold joint interface conditions.

In addition to cracking of the shear interface, post-peak spalling of the concrete cover was observed adjacent to the shear plane crack for many specimens. For specimens with a cold joint, the dilation of the crack along the shear plane was larger for specimens with a roughened interface than for corresponding specimens with a smooth interface.

Figure 3. Test setup and instrumentation. Note: DC-LVDT = direct-current linear variable displacement transducer.

Secondary confinement

system

Primary confinement

system

Strain gauge data

acquisition

Slip DC-LVDTs

DilationDC-LVDTs


The slate and clay all-lightweight concrete cold joint interface specimens (Fig. 7) also exhibited similar initial stiffness. A larger peak shear force was observed in the specimens with a roughened interface (dashed lines in the figure) than the corresponding smooth interface specimens (solid lines in the figure). This higher peak shear force was accompanied with quasi-brittle post-peak behavior. The post-peak shear force did not appear to be affected by the shear plane surface preparation.

Figure 8 plots representative V-slip relations for the sand-lightweight concrete specimens with varying reinforce-ment ratios ρ. The slate sand-lightweight specimens with a roughened interface in Fig. 8 (dashed lines in the figure) had larger peak shear force than corresponding specimens with a smooth interface (solid lines in the figure) and the same ρ. The same trend was observed for the clay sand-lightweight specimens in Fig. 8, though the difference in shear force between corresponding rough and smooth interface specimens is more pronounced for the slate aggregate specimens than the clay aggregate specimens. As a general trend, the slate sand-lightweight concrete specimens had a larger peak shear force than correspond-ing clay sand-lightweight specimens with similar interface condition and ρ, with two exceptions, S-CL-CJ-S-13-2 and S-CL-CJ-S-22-1.

As mentioned, six of the clay sand-lightweight speci-mens failed due to concrete splitting and loss of bond instead of shear failure along the intended shear plane. This behavior is observed in Fig. 8, which shows a sharp drop-off in applied shear force after the peak

than the corresponding smooth interface specimens (solid lines in the figure) in that applied load dropped off sharply after the peak shear force was achieved, though the residual strength was similar to that of the corresponding specimens with a smooth interface.

Figure 7 shows that the shale all-lightweight concrete monolithic interface specimens had similar initial stiff-ness. The peak shear force did not appear to be influenced by precracking of the specimens. However, the uncracked specimens exhibited more quasi-brittle post-peak behavior compared with the precracked specimens.

Figure 4. Typical cracking of specimens.

Specimen S-CL-CJ-R-13-1 roughened interface shear crack

Specimen S-SL-CJ-S-22-2 smooth interface shear crack

Specimen S-CL-CJ-17-R-2 splitting crack

Interface cracking

Interface cracking

Splitting crack

Figure 5. Applied shear force-slip relations for normalweight concrete specimens with shear friction reinforcement ratio ρ of 0.013 and monolithic interface. 1 in. = 25.4 mm; 1 kip = 4.448 kN.


for specimens with a monolithic uncracked interface and with a monolithic precracked interface. The ultimate shear stress of specimens with a precracked interface was lower than that of specimens with an uncracked interface.

The ultimate shear stress vu is plotted versus concrete unit weight for the specimens with ρ of 0.013 and a cold joint interface (Fig. 9). Distinction is maintained between specimens with different interface conditions and different lightweight aggregates. The data are supplemented with data from the N-5, S-5, and A-5 series reported by Shaw and Sneed,10 which had a reinforcement ratio ρ of 0.013, expanded shale lightweight aggregates from the same producer, and the same nominal compressive strength of concrete as the specimens in the present study. Trend lines are also plotted for each lightweight-aggregate material and include data for the normalweight concrete specimens from the N-5 series by Shaw and Sneed. For the rough-

shear force is achieved for specimens S-CL-CJ-R-17-1 and S-CL-CJ-S-22-1.

Shear strength

Table 6 summarizes the peak values of shear force and shear stress measured for each test specimen. In the table, Vu is the ultimate (peak) shear force measured during testing, and ultimate shear stress vu is the ultimate shear force Vu divided by the area of the shear plane Ac (49.5 in.2 [31,900 mm2]).

The ultimate shear stress vu is plotted versus concrete unit weight for the specimens in Table 6 with ρ of 0.013 and a monolithic interface, maintaining the distinction between specimens with an uncracked and a precracked interface (Fig. 9). Trend lines in Fig. 9 show that the shear friction strength of concrete increases with increasing unit weight

Figure 6. Applied shear force-slip relations for sand-lightweight concrete specimens with shear friction reinforcement ratio ρ of 0.013. Note: 1 in. = 25.4 mm; 1 kip = 4.448 kN.

Monolithic interface Cold joint interface

Figure 7. Applied shear force-slip relations for all-lightweight concrete specimens with shear friction reinforcement ratio ρ of 0.013. Note: 1 in. = 25.4 mm; 1 kip = 4.448 kN.



interface, where the shear transfer may be attributed to cohesion and dowel action of the reinforcement. These re-sults are supported by those from Shaw and Sneed,10 which showed that the shear transfer strength across a smooth interface was not influenced by concrete unit weight. However, Shaw and Sneed also showed that shear transfer strength of lightweight-aggregate concrete with a smooth interface condition increases with increasing concrete strength, where a higher compressive strength is associated with a higher cementitious-materials content, resulting in increased cohesion.

As expected, Fig. 9 also shows that the average values of vu for specimens with a roughened interface are larger than those with a smooth interface for the same concrete compressive strength, lightweight-aggregate material, and unit weight. Figures 6 and 7 also show this relationship.

ened interface condition, the trend lines show an increase in the ultimate shear stress with increasing unit weight. In general, ultimate shear stress values for the expanded clay aggregate specimens were slightly lower than those of specimens with expanded shale and expanded slate. For the smooth interface condition, the trend lines have a slightly negative slope. Values of ultimate shear stress are not nor-malized, and the compressive (and tensile) strengths of the different concretes have some slight variations, which may explain the negative slope. In general, the trend lines for the specimens with a smooth cold joint interface condition show that specimens with the same concrete compressive strength had nearly the same ultimate shear stress vu (ap-proximately 800 psi [5500 kPa]) irrespective of concrete unit weight (concrete type) and lightweight-aggregate material. This suggests that aggregate material does not play a role in the shear transfer strength across a smooth

Figure 9. Comparison of ultimate shear stress vu versus concrete unit weight for specimens with shear friction reinforcement ratio ρ of 0.013. 1 ft = 0.305 m; 1 lb = 0.454 kg; 1 psi = 6.895 kPa.


Figure 8. Applied shear force-slip relations for representative sand-lightweight concrete specimens with cold joint interface and different shear friction reinforce-ment ratio ρ. Note: 1 in. = 25.4 mm; 1 kip = 4.448 kN.

Slate lightweight aggregate Clay lightweight aggregate


also an increase in ultimate shear stress with increasing ρfy, yet values of vu level off at high values of ρfy (after approximately 1200 psi [8.3 MPa]). As mentioned, all four clay aggregate sand-lightweight specimens with ρ of 0.022, as well as the two clay sand-lightweight roughened specimens with ρ of 0.017, failed due to splitting rather than shear. For the specimens with slate aggregate and a roughened interface, Fig. 10 also shows that the values of vu tend to level off at the largest values of ρfy. Again, these specimens exhibited some small flexure and splitting cracks, but the main mechanism of failure was shear along the shear plane.16 This figure also shows the recurring trend of the series with a roughened interface condition (solid markers in the figure) having larger values of vu than cor-responding specimens with a smooth interface condition (hollow markers in the figure).

Analysis

Figures 11 to 13 compare the ultimate shear stress vu of the sand-lightweight and all-lightweight concrete speci-mens with Eq. (1) and design provisions in ACI 318-14 and the PCI Design Handbook for a monolithic interface condition, roughened interface condition, and smooth interface condition, respectively. In comparisons with Eq. (1), the strength reduction factor ϕ was taken as 1.0 because the loads and material properties were known. For a monolithic interface condition, Eq. (1) is limited to 800 psi [5.5 MPa] in ACI 318-14 and to 1000λ psi [6.9λ MPa] in the PCI Design Handbook (Table 1). For a roughened interface condition (1⁄4 in. [6 mm] amplitude) corresponding to case 2, Eq. (1) is limited to 800 psi in ACI 318-14 and to 1000λ psi in the PCI Design Hand-book. For an interface that is not intentionally roughened, corresponding to case 3, Eq. (1) is limited to 800 psi in ACI 318-14 and to 800λ psi (5.5λ MPa) in the PCI Design Handbook. In Fig. 11 to 13, values of λ were taken as 0.85

The increase in ultimate shear stress for specimens with a roughened interface is attributed to increased surface interaction resulting from the irregular profile and the separation (dilation) that must be achieved to overcome the interlock of the shear interface.

Figure 10 plots the ultimate shear stress vu of the sand-lightweight concrete specimens versus ρfy. The data are supplemented with the S-5 series by Shaw and Sneed10 with the same expanded shale aggregate and nominal compressive strength of concrete as the specimens in the present study. A distinction is maintained between different lightweight aggregates and interface condition. Speci-mens that failed in splitting are identified in the figure. For each smooth interface series plotted, Fig. 10 shows an increase in ultimate shear stress with increasing ρfy. For the roughened interface specimens in Fig. 10, there is

Figure 11. Comparison of ultimate shear stress vu with Eq. (1) for specimens with a monolithic interface. Note: fy = yield stress of reinforcement; λ = lightweight modification factor; µ = coefficient of friction; ρ = shear friction reinforcement ratio. 1 psi = 6.895 kPa

Sand-lightweight concrete specimens All-lightweight concrete specimens

Figure 10. Comparison of ultimate shear stress vu versus ρfy for sand-light-weight concrete specimens with different cold joint interface conditions. Note: fy = yield stress of reinforcement; ρ = shear friction reinforcement ratio. 1 psi = 6.895 kPa.


provisions in ACI 318-14 and the PCI Design Handbook for case 1, 2, and 3 interface conditions in Table 1 are conservative for the sand-lightweight and all-lightweight specimens in this study, even using the measured values of fy. If the maximum specified value of 60,000 psi [420 MPa] for fy is used instead, the points shift slightly to the left in each graph, which would yield even more conservative results with respect to Eq. (1). These results are significant because they are among the first in the literature that can be used to validate, with physical test data, shear friction design provisions for sand-lightweight and all-lightweight concrete with a nonmonolithic interface condition.

Previous researchers9,23 have noted that the simplified mod-el in Eq. (1) does not account for variations in concrete strength, particularly the higher shear transfer strengths observed with high-strength concretes. Early experimental studies on shear friction included specimens with con-

and 0.75 for sand-lightweight and all-lightweight concrete, respectively. Values of fy were the measured values. Speci-mens that failed in splitting are identified in the figures.

Figure 11 shows that the ultimate shear stress of each sand-lightweight and all-lightweight concrete specimen with a monolithic interface condition (case 1 in Table 1) is larger than the value computed using Eq. (1) for both ACI 318-14 and the PCI Design Handbook. Similarly, Fig. 12 shows that the ultimate shear stress of each sand-lightweight and all-lightweight specimen with a roughened interface (case 2 in Table 1) was larger than the value computed by Eq. (1) for both ACI 318-14 and the PCI Design Hand-book. Figure 13 shows that the ultimate shear stress of each sand-lightweight and all-lightweight specimen with a smooth interface (case 3 in Table 1) was larger than the value computed by Eq. (1) for both ACI 318-14 and the PCI Design Handbook. Therefore, the shear friction design

Figure 12. Comparison of ultimate shear stress vu with Eq. (1) for specimens with a roughened interface. Note: fy = yield stress of reinforcement; ρ = shear friction reinforcement ratio. 1 psi = 6.895 kPa.


Figure 13. Comparison of ultimate shear stress vu with Eq. (1) for specimens with a smooth interface. Note: fy = yield stress of reinforcement; ρ = shear friction reinforcement ratio. 1 psi = 6.895 kPa.



(case 2 in Table 1) is given in Eq. (2), which is presented in lb-in. units:

when

A f N

K Avf y x

c+( )≥ 1

1 45.

the nominal shear strength is calculated using Eq. (2a)

V A K A f N K f A K An c vf y x c c c+( )1 2 30 8. '

(2a)

when

A f N

K Avf y x

c+( ) 1

1 45.

the nominal shear strength is calculated using Eq. (2b)

V A f Nn vf y x= +( )2 25.

(2b)

where

Nx = force normal to the shear interface, where compres-sion is positive and tension is negative

K1 = constant = 250 psi for sand-lightweight concrete or 200 psi for all-lightweight concrete

K2 = constant = 0.2 for sand-lightweight concrete and all-lightweight concrete

K3 = constant = 1200 psi for sand-lightweight concrete and all-lightweight concrete

Figure 14 compares the ultimate shear stress vu of the sand-lightweight and all-lightweight concrete specimens with a roughened interface condition with Eq. (2). The data are supplemented with those from Shaw and Sneed.10 Fig-

crete compressive strengths ranging from approximately 2500 to 6000 psi (17 to 41 MPa). Recent studies by Kahn and Mitchell9 and Mattock23 examined the shear transfer strength and behavior of specimens with different concrete strengths, including high-strength concretes. Kahn and Mitchell extended the available shear friction test data to very-high-strength normalweight concrete (approximately 6800 to 17,900 psi [47 to 123 MPa]) with a monolithic or rough—that is, not intentionally roughened, but reported to be approximately 1⁄4 in. (6 mm)—interface and proposed a single design equation applicable to normalweight concrete with a monolithic or roughened interface condition that accounts for the higher shear strengths of high-strength concrete. However, their equation did not consider light-weight concretes or smooth interface conditions.

Mattock evaluated the test data available in the literature and proposed a set of modified shear friction design equa-tions that considered the effects of high-strength concrete and included the four interface conditions in Table 1 as well as the effects of sand-lightweight and all-lightweight concrete. However, no experimental data with sand-lightweight and all-lightweight concrete combined with a roughened or smooth interface condition were used to validate Mattock’s proposed equations because none were available at that time.

For monolithic or roughened interface conditions, both studies9,23 proposed an equation that includes separate terms for bond (in terms of fc

' ) and friction (in terms of ρfy). Because Mattock’s study also included provisions for sand-lightweight and all-lightweight concrete with roughened or smooth interface conditions, his proposed equations were evaluated against the experimental results in this present study. Mattock’s equation for computing the nominal shear strength of “concrete placed against hard-ened concrete with its surface intentionally roughened”

Figure 14. Comparison of ultimate shear stress vu with Eq. (2) for specimens with a roughened interface. Note: fc' = 28-day concrete compressive strength; fy = yield

stress of reinforcement; K1 = constant; K2 = constant; K3 = constant; ρ = shear friction reinforcement ratio. 1 psi = 6.895 kPa.



ure 14 shows that Eq. (2) is in agreement with the experi-mental data for both sand-lightweight and all-lightweight concrete.

For the “concrete placed against hardened concrete not intentionally roughened” condition (case 3 in Table 1), Mattock23 proposed Eq. (3), which is consistent with the ACI 318-14 provisions (presented in lb-in. units):

. .V A f A f An vf y c c c0 6 0 2 800'λ= (3)

Equation (3) is also consistent with the PCI Design Hand-book provisions using the coefficient of friction approach in Eq. (5-32a), with the exception of the limits on Vn equal to Vu/ϕ, which are multiplied by the lightweight modifica-tion factor λ in the PCI Design Handbook (Table 1). In Mattock’s23 study, Eq. (3) was validated against experimen-tal data that included normalweight concrete push-off spec-imens with a smooth interface that either were precracked or had a broken bond. Accordingly, these test results should represent a lower bound condition of the shear transfer strength. Mattock pointed out that the shear strength of these specimens was equal to the shear yield strength of the reinforcement perpendicular to the interface (hence the coefficient 0.6 in Eq. [3]) and that true shear friction cannot be developed in the absence of interfacial roughness. Given this reasoning, however, it is not clear why the lightweight modification factor λ was included in Eq. (3) (or in the ACI 318-14 or PCI Design Handbook provisions), especially in the absence of test data on lightweight concretes with a smooth interface condition.

To further examine this issue, Fig. 15 compares the ultimate shear stress vu of the sand-lightweight and all-lightweight concrete specimens with a smooth interface condition with Eq. (3), but with λ taken as 1.0—that is, no influence of concrete type. Data are supplemented with test results on sand-lightweight and all-lightweight

concrete cold joint smooth interface specimens from Shaw and Sneed.10 Results in Fig. 15 illustrate that the values of ultimate shear stress were larger than values computed by Eq. (3), with λ taken as 1.0, for the sand-lightweight and all-lightweight specimens with a smooth interface. These results, along with the previous reasoning, question the need for the lightweight modification factor λ in the friction coefficient in the case 3 interface condition, which is also supported by the results in Fig. 9. Thus, the results of this study suggest that the coefficient of friction µ can be taken as 0.6 (not as 0.6λ) for concrete placed against hardened concrete not intentionally roughened in the shear friction provisions in ACI 318-14 and the PCI Design Handbook.

Recommendations

The authors recommend that the coefficient of friction µ be taken as 0.6 (not as 0.6λ) for concrete placed against hard-ened concrete not intentionally roughened in the shear friction provisions in ACI 318-14 and the PCI Design Handbook.

Conclusion

This paper presents the results from the second phase of an ongoing investigation of the direct shear transfer across an interface of lightweight-aggregate concretes. Test results of 52 push-off specimens were described in this paper to in-vestigate the applicability of the shear friction concept for lightweight-aggregate concretes with different lightweight aggregates and different interface conditions on the shear interface. These results help fill gaps in the literature with respect to sand-lightweight and all-lightweight concretes and interfaces of concretes cast at different times—that is, cold joint conditions. Based on the results of this study, the following conclusions are made:

• The ultimate shear stress of specimens with the same reinforcement ratio (ρ equal to 0.013) and a mono-lithic uncracked or precracked interface increased with increasing unit weight. The ultimate shear stress of cold joint specimens with an intentionally roughened interface increased as the unit weight of concrete increased. The ultimate shear stress of cold joint specimens with a smooth interface appeared to be independent of type or unit weight of concrete.

• The ultimate shear stress of specimens with the same reinforcement ratio (ρ equal to 0.013) and a cold joint roughened interface was higher than that of corre-sponding cold joint smooth interface specimens with the same lightweight-aggregate material.

• The sand-lightweight concrete specimens with a monolithic or cold joint roughened interface condi-tion achieved a higher ultimate shear stress than the all-lightweight concrete specimens with the same lightweight-aggregate material and reinforcement ratio.

Figure 15. Comparison of ultimate shear stress vu with Eq. (3) for specimens with a smooth interface and lightweight modification factor λ of 1.0. Note: fy = yield stress of reinforcement; ρ = shear friction reinforcement ratio. 1 psi = 6.895 kPa.


• A preexisting crack reduced the ultimate shear stress of normalweight, sand-lightweight, and all-lightweight concrete specimens relative to the corresponding un-cracked monolithic specimens.

• The ultimate shear stress of specimens with a roughened interface appeared to be influenced by the type of lightweight-aggregate material. The ultimate shear stress of lightweight concretes made with expanded slate aggregate was higher than that of lightweight concretes made with expanded clay aggregate for roughened interface specimens. The ultimate shear stress of specimens with a smooth interface appeared to be independent of lightweight-aggregate material.

• Six of the sand-lightweight specimens with clay aggregate and a high reinforcement ratio (ρ equal to 0.017 or 0.022) failed due to concrete splitting prior to shear failure, which was attributed to the low tensile strength of concrete.

• The ultimate shear stress of the sand-lightweight specimens with clay and slate aggregates increased with increasing reinforcement parameter ρfy; however, particularly for specimens with a roughened interface, the ultimate shear stress values leveled off after ρfy ap-proximately equaled 1200 psi (8.3 MPa).

• Shear strengths computed by ACI 318-14 Eq. (22.9.4.2) and PCI Design Handbook Eq. (5-32a) using the coefficient of friction µ approach were con-servative for the sand-lightweight and all-lightweight monolithic and cold joint specimens in this study and the Shaw and Sneed10 study. In other words, the use of λ in ACI 318-14 Eq. (22.9.4.2) and PCI Design Hand-book Eq. (5-32a) provided conservative designs for all-lightweight aggregates included in both studies. The use of the alternative effective coefficient of friction µe approach in Eq. (5-32b) of the PCI Design Handbook will be discussed in a future paper by the authors.

• The shear friction design equations proposed by Mat-tock23 were in good agreement with the experimental test data for the sand-lightweight and all-lightweight roughened interface specimens in this study and the Shaw and Sneed10 study.

Acknowledgments

This research was conducted with the sponsorship of PCI and the American Concrete Institute Concrete Research Council. Lightweight aggregates were donated by Buildex Inc., STALITE, and Trinity Lightweight. Longitudinal reinforcing steel bars used in this work were provided by Ambassador Steel Corp. and the Concrete Reinforcing Steel Institute. The authors wish to thank Neal Anderson of Simpson Gumpertz and Heger Inc., Roger Becker of PCI,

Reid Castrodale of Castrodale Engineering Consultants PC, Harry Gleich of Metromont Precast, Neil Hawkins of the University of Illinois, and Larbi Sennour of the Con-sulting Engineers Group Inc., who served as advisors to this project; their assistance and input are greatly appreci-ated. Additionally, the authors gratefully acknowledge the assistance provided by Metromont Precast.

References

1. ACI (American Concrete Institute) Committee 318. 2014. Building Code Requirements for Structural Concrete (ACI 318-14) and Commentary (ACI 318R-14). Farmington Hills, MI: ACI.

2. PCI Industry Handbook Committee. 2010. PCI Design Handbook: Precast and Prestressed Concrete Insti-tute. MNL-120. 7th ed. Chicago, IL: PCI.

3. Anderson, A. R. 1960. “Composite Designs in Precast and Cast-in-Place Concrete.” Progressive Architecture 41 (9): 172–179.

4. Hofbeck, J. A., I. O. Ibrahim, and A. H. Mattock. 1969. “Shear Transfer in Reinforced Concrete.” Journal of the American Concrete Institute 66 (2): 119–128.

5. Mattock, A. H., and N. M. Hawkins. 1972. “Shear Transfer in Reinforced Concrete—Recent Research.” PCI Journal 17 (2): 55–75.

6. Mattock, A. H. 1976. “Shear Transfer Under Mono-tonic Loading Across and Interface Between Con-cretes Cast at Different Times.” Report SM 76-3. Seattle, WA: University of Washington Department of Civil Engineering.

7. Mattock, A. H., W. K. Li, and T. C. Wang. 1976. “Shear Transfer in Lightweight Reinforced Concrete.” PCI Journal 21 (1): 20–39.

8. Walraven, J., and J. Stroband. 1994. “Shear Friction in High-Strength Concrete.” In Shear in Reinforced Con-crete—Volume 1 and 2, 311–330. SP-42. Farmington Hills, MI: American Concrete Institute.

9. Kahn, L. F., and A. D. Mitchell. 2002. “Shear Friction Tests with High-Strength Concrete.” ACI Structural Journal 99 (1): 98–103.

10. Shaw, D., and L. Sneed. 2014. “Interface Shear Trans-fer of Lightweight Aggregate Concretes Cast at Differ-ent Times.” PCI Journal 59 (3): 130–144.

11. Birkeland, P. W., and H. W. Birkeland. 1966. “Con-nections in Precast Concrete Construction.” Journal of the American Concrete Institute 63 (3): 345–368.


24. ASTM Subcommittee C09.20. 2013. Standard Speci-fication for Concrete Aggregates. ASTM C33/C33M. West Conshohocken, PA: ASTM International.

25. ASTM Subcommittee C09.21. 2014. Standard Speci-fication for Lightweight Aggregates for Structural Concrete. ASTM C330/C330M. West Conshohocken, PA: ASTM International.

26. ASTM Subcommittee C09.60. 2014. Standard Test Method for Air Content of Freshly Mixed Concrete by the Volumetric Method. ASTM C173/C173M. West Conshohocken, PA: ASTM International.

27. ASTM Subcommittee C09.60. 2012. Standard Test Method for Slump of Hydraulic-Cement Concrete. ASTM C143/C143M. West Conshohocken, PA: ASTM International.

Notation

Ac = area of concrete shear interface (ACI 318-14)

Acr = area of concrete shear interface (PCI Design Handbook)

Avf = area of shear reinforcement across shear plane

Ec = modulus of elasticity of concrete

fc' = 28-day concrete compressive strength (also at test day)

fct = tensile strength of concrete, measured by splitting tensile strength

fy = yield stress of reinforcement

K1 = constant = 250 psi for sand-lightweight concrete or 200 psi for all-lightweight concrete

K2 = constant = 0.2 for sand-lightweight concrete and all-lightweight concrete

K3 = constant = 1200 psi for sand-lightweight concrete and all-lightweight concrete

Nx = force normal to the shear interface, where compres-sion is positive, and tension is negative

vn = nominal shear stress

vu = ultimate shear stress

vu,ave = average value of vu for the two replicate specimens

V = applied shear force

12. ASTM Subcommittee C09.20. 2009. Standard Test Method for Bulk Density (Unit Weight) and Voids in Aggregate. ASTM C29/C29M-09. West Conshohock-en, PA: ASTM International.

13. ASTM Subcommittee C09.20. 2015. Standard Test Method for Relative Density (Specific Gravity) and Ab-sorption of Coarse Aggregate. ASTM C127-15. West Conshohocken, PA: ASTM International.

14. ASTM Subcommittee C09.20. 2015. Standard Test Method for Relative Density (Specific Gravity) and Absorption of Fine Aggregate. ASTM C128-15. West Conshohocken, PA: ASTM International.

15. Krc, K. 2015. “An Investigation of Shear-Friction of Lightweight Aggregate Concretes.” MS thesis, Missouri University of Science and Technology, Rolla, MO.

16. Wermager, S. 2015. “Shear-Friction of Sand-Light-weight Clay and Slate Aggregate Concretes with Varied Reinforcement Ratios.” MS thesis, Missouri University of Science and Technology, Rolla, MO.

17. ASTM Subcommittee C09.60. 2014. Standard Test Method for Density (Unit Weight), Yield, and Air Content (Gravimetric) of Concrete. ASTM C138/C138M-14. West Conshohocken, PA: ASTM International.

18. ASTM Subcommittee C09.61. 2014. Standard Prac-tice for Use of Unbonded Caps in Determination of Compressive Strength of Concrete Cylinders. ASTM C1231/C1231M-14. West Conshohocken, PA: ASTM International.

19. ASTM Subcommittee C09.61. 2011. Standard Test Method for Splitting Tensile Strength of Cylindrical Concrete Specimens. ASTM C496/C496M-11. West Conshohocken, PA: ASTM International.

20. ASTM Subcommittee C09.61. 2014. Standard Test Meth-od for Static Modulus of Elasticity and Poisson’s Ratio of Concrete in Compression. ASTM C469/C469M-14. West Conshohocken, PA: ASTM International.

21. ASTM Subcommittee A01.05. 2015. Standard Specifi-cation for Deformed and Plain Carbon-Steel Bars for Concrete Reinforcement. ASTM A615/A615M. West Conshohocken, PA: ASTM International.

22. ACI Committee 408. 2003. Bond and Development of Straight Reinforcing Bars in Tension. Farmington Hills, MI: ACI.

23. Mattock, A. H. 2001. “Shear Friction and High-Strength Concrete.” ACI Structural Journal 98 (1): 50–59.


µ = coefficient of friction

µe = effective coefficient of friction

ρ = shear friction reinforcement ratio = Avf/Ac

ϕ = strength reduction factor

Vn = nominal shear strength

Vu = ultimate shear force

w/c = water-cement ratio

λ = modification factor reflecting the reduced mechanical properties of lightweight concrete, relative to normal-weight concrete of the same compressive strength

About the authors

Lesley H. Sneed, PhD, PE, is associate professor and Stirrat Faculty Scholar in the Department of Civil, Architectural, and Environmental Engineering at the Missouri University of Science and Technology in Rolla, Mo.

Kristian Krc, EIT, is a master’s student in structural engineering in the Department of Civil, Architec-tural, and Environmental Engineer-ing at the Missouri University of Science and Technology.

Samantha Wermager, EIT, is a master’s student in structural engineering and Chancellor’s Fellow in the Department of Civil, Architec-tural, and Environmental Engineer-ing at the Missouri University of Science and Technology.

Donald Meinheit, PhD, PE, SE, is a member of the PCI Research and Development Council and was the chair of the PCI Industry Advisory Committee providing input to the researchers at the Missouri University of Science and Technol-ogy.

Abstract

This paper presents the results from the second phase of an ongoing investigation of the direct shear transfer

across an interface of lightweight aggregate concretes. The lightweight concretes were made with differ-ent lightweight-aggregate materials (expanded shale, expanded clay, or expanded slate). The second phase of the experimental investigation included 52 push-off specimens. Test variables included concrete type (normalweight, sand-lightweight, or all-lightweight), lightweight-aggregate material, surface preparation of the shear interface, reinforcement ratio, and crack interface condition. Applied shear force–slip relations are presented and discussed. Peak shear forces are also compared. Current shear friction design provisions in ACI 318-14 and PCI Design Handbook are examined. Shear strengths computed by ACI 318-14 Eq. (22.9.4.2) and PCI Design Handbook Eq. (5-32a) using the coef-ficient of friction µ approach were conservative for the sand-lightweight and all-lightweight monolithic and cold joint specimens in this study. Revisions are proposed for shear friction design of a smooth interface condition.

Keywords

All-lightweight concrete, coefficient of friction, con-nection, interface condition, push-off specimen, sand-lightweight concrete, shear friction.

Review policy

This paper was reviewed in accordance with the Precast/Prestressed Concrete Institute’s peer-review process.

Reader comments

Please address reader comments to journal@pci .org or Precast/Prestressed Concrete Institute, c/o PCI Journal, 200 W. Adams St., Suite 2100, Chicago, IL 60606. J

interface shear transfer of lightweight-aggregate ... journal/2016/march-april/interface...

Documents