# ACI STRUCTURAL JOURNAL

Title no. 107-S05

TECHNICAL PAPER

A New Formula to Calculate Crack Spacing for Concrete Plates
by E. Rizk and H. Marzouk
Different codes have different formulas to calculate crack spacing and crack width developed in flexural members. Most of these formulas are based on the analysis of results of tested beams or one-way slabs. Crack control equations for beams underestimate the crack width developed in plates and two-way slabs. It seems that little attention has been paid in determining the crack spacing and width in reinforced concrete plates. The behavior of reinforced concrete plates and two-way slabs is different from beams or oneway slabs; therefore, the methods developed for beams cannot be directly applied to plates and two-way slabs. In this paper, a new analytical equation is proposed for calculating the crack spacing for plates and two-way slabs. A special focus will be given to thick concrete plates used for offshore and nuclear containment structures. The proposed equation takes into account the effect of steel reinforcement in the transverse direction through the splitting bond stress. The new equation provides good estimates for crack spacing in plates and two-way slabs with different concrete covers. Eight full-scale two-way slabs were designed and tested to examine the effects of concrete cover and bar spacing of normal- and highstrength concrete on crack spacing. The different code expressions are evaluated with respect to the experimental results.
Keywords: bond stress; crack spacing; plate; transverse reinforcement.

slabs. An accurate estimate of the crack spacing and crack width of thick concrete plates used for offshore and nuclear power plant structures can result in the reduction of steel reinforcement. The saving of steel reinforcement to satisfy the crack width limitations can be estimated in millions of dollars for a single project (for example, Hibernia oil platform). The proposed equation combines the known bond stress effect with the contribution of splitting bond stress in the transverse direction due to the action of two-way slabs. The proposed equation gives a good estimate for crack spacing in plates and two-way slabs with concrete covers (Cc < 2.5db). The proposed method can also be modified and used for plates with thick concrete covers (C = 2.5 – 5.0db). PREVIOUS RESEARCH Crack width models clearly illustrate that the crack spacing and width are functions of the distance between the reinforcing steel. Therefore, crack control can be achieved by limiting the spacing of the reinforcing steel. Maximum bar spacing can be determined by limiting the crack width to acceptable limits. Crack spacing The Canadian offshore code CSA-S474-042 provides the following expression for calculating the average crack spacing. This is the same equation used by the Norwegian code NS 3473E3 Sm = 2.0(Cc + 0.1S) + k1k2dbe hef b/As ′ (1)

INTRODUCTION Crack-control equations for beams underestimate the crack width developed in plates and two-way slabs.1 The behavior of reinforced concrete plates and two-way slabs is different from that of one-way slabs and beams; hence, the methods developed for beams cannot be directly applied to plates and two-way slabs. The expression for crack spacing is based on the beam theory in several codes, such as the Canadian offshore code CSA-S474,2 Norwegian Code NS 3473E,3 and the European CEB-FIP4 model code. With the extensive use of thick concrete plates and slabs with thick concrete covers for offshore and nuclear containment structures, the development of new formulas is needed to predict crack spacing and width for plates and two-way concrete slabs. This paper examines the different approaches and codes for estimating the crack spacing. Little attention has been paid in determining the crack spacing and width in reinforced concrete thick plates. A lack of available research data on the prediction of crack properties results in unnecessary overdesign of steel reinforcement to satisfy conservative crack requirements in codes for offshore structures. This investigation presents a unique experimental work. Crack spacing of eight reinforced concrete specimens relevant to the offshore structures had been examined. Furthermore, the results were evaluated with regard to the available codes. RESEARCH SIGNIFICANCE This paper provides a rational method for designers to calculate crack spacing for plates and two-way concrete ACI Structural Journal/January-February 2010

where Sm is the average crack spacing, mm; Cc is concrete cover, mm; S is bar spacing of outer layer, mm; k1 is the coefficient that characterizes bond properties of bars; k1 = 0.4 for deformed bars; and k1 = 0.8 for plain bars—this is related to the deformed rips on reinforcing bars; k2 is the coefficient to account for strain gradient; k2 = 0.25(ε1 + ε1)/2ε1, where ε1 and ε2 are the largest and the smallest tensile strains in the effective embedment zones; dbe is the bar diameter of outer ′ layer, mm; hef is the effective embedment thickness as the ′ ′ greater of a1 + 7.5dbe and a2 + 7.5dbe but not greater than the tension zone or half slab thickness, mm; a1 and a1 are the distances from the centers of the bars to the surface of the concrete, mm (refer to Fig. 1); b is the width of the section, mm; and As is the area of reinforcement within the effective embedment thickness (mm2).

ACI Structural Journal, V. 107, No. 1, January-February 2010. MS No. S-2008-254.R3 received February 13, 2009, and reviewed under Institute publication policies. Copyright © 2010, American Concrete Institute. All rights reserved, including the making of copies unless permission is obtained from the copyright proprietors. Pertinent discussion including author’s closure, if any, will be published in the NovemberDecember 2010 ACI Structural Journal if the discussion is received by July 1, 2010.

43

NL. the clear concrete cover in plates and two-way slabs is nearly constant (20 mm [0. σs2 is the steel stress at crack.ϕ s 4τ bk For stabilized cracking ϕs l s. 44 where the terms under the square root are collectively termed the grid index. Crack control in plates and two-way slabs The cracking width in plates and two-way slabs is controlled primarily by the steel stress level and spacing of the reinforcement in two perpendicular directions. ksi. Desayi and Kulkarni7 developed an approximate method to predict the maximum crack width in two-way reinforced concrete slabs.E. John’s.000 S = ----------------. For concentrated loads or reactions or when the ratio of short to long span is less than 0. mm.). and finite element analysis. (2) σ s2 – σ sE l s. which occur within this length.5. β = 1. A reevaluation of cracking data5 provides a new equation based on the physical phenomenon for the determination of the flexural crack width of reinforced concrete members.60 f y.8 × 10–5) for uniformly loaded restrained two-way action square slabs and plates.1 × 10–5 is applicable. ACI 318-996 requires the spacing of reinforcement closest to a surface in tension not exceed 95. mm. Canada. ON. Alternatively.0[Cc + 0. Term B relates to the type of bar. ef).ef is the effective reinforcement ratio (As/Ac. which is between that stage and the ultimate load. As a result of the two-way action of the slabs. He received his BSc and MSc from Menoufiya University in 1999 and 2005. Meanwhile. τbk is the lower fractile value of the average Fig. The stress in the steel caused by steel strain will be reduced due to the bond stress τbk between the steel and surrounding tensile concrete. Frosch (1999) introduced this phenomenon into a new expression that was adopted by ACI 318-99. although it varies between 1. a value of k = 2.– 2. mm. offshore design.5 in. contribute to the width of the crack. Ac. MPa. He received his MSc and PhD from the University of Saskatchewan.35). Egypt. an increase in loading will result in an increase in steel strain. max = 2 --------------------. Saskatoon. and type of stress (B = k1k2d be hef b/As). ρs. Steel and concrete strains.5. and is an Assistant Lecturer at Menoufiya University. The spacing of cracks formed in Direction X can be calculated using the following formula ACI Structural Journal/January-February 2010 . For span aspect ratios less than 0. His research interests include structural and material properties of high-strength and lightweight high-strength concrete.5C c fs (5) where fs is the calculated stress in reinforcement at service load = unfactored moment divided by the product of steel area and internal moment arm. Canada. MPa.1S]).l s. and 213. fs can be taken as 0. MPa. The crack spacing provided in Eq. Lightweight Aggregate and Concrete. ′ The CEB-FIP4 model code crack spacing expression is recommending a different approach compared to other codes (CSA2 and NS3).6 × 10–5. He is a member of ACI Committees 209. In addition. creep.6 For crack control in beams and one-way slabs. Analysis of data on cracking in plates and two-way slabs1 has provided the following equation for predicting the maximum crack width w = kβf s I (6) The CEB-FIP4 model code provides the following expression for calculating the average crack spacing for stabilized cracking S rm 2 = -. His research interests include cracking and minimum reinforcement of offshore structures and shear strength of two-way concrete slabs. respectively.75 but larger than 0. and Srm is the average crack spacing. and k is a fracture coefficient (k = 2. the diameter. k =1. 1—Effective embedment thickness (effective tension area). C c is the clear cover from the nearest surface in tension to the flexural tension reinforcement. whereas it is a major variable in the crack control equations for beams. The researchers calculated the maximum crack width based on an estimation of the crack spacing at any given stage of loading. Therefore. Canada. the bars in the perpendicular direction can be assumed to bear against the concrete surrounding them.8 in. max = ----------------3. max 3 (4) where ls.25 (chosen to simplify calculations. the bond effect of CEB-FIP4 is treated in a different manner. and fs is the actual average service-load stress level or 40% of the specified yield strength fy. Rizk is a PhD Candidate and Research Assistant at Memorial University of Newfoundland. This study shows that previous crack width equations are valid for a relatively narrow range of covers up to 63 mm (2. ef is the effective area of concrete in tension limited by slab width and height equal to the lesser of 2. This will cause an elongation of the reinforcing bar in which the bar ribs will tend to move toward the nearest crack relative to the surrounding concrete. instead of using the factor k1 to account for the bond effect. Term A is a function of concrete cover and bar spacing (A = 2. (1) can be divided into two terms. ef (3) (2) bond stress. and S is the center-to-center spacing of the flexural tension reinforcement nearest to the surface of the extreme tension face.max is the length over which slip between the steel reinforcement and concrete occurs (approximating crack spacing in stabilized cracking).20 and 1. when the stretching of bars in Direction X and the concrete surrounding them is considered. ACI member H.5 (c + φ/2) or (h – c)/3 (mm2). the CEB-FIP4 model code uses the bond stress τbk directly in the expression as shown in Eq. St. Toronto. σsE is the steel stress at point of zero slip. SK. For a cracked reinforced concrete section.]) for most interior structural slabs. Creep and Shrinkage in Concrete.6ρ s. Shibin el Kom. Marzouk is Chair of the Civil Engineering Department at Ryerson University. ϕs is the bar diameter.

were studied. The structural behavior.1 Marzouk and Hossin8 tested eight square full-scale specimens to investigate the crack width and spacing of high-strength concrete slabs. 2(a) for a unit width of the slab in Direction Y ACI Structural Journal/January-February 2010 The constant kt accounts for the distribution of tensile stress in Section 1-1 on the effective area of concrete Actx and fctm is the mean tensile strength value of the concrete that is calculated according to the CEB-FIP4 model code. with regard to the deformation and strength characteristic of high-strength concrete slabs of various thicknesses and different reinforcement ratios (0.40 to 2. The CEB-FIP 4 model code (Table 3. splitting stress and tensile stress over a section: (a) plan of two-way plate. A sufficient bond force is developed at this location (x = smx). 2(b). kt is the constant to account for the distribution of tensile stress. Stretching the bars in Direction X with the concrete surrounding the bars will result in another crack at a Distance x = smx. which considered the bond force on the bars and anchorage forces due to cross wires. Variation of the bond stress between these two zero-points (between the midpoint and the nearest cracked section) were established by many researchers9. it is further assumed that the peak bond stress occurs at the midsection between the two zero points. The proposed method gives a large coefficient of variation.0 f c′ ( MPa ) (10) Fig.πd bx fbo smx n x + 3 (8) contribution of transverse steel reinforcement = k t f ctm A ctx where fb is the bond strength. The number of bars per unit width in X direction is nx. as a result of the two-way action of the slabs. fbb is the bearing stress. The diameter of the bars in Direction X is ϕ1 and the spacing between bars is S1.9. five high-strength concrete slabs. 2—Distribution of bond stress.k t ft A ct1 α 1 = -----------------------------------------------------------( πφ 1 k b f b/s 1 ) + ( φ 2 f bb /s 2 ) (7) 2 -.1. kb is the constant to account for the surface characteristics of the bar and the distribution of bond stress. zero bond stresses at the two cracked sections and at the midpoint can be assumed. The resulting bond stress distribution closely agrees with the experimental observations. the coefficient μ is taken equal to unity and. is just large enough to induce a maximum tensile stress equal to the tensile strength of concrete. stretching the bars in a perpendicular direction results in splitting circumferential forces in Direction X. with a parabolic variation. Longitudinal steel reinforcement (loading direction) The equilibrium forces acting on a concrete Section 1-1 in Direction X are shown in Fig. and three normal-strength concrete slabs.10 based on experimental results. 45 . Bond stress distribution For a concrete section between two successive cracks in a tensile test specimen. the diameter is ϕ2 and the spacing between bars is S2. The researchers7 presented an equation to predict crack width in two-way slabs. and the peak bond strength is fbo. In Direction Y. At the same time. calculated using the CEB-FIP4 Model Code equation. hence. together with the splitting stresses along the transverse bars. ANALYTICAL MODEL The presented theoretical model for calculating crack spacing for two-way slabs combines the known effect of bond stress with the splitting bond stress in the transverse direction. and (b) cross section of two-way plate. and ft is the tensile strength of the concrete. compared to the more practical equation used by Nawy and Blair. These two assumptions greatly simplify the mathematical formulation in calculating the bond stress.68%). fbo is calculated as follows f bo = 1. Figure 2 shows a cross section of a slab and the layout of reinforcement in the Directions X and Y.10 The resulting parabolic bond stress distribution between two successive flexural cracks is shown in Fig.1) provides the following expression for calculating peak bond stress for confined and unconfined concrete for different bond conditions f bo = μ f c′ ( MPa ) (9) For cases where failure is initiated by splitting of the concrete (unconfined concrete). In the present research. which is due to the action of two-way slabs. which.

5db). 3(a) and (b). Acty. (10) can be written as follows f bo = 0.πd by f bo n y 3 (17b) where K is the ratio of the average tensile stress to the maximum tensile stress and equals 0. In the proposed expression.– 1⎞ f ctm ⎝ db ⎠ (13) where cs is the diameter of the effective embedment zone where the reinforcing bar can influence the concrete bond. 46 Equations (17a) and (17b) give the crack spacing in Directions X and Y. The length of the effective embedment zone ly is taken equal to the slab unit width.⎞ f ctm l y ⎝2 2⎠ 2 (12) The different components of the right hand side of Eq. and fbo must be estimated. The proposed model suggests that increasing bar diameter db will result in decreased crack spacing and. this assumption has been found to provide more consistent values compared to the experimental results.67 d by fctm s mx = ------------------------------------------------------2 -.5. 3(a). (10) is only valid for concrete covers equal to or less than 2. Fig. 2(a)).75 for the coefficient μ will be more consistent. (14) can be written as follows for a unit width in the Y direction Contribution of transverse steel reinforcement = 0. A rearrangement gives cs P = 0.= K ⎛ ---s – ---. For equilibrium in the transverse direction in a prism with a length equal to ly Pd b l y c db -----------.tly (14) This is due to the fact that such plates act as cross sections that contain two separate materials: a reinforced concrete part and a plain concrete part. shown in Fig. achieving required crack control. (14) can be estimated in the following: The splitting bond stress fsp.75 f c′ ( MPa ) (11) arbitrarily taking cs = (3. In Fig.0dby. so Eq. 3—Stresses in circular concrete prism subjected to bond stresses: (a) parabolic stress distribution.33 for the parabolic stress distribution. (8).πd bx f bo n x 3 (17a) Similarly.0.33 ⎛ ---. Actx. hence. Transverse steel reinforcement and splitting bond stress The contribution of the transverse steel reinforcement is considered through splitting bond stress. values of kt. K equals 0. To use the previous expression. The value kt is the ratio of the average tensile stress area to the actual tensile stress area within the effective embedment thickness hef.0 – 3. decreased crack width.5db (Cc ≤ 2. (b) triangular stress distribution.67 d bx fctm s my = ------------------------------------------------------2 -. Cc is the clear concrete cover. For a triangular stress distribution. tensile stress on the concrete is assumed to be uniformly distributed and. for plates and two-way slabs with thick concrete covers greater than the radius of the effective embedment zone (Cc > 2. kt can be taken as equal to unity.5db). The constant kt is a tensile stress factor that depends on the distribution of tensile stress on concrete areas Actx and Acty. it was found that a value of 0.11 The contribution of the transverse splitting bond can be estimated by considering the equilibrium of forces acting on concrete to the left and right of Section 1-1 (Fig. hence. cause a pressure p on a portion of the cross section of the prism. Eq.0db.33(3dby – dby)fctm (15) The contribution of the splitting bond stress determined from Eq. Splitting is assumed to occur when the maximum stress is equal to the tensile strength of the concrete fctm.67 to 1.5db and less than 5.33 ( 2d by )f ctm = k t fctm A ctx 3 (16) The crack spacing formed in Direction X can be estimated as follows k t fctm A ctx – 0.7 For thick plates and twoway slabs with thick concrete covers that are greater than 2. hence. kt could be assumed equal to 0. hence. representing the equilibrium forces in Direction X to determine the crack spacing as follows 2 -. For thick slabs. (15) can be substituted into Eq. The diameter of the effective embedment zone cs = 3. fctm. respectively. 3(c). which is also known as the diameter of the splitting cylinder. and db is the bar diameter. increasing the number of bars (decreasing bar spacing) will result in decreasing crack spacing and. Therefore. tensile stress distribution on the ACI Structural Journal/January-February 2010 . and (c) diameter of effective embedment zone. Consider a concrete cylindrical prism with a diameter of cs (diameter of effective embedment zone) containing a bar with a diameter of db. for plates and two-way slabs having concrete covers of (Cc < 2. the distribution of these stresses has been previously assumed to be parabolic. and the unit width of the slab in transverse Direction Y Contribution of transverse steel reinforcement = K(cs – dby)fsp. The radial components of the forces on the concrete.t can be assumed to be equal to fctm. Also.πd bx f bos mx n x + 0.5) db.It should be noted that Eq. the spacing of cracks formed in Direction Y can be estimated as follows k t fctm Acty – 0.5db). at a given stage of loading. as shown in Fig. This is equilibrated by tensile stresses in the concrete on either side of the bar. the tensile stress distribution within the effective embedment thickness is trapezoidal and.

Fig.]).67 for the coefficient kt was found to be more consistent for beams and one-way slabs.5 in.5 0. For cases where failure is initiated by shearing of the concrete between the ribs (all other bond conditions). Test procedure A typical cross section of the tested specimen is shown in Fig. 5 and No. The tested slab was placed in the frame in a vertical position. kt can be taken equal to 0.]).]). fbo is calculated as f bo = 1.1. The tested slabs were carefully inspected at each load step. The constant kt is a tensile stress factor. MPa mm No. and HS is high-strength slabs. with different bar spacing.0 0.35 217. Then. spacing s.75 in. The cracks were marked manually after mapping all the cracks on the specimen. a value of 0. and different bar sizes (15M and 25 M [No.0 64. crack gauges were installed using epoxy glue on the tension surface of the slab and left for 1 hour to enable the epoxy to dry. The slabs’ thicknesses ranged from 150 to 400 mm (6 to 16 in. A total of eight concrete slabs were tested.0 kN (10 kips). as recommended by Marzouk and Hussein.). slab thickness. c mm mm mm ρ. according to the CEB-FIP4 model code provisions (Table 3.]) with the same 70 mm (2.0 10 15 15 15 25 25 35 35 210 240 368 368 368 368 289 217 45 40 60 60 70 70 70 70 150 200 250 250 300 300 350 400 105.]). In the present research. was designed to investigate the effect of other modes of failure on crack spacing. The effect of concrete strength on crack spacing was not considered because it has a small influence on the crack spacing in Series I.5 0. The group was made of two slabs designated as NS1 (h = 150 mm [6 in. 1 mm = 0. are assumed to be Actx = hefxb Acty = hefyb (18a) (18b) where hef is effective embedment thickness (shown in Fig. Crack mapping of Table 1—Details of test specimens Compressive Bar Concrete Slab Bar Steel Series Slab strength size. 1) as the greater of a1 + 7. The second group (Series II) was designed to investigate the effect of concrete cover. different concrete cover thicknesses (60 and 70 mm [2. 4—Details of typical test specimen HSC2 (thickness 300 mm [12 in. 4.π d b f bon 3 (20) (19) had different slab thicknesses (250 and 300 mm [10 and 12 in. cover Cc. The slabs of this group ACI Structural Journal/January-February 2010 44.73 217.44 312.concrete is assumed to be trapezoidal and. heavy. mm. The selected values for the proposed experimental testing are typical for the possible use in Canadian offshore applications. All the specimens of Series I and Series II were designed to fail in flexure. 47 . as discussed by Osman et al.73 262. which depends on the distribution of tensile stress on concrete areas Act. and bar spacing for normal.48 152.5 1. The specimen was initially loaded up to 10% of the ultimate load. The slabs had the same concrete cover (Cc = 40 mm [1.0 70.12 The third group (Series III). 4.5 0. ratio * f ′ . The test slabs were simply supported along all four edges with the corners free to lift. which is a typical practice for offshore structures.5d′be but not greater than the tension zone or half slab thickness.]) and NS5 (h = 400 mm [16 in. however.6 in. mm. and b is the width of the section. and corresponding change in steel ratio on the average crack spacing. reinforced specimens HS3 (h = 350 mm [14 in.5 1.4 and 2. this assumption was found to be more convenient. Depth. The values of Actx and Acty.2 35.5d ′be and a2 + 7. which are the effective stretched area of concrete in the X and Y direction. Crack spacing for beams and one-way slabs The proposed equation can be used to calculate the crack spacing for beams and one-way slabs by modifying the peak bond strength fbo.1).35 182.% mm I NS1 NS2 NS3 II HS1 NS4 HS2 III *NS where n is the number of bars per unit width.5 0.25 f c′ ( MPa ) The crack spacing can be estimated as k t f ct A ct s m = --------------------2 -. no. The specimens in this series were designed to fail under pure punching failure mode. The first group (Series I) was designed to investigate the effect of concrete cover and bar spacing on the crack width. EXPERIMENTAL INVESTIGATION The variables considered in the current investigation are the concrete cover.0 40.]).]) and NS2 (h = 200 mm [8 in.13 The third group includes two thick.and high-strength concrete. concrete strength.5 0. The details of a typical test specimen are shown in Fig. The load was released and then reapplied at a selected load increment of 44.67.54 182.4 40. Five normal-strength concrete slabs (NS) and three high-strength concrete slabs (HS) were selected for the experimental investigation of the cracking behavior study as detailed in Table 1.58 HS3 NS5 is normal-strength slabs.75 in.7 50.7 65.0394 in.) thick concrete cover and heavy reinforcement ratio. The test slabs were classified into three series. thickness. Note: 1 MPa = 145 psi. 8]) but with the same big bar spacing of 368 mm (14. hence.) and were designed to examine the effect of slab thickness on the cracking behavior.

the crack pattern observed prior to punching consisted of one tangential crack. This failure can be classified as flexure failure.7 in. 5 to 7.7% to fail under flexure mode. It is interesting to point out that the average crack spacing almost equal to 253 mm (10 in. These photographs were inserted in a computer aided AutoCAD software drafting package on a two-dimensional grid with a scale of one to one. and the sustained load level.). For the slabs failing in flexure (NS1. and two specimens (NS4 and HS2) reinforced with medium steel reinforcement ratio 0. and (b) NS2 (thickness 200 mm [8 in. the spacing between bars s.75 in. the initial observed cracks were first formed tangentially under the edge of the column stub.7 to 10. The measured crack width and spacing were also compared to calculated crack width and spacing. Frosch et al. The primary variables evaluated in the study were the spacing of the reinforcement and the epoxy coating thickness. Also.3 in. In all slabs. the first crack forms along the reinforcing bar and passes through the slab center or close to the slab center.the specimen was depicted by means of photographs at each stage of loading throughout the experiment. Series II included two specimens (NS3 and HS1) reinforced with low steel reinforcement ratio 0. the number of reinforcing bars. followed by radial cracking extending from the column edge toward the edge of the slab. The parameters varied in the tests were the reinforcing bars type. and the sustained load level. and the spacing was measured and averaged using the software.) thick concrete cover.). NS2. A comparison between beam Series 1 and 2 (Table 2) demonstrates that increasing the clear concrete cover increases the average crack spacing. the crack spacing increased from 201 to 221 mm (8 to 8.).75 in.58% with the same 70 mm (2. It was found that for all the specimens. 5—Crack patterns of Series I: (a) NS1 (thickness 150 mm [6 in. The steel strain can be determined at any loading by determining the neutral axis and assuming linear stress distribution. For the slabs failing by flexurepunching. The crack width can be calculated by multiplying the crack spacing by the steel strain. followed by radial cracking extending from the column. This is because the crack spacing srm is inversely proportional to the effective reinforcement ratio ρeff. the concrete cover Cc. Cracks were retraced on the computer using AutoCAD. roughly at the column outline. Test results of Series II indicated that increasing the concrete cover from 60 to 70 mm (2. The yield strain was measured at a location 150 mm (6 in. increasing the tensile reinforcement area decreases crack spacing and reduces crack width (because crack spacing is inversely proportional to the effective reinforcement ratio). Increasing the bottom cover increases the effective tension area of the concrete and decreases the effective reinforcement ratio.]). and HS1). Test results The first crack of each specimen was visually inspected and the corresponding load was recorded as the first crack load. It was noticed that increasing the concrete cover resulted in increased crack spacing. DISCUSSION Verification of proposed model A total of 12 simply-supported beams and one-way slabs were subjected to constant sustained service loads for a period of 400 days by Gilbert and Nejadi. The second crack forms along the perpendicular reinforcing bar in the other direction. which produced a stress in the steel reinforcing bar equal to 400 MPa (58 ksi). Series III included two specimens (HS3 and NS5) reinforced with heavy steel reinforcement ratios 1.). The test results of Series I (Specimens NS1 and NS2) and Series II indicated that as the bar spacing is increased from 210 to 240 mm (8.3 to 9. radial cracking extending from the column was the most dominant crack pattern. 48 . tools. which results in a larger crack spacing. flexure yield lines were well developed.]).44 and 1. All specimens had the same bar spacing 368 mm (14.35%.4 to 2. Major conclusions derived from this investigation include: ACI Structural Journal/January-February 2010 Fig.5 in. NS3.5 in.) was much smaller than the bar spacing. In all tested slabs.) from the center of the slab. The yield steel strain was recorded at a value of 2000 με. The value of 2000 με was suggested based on experimental observations of the stress-strain curve of a single reinforcing bar.14 The parameters varied in the tests were the shape of the section b/h. Test results revealed that crack control (crack width) can still be achieved by limiting the spacing of the reinforcing steel despite using thick concrete cover.) increased the crack spacing from 245 to 261 mm (9.15 tested 10 one-way slabs to determine the effects of bar spacing and epoxy coating thickness on crack width and spacing. respectively. the crack pattern observed prior to punching consisted of almost no tangential crack. Failure patterns of the tested slabs are shown in Fig. the spacing between bars s.

the overall average theory/test ratio was 1.]).and two-way concrete slabs.15 with different concrete covers and different bar spacing. and concrete cover. tests other than those of the authors). Fig.. as well as the fact that the tests themselves are one-to-one scale models of the prototype and the inevitable scatter of test results in concrete behavior.8 with different concrete strengths. and the results are shown in Tables 2 to 4 and include 12 test results of Gilbert and Nejadi.spacing of reinforcement significantly affected the width and spacing of cracks.149. 6—Crack patterns of Series II: (a) NS3 (thickness 250 mm [10 in. the spacing of primary cracks decreased and the number of primary cracks increased. For the proposed model. and as the reinforcement spacing decreased. the model was applied to predict the average crack spacing of normalweight concrete test slabs reported in the literature (that is. the theoretical model developed herein is an excellent representation of the physical behavior of tested specimens. different concrete covers. bar spacing. It is also worth emphasizing that the slabs analyzed and presented in Tables 2 to 4 cover many variables that influence crack spacing. To verify the validity of the new proposed model. The geometry of test slabs. the model has been applied to 30 tests to predict the average crack spacing of beams and one. and different bar spacing. and 8 test results of Marzouk and Hossin. giving strong support to the ability of the proposed model to evaluate the average crack spacing in tested slabs.028 with a standard deviation of 0. ACI Structural Journal/January-February 2010 49 . and (d) HS2 (thickness 300 mm [12 in. Bearing this in mind. the analysis.14 10 test results of Frosch et al. such as concrete strength. In this paper. (c) NS4 (thickness 300 mm [12 in.]).]). The results indicate that a very good correlation exists between theoretical and measured average crack spacing values and between theoretical and calculated average spacing values using CSA-S474-04 and NS-92 codes were very close to the experiments with approximately 5% error. (b) HS1 (thickness 250 mm [10 in.]).

Beam 1-a Beam 1-b Beam 2-a Beam 2-b Beam 3-a Beam 3-b Slab 1-a Slab 1-b Slab 2-a Slab 2-b Slab 3-a Slab 3-b Concrete cover Cc .]).0394 in.Table 2—Comparison between calculated crack spacing values using code formulas with measured experimental values for test specimens by Gilbert and Nejadi14 Slab no. mm 190 190 190 190 127 127 178 178 119 119 89 89 New proposed model.8 Figures 8 and 9 indicate that the crack spacing values estimated using the presented model. Figures 8 and 9 show a comparison in the calculations for crack spacing between the new proposed model and different codes with the measured experimental values by Marzouk and Hossin. Note: 1 MPa = 145 psi. the calculated average crack spacing was ACI Structural Journal/January-February 2010 Fig. Fig. The codes’ expressions for crack spacing are based on the beam theory. mm 179 179 151 151 105 105 184 184 124 124 100 100 CEB. and (b) NS5 (thickness 400 mm [16 in. mm 173 173 157 157 105 105 177 177 118 118 88 88 Experimental results. and HS is high-strength slabs. mm 40 40 25 25 25 25 25 25 25 25 25 25 Height h. whereas the presented model is rational because it is based on twoway action. CSA-04.]). Tables 2 to 4 show a comparison between the calculated values of crack spacing with the measured experimental values reported by different researchers. For bar spacing greater than 300 mm (12 in. mm 192 186 149 163 109 104 131 128 92 131 89 117 * NS is normal-strength slabs. mm 150 150 180 180 90 90 308 308 154 154 103 103 fc′ . 1 mm = 0. 50 . the Norwegian code NS 3474 E3 and the Canadian offshore code CSA-S474-042 overestimates the average crack spacing by approximately 33%. NS-92. MPa 36 36 36 36 36 36 36 36 36 36 36 36 NS/CSA. 8—Comparison of crack spacing equations at 150 mm (6 in. mm 348 348 333 333 333 333 161 161 161 161 161 161 Bar spacing s. Comparison of experimental versus theoretical estimates of crack spacing Table 5 shows a comparison between the calculated values of crack spacing with the measured experimental values.15 Analysis of the results given in Tables 2 to 4 indicates that the new proposed model provides good estimates for crack spacing in slabs having small and thick concrete covers. 7—Crack patterns of Series III: (a) HS3 (thickness 350 mm [14 in. In general.14. and EC2-04 codes were very close to the experiments with approximately 5 to 9% error.) bar spacing.).8.

0 64. the CEB-FIB4 model code underestimates the average crack spacing by approximately 31%. mm 118 177 236 355 118 177 236 355 177 177 New proposed model. higher than test results.9 NS/CSA. 9—Comparison of crack spacing equations at 250 mm (10 in.4 40. mm 125 165 186 146 167 107 204 228 CEB. The proposed method takes into consideration the effects of steel bond in the loading direction and the ACI Structural Journal/January-February 2010 Fig. mm 201 221 245 263 261 246 264 250 Note: 1 MPa = 145 psi.6 44.0 33. mm 77 68 63 100 125 56 204 182 New proposed model. mm 150 150 150 200 250 100 240 240 fc′ .0394 in. the crack spacing increased theoretically and experimentally.8 46. A new analytical expression is recommended for plates and two-way slabs with longitudinal and transverse reinforcements.) bar spacing. mm 248 234 320 361 273 304 273 226 Experimental results.7 65. mm 167 213 260 352 167 213 260 352 213 213 CEB. Table 5—Comparison between calculated crack spacing values using code formulas with measured experimental values Slab no.7 46.4 44.0394 in. 1 mm = 0.0394 in. mm 152 229 305 457 152 229 305 457 229 229 fc′ . The behavior of reinforced concrete plates and two-way slabs is different from the behavior of the one-way beams.4 45. mm 45 40 60 60 70 70 70 70 Slab thickness. B-6 B-9 B-12 B-18 E12-6 E12-9 E12-12 E12-18 E6-9 E18-9 Concrete cover Cc . mm 126 187 188 182 184 111 225 230 Experimental results.5 70. For bar spacing less than 250 mm (10 in. NSC1 HSC1 HSC2 HSC3 HSC4 HSC5 NSC2 NSC3 Concrete cover Slab thickness. and as both the concrete cover and bar spacing increased.2 70.1 45. MPa 44.0 NS/CSA. mm 136 211 282 411 136 206 278 414 207 207 Experimental results.0 66. Cc .5 47. SUMMARY AND CONCLUSIONS Most of the available expressions for estimating the crack spacing and width are based on beams and test results for one-way slabs.7 61.2 35. mm 203 203 203 203 203 203 203 203 203 203 Bar spacing s.0 70.15 Slab no. Table 4—Comparison between calculated crack spacing values using code formulas with measured experimental values for test specimens by Frosch et al. mm 46 46 46 46 46 46 46 46 46 46 Height h.7 46. mm 150 200 250 250 300 300 350 400 Bar spacing s. 1 mm = 0.0 NS/CSA. 51 . mm mm 30 200 50 200 60 200 30 200 30 200 30 150 30 200 40 150 Bar spacing s. MPa 46.7 50. mm 211 216 341 341 331 331 276 252 CEB.). mm 134 171 185 163 172 120 223 239 Note: 1 MPa = 145 psi. mm 137 176 279 279 225 225 160 145 New proposed model.0 35.0 68. NS1 NS2 NS3 HS1 NS4 HS2 HS3 NS5 Concrete cover Cc . mm 210 240 368 368 368 368 289 217 fc′ . MPa 35. 1 mm = 0. compared to the one measured during testing.4 46. mm 175 229 249 310 170 226 257 338 203 188 Note: 1 MPa = 145 psi.Table 3—Comparison between calculated crack spacing values using code formulas with measured experimental values for test specimens by Marzouk and Hossin8 Slab no.0 34.

H. Comité Euro-International du Béton-Fédération de la Précontrainte (CEB-FIP).. and Hussein..” FHWA/ IN/JTRP Report No. Steel Stress. 6. 3.. Oslo. Osman. Nov. 65.t = splitting bond stress = yield stress of steel fy h = section height hef = effective embedment thickness k1 = coefficient that characterizes bond properties of bars Ac.). 88. 1042 pp. and Hossin. No. Memorial University of Newfoundland.. IN.” Research Report RCS01.. G.5 – 5. it can be speculated that the crack spacing behaves randomly. 12.” ICE Proceedings. School of Civil Engineering. 3. pp. pp. 2.. MI. Memorial University of Newfoundland. Reinforced Concrete: Mechanics and Design. Mississauga. 701-713. 251-259. “Building Code Requirements for Structural Concrete (ACI 318M-99) and Commentary (ACI 318RM-99). Nawy. NS 3473 E (English translation).5db). F. and two-way slabs after reducing one third of the tensile stress constant kt. Oct..” CSA-S474-04. 7. “Study of the Transfer of Tensile Forces by Bond. “Experimental Investigation on the Behavior of High-Strength Concrete Slabs.. R-435. 1. 1999. Marzouk. 159 pp. ON. 10. No. R. P. May-June 1984. D. University of New South Wales. India. 3. 96.” UNICIV Report No. first Canadian edition. John’s.max = length over which slip between steel and concrete occurs. 1991. Sincere thanks are extended to Capital Ready Mix Ltd. “Another Look at Cracking and Crack Control in Reinforced Concrete. and Nejadi. Organ. May-June 1999. No.. 2003. Curtis. This is due to the fact that such slabs act as cross sections that contain two separate materials. pp. contribute to width of crack S = center-to-center spacing of flexural tension reinforcement nearest to surface of extreme tension face Smx = crack spacing for cracks normal to x reinforcement = crack spacing for cracks normal to y reinforcement Smy = average stabilized crack spacing Srm ρs. steel and concrete strains.0db). J.. 1992.. V.” ACI JOURNAL. V. Canadian Standards Association (CSA). The authors would also like to thank M. 369 pp. 123. 13. School of Civil and Environmental Engineering. Gilbert. 7-4-50. 2000. “Behaviour of HighStrength Lightweight Concrete Slabs under Punching Loads. 14. No.. West Lafayette.. Marzouk. 15. V. 97. Department of Civil Engineering. Prentice Hall. ACKNOWLEDGMENTS The authors are grateful to the Natural Sciences and Engineering Research Council of Canada (NSERC) for providing the funds for the project. No. Model Code 1990. Newfoundland. S. June 1976. S.” ACI Structural Journal. Indian Institute of Science. A. M. P. and Helmy.” Norwegian Council for Building Standardization. St. Shah. M.” ACI Structural Journal. Desayi. 1997. 343-349. 2007.). Bulletin D’Information.” ACI Structural Journal. ACI Structural Journal. 79 pp. 61. 1968. V... • For bar spacing greater than 300 mm (12 in. R. “Relationship of Bond Stress. Canada. No. and Slip in Reinforced Concrete. despite using thick concrete covers. “Crack Analysis of Reinforced Concrete Two-Way Slabs. Faculty of Engineering and Applied Science. Frosch. I. “Investigation of Bridge Deck Cracking in Various Bridge Superstructure Systems. S. 825-836. Blackman. and Radabaugh. 492. D. 2004. the CEB-FIB4 model code estimated average crack spacing smaller than the one measured during testing. Sincere thanks are due to M.” American Concrete Institute. • The proposed method can be used for thick concrete covers (Cc = 2. American Society of Civil Engineers. Norway. R. however. and the technical staff of the Structural Engineering Laboratory. “Concrete Structures. 8. 4.. V. the calculated average crack spacing was higher than the measured one from test results.ef = effective reinforcement ratio. “Determination of Maximum Crack Width in Two-Way Reinforced Concrete Slabs. Bangalore. 203-305. pp.. “An Experimental Study of Flexural Cracking in Reinforced Concrete Members under Sustained Loads. MacGregor. ACI Committee 318. and Andonian. pp. C. Lausanne. Pike. Canada. and Kulkarni. 437-442. ON. Jiang. = coefficient to account for strain gradient k2 = tensile stress factor kt ls.contribution of the splitting bond stresses for the transverse steel. • The analytical and experimental investigations revealed that the average crack spacing is increased as the bar spacing or the concrete cover is increased for the specimens with low reinforcement ratio that fail in flexure. Norwegian Standard. Canada. In general. A. E. A. D. No. Proceedings V. pp. Marzouk. for their assistance during the preparation of the specimens and during testing. 10. which occur within this length. Scarborough. Purdue University. 6.. File No. 9. 59 pp. B. H. V.. and Bartlett... for providing the concrete for this project. 2004. May-June 2000. No.-Dec. R. H. Farmington Hills. 462 pp. Australia. 5. The main conclusions can be summarized as follows: • The proposed method gives a good estimate for crack spacing in plates and two-way slabs with concrete covers equal to (Cc < 2. 81. 68 pp... plates.498. • For bar spacing less than 250 mm (10 in. C-36-56YY. • Test results revealed that crack control can still be achieved by limiting the spacing of the reinforcing steel. NL. 2. NOTATION area of concrete symmetric with reinforcing steel divided by number of bars As = area of reinforcement within the effective embedment thickness b = width of section = clear cover from nearest surface in tension to flexural tension Cc reinforcement cs = diameter of effective embedment zone where reinforcing bar can influence concrete bond d = effective depth to centroid of tensile reinforcement d c be = equivalent bar diameter of outer layer of bars ′ fbo = maximum bond strength = uniaxial compressive strength of concrete (cylinder strength) fc ′ fctm = mean value of concrete tensile strength at time that crack forms fsp. pp. 79-85.0db . For two-way slabs with concrete covers larger than 5. “Crack Control in Reinforced Concrete Structures. Switzerland. Hossin for providing his test data.ef = 52 ACI Structural Journal/January-February 2010 . 286 pp. 3. “Offshore Concrete Structures. Frosch. Design Rules. Sydney.” Journal of Structural Engineering. S. Kankam. 11. and equals area of steel considered divided by area of effective zone where concrete can influence crack widths ρs2 = reinforcement stress at crack location = steel stress at point of zero slip ρsE = lower fractile value of average bond stress τbk REFERENCES 1. the Norwegian code NS 3474 E3 and the Canadian offshore code CSA-S474-042 overestimates the average crack spacing by approximately 33%.

. Further reproduction prohibited without permission.Reproduced with permission of the copyright owner.