Experimental Investigation of Shear Behavior of Deep RC T-beams Under Indirect Loading

study, performed the statistical the and managed study and ABSTRACT The main goal of present study is to experimentally and analytically investigate the behavior of indirect loading, deep flanged reinforced concrete (RC) beams. The load is applied via shear on the side arms of the beam with bearing plates. Eighteen reinforced concrete deep T- beams designed to fail in shear. The beams without web reinforcement were tested under indirectly loading conditions. The beams were divided into three groups according to the ratio of shear span to effective depth. The specimens had different flange depth and flange width in order to investigate the effects of flange dimensions. The behavior of beams was observed; cracking load, ultimate loads, concrete strain, deflections and crack widths. Experimental results indicate that the indirectly loaded deep beams can carry additional loads after diagonal cracking. numerically Ultimate Loads carrying capacity with strut-and-tie prediction results ultimate performance of the beams conformed with 3D non-linear finite element analysis that involves discrete reinforcement modeling. This modeling process is performed by using ANSYS 12.1 the prediction results of ultimate shear capacity for Ansys models are agreeing with the experimental finding. Furthermore, the experimental results were compared with prediction data used the equation for ordinary beams that recommended by ACI Building Code (318-08). It was concluded that the ACI Code equation underestimated the concrete shear strength of the single span indirectly loaded Flanged deep beams.


INTRODUCTION
Nowadays, many building codes concerned with Shear design provisions distinguish between regions in normal size beams, relating to the shear span to effective depth ratio.
Regions in beams with short spans are categorized as D-regions where the D refers to disturbed or deep when the portion of a member have a distance equal to the member height or effective depth from a force discontinuity or a geometric discontinuity as defined by ACI building Code 318R-08 [1]; in which the load is transferred by the arch action mechanism and the distribution of strain is nonlinear across the section of the beam. And B-region where the B refers to beam or Bernoulli, in which the load is carrying by beam action and distribution of strain is linear across the section as well as ACI building Code 318R-8 defined deep beams as the members loaded on one face and supported on the opposite face so that compression struts can develop between the loads and the supports, and have either with clear span equal to or less than four times the overall member depth, or regions of beams loaded with concentrated loads within twice the member depth. And it is recommended taking into account the nonlinearity of strain distribution and lateral buckling in the design of deep beams. Also in the longer shear spans, the Regions in reinforced concrete beams where the shear span is less than 2h defined as D-region. Park [2] defined the beam which has a shear span to depth ratio less than 2 of simply supported beams, or less than 2.5 for any span of a continuous beams, as deep beam.
The behavior of deep beams is different from that of common flexural members; the behavior of deep beams is governed by shear rather than flexure since a large portion of compressive force is directly transferred to supports, and the shear strength is much greater than that of ordinary beams. Deep or short beam grow inclined cracks showed by Syroka [3] and tan [4], capable to carry further load by compressive strut through arch action.
In many instances reinforced concrete slabs together with the beams behaves T-sections. the Stronger compression zones provided with flanges in T-sections observed by Kalyanariman et al. [5] and Kong [6] generally precludes shear, compression failure in T-beams and as a result improves the shear strength capacity. The shear strength of flanged deep beams is a function of many factors such as concrete compressive strength, main and web reinforcement, slenderness (shear span/effective depth ratio), loading and supporting conditions, flange size and beam depth. According to the Fereig and Smith [7], Heywood [8], Simply supported deep beams may be classified according to loading and supporting conditions as directly loaded (forces applied on the top and reactions on the underside), and indirectly loaded beams (loads applied to the sides of members) as in the headstocks supporting beams on nibs' side face. Fig. 1 shows the direct and indirect loaded.
The tested short shear span beams by Ferguson [9], Taylor [10], and Neville and Taub [11] deals with indirectly loaded (shear span-to effective depth ratios of 1.35, 1.64, and 2.09) conclude that the large portion of the additional strength associated with small shear span-to-effective depth ratios had been lost. While Kalyanariman et al. [5] observed that the tensile stress in steel was distributed nonuniformly over the length before diagonal tension cracks were formed in single span deep beams with flanges were indirectly loaded via side arms where the shear span to depth ratio of tested specimens were 1.0 and 1.35. In addition, the tensile stress gradually became uniform over the length of steel bars after diagonal tension cracks were formed attributed to the tied arch action commonly encountered in deep beams.

Fig. 1. Dir
Test results showed that indirectly loaded deep T-beams could have as much as 70% higher shear strength than that of indirectly loa rectangular beams related to the influences of the loading and supporting conditions observed by Zsutty [12]. However, investigation of the performance of indirectly loaded reinforced concrete deep beams is quite lacking and needs further researches concerned wi performance of reinforced concrete deep beams with flanges subjected to indirectly loading. As well as no data have been reported on flanges size effects on shear capacity, deformation and failure manner of indirect loaded deep beams.
The main object of this paper is to investigate the structural behavior of indirectly loaded single span RC deep beams with flanges to evaluate the influence of flange dimensions on structural behavior of the beams. The study consists of an experimental work in which eighteen reinforced concrete deep T-beams were tested with different flange size and different shear span to effective depth ratio (a/d). All beams were reinforced with the same amount and arrangement of tensile steel, without shear reinforcement and the shear span to effective depth ratio (a/d) was varied as (1.0, 1.4, and 1.8). During the testing period, ultimate loads, crack width, deflection, and strains are recorded. Tied arch will be formed after the appearance of diagonal cracking. An important part load is transmitted directly from point of applied load to support by a diagonal compressive strut. While the horizontal compression stress in concrete and the tensile stress in the main reinforcement have to equilibrate the load. The geometry of this technique is clearly depends on location of loads and reactions.

Fig. 1. Directly or indirectly loaded deep beam
Test results showed that indirectly loaded deep beams could have as much as 70% higher shear strength than that of indirectly loaded deep rectangular beams related to the influences of the loading and supporting conditions observed However, investigation of the performance of indirectly loaded reinforced concrete deep beams is quite lacking and needs further researches concerned with the performance of reinforced concrete deep beams with flanges subjected to indirectly loading. As well as no data have been reported on flanges size effects on shear capacity, deformation and failure manner of indirect loaded deep beams.
ject of this paper is to investigate the structural behavior of indirectly loaded single span RC deep beams with flanges to evaluate the influence of flange dimensions on structural behavior of the beams. The study consists of an eighteen reinforced beams were tested with different flange size and different shear span to effective depth ratio (a/d). All beams were reinforced with the same amount and arrangement of tensile steel, without shear shear span to effective depth ratio (a/d) was varied as (

EXPERIMENTAL INVESTIGATION
The tested beams having the span to depth ratio of 1, 1.4 and 1.8 were designated as group 1 (GRgroup 2 (GR2) and group 3 (GR3) respectively and each group consisted of six beams.

Casting of Beams
Wood forms were used in the fabrication of the beams. The inside of the forms was lubricated before the reinforcement cage was placed in position, the concrete mix was placed in three approximately equal layers and each layer compacted by means of poker vibrator surface of the beam was finished level. The beams were moist cured by water for seven days under damp canvas and then stored in the laboratory to testing. 18

EXPERIMENTAL INVESTIGATION
The tested beams having the span to depth ratio re designated as group 1 group 2 (GR2) and group 3 (GR3) respectively and each group consisted of six Wood forms were used in the fabrication of the beams. The inside of the forms was lubricated before the reinforcement cage was placed in position, the concrete mix was placed in three approximately equal layers and each layer ed by means of poker vibrator, the top surface of the beam was finished level. The beams were moist cured by water for seven days under damp canvas and then stored in the laboratory to testing. 18 simply supported single span concrete deep beams divided into three series are casting. Each group consists of six Specimens. All beams of T-cross section with web thickness b w of (120 mm) and flange width b f of (240 mm, 300 mm and 360 mm) and flange depth h f of (60 mm and 90 mm).
The specimens have rectangular central transverse intersecting member with the same overall depth of the beams, (120 mm) width and length equal the flange width of the beams. Beams are tested to failure under indirectly point loads. The details and dimensions of the beams are given in Fig. 2.
All beams are single span and without shear reinforcement. The beams were designed to fail in shear, to satisfy this type of failure a suitable reinforcement was provided [13]. The design compressive strength was obtained in accordance with ASTM-C-39 [14] on cylindrical specimens with dimensions of 300 mm (height)×150 mm (diameter). The value of ƒ c` is obtained from the mean values of three concrete cylinders cured under the same conditions and tested on the same day as the specimens. After curing of 28 days all cylindrical specimens were tested. The measured compressive strength of the employed concrete is 25MPa and slump of about 120 mm. Two types of reinforcement are used in the specimens, namely, deformed steel for longitudinal reinforcement and smooth round steel for flange reinforcement and stirrups. The main beam is reinforced with 12 mm diameter deformed steel bars with yield strength of 330 Mpa and anchored behind the support with sufficient length [3] and with 90 degree bent in order to prevent the anchorage zone failure.
The intersecting beam is reinforced with 16 mm diameter deformed steel bars with yield strength of 330 MPa. Stirrups of 6mm diameter with yield strength of 450 MPa. These are average values achieved from 3 bars of 600 mm length for each type of steel bar. Bearing plates of (120 mm width, 10 mm thickness and 100 mm length) were seated at loading and supporting points in order to avoid concrete local crushing failure as shown in Fig. 5. The specimen notation is typically written as `TB1Gr1', where TB1 is the number of beams in the series of (TB1, TB2, …., TB6); and character (Gr1, Gr2, and Gr3) was definitely the three beam's series of different a/d ratios.

Testing of Beams
Two days before testing, the beams were painted to facilitate tracing the cracks. Aluminum discs with a 10 mm diameter and central hole of 1.5 mm diameter are used to measure the surface strain. The positions of the discs, points are marked and the discs are attached using an epoxy resin as shown in Fig. 3.

Fig. 3. Demic disc arrangement in tested beams
Typical arrangement and experimental setups are shown in Figs. 4 and 5.

Fig. 4. Test setup
A steel spreader beam is used to affect two equal point loads. It is seated on two pins support which placed on top of the beams. The support in beams' ends has two types; roller and pin. The bottom spreader consisted of two universal steel I-beams with a depth of 200 mm. Torsee's Universal Testing Machine (UTM) with capacity of 2000 kN was used to apply the load. Mid span deflections were measured by (0.01 mm) accuracy dial gauge (ELE type) and (30 mm) capacity placed underneath the center of the bottom face of specimens. A mechanical type Demec Gauge (Lc= 0.002 mm) was used to measure surface concrete strain at every stage of loading. The beams were loaded from the top via the intersecting member at the center of beams span. Load was applied incrementally. At each load increment, the total applied load on the beam, mid-span deflection, and crack width have been measured. The crack width is measured by using a hand microscope and plotted and marked. A test was terminated when the total load on the specimens started to drop off.

Behavior of beams under loading and crack pattern
All the tested beams failed in shear. No local failure due to crushing of concrete under the load or over the supports was observed. End anchorage of the main bars functioned properly during testing and did not affect the mode of beam failure. Crack patterns after failure are shown in Figs. 6, 7 and 8.

Behavior of group (1) a/d = 1.0
It was noticed that, the inclined cracks in the shear span of indirectly loaded T-beams occurred suddenly with a thud at mid depth along the line between the applied load and supports. As the load increased, the major inclined crack became wider and extended towards the load and supports then extended through the compression flange near the failure.

Behavior of group (2) a/d=1.4
Few fine flexural cracks was gradually noticed at the bottom of the beam near mid span, but they still with so small crack width in spite of increasing the load. By increasing the load the numbers of cracks increased and development of major diagonal cracks between the load and each support.

Behavior of group (3) a/d=1.8
The behavior of this group was similar to previous groups, however the number of flexural cracks extend upwards with increasing load but still of less width, until shear failure by diagonal cracks. For all indirectly loaded T-beams were observed that more diagonal cracks appear before failure parallel to the strut line between load and support, but almost failure occurs by the major diagonal crack between load and support. For all groups, deflection at mid-span can be regarded as a measure of ductility, increased with increasing the a/d ratio.

Inclined cracking and ultimate load
Inclined cracking and ultimate load are listed in Table 1. Inclined cracking load is defined as the load at which the first major inclined crack appears in the shear span. The ratios of inclined cracking to ultimate load ranged from (0.36 to 0.67) with an average value of (0.52) and standard deviation of (0.08). The variation in a/d ratio seems to have a significant effect on the ultimate load. From Table 1 it can be shown from value of non-dimension (R) that the beams with the same flange width and 90 mm flange depth exhibited higher ultimate load than with 60 mm in the similar (a/d) ratio. Beams TB2,TB4, and TB6 of group (1) showed 37%, 38% and 10% higher ultimate loads than beams TB1, TB3 and TB5 of group (1); beams TB2, TB4 and TB6 of group (2) showed 10%, 15% and 26% higher ultimate loads than beams TB1, TB3 and TB5 of group (2); beams TB2 and TB4 of group (3) showed 11% and 10% higher ultimate loads than beams TB1 and TB3 of group (3). The up rise in ultimate loads of higher flange depth beams is attributed to the strong compression zone provided by increasing depth of the flanges. However the effect of flange width and flange depth on the ultimate loads is small as (a/d) ratio increase.

Effect of (a/d) ratio
The inclined cracking and the subsequent failure of reinforced concrete deep beams are strongly affected by the relative magnitudes of the shearing stress and flexural stress [15]. This effect may be conveniently considered as a function of (a/d) ratio. The distribution of stresses in the indirect loading beams after forming of inclined crack revealed a tendency towards the arch action, but not pure arch action since the force in the tension reinforcement was not constant through the span [7]. From Table 1

Deflection
Typical load-mid span deflection curves for beams in groups 1, 2, and 3 are shown in Figs 9, 10 and 11, respectively. It can be noticed that the formation of the first major inclined crack significantly reduced beams stiffness and tends largely to cause the change the slope of the curves. The amount of deflection is increased considerably with the span/effective depth ratio. The beams with 90 mm flange depth exhibits smaller deflection than beams with 60 mm flange depth with the same a/d ratio.

Effect of flange depth
It can be seen from Table 1 that the maximum loads are increased as the flange depth increased (with the same a/d ratio). This increase is attributed to the strong compressive zone provided by an increase in the flange depth. whereas, beams TB2Gr1, TB4Gr1, TB6Gr1 exhibit 34%, 3% and 9% higher first cracking load and 18%, 15% and 20% higher ultimate load compared with corresponding beams TB1Gr1, TB3Gr1 and TB5Gr1. In group 2 and group 3 the variation of the flange depth significantly affects the ultimate load but less for cracking loads. The load vs. mid-span deflection curves for beams of the same span and flange width show that mid-span deflection is increased as the flange depth is decreased as shown in Figs. 9, 10, and 11.

Effect of flange width
From Table 1 can be seen that the effect of flange width is small on ultimate load -carrying capacity f beams. Curves of mid-span deflection vs. span load showed that the increase of flange width leads to an increase in maximum deflection of beams with similar (a/d) ratio.  It's observed that the measured surface strain along the diagonal line between the upper portion of the center line of the beam and the support plate of the main beam is different in beams, because the measured strain were affected by the existence of any crack within the gauge length of the mechanical strain gauge used to measure the strain, therefore, measurements present the concrete strains and crack width.

Inclined crack width
For each load increment, crack width of the major inclined crack propagates at mid-depth of the beam diagonally between load and support is measured by using a hand microscope. Loadcrack widths relationships for groups 1, 2 and 3 are shown in Fig. 13 (a, b, and c). It is concluded that: (1) At the same load level, the beams with higher flange depth showed smaller crack width than the beams with smaller flange depth in the same a/d ratio.
(2) Flange width exhibits a small effect on crack width. (3) Crack width for tested beams showed a clear tendency to increase with increasing the (a/d) ratio.

General
The strut-and -tie model (STM) was used by Schlaich et al. [16] as an alternative design procedure. Schlaich et al. [16] distinguish between: B region, where B is referring to Bernoulli concepts; and D region, where D refers to discontinuity. In D region the strain distribution is nonlinear, whereas the distribution in B region is linear. According to the ACI code, D region extends a distance equal to the member depth away from any discontinuity, such as change in cross section (geometric discontinuity) or the presence the constricted loads (static discontinuity). Deep or short beams are a typical example of D region. Generally, The Ultimate Shear strength, Vu, of RC deep beams can be determined by STM. ACI 318 building code [1]; The Canadian standard CSA-A23.3-3-04 [17] and the Euro code [18] induces to use the STM for designing the deep beams with strut and tie method in which the flow of internal forces is modeled; the concrete in uniaxial compression represented by concrete compressive struts, main reinforcement represents the tension ties and nodal zones where the struts and ties meet, are representing the regions of concrete subjected to multidirectional stresses. There is more than one choice of strut and tie model for the same problem in order to transmit the loads through the D regions to support. The two most important factors affecting the type of load transfer mechanism are the shear span-depth ratio (a/d) of the beams and the amount of transverse reinforcement [19].

Direct STM
Direct STM means the shear force is carried from the load points toward the supports directly by major compression struts. Fig. 14 shows a typical strut and tie model for two concentration point loads. It can be observed that the model is statically determinate truss containing three components: The concrete strut, the nodal zones and the tie strength of main reinforcing bars. The geometrical notations of the strut-tie method can be shown in Fig. 14. . 14. (a, b)Typical strut-tie model of deep beam

Inclined strut
The upper and lower widths of the inclined strut are determined according to the following equations with respect to Fig. 14. Where f ce is the compressive strength of the concrete in the strut, A cs is the cross area at one end of the strut, and recommended by the ACI code. A value of for the top horizontal strut, β s =0.75 or 0.6 for the inclined strut depending on the existence of web reinforcement.

Reinforcing tie
The effective tie width w t is usually taken as twice the distance from the extreme tension fiber to the centroid of the main reinforcement of the beam. ACI 318-08 code recommended a practical upper limit of the tie width corresponding to the width in a hydrostatic nodal zone, calculated as: is the compressive strength of the is the cross-sectional area at one end of the strut, and β s as recommended by the ACI code. A value of β s =1 =0.75 or 0.6 for the inclined strut depending on the existence of web is usually taken as e from the extreme tension fiber nforcement of the 08 code recommended a practical upper limit of the tie width corresponding to the width in a hydrostatic nodal (7) is the strength of the tie and f ce is the concrete effective compressive strength in a static nodal zone. The strength of the tie, Where A s and f y are the area and yield strength of tension steel reinforcement.

Nodal zones
The compression strength of a nodal zone, calculated such that: where A nz is the smallest of: (1) The area of the face of the nodal zone on which the force acts, taken perpendicular to the line of action of th force; (2) The area of a section through the nodal zone, taken perpendicular to the line of the resultant force on the section, compressive strength of the concrete in the nodal zone and β n is the efficiency factor. The Procedure of calculation begins by calculating the compression block's depth angle θ, area of the strut which is the smaller of the top and bottom nodes, which is used to predict the strut capacity in which the shear force depend on. Furthermore, the tension force in the main steel reinforcement is checkered comparably with force found from the estimated of shear force.

FINITE ELEMENT MODEL
FE analysis is done using ANSYS software 12.1 [20]. According to the symmetry of the boundary conditions, materials, and loading, a quarter of the full beam was modeled to reduce the number of element and save much comput The model considered to have dimensions, geometry, and configuration with the same as the tested specimens. The element types of the , 2016; Article no. BJAST.25264 are the area and yield strength The compression strength of a nodal zone, F nn , is (9) is the smallest of: (1) The area of the face of the nodal zone on which the force acts, taken perpendicular to the line of action of the The area of a section through the nodal zone, taken perpendicular to the line of the resultant force on the section, f ce is the compressive strength of the concrete in the nodal is the efficiency factor. The Procedure of calculation begins by calculating the compression block's depth ɑ 1 , inclination , area of the strut which is the smaller of bottom nodes, which is used to predict the strut capacity in which the shear force depend on. Furthermore, the tension force in the main steel reinforcement is checkered comparably with force found from the estimated ain curve for concrete in FE analysis is done using ANSYS software 12.1 . According to the symmetry of the boundary conditions, materials, and loading, a quarter of the full beam was modeled to reduce the number of element and save much computation time. The model considered to have dimensions, geometry, and configuration with the same as the tested specimens. The element types of the model are: Structural element SOLID 65 for concrete material to personification non of brittle material; Element LINK 8 for steel reinforcement; Element SOLID 45 for steel plate in the support and area of applied load. The selected element types are used by Wolanski [21], Kachlakev [22], Mohammed Hemmaty [24] for material model for analysis the structural stress based on the formulation developed by Williams and Warnke transverse shear coefficient is used by Hemmaty [24] in this paper used as (β t =0.5, Fig.15 shows the typical tensile stress uniaxial compressive relation for concrete The simplified Uniaxial compressive stress curve for compressed concrete is constructed in this paper by numerical expressions consist Equation 10 used by Desayi [27].  Fig. 17 shows the FE model of one quarter of the beam and mesh. The mesh size used in this paper is (40mm). The Poisson's ratio for concrete and steel used in this study are 0.2and 0.3respectively. Fig. 18 shows the idealized concrete response in tension tensile stress-strain which modeled as linear elastic up to a tensile rupture strength off ( 0.62√ƒc') [31]. E s is 200000MPa. The relation of uniaxial stress-strain for steel reinforcement is idealized as a bilinear curve as elastic behavior with strain hardening, which is assumed to be identical in terms of both tension and compression as per the European Committee's [32] criterion of von Mises plasticity yielding, as displayed in Fig. 19. The concept of the 3D failure of concrete surface used by Hawileh is taken into consideration in this study as shown Lafta and Ye; BJAST,15(5): 1-19, 2016;Article no. 14 element SOLID 65 for concrete material to personification non-linearity al; Element LINK 8 for steel reinforcement; Element SOLID 45 for steel plate in the support and area of applied load. The selected element types are used by Wolanski , Mohammed [23] and for material model for analysis the structural stress based on the formulation developed by Williams and Warnke [25]. The ear coefficient is used by Hemmaty =0.5, β c =0.95). 15 shows the typical tensile stress-strain and uniaxial compressive relation for concrete [26]. The simplified Uniaxial compressive stress-strain curve for compressed concrete is constructed in this paper by numerical expressions consist (10) the stress at any strain ε , ε is the is the strain at ultimate [28] achieved first point in the curve, which start at 0.30ƒ c ' . Points 2, 3, and 4 are finding from Equation 10. 16 shows the simplified relationship of compressive uniaxial strain used in the stress analysis [29] to help with the convergence of nonlinear solution 17 shows the FE model of one quarter of the beam and mesh. The mesh size used in this paper is (40mm). The Poisson's ratio for concrete and steel used in this study are 18 shows the idealized concrete response in tension [30]. The strain which modeled as linear elastic up to a tensile rupture strength off ( f r = is 200000MPa. The relation of strain for steel reinforcement is idealized as a bilinear curve as elastic-plastic in hardening, which is assumed to be identical in terms of both tension and compression as per the European Committee's criterion of von Mises plasticity yielding, as 19. The concept of the 3Dsurface used by Hawileh [33] is taken into consideration in this study as shown in Fig. 20. Newton-Raphson procedure is into account for applying load which is subdivided into increments of series loads that are applied in several steps as shown in

Ultimate Shear Capacity
The Ultimate Shear force capacity of test results obtained from 18 deep reinforced concrete T Beams without web reinforcement are compared with predictions used strut-and-tie model based on ACI code [13] recommendation, as well as to FE Ansys result. Fig. 22  Raphson procedure is taken into account for applying load which is subdivided into increments of series loads that are applied in several steps as shown in Fig. 21.   differ slightly from the experimental curves; this because of, First, existing the micro cracks in the tested concrete specimens which formed through drying shrinkage in the concrete and/or beam handling. On the other hand, the FE models do not deal with micro cracks. Second, the bonding of concrete and steel as perfection in the FE model. However, this supposition is not true for  Fig. 24 presents the patterns of cracks appears in ANSYS models for comparison with the tested beams. the study's results confirm the FE analysis and experimental results presented by Özcan et al. [34].

CONCLUSION
In this paper, the total of 18 RC Deep T-Beams subjected to indirect points via side arms loading. The general behaviors of beams were examined. The effects of the test parameters such as a/d ratio, flange width and flange depth on the measured Ultimate load capacity of the beams are investigated. The main remarks of the study achieved as the following: • All the beams fail in shear due to diagonal cracking which leads to the splitting of the beams along the diagonal crack. • The crack pattern showed that the webs of all beams functioned as simple struts between loads and supports. • Experimental findings showed that the increasing in flange depth (h f ) of the Tbeams in the same (a/d) ratio result in an increased the ultimate load -carrying capacities. From value of non-dimension (R)can be noticed that the beams with the similar flange width and 90 mm flange depth exhibited higher ultimate load than the beams with (h f ) of 60 mm in the same (a/d) ratio. Beams TB2,TB4, and TB6 of group (1) showed 37%, 38%and 10% higher ultimate loads than beams TB1, TB3 and TB5 of group (1); beams TB2, TB4 and TB6 of group (2) showed 10%, 15% and 26% higher ultimate loads than beams TB1, TB3 and TB5 of group (2); beams TB2 and TB4 of group (3) showed 11% and 10% higher ultimate loads than beams TB1 and TB3 of group (3). The same tendency is observed in the case of the respective deflections. • The mid-span deflection of the indirectly loaded flanged deep beams are small and less than the permissible deflection specified by the ACI Building Code and does not cause any problem at service load (P u /1.5). • Both inclined cracking load and ultimate load have a tendency to increase with decreasing the shear span to effective depth ratio below 1.8. An increase of about (50%) in inclined cracking load and (78%) in ultimate load of beam TB5Gr1 and TB5Gr3 is obtained by reducing the (a/d) from 1.8 to 1. • Inclined crack width is increasing with an increase in (a/d) ratio, as well as it is smaller than for beam with (h ƒ ) of 90 mm than for beam with (h ƒ ) of 60 mm in the same (a/d) ratio. • The surface concrete strain is linear at early stage of loading but it becomes nonlinear at higher loads.