Digital Image Correlation analysis is performed to obtain measurements of the crack kinematics of four reinforced concrete beams without shear reinforcement. The results are used to implement the critical shear displacement theory. This new theory proposes that the opening of the critical inclined crack can be considered as a lower bound for the shear capacity. The unstable opening of the critical inclined crack is triggered when the shear displacement of an existing flexural crack reaches a critical value. Therefore, the critical shear displacement is used as a failure criterion.
The results indicate that digital image correlation is an adequate technique to obtain the refined crack kinematics since it permits to continuously track the evaluation of the crack even an instant before failure. This allows identifying the beginning of the opening of the critical inclined crack and the critical shear displacement value. A critical load representing the initiation of the irreversible damage is found and a crack width is linked to be applied in a proof load test.

Shear failure is one of the most critical failure modes of reinforced concrete members, especially for those without shear reinforcement. Despite the extensive research programs in the last decades, there is still no general agreement on a rational theory to assess their shear capacity. In the recent years, researchers have focused on developing mechanical models for shear design that are based on a predefined crack pattern and kinematics assuming that the shear force can be transferred through a critical shear crack by various shear transfer mechanisms. These mechanical models need to be validated by detailed kinematic measurements taken from experiments. This information could not be acquired before, however, in the recent years, new measurement technology has developed quickly, amongst others, Digital Image Correlation (DIC) provides new opportunities to obtain the crack pattern and kinematics through the displacement field of the whole surface of the target specimen. In this research, the possibility of using DIC measurements to apply a detailed analysis on the contribution of the shear transfer mechanisms (uncracked concrete, aggregate interlock, dowel action, and arch action) is explored in order to obtain a better understanding of the shear failure process. The analysis is based on ten representative tests on reinforced concrete beam specimens with a height of 1200 mm. The tests are selected from an experimental program designed to study the shear behaviour of reinforced concrete slab strips without shear reinforcement. A new algorithm is developed to automatically determine the contributions of the different shear transfer mechanisms along a crack from the displacement field obtained by DIC measurements. A comparison between the experimental results and the sum of the calculated contributions yield to a reasonable agreement, with an error of 40% when the shear failure is presented just after the formation of the flexural shear crack. With the help of the new algorithm, new insights on the three different shear failure modes observed in the experiments (flexural shear failure, shear compression, and dowel failure) are discussed. Flexural shear failure is attributed to the loss of aggregate interlock when a sudden increase of the shear displacement of the critical shear crack is observed. However, such decrease occurs earlier before the actual failure occurs. Therefore, the actual shear failure mechanism should be studied at an earlier stage. In a shear compression failure, the arch action turns out to be dominating. For members with large depth and exceptionally low reinforcement ratio (< 0.3%), a different failure mode may occur: dowel failure. Which is defined by the opening of the secondary branch of a major flexural crack along the tensile reinforcement. In the three failure modes, the importance of the crack opening in vertical and longitudinal direction is demonstrated, since the increase of them directly result in the drop of shear force that can be transferred through aggregate interlock. Finally, the results are a valuable input to further improve the Critical Shear Displacement theory. It is suggested that the assumed simplified crack profile is modified to a crack with an angle between 60° to 70° to represent more accurately the results and to allow for larger shear displacements. It is demonstrated that an increase of the shear displacement is critical to triggering the shear flexural failure as proposed by this theory and the actual value of the critical shear displacement is larger than the proposed value based on regression analysis.