Postpeak properties of high strength concrete cylinders in compression and reinforced beams in shear

Postpeak behavior of concrete in compression is significant because of its underlying use in design codes. Studying the postpeak behavior of longitudinally reinforced concrete beams can provide insight into shear failure. To obtain stable test control during the postpeak, novel closed-loop test techniques were devised which use combinations of multiple signals. To accurately study crack propagation for beams failing in shear, a computer vision technique was employed.; An initial study using circumferential expansion feedback control of normal, medium, and high strength concrete (35, 60, and 100 MPa) cylinders in compression was performed to examine the effects of various testing procedures on measured stress-strain curves. A second study investigated localization in compression by testing normal and high strength concrete cylinders with height to diameter ratios ranging from 2.0 to 5.5. For this study, circumferential expansion could not be used as the feedback since the failure zone location cannot be determined prior to the test. A unique closed-loop feedback control method was implemented which subtracts a portion of the elastic displacement by combining displacement and force. Stable control is maintained even when severe snapback is experienced due to localization. Tests showed that localization caused significant changes in postpeak behavior. A localization model was developed which separates the response of the bulk concrete from the failure zone. Postpeak compression fracture energy of the failure zone did not change with specimen length and was only slightly larger for high strength concrete than for normal strength.; A feedback control method was developed which controlled shear crack propagation in longitudinally reinforced concrete beams by combining beam displacement with distributed measurements of the vertical shear crack opening. A computer vision technique, based on normalized image cross correlation, was applied to monitor and quantify shear crack propagation. A series of beams was tested to study influences of reinforcement ratio, shear span ratio, and concrete strength. In all beams, the initial indication of failure was attributed to loss of bond as a result of combined pull-out and dowel action. The final failure mechanisms and postpeak behaviors were highly dependent on test variables.