Prediction of concrete anchor loads by finite element methods for flexible baseplates

To determine the accuracy of predicting anchor loads for flexible baseplates by finite element analysis, tests were conducted at The University of Tennessee in the Fall of 1983. The test results are compared to the analysis by finite element methods for possible correlation. Tests were conducted on five bolt patterns which included four, six, and eight bolt groups. Also investigated are the effects of a grouted baseplate compared to a baseplate fastened directly to the concrete surface, as well as the effects of bolt preload. A uniaxial moment was induced to the bolt patterns through a cantilevered tube section welded to the plate. Plate thicknesses of 1/2, 3/4, and 1 inch were used. Test data were recorded by an automatic data acquisition system capable of reading 22 channels of input which included bolt loads, plate stresses, applied load, and deflection.

Results indicate that maximum anchor loads for grouted baseplates can be reasonably approximated by finite element analysis; however, finite element analysis does not always account for redistribution of loads due to plate flexibility. The finite element program can not model the irregular surface conditions for nongrouted baseplates and subsequent distribution of load to the anchors. There also appears to be an inability to adequately model preloaded anchors, especially for the thinner plates, by the finite element program.