Vibration of Hollow Core Concrete Elements Induced by Walking

Sammanfattning: Historically, traditional concrete floors have performed well with regard to vibration serviceability. This is much due to their heavy weight. However, the use of stronger concrete materials and prestressing has resulted in slender cross sections and the possibility to build long-span floor elements. The combination of long span and relatively light weight means that the floor element is more sensitive to vibrations. This thesis investigates vibration in hollow core concrete elements induced by human walking. One of the objectives of the thesis was to establish a maximum span with respect to vibration serviceability for the smallest HD elements in the series, HD/F 120/20. Another objective was to investigate how the dynamics of a floor structure is affected when different types of connections are used, when different spans are used, or when a concrete topping is cast. An experimental floor structure made of three hollow core elements was built in a laboratory. The floor was simply supported and the span of the floor was 8 m, which is within the recommended limits. Subjective tests were performed before and after a concrete topping was cast. The results from the subjective tests showed that a large majority of the test persons found the experimental floor structure unacceptable with regard to vibrations induced by another person walking. The results from the subjective tests indicated that a concrete topping improved the vibration performance slightly, but the vibrations were still classified as clearly perceptible or strongly perceptible. Accelerometers were used to measure the accelerations of the slab induced by a number of different persons walking, one at a time. For offices many researchers propose a single person walking to be the governing load case when checking the vibration serviceability of a floor structure. This is also the load case that was used in this study. A limitation of this master thesis is however that only one walking line was examined. From the measurements the dynamic properties of the test floor, such as damping and natural frequencies, could be determined. The improvement of the vibration performance of the experimental floor structure after a concrete topping was cast could also be seen in the measurements; in this case in the form of slightly reduced average values of overall weighted acceleration. The measured acceleration magnitudes were also evaluated according to the former ISO standard ISO 2631-2:1989 and a method proposed by Talja & Toratti. Both methods indicated the same as the subjective tests: the experimental floor structure is unacceptable with regard to vibrations. When the vibration signals were filtered and plotted in 1/3 octave bands it could be seen that the highest acceleration magnitudes were found in the frequency interval of maximum human sensitivity. A finite element model of the experimental floor structure was built in Abaqus. In order to decrease computational time and cost of each analysis the three-dimensional structure was replaced by a shell with the same stiffness properties and density as the hollow core elements. The results from the measurements were used to validate the FE-model of the floor, and then a number of simulations of different boundary conditions and different spans were performed. During the simulations it was found that the frequency content of the applied load function affected the resulting accelerations significantly. This sensitivity made it difficult to draw any clear conclusions about the maximum possible span, based on only the three different load functions that were used. However, it was concluded that supports with some rotational resistance, or a shorter span will improve the vibration performance. The calculations also showed that there is a probability of adverse comment in the case of simply supported slabs with spans of 6 or 7 m. In order to investigate if the problem with annoying vibrations in HD elements is widespread, a handful of interviews were performed with structural designers. Only one of the interviewed persons had come into contact with a case where people were complaining about annoying vibrations induced by human walking. However, one must keep in mind that only structural engineers were interviewed, and the picture might have been another if occupants of different office buildings had been interviewed.