Undvikande av sprickor i industrigolv : Metodutveckling

A known problem in concrete structures, such as industrial floors, is temperature and foremost shrink induced cracks. The consequences, if these aren’t regarded properly, can lead to costly repairs and to reduced durability and function. Existing technologies could help explain the causes to the cracks origin in a better way and thereby be avoided. These must be combined properly when crack analysis today often is very uncertain since the included external and internal parameters are not properly regarded. In addition, there is a need to calculate crack widths if the cracking risk is too high. From a preliminary model for shrinkage based on moisture levels expressed of water content applies the dehydration program BI Dry, where dehydration profiles obtained for a cross-section with time. The program takes into account the concrete's internal mechanisms and through this the total shrinkage is regarded. On the basis of the preliminary shrinkage model the dehydration profiles are calculated in to fictive temperature movement at the import to the two-dimensional crack analysis program ConTeStPro. At the simulation in ConTeStPro take the imported temperature into count and thus the total shrinkage, autogenous shrinkage and drying shrinkage. From the results can the simulated element curvature and external shrinkage be obtained. Thru this the model for shrinkage can be adjusted in regard to the measured data received from the test sample of which a more accurate model for shrinkage can be obtained. With an improved shrink model performs calibration of the material model for the three-dimensional-based FEM software ANSYS in terms of ConTeStPro, whose material model is based on laboratory experiments. Dehydration profiles from well-defined scenario is calculated in BI Dry and then exported to ANSYS respectively ConTeStPro, at the exportation to ANSYS the fictive temperature movement is calculated separately in Excel. Comparison of results from the stress distribution is carried out in which ANSYS material model is adjusted and calibrated so that the same stress levels as theoretically possible obtains. With more accurate material model in ANSYS the crack distribution of well known scenarios eventually be explained. Currently, this method cannot be performed because the crack analysis program ConTeStPro suffered complications. Attempts were made to circumvent the problems that arose, but without success. This contributed that the preliminary shrinkage model could not be calibrated correctly therefore it became approximate. The shrinkage model comparison was performed by a typical case, concrete slab on ground, between Eurocode 2, The Swedish Concrete Handbook - Materials and the approximate shrinkage model, the results show that the approximate shrinkage model provides much greater and faster development of shrinkage. After which a method trial of a typical case would be exported from BI Dry to ANSYS to verify any emerging problems and interface issues. Due to delays in the thesis there were no time resources left to perform such an experiment which contributed to that the results section of this thesis doesn’t appear. It is very important to understand the consequences that can emerge from changing a concrete recipe. Has crack formation occurred, it is afterwards difficult to evaluate the reason for this only by studying the fracture pattern. Through increased and more continuous studies on samples can updated material models and data be available. Some material data in ConTeStPro and BI Dry is missing today for some concrete classes in which engineering assumptions must be made in the programs. Better material models in the programs makes that they could be regarded in a more proper way. Swedish standard for measuring the shrinkage of samples should be developed to correspond to real conditions. This also applies to measurement techniques and tools to measure the internal deformations and stresses.Validerat; 20120301 (anonymous