Quantifying the Non-Linear Hydraulic Behavior of Pervious Concrete

摘要

Results are presented for the intrinsic permeability and form drag coefficient (C_F) for a set of nominally identical cylindrical pervious concrete specimens. Specimens were tested at a range of flow rates in a constant head permeameter. The mean and standard deviation of the permeability and C_F show substantial variability. This statistical characterization of hydraulic performance was used to model the behavior that would be observed using other standard methods for characterizing hydraulic performance of porous pavement mixtures. Results indicate that falling head, constant head, and infiltration tests have the potential to significantly underestimate the permeability of porous pavement mixtures. Results from modeled falling head tests showed that this test results in permeability estimates up to a factor of ten lower than the actual permeability and are uncorrelated to the actual permeability. For high permeability pavement mixtures, falling head tests will only provide adequate values of permeability if run at heads so low that measurement resolution is a problem. Constant head tests can be used, though model results indicate that, even for very small hydraulic gradients, the head versus flow rate relationship is quadratic. As such, C_F must be calculated. Failure to do so leads to significant underestimation of the permeability. For most applications, the permeability of a well-designed and placed pervious concrete is more than adequate and the improvement in accuracy gained by fully characterizing both the permeability and C_Fof a pavement is not necessary. However, for poorly constructed pavements, or pavements that need to infiltrate large amounts of run-on from adjacent impervious areas, the additional accuracy may be important. The additional level of accuracy is also important for research into improving design and construction of pervious concrete where comprehensive and accurate characterization of test specimens is needed for statistically significant comparisons between different design specimens.