Filter cake washing is a critical process in mechanical solid-liquid separation, particularly in industrial applications where high purity, efficiency (time, equipment, wash liquid demand, ..) and process stability and predictability are required. Additionally, there are specific boundary conditions imposed on the washing process, since it is usually an integral part of the process chain cake formation – pre-deliquoring – cake washing – post-deliquoring on the same filter. As a consequence, there are a variety of process options with respect to stage numbers, type of wash fluid, etc. In the course of the washing process, a wash fluid penetrates the filter cake and displaces the pore liquid that contains dissolved substances (impurities, products) to be removed from the particle system. In contrast to pure de-liquoring, reductions in the concentration of the dissolved substance in the filter cake of 2-3 orders of magnitude are possible under optimum conditions. Especially under well-defined laboratory conditions, these values are achievable and can be described qualitatively and quantitatively with knowledge of the dominant mass transfer mechanisms. It can be shown that an extended dispersion model has proven to be suitable for this and allows an easy parameter estimation. However, these washing results are difficult to achieve in industrial washing processes, sometimes with serious consequences for the design and operation of a filter. One reason for this is that scale-dependent effects can usually be neglected in laboratory tests. Scale-dependent effects include deviations of all kinds from the well-defined laboratory experiment. The examples of “deviation of the cake geometry”, “segregation within the filter cake” and “wash liquid application” are used to demonstrate how even seemingly small deviations have a massive effect on the course of the washing process. Model considerations and simulations are presented and validated with experiments, which show the quantitative effects of these imperfections and which optimization potentials arise. The aim of the presentation is to demonstrate how to set up a path between targeted laboratory tests and the industrial implementation of a washing process, thus pushing the limits of understanding and reliable application of washing processes.