The realisation of fuel cell stacks with maximal power density combined with low production costs requires a high level of optimisation of all contributing phenomena. In addition to several other optimisation factors, an optimal flow channel design plays a central roll, as the occurance of liquid water in the cell can be the cause for a significant power drop. An acceptable improvement of existing models without a detailed experimental investigation of the underlying cause-effect relations between operating point, power density, two-phase flow regime and power drop proved to be extremely difficult. The objective of this project ist twofold. To advance beyond the present level of model developement, the first part builds upon the extensive experimental collection of reference data for characterising and cataloguing the relevant two-phase flow regimes that is needed to model the underlying causalities. In the latter part of the project, the collected data is used to develop improved liquid water models, which are validated towards improved ability of the prediction of flow regimes inside the flow channels.