Abstract A unique large scale pilot plant of the CellFlux thermal energy storage concept is experimentally investigated. This storage concept consists of a regenerator type thermal energy storage volume, which is coupled to a finned tube heat exchanger by a circulating intermediate working fluid. The system investigated in this work operates at a temperature of 390 °C and uses air as intermediate working fluid which is conveyed by a centrifugal fan. The storage volume has a bed length of over ten meters and is of a novel design, where the air flows in horizontal direction. Since this approach could cause a flow maldistribution, a thorough analysis is of major interest for the accuracy of subsequent numerical simulations. The experiments reveal that the mass flow along the centerline can be up to 20% higher than the mean bulk flow. A significant maldistribution between top and bottom area, however, is not observed. As an alternative to the typically used rock filling, the storage volume is equipped with standard hollow bricks. These bricks are cost effective but do not have a well-defined shape. Thus, the predictability of the pressure drop by correlations found in the literature is unclear. It turns out that the measured pressure drop is evenly distributed in axial flow direction but generally higher than expected from the assumption of pure channel flow. Further experiments are conducted to validate the heat capacity of the bricks and to derive a correlation for the inner heat transfer between bricks and storage walls. Eventually, the aim of the experimental investigation is a general proof of concept as basis for the numerical investigation. Thus, all specifications of the plant and the storage material are provided. The plant is analyzed towards plausibility of heat losses, showing that heat losses can be predicted well within the given uncertainties.