Abstract
Representative Elementary Unit (REU) simplification has become a standard approach for CFD modeling of structured packings; however, its reliability in predicting dry pressure drop remains a critical yet underexplored issue. In particular, the sensitivity of hydraulic performance to internal geometric parameters is often overlooked in the current literature. This study addresses this gap by calibrating an REU model for Mellapak 500.Y against the semi-empirical correlation of Stichlmair et al. (1989) and experimental data from Tsai (2010). Single-phase simulations were conducted using the SST 𝑘−𝜔 turbulence model over a gas-load range of 0.72–4.32 Pa0.5. Grid independence was established at 0.53 million cells per REU. A domain-sensitivity analysis using one, four, and six REUs yielded deviations below 10%, confirming that a single-unit domain provides sufficient accuracy for dry-flow calibration. The results show that the nominal 2.00 mm sheet spacing leads to substantial underprediction of pressure drop, whereas an effective spacing of 0.24 mm brings the CFD predictions into agreement with Tsai’s measurements within a ±20% error band and consistent with the Stichlmair correlation across the operating window. These findings identify the effective sheet spacing as the dominant geometric calibration parameter and provide a reproducible, computationally efficient baseline REU for subsequent hydrodynamics simulations in Mellapak 500.Y.