Abstract:
This study investigates the thermal–hydraulic performance of an additively manufactured heat exchanger (AMHE) operating in a nitrogen-nitrogen counter flow open loop. The AMHE, consisting of ten semicircular mini channels with diverging inlets and converging outlet headers for both hot and cold fluids, was 3D printed using the Selective Laser Melting (SLM) technique with AlSi10Mg. The rough surface of its internal channels is characterized by using a cut sample with Field Emission Scanning Electron Microscopy (FESEM) images and a surface profilometer. An open-loop experimental test facility was developed to evaluate AMHE performance. Experiments are conducted by varying balanced mass flow rates (1.11 to 4.44 kg/h) and hot inlet temperatures (324.9 to 353.0 K). Balanced mass flow rate, temperature, and pressure measurements were recorded at steady state, and heat transfer rates and channel pressure drops were calculated. AMHE achieved a maximum power density of about 125.4 kW/m3 at a low log mean temperature difference (LMTD) of 6.5 K in a counter-flow arrangement. The experimental results were compared with standard ∊-NTU correlations available in the literature and showed agreement within 1 %. We noted that the effectiveness and entropy generation increase, and axial conduction decreases with an increase in balanced flow rates. A multivariable regression model was developed to predict the experimentally obtained heat transfer rate and pressure drops within a 2 % error limit and used to predict the effect of various operating conditions. Parametric results showed that increasing the balanced flow rate and hot inlet temperature enhanced the heat transfer rate by a factor of about 5, with the corresponding pressure drop rising by up to a factor of 10. This novel combined experimental and multivariable regression approach provides practical predictive correlations for gas-to-gas mini-channel heat exchangers, compensates for input variations, and enables reliable performance estimation under varied operating conditions, offering a valuable contribution for future design and optimization.