Conference Year
2018
Keywords
small-diameter, heat exchanger, correlation, heat transfer, air-side
Abstract
This paper examines a recent effort to experimentally test copper-tube, aluminum-fin heat exchangers (HXs) to validate CFD-based correlations for air-side thermal-hydraulic performance of enhanced fins with small-diameter tubes (Sarpotdar et al., 2016). Sample HXs with 5 mm diameter tubes were prepared from four manufacturers and covered a range of unique slit- and louver-fin geometries, fin densities, and tube rows. Experimental data was collected for heat exchangers tested in the dry condition using water as the tube-side working fluid and air as the external fluid with velocities ranging from 1-4 m/s. Rather than following the conventional ε-NTU-based data reduction procedure outlined by Wang (2000), a finite-control-volume-based heat exchanger modeling software was used to reduce heat transfer coefficients from experimental data. The presented experimental results serve as a reference for the air-side performance of these 5 mm slit and louver fin HX geometries, but also validate the predictions of a CFD-based correlation that covers a much larger design space, including tube diameters from 3-5 mm. Correction factors were applied to the existing correlations based on experimental findings and the final correlations were able to predict 100% of all observed air-side pressure drops and 98% of observed heat transfer coefficients with 20% error less. The findings suggest that the CFD-based approach is suitable for developing correlations for air-side heat transfer and pressure drop. This method is highly desirable because of its ability to simulate a wide-ranging design space that could not be feasibly evaluated experimentally, including heat exchanger geometries that cannot be manufactured at present. However, some deviations between CFD predictions and experimental observations are evident and the work underscores the importance of experimental validation and tuning. It is hypothesized that factors which are not accounted for in the CFD model, including thermal contact resistances between fin and tube, may contribute toward the CFD correlation’s consistent over-prediction of observed heat transfer.