Selecting superior fin geometry among four suggested geometries for shell and helically coiled finned tube heat exchangers with numerical simulation and experimental validation

In this paper, the authors investigated the shell and helically coiled finned tube heat exchanger numerically and experimentally. At first they investigated the shell and helically coiled finned tube heat exchanger with simple annular fins for the first time. They considered nine cases for experimental investigation, in which three different fluid flow rates are considered for each side. Then they studied the geometry numerically using Ansys Fluent 18.2 software. After validating the numerical model, the authors suggested four new fin geometries. They studied all the geometries numerically and selected the stepped fin geometry as the superior geometry. The fluid flow inside the tube in this work is in the turbulent flow range and the realizable k -epsilon turbulence model is used. The authors proposed correlations using the response surface methodology and Taguchi method to predict the mean Nusselt number on the shell and tube sides. The presented correlations have good accuracy for calculating the Nusselt numbers (maximum 10% error). The results show that the mean Nusselt number of the coil side is the same in all geometries, and the Reynolds number of each side has almost no effect on the mean Nusselt number of the other side, and with increasing the fin pitch, the mean Nusselt number of the shell side decreases. The mean Nusselt number of the shell side is 80.79% higher for the stepped fin geometry (superior geometry) compared to the simple annular fin at best.

Automated summary

In this study, the authors investigated the performance of a shell and helically coiled finned tube heat exchanger using numerical and experimental methods. They examined the heat exchanger with simple annular fins for the first time and proposed four new fin geometries, selecting the stepped fin geometry as the superior design. The authors developed correlations to predict the mean Nusselt number on both the shell and tube sides, with good accuracy (maximum 10% error) for calculating the Nusselt numbers. The results showed that the mean Nusselt number of the coil side was consistent across all geometries, and the Reynolds number had little effect on the mean Nusselt number on the other side.