Condensation heat transfer and pressure drop correlations in plate heat exchangers for heat pump and organic Rankine cycle systems

An accurate prediction method of flow condensation heat transfer in plate heat exchangers is important for the design of condensers. The existing prediction methods are semi-empirical correlations, whose applicability highly rely on the range of original test data used for the correlation development. In this study, prediction methods of frictional pressure drop and heat transfer coefficient of flow condensation in a plate heat exchanger which were tailored for the applications in heat pump and organic Rankine cycle systems, were developed. We conducted a comprehensive experimental investigation, building a database including 283 data points for both pressure drop and heat transfer. We tested seven working fluids, classified by hydrofluorocarbons - R134a, R236fa and R245fa, hydrofluoroolefins - R1234ze(E) and R1233zd(E), and hydrocarbons - propane and isobutane, for a wide range of condensation temperatures varying from 30 degrees C to 90 degrees C. Two dimensionless numbers, the density ratio of liquid phase to vapour phase and the Bond number, were involved for developing the new prediction methods. The paper presents data for higher temperatures than those presented in previous works, and novel analysis and prediction methods. The new correlations achieve an improved prediction, enabling an 8.9% and a 10.3% mean absolute percentage deviation for heat transfer and pressure drop data, respectively.

Automated summary

In this study, tailored prediction methods for flow condensation in plate heat exchangers were developed for the applications in heat pump and organic Rankine cycle systems. A comprehensive experimental investigation was conducted, including 283 data points for pressure drop and heat transfer. New correlations utilizing two dimensionless numbers achieved an improved prediction with mean absolute percentage deviation of 8.9% for heat transfer and 10.3% for pressure drop data.