Predictive method for flow condensation heat transfer in plain channels

The flow condensation heat transfer (FCHT) in plain channels has a variety of applications in many industrial sectors, and thus it is important to predict FCHT coefficients accurately. This paper compiled a large compound FCHT database containing 5607 data points and 30 fluids and presented a strategy for developing a new correlation of FCHT coefficients. The parameter range of the database is the hydraulic diameter D = 0.493–20 mm, vapor quality x = 0.003–0.992, mass flux G = 24–1533 kg/m2 s, heat flux q = 2.9–422 kW/m2, reduced pressure pr = 0.04–0.95, liquid Prandtl number Prl = 1.7–8.5, liquid Reynolds number Rel = 11.6–5.3 × 104, and gas Reynolds number Reg = 75.1–9.1 × 105. Based on the database and strategy, a new general correlation with substantially better accuracy was developed, which is applicable to plain channels of various sizes and a broad operational parameter range. It predicts the database with a mean absolute deviation (MAD) of 14.1%, while the best existing ΔT-independent correlation predicts the database with an MAD of 20.2%. The applicability of the new and 38 existing correlations to individual fluids was assessed. The new correlation performs best for 8 of the 14 fluids that have more than 50 data points in the entire database, while the most accurate existing one performs best only for 2 of them. The Fang number Fa plays an important role in the new correlation, implying that it relates to the fundamental mechanisms of both boiling and condensation heat transfer.