Summary

The goal of this work is to study the capillary clogging phenomenon in refrigeration systems employing HFC134a as refrigerant and polyolester (POE) as lubricant. An experimental apparatus, fully described in this work, was specially designed, constructed and calibrated. An experimental actorial design was run to investigate the effect of the following variables in the following ranges on the drop of refrigerant mass flow rate through the capillary tubes: pressure (11bar and 14bar), subcooling (7.5ºC and 12.5ºC), internal diameter (0.57mm and 0.79mm) and length of the capillary tube (3m and 4m), oil concentration (0.5% and 1.0%) and acid type (linear and branched). It was found that the capillary tubes were more easily blocked with linear oils than with branched oils. On the average the refrigerant mass flow rate dropped 27% with the RL7H oil (linear) and 15% with the α8E oil (branched). Measurements using the standardized nitrogen flow technique were also performed, before and after the experiments with HFC134a. The nitrogen flow rate drop was always lower than the one obtained with HFC134a. A numerical model to simulate the pure refrigerant flow and also mixtures of refrigerant-oil flow, with and without oil adsorption, was also developed. It was found that the adsorption of oil ester is governed mainly by the molecular diffusion of the oil in the refrigerant and by the shear stress acting on the capillary tube wall. The model predictions for the pure refrigerant mass flow rate were in a reasonable agreement with the experimental data, with all the data points within an uncertainty band of ±10%. It was also found that the oil effect on the refrigerant mass flow rate was negligible (less than 0.2%). The complete model (refrigerant + oil+ oil adsorption) predictions were compared with experimental data, and it was found that from the 16 data points, 10 (62%) were within an acceptable error band of ±20%.

Material for download