The present work investigate the effect of droplets entrainment on critical gas velocity, using the liquid film reversal model from Barnea (1986), Luo et al. (2014) and Shekhar et al. (2017). Especial attention was given to the onset of liquid loading in gas well. Experimental and field data were considered for model evaluation. Field data were taken from published data (Turner et al., 1969, Belfroid et al., 2008 and Veeken et al., 2010). Experiments were performed at the multiphase laboratory (EPT-NTNU) in an upward inclinable pipe. The test section was 6 m long and 60 mm ID. Inclination angles varied from 30 ? to 70 ? from horizontal. The fluids used were air and water. Measurements included fluid velocities and reversal point. High-speed video cameras were used to record the flow regime transition (slug to annular) present in the system. Prediction using the film reversal models revealed that the model over-estimate the critical gas velocity compared to results where entrainment is neglected.
Most gas wells produce liquid as co-produced fluid during well production. Liquid flows along with the gas core as droplets or liquid film on the tubing wall. At the beginning of the production, the gas rate is sufficient to carry all the produced fluid to the surface. However, the declining on the reservoir pressure, the gas production rate decreases until the current gas velocity is insufficient to lift the liquid to the surface. Once this condition is establish, fraction of the liquid starts to flow counter-current to the gas core and accumulates at the bottom of the well, creating a static column of liquid. This accumulation causes backpressure against the formation, which affect the production capacity of the well, making the well produce at unstable flow condition. If the well keep producing at unstable condition, it may lead to a premature abandonment of the well or in some case to wrong well test calculations due to slugging or churning of the liquid.