Abstract
Natural gas wells face flow pattern changes and liquid accumulation issues as their pressures decline. This issue, if not addressed properly, hinders the production, and may eventually kill the well. The onset of the liquid accumulation corresponds to the beginning of liquid fallback in the tubing and the transition from annular to churn flow. In this experimental study, the impact of partial tubing restrictions, namely insert rings, on liquid loading is assessed. Partial restrictions in the tubing, such as insert rings, are believed to amplify droplet entrainment and interfacial shear, and hence delay the liquid loading. This theory is tested experimentally in this study. Two-phase tests are performed using air and mineral oil in a flow loop with a vertical length of 25 ft, made from 2-inch ID clear acrylic pipe. The tests are carried out in room temperature and atmospheric pressure. Differential pressure transducers, quick-closing valves, and a high-quality camera are used to measure the pressure gradient, liquid holdup, and flow pattern for each test. Three liquid rates, corresponding to gas wells are tested. Gas rates are varied to cover a wide range from annular to slug flow. The tests are conducted without inserts and repeated with three insert sizes. In addition, the number of inserts along the test section is changed to test the impact of spacing on the effects provided by inserts over the flow.
The objective of this study is to quantify the effects of restrictions with various sizes and their spacing on liquid lifting. It is found that the inserts enhance liquid lifting by two mechanisms, liquid fallback prevention and droplets generation. The latter is explained by the collision of the upward flowing liquid with the inserts and the liquid bridge rupture by the gas flow. At the same time, the inserts restrict the effective flow diameter and cause an increase in frictional losses. The experimental results suggest there is an optimal region for the use of inserts, in which liquid lifting is enhanced with minimal increases in frictional losses. This region varies slightly depending on the flow rates and it is found mostly in the churn flow region. Within this window, the pressure drop decreases by up to 50% as a result of adding the inserts to the tubing. Also, it is observed that decreasing the number of the inserts helps reduce the frictional pressure losses, while maintaining the reduction in liquid holdup. This is a cheap, passive, and efficient method for liquid unloading of gas wells. The tubing joints could play the role of the inserts with an appropriate design, making this technology easily applicable.