Liquid Lifting From Natural Gas Wells: Tubing-Casing Junction

Abstract

Effective removal of water and hydrocarbon liquids from gas wells can maintain and increase gas production.In this paper, results of flow loop tests on transport of liquid from casing to tubing is reported for flow rates ranging from 50% below to 50% above the critical gas flow rate for the tubing.The results show that liquid transport is severely constrained at the tubing-casing junction because the gas flow rate is usually much below the casing critical flow rate.To resolve the tubing-casing bottleneck, six approaches were tested: constrictions in the tubing, a vortex-generating device in the tubing, an assembly of baffles in the casing, a ball pack in the casing, a closed-end pipe in the casing, and foam-generating surfactants.The first two approaches provided no benefit.The last four approaches enhanced liquid lifting, in some cases by a factor of 10 or more.All approaches that increase liquid transport from the casing to the tubing also increase pressure gradient.Approaches that boost liquid transport rate with a small increase in pressure gradient are most desirable. Introduction When initially completed, many natural gas wells are capable of lifting liquids to the surface.But, with depletion of the reservoir pressure, there comes a time when liquids can no longer be lifted to the surface and they begin to accumulate in the bottom of the well, dramatically inhibiting or stopping gas production.The water that accumulates in these wells may be liquid water produced from the gas bearing formation or vaporized water, which condenses as it travels up the well, and then falls back to the bottom of the well.Water and hydrocarbon liquids that accumulate in gas wells decrease productivity by increasing back pressure on the reservoir and by forming water blocks in the near well rock formation.[1]To maintain or increase gas productivity, these liquids must be removed.A key factor for water removal is the location of the end of the tubing in the casing relative to the liquid level in the casing and to the various gas-bearing formations that have been completed. In this paper, the focus of attention is transport of liquids to the tubing from the casing below the end of the tubing.The work of Turner et al.[2] with the later adjustment by Coleman et al.[3] provides a good starting point for understanding the problem.Turner et al. proposed an expression for estimating the critical flow rate at which liquids can be lifted up the well by a flowing gas stream.The critical flow rate is the product of the cross-sectional area of the conduit and the critical velocity: (1) in which M is between 0.57 (as recommended by Coleman et al. for pressures below 500 psia) and 0.68 (as recommended by Turner et al. for pressures near 2000 psia) for critical velocity in units of ft/sec, liquid density Pl and the gas density Pg in units of g/cm[3], and gas-liquid interfacial tension slg in units of dyne/cm. For conditions typical of many mature gas wells in the US, the critical rate is between 300 and 600 Mscf/day for flow in 2 3/8-inch tubing (which has an ID of about 2 inches). With casing IDs of 4 or more inches, the critical flow rate in the casing is four or more times the tubing critical rate.More than 90% of US gas wells produce below the critical flow rates for tubing; as a result, liquid accumulates in the casing below the tubing. Moving liquids from the casing to the tubing requires impractically high rates with conventional equipment in the well bore.In this study, we looked at several alternatives for enhancing lifting of liquids near the tubing critical flow rate using a flow loop to test the designs.Although some of these alternatives are already in commercial use, it is easy to test variations in the flow loop to see if commercial applications are anywhere near optimum design.In the remainder of the paper, the flow loop operation will be described, followed by presentation of results and discussion of questions inspired by the results.