scope:

WET GAS DEFINITION AND CLASSIFICATIONS

Defining wet gas is not an easy task. Historically multiphase flow where gas volume fractions (GVF) have exceeded 90% or 95% has been called wet gas. However, GVF is based on volumetric flow rates at actual conditions in the pipe, and doesn't account for relative differences in the gas and liquid densities. Since many successful devices used for wet gas measurement employ differential methods that are strongly affected by the densities of the gas and liquid relative to one another, the Lockhart-Martinelli parameter is often utilized in defining the boundary between wet gas and other multiphase flow. The Lockhart-Martinelli parameter is defined as

where Ql and Qg are the liquid and gas mass flow rates, and and are the densities of liquid and gas at meter conditions. Since mass flow is volumetric flow multiplied by density, we can also define the Lockhart-Martinelli parameter of the wet gas flow in terms of actual volumetric flow rates Qvl and Qvg.

Based on experience gained in flow loop tests, it has been suggested that when the Lockhart-Martinelli parameter for a fluid remains below about 0.35, its behavior is such that many common methods employed for wet gas flow measurement work as they have been designed. Above this boundary these methods may begin to break down and cannot be counted on to yield reliable answers.

The magnitude of the effort that a producer should expend to estimate liquid hydrocarbon production should reflect its importance relative to the produced gas based on its mass flow rate. There will be a class of wet gas where the mass flow of liquid hydrocarbons is insignificant relative to that of the hydrocarbon gas. This shall be called Category 1 wet gas. There will also be a class of wet gas in which the liquid hydrocarbon mass flow is of sufficient magnitude to warrant its careful measurement and recovery. This shall be called Category 2 wet gas. The boundary between the two will normally be at a point where the mass flow rate of the hydrocarbon liquid is 5% of that of the gas.

LIQUID HYDROCARBON MEASUREMENT

A central problem that must be addressed for those using wet gas meters is the determination of the liquid hydrocarbon flow rates of a well stream. A key issue is that water and hydrocarbon liquids co-exist in the liquid phase of the stream. Furthermore, the liquid measured by the wet gas meter may contain injected chemicals (hydrate inhibitor, corrosion inhibitor, etc.), in addition to the condensate, oil, and/or water. In either case discussed below, the volume of injected chemicals flowing through the wet gas meter must be known and input to the computations.

Dependent on whether the wet gas that a particular well is expected to produce is Category 1 or Category 2, the effort to estimate liquid hydrocarbon flow rates will range from very little to very much. The general procedure will be as follows:

1. Determine if there is an online method of measuring water volume fraction available that can be used in the application.

2. Obtain and analyze a sample of the reservoir fluids for each well prior to the onset of normal production. Determine the gas-oil ratio (GOR) of each.

3. For Category 1 Wet Gas, an average GOR may be utilized across all producing wells in the system.

4. Using the GOR derived from these samples and adjusted to each allocation meter's conditions, apply these factors to the gas production for each well to obtain the liquid hydrocarbon production for each.

5. For Category 2 Wet Gas, if the liquid hydrocarbon imbalance grows beyond a predetermined threshold, one of two avenues must be pursued:

a. Actions must be taken to remedy the imbalance. This could involve acquiring a new sample from a well or all wells in the system, or re-estimating the GOR from secondary data sources. Strategies for doing this are considered in Chapter 7 on Abnormal Operations,

or

b. A justification acceptable to all interested parties must be made to explain why choosing (a) is not appropriate.

In the general case, a project will consist of a combination of Category 1 and Category 2 wells, therefore the plans for production must account for this.

SCOPE SUMMARY

Until a better alternative is found, liquid hydrocarbon measurement will be accomplished by utilizing whatever sampling information is available to determine the well's water volume fraction and GOR. Dependent on the degree of difficulty in obtaining the sample and on the importance of the liquid hydrocarbon production, repeating this activity to obtain new information on the fluid properties may be done infrequently. Although an operator will certainly have a production sample acquired from each well at its startup (i.e., from a wireline sample-taking tool, or from the flow back to the completion rig) unless the system falls out of balance, there is no requirement to take further samples.

Another problem that must be addressed is the fact that the conditions at the subsea meter will be quite different from those at the reference measurement point at the host processing facility. PVT analyses must be applied to account for phase changes incurred due to the tieback flowline length and differential water depth, as well as any other changes in pressure and temperature that might alter the phase state of the fluid. This will affect both the liquid and gas measurements, and will increase the difficulty of the task. This whole subject of mass transfer between phases and its effect on measurement uncertainty is addressed in Appendix A.