MASS MEASUREMENT OF NGL MIXTURES MORE ACCURATE THAN VOLUMETRIC
Gerald C. Armstrong
Phillips Pipe Line Co.
Odessa, Tex.
Virtually all hydrocarbons are bought and sold by volume, i.e., gallons, barrels, standard cubic feet per meter, liters, etc. But variations in component content and density make it difficult to measure NGLs in this conventional manner.
Measuring NGLs first by mass measurement is more accurate. This measurement can then be converted to volumetric terms for commercial transactions.
COMPONENT VARIATIONS
Natural gas liquids are generally described by use of analyses performed by chromatography. Fig. 1 shows a printout of a typical analysis. Notice the sample is made up of several single components, in addition to hexanes and C7+. The last two contain several components each.
The actual makeup of a given NGL sample is characteristic for that particular sample. Another sample, having the same smell, color, specific gravity, etc., may not, indeed probably won't, have the same component percentages.
Table 1 shows 10 analyses, each having 0.475 sp gr. The samples all probably had the same odor and color. These are actual analyses.
Notice the ethane (C2) content varies from 36.73% to 41.69%. That is almost 5% of the total sample. The density of C2 is 2.97 ppg.
The propane content varies from 28.72% to 34.27%. The density of propane is 4.23 ppg.
These variations in component content and density, therefore, make measuring NGLs in the conventional manner difficult.
VOLUMETRIC MEASUREMENT
Liquid measurement by volume employs a volume correction factor (VCF) to adjust the measured volume from operating conditions to base conditions of 60 F. and equilibrium pressure. These base conditions are industry-accepted "standard conditions."
The adjusted measured volume is called "equivalent" or "net volume." The VCF uses the measured or observed temperature and specific gravity to generate the adjustment for thermal expansion.
Tables are available to provide VCFs as a function of specific or API gravity of the product and the temperature at which it is measured. The equivalent of net volume at base conditions is equal to the metered volume multiplied by the VCF.
This volumetric method of measurement is accepted by the liquid producing, processing, and transporting industries in the U.S. and other countries. It is best for crude oils or refined or commercial grade products because their behavior is predictable.
Volumetric measurement assumes the liquid is a pure compound, with a particular specific gravity, and that it will expand and contract predictably with a change in temperature.
NGLs are mixtures of hydrocarbon compounds, many of which must be pressured to keep them in a liquid state at the measured temperature. The physical behavior of NGLs is not very predictable.
Two NGL streams may have the same specific gravity but differ greatly in composition (Table 1).
The different compositions cause the two streams to behave differently with a change in temperature. They will not follow the same thermal expansion pattern as predicted by any given VCF.
Therefore, using a volume correction factor that depends on the specific gravity to correct for thermal expansion will be inaccurate for these streams.
VOLUME SHRINKAGE
Another factor which must be taken into account for measurement of NGL streams is volume shrinkage.
To understand volume shrinkage, recall that NGLs are mixtures of hydrocarbons.
Very often, small amounts of methane and significant amounts of ethane are mixed with propanes, butanes, pentanes, and heavier hydrocarbons. The lighter hydrocarbons, with much smaller (mass) molecules, tend to mesh into the molecular structure of the more massive molecules of the heavier hydrocarbons.
A simple analogy may be made to a mixture of gravel and sand. If the sand grains penetrate and fill the void spaces between the larger particles of gravel, then 1 gal of sand and 1 gal of gravel will obviously not produce 2 gal of mixture.
The molecules of methane and ethane, when mixed with the heavier hydrocarbons, do a similar thing. Thus, these mixtures, have less volume than if all components were measured separately.
One more problem with using volumetric measurement for NGLs is compressibility.
The VCF assumes that, if a liquid is at or above its equilibrium pressure, it is in a liquid state and incompressible. The only problem with that line of reasoning is that in their liquid states, NGLs are compressible. Most pure liquid compounds are not compressible to any great extent, but hydrocarbon mixtures are.
One must conclude that the volume metered into a pipeline system will be different from the metered volume exiting the same system if the pressures are different.
There are, then, three important reasons for not depending on volumetric measurement for metering of NGLs:
- The VCF which corrects for thermal expansion is a function of specific gravity (or API gravity) instead of the stream composition.
- Volume shrinkage due to molecular mixing cannot be corrected for.
- There is no complete correction for compressibility.
A typical volumetric flow measurement calculation is shown in Equation 1 (box).
MASS MEASUREMENT
Because volumetric measurement is not accurate for metering NGLS, industry has sought and found a better method of measurement.
When volume shrinkage occurs, the density increases. We have no problem measuring volume (under operating conditions) and no problem measuring density. Consequently, if we can come up with an accurate composition of a given product, the volume of individual components in a measured stream can be determined accurately.
Mass is an absolute measure of the quantity of matter. A pound of one component added to a pound of another component always produces 2 lb of the mixture.
Referring to the gravel and sand, remember that when a gallon of each was mixed, the resulting mixture was not 2 gal but somewhat less. By weight (mass), however, we don't lose.
Assume the gallon of sand weighs 22 lb and the gallon of gravel weighs 16 lb. When mixed together at the same elevation, the mixture will weigh 38 lb regardless of the resulting volume.
Mass measurement is not affected by temperature, pressure, or volume shrinkage. It eliminates the need to compensate for compressibility and nonideal mixing because correction factors are not required to convert the measurement to standard conditions.
Chapter 14, Section 7 of the API measurement standards documents the mass measurement method.
For a measurement of liquid mass, the following are necessary:
- A volumetric measurement at operating conditions
- A density measurement at operating conditions
- An accurate analysis of a composite sample of the stream for the duration of measurement.
The mass flow is the product of the volume multiplied by the density. The composite sample is analyzed by chromatograph to determine the weight-percent of each component of interest,
The equivalent volume of each component at standard conditions is the product of the total mass multiplied by the weight-percent of the component fraction divided by the density of the pure component at standard conditions.
The volumetric measurement is obtained from any basic volume meter. Typical are turbine meters, positive displacement meters, orifice meters, magnetic meters, or vortex-shedding meters.
The density in mass-per-unit volume is generally obtained from a frequency-type densitometer, either vibrating reed or sonic tube. Analog-output density meters are used, but they do not typically provide the accuracy of frequency types.
The composite sample is normally pumped directly from the flowing stream or from a representative slip stream into a large sliding-piston cylinder kept under a pressure higher than the line pressure.
Line pressure should be 150% of the vapor pressure of the stream at flowing temperature.
The pump is operated so that the amount of sample is directly proportional to the volumetric flow rate or the mass rate.
PREVENTING INACCURACIES
Precautions which must be taken to prevent measurement inaccuracies include the following:
- Temperature-compensated flow meters must not be used in mass measurement.
- The flow meter and the densitometer must be at the same temperature and pressure.
- The stream must be all liquid. Cavitation or flashing will invalidate the measurement.
- The measurement and calibration proving must be done at normal flowing conditions.
- The composite sample must be representative of the quantity of product measured, i.e., the same temperature, pressure, and composition.
- Transfer of the sample and the chromatographic analysis must be accurate.
The typical basic mass-measurement facility (Fig. 2) consists of the following components:
- Turbine meter (or other volumetric flow meter)
- Densitometer
- Back-pressure regulator
- Sampling system
- Proving connections for turbine meter and densitometer totalizer (usually a microprocessor)
- Check valve to prevent back flow (system can increase in both directions).
In addition, a chromatograph capable of extended analysis, usually dual column-dual detector, is needed (GPA 2186-86).
A typical mass-measurement flow calculation is shown in Equation 2.
Once the mass measurement is complete and the composite sample is collected, an analysis of the sample is necessary. The analysis will look something like that shown in Fig. 1.
Once this analysis is obtained, we know the percentage by weight, hence the exact weight of each component in the shipment.
Remember that hydrocarbon fluids are sold by volume, however, not by weight. Now, we need to figure out how many gallons, barrels, etc. equate to each component weight fraction.
This procedure is shown in Equation 3.
See Fig. 3 for an example of a mass conversion to volume.
Copyright 1990 Oil & Gas Journal. All Rights Reserved.