HYDROSTATIC TANK GAUGES ACCURATELY MEASURE MASS, VOLUME, AND LEVEL

Frank J. Berto
Consulting engineer
San Anselmo, Calif.

Hydrostatic tank gauging (HTG) systems accurately measure the hydrostatic pressure of material in atmospheric and pressurized storage tanks, and use the results to determine the mass, liquid density, volume, and level of the material in the tank. The systems are suitable for automatic tank-gauging applications.

Many HTG systems have been installed on storage tanks around the world, and the American Petroleum Institute (API) and International Standards Organization draft standards on automatic tank gauging (ATG) systems include HTG'S. Therefore, an examination of the operating principles and accuracies and inaccuracies of HTG's is in order.

Seven years ago, Exxon Co. issued detailed specifications describing a new method for measuring the mass, volume, or level of the liquid in a storage tank. Three instrument companies developed hydrostatic tank gauges to meet the specifications. Today, there are approximately 2,000 HTG's in service.

Because HTG's measure hydrostatic pressure rather than level, they differ from other tank-measuring methods. Mass, density, volume, and level are derived from the hydrostatic pressure measurements.

OPERATING PRINCIPLES

An HTG on an atmospheric tank consists of two precision pressure sensors and a resistance temperature detector (RTD).

Pressurized tanks require a third pressure sensor to measure the pressure in the vapor space. Fig. 1 shows the components in a typical HTG system.

One pressure sensor (P1) is mounted on the shell of the tank, just off the bottom. The second pressure sensor (P2) is mounted about 8 ft above the first sensor.

The P, sensor measures the total hydrostatic pressure of the liquid in the tank. The bottom hydrostatic pressure multiplied by the tank area gives the mass of liquid in the tank:

Mass = P1 x Tank Area (from tank strapping table)

The difference between the P1 and P2 pressures permits calculation of the liquid density at the storage temperature measured by the RTD:

Density = (P1 - P2) x G/H

where: G is the gravity coefficient and H is the distance between P1 and P2.

The temperature sensor permits calculation of standard density at 60 F. by using the appropriate API coefficient for the storage temperature measured.

Dividing the mass by the standard density gives the standard volume:

Standard volume = Mass/Density at ref. temperature

Dividing the mass by the liquid density and the tank area gives the tank level:

Level = P1/Density

The pressure and temperature signals are connected to a hydrostatic interface unit (HIU) that is mounted near the transmitters. The HIU contains the logic to calculate the liquid mass, density at the liquid temperature, standard density at 60 F., volume at storage temperature, and standard volume at 60 F.

The HIU also has microprocessor memory to retain a set of tank capacity tables for the tank.

NECESSARY DEVELOPMENTS

Three developments during the last 10 years have made HTG's feasible:

Accurate, stable, micro-processor-equipped (smart) pressure transmitters
Digital output signals
Inexpensive microprocessor logic.

Smart pressure transmitters use a variety of sensing mechanisms. One vendor uses a vibrating wire where the frequency of the vibrations changes with tension caused by pressure.

Another uses differential capacitance where the difference in capacitance is proportional to pressure, and a third uses a precision strain gauge where strain is proportional to pressure.

To obtain the high precision required for HTG service, the smart pressure transmitters contain a microprocessor, and they are temperature characterized and electronically linearized. They have fixed span, and the stable accuracy is typically 0.05%.

Smart HTG transmitters have either digital or frequency outputs. With a digital output, the only limit to accuracy of the signal transmission is the number of bits (typically 32) in the digital word.

Digital output avoids problems with the repeatability of analog-to-digital converters and problems with field calibration of analog signals.

The microprocessor circuits in the HIU were the final breakthrough. Essentially, perfect calculation accuracy is provided so that the total system accuracy is limited only by the accuracy of the pressure transmitters.

The system allows a number of corrections that can be input into the HIU. The height of the P, transmitter off the bottom of the tank, and the distance (H) between P1 and P2 can be input to the system (Fig. 1).

Correction for the change in H due to ambient temperature changes can also be input. (Some systems include a tie bar between the two transmitters to provide a more predictable spacing.)

Correction for the expansion of the tank diameter due to ambient temperature variations can be entered. This correction and the effects of roof weight and tank deadwood are entered with the tank capacity tables.

API temperature and gravity conversion constants and correction for geographic variation in the earth's gravity can also be entered.

Depending on the system, the characterization constants for the pressure transmitters may be integral to the transmitter microprocessors, or the constants may be entered into the HIU. The HIU typically includes a local display of tank level and temperature. It also contains transmitter diagnostic routines and system alarms.

HTG ADVANTAGES

Compared to manual tank gauging (hand dipping), all types of automatic tank gauges provide improved manpower utilization, safety, convenience, repeatability, timely measurement, and freedom from human errors. The three most popular ATG's are float-operated ATG'S, servo-powered ATG'S, and radar ATG'S. Compared to these systems, HTG's have several advantages.

HTG's measure the liquid from the bottom of the tank up rather than from the top down as the other systems do. That avoids errors caused by movement of the top reference point. In addition, grade-mounted gauges such as HTG's are less expensive to install, calibrate, and maintain.

The typical HTG smart pressure transmitter has no mechanical span or range adjustments. It is factory calibrated and compensated for ambient changes.

Zero and elevation calibration is done digitally, and drift rates are low. The transmitters are provided with a port for use with a hand-held terminal for field diagnosis, resulting in easier maintenance.

HTG pressure transmitters are usually mounted on 1 1/2-in. or 2-in. flanged connections (Fig. 2). The pressure transmitter body and the RTD thermowell are the only instrument parts that contact the stored liquid, and they are designed for zero corrosion. Therefore, HTG's are good for viscous or corrosive products like asphalt, acid, and caustic.

HTG pressure transmitters are mounted with a valve between the liquid and the transmitter so that they can be easily zeroed or replaced. The external mounting lends itself to hot tapping on existing tanks. The existing tank thermowell can often be used for the HTG temperature sensor.

The combination of external mounting and a highly reliable transmitter makes HTG particularly suitable for LPG service. LPG tanks have been a particularly difficult application for automatic tank gauges because the instrumentation has to operate both inside and outside of a pressurized tank.

HTG's provide a precision RTD mounted between the two pressure transmitters and close connected to a temperature converter in the locally mounted HIU. Long runs of RTD wires are thus avoided.

A single RTD is all that is normally needed for the HTG installation. Even if the tank is thermally stratified, the mass determination is not affected by the thermal stratification.

ATG's that directly measure tank level require multiple temperature sensors to avoid errors associated with thermal stratification.

HTG's provide good measurements during tank filling and draining. The HTG's mass and standard volume measurements are not affected by level errors caused by jerky roof movement, release of entrained vapor, or operation of tank mixers.

HTG's eliminate the need to manually sample the product in the tank to obtain the density of the stored product. The density is that of the lowest 8-ft slice of liquid in the tank.

Manual sampling for density determination is time consuming, and the gauger is exposed to liquid vapors during the sampling. Also, it is difficult to obtain representative density samples from a tank because of the loss of light ends from the sample.

Monitoring the density readout can prevent transfer of off-specification product. A high-density alarm can be used to monitor water buildup in the bottom of the tank.

HTG's provide a measurement of product mass in the tank. Even though most custody transfer and oil accounting in the U.S. is based on volume rather than mass, many users need to know the mass of the liquids in their tanks.

With conventional ATG'S, liquid mass is determined by multiplying the standard volume by the standard density. Determination of the standard density requires sampling the tank and analyzing the samples in a laboratory.

Determining the standard volume requires measuring the average tank temperature. Mass accuracy determined with ATG's is normally much worse than volume accuracy because tank average density and tank average temperature are difficult to measure accurately.

Mass-based measurement provides advantages for products like asphalt, ammonia, or LPG, which are often sold by weight. Batch processing, refinery mass balances, leak detection, and stock loss control are much easier with mass-based HTG'S.

Most HTG's will be installed to replace existing ATG'S, and initially, they will be used to measure level. But when operators realize that accurate, reliable mass measurements are available, they likely will gradually convert to mass-based measurement.

HTG ACCURACY

Compared to float-operated, servo-operated, or radar automatic tank gauges, HTG's are more accurate in several respects. Bottom-up measurement with HTG avoids inaccuracy caused by movement of the top reference point,

Accurate measurement with top-down ATG's requires a properly supported gauging well. If all ATG's were mounted on properly supported gauging wells (stilling wells or gauge pipes), top-down measurement would cause a relatively minor measurement error. But in the U.S., ATG's are rarely mounted on gauging wells. Top-down measurement without gauging wells typically has errors of 1 in., regardless of the accuracy of the ATG. And the cost of installing a gauging well in an existing tank far exceeds the cost of an ATG or an HTG.

HTG's directly determine mass from the hydrostatic pressure measurement and derive standard volume and level from the mass value, requiring only a single-point temperature measurement. Level-based tank gauges require multiple temperature sensors for the same accuracy.

Because HTG's have no moving parts to stick or wear, periodic calibration is unnecessary, and accuracy remains constant over long periods of time.

Most tank-capacity tables are calculated for oil at 72 F. When the liquid in the tank is hotter or colder, the actual tank volume determined from level measurements will be in error. HTG's automatically calculate a correction for the thermal expansion of the tank and more accurately determine mass, density, volume, and level.

HTG INACCURACIES

The major inaccuracies of HTG's occur because volume and level measurements are determined from the mass measurement. To calculate the tank volume and level, the HTG assumes that all of the liquid in the tank is the same density as the 8-ft slice between the P, and P2 transmitters.

If the tank is thermally stratified or density stratified, the level will be in error. Where stratification is present, the level error is greatest with a full tank and zero when the tank level is at the 8-ft level.

Thermal stratification errors are minimized if the entire contents of the tank are at the same API gravity. The standard volume, standard density, and mass will be correct, but the level will be in error.

For instance, in a tank filled to a 40-ft level with the bottom half at 60 F. and the top half at 70 F., an HTG will indicate a level about 1 1/4 in. below the actual level.

Most tanks store liquids with fairly uniform API gravities. And where density is not uniform, tank mixers are usually provided.

In applications where density is not uniform because of inadequate mixing, HTG provides accurate mass measurement but less accurate standard volume and level measurement. All ATG'S, including HTG'S, have volume and standard density errors with density stratified tanks, but the errors will be worse with HTG.

For instance, with a full 40-ft tank with the bottom half at 38 API gravity and the top half at 40 API gravity, an HTG will indicate a level about 3 in. below the actual level. The standard volume measurement will have a similar error.

Metrology experts know that system accuracy is never better than the least accurate element in the system. For HTG'S, the pressure transmitter is that element.

Typical specified accuracy by pressure transmitter manufacturers is 0.05% calibrated accuracy. Therefore, the best possible accuracy in the level measurement is 1/4 in. But for most HTG applications, repeatability and long-term stability are more important than absolute accuracy.

Another source of inaccuracy results from field calibrating HTG's against manual level measurements. This is like calibrating a micrometer with a yardstick.

This is done because most tank measurement in the U.S. is volume-based rather than mass-based. Therefore, most HTG's are used for level measurement.

Level measurements are limited in accuracy because of thermal or density stratification. Mass measurements by HTG's are not affected by stratification.

The next HTG development will be hybrid systems with additional pressure transmitters and level switches that compensate for thermal or density stratification to minimize level error.

The draft API standard on automatic tank gauging (Chapter 3.1b) covers the use of HTG's for level measurement. This standard is in the final balloting phase.

In Europe, ISO is working on a draft standard covering the use of HTG's for mass measurement.

BIBLIOGRAPHY

Neesbye-Hansen, O., "Accuracy of oil-custody transfers can be improved," OGJ, Jan. 3, 1983, p. 97.
Robinson, C., "Hydrostatic tank gauging, What it is, Where it's used, What's available," InTech, February 1988.
Early, P.L., "New HTG Technology Solves Old Tank Gauging Problems," Hydrocarbon Processing, November 1988.
Sivaraman, S., and Holloway, C.J., "Method measures cylindrical storage-tank reference height variations," OGJ, Dec. 12, 1988, p. 50.
Berto, Frank J., "Methods for volume measurement using tank gauging devices can be error prone," OGJ, Mar. 13, 1989, p. 57.
Mej, Kenneth W., "Automatic tank gauges can be used for custody transfer," OGJ, Nov. 13, 1989 p. 81.