Downhole probes evaluate cavern integrity

N. Graeme Crossley
TransGas Ltd.
Regina, Sask.

• McAllister pressure, temperature probes [.pdf file]
• Inventory verification [12040 bytes]

Obtaining natural-gas storage caverns' pressures and temperatures with downhole probes has allowed TransGas Ltd., Regina, to monitor and evaluate cavern integrity. TransGas has more than 5 years' experience with the devices.

The acquired data have also helped determine gas-in-place inventory and confirm and assess changes in spatial volumes. These changes may have resulted from cavern creep (shrinkage or closure) or downhole abnormality such as fluid infill or collapse of the side walls or roof.

This first of two articles presents background and many of the details and lessons to date of TransGas' cavern gas-storage probe program; the conclusion describes a specific storage site with some results.

Individual monitoring

Cavern-integrity testing or certification has generated much discussion in recent years. Precise knowledge of natural-gas inventory allows more-accurate financial records and/or operations to customers.

Finding a suitable measuring device for determining actual inventory or gas-in-place volumes in bedded salt solution-mined caverns has posed a problem for TransGas.

The company operates 24 bedded-salt caverns, of which 18 were solution mined exclusively for sweet natural-gas storage,¹ and 6 are converted bedded-salt LPG caverns.²

The surface piping for all these caverns was not initially set up to facilitate measurement of gas movement in and out of individual caverns.

The search for suitable measurement devices started in the late 1980s. Traditionally, TransGas has only bulk-metered the gas flow at the plant inlet and outlet with orifice measurement at normal pipeline pressures, whether the facility consisted of a single cavern or groups of caverns.

Caverns tend to be operated in groups rather than individually. A facility with more than one cavern was treated as a single entity or one large cavern.

Because no individual cavern measurement existed, it was not known how much gas went into or came out of an individual cavern.

Individual bi-directional wellhead turbine meters were tried selectively in the 1980s but were soon discarded because the fluids and particulate matter typically produced from a cavern damaged the turbine rotor and rendered gas measurement inaccurate.

Other companies have experienced similar equipment failure in attempts at high-pressure wellhead measurement.

Orifice metering was considered unsuitable for wellhead measurement because frequent adjustments are needed to accommodate a wide range of flows from the cavern. TransGas' yard and plant piping, which was typically up to a quarter mile or more away from the individual caverns, was not originally constructed for ease of modification into a manifold system. This would have made individual cavern metering economical and successful.

TransGas felt that metering individual caverns' volumes was important not only for accurate inventory auditing but also for ensuring health and safety and determining caverns' conditions at any time. The company believed one device could possibly be used for several purposes.

TransGas decided to install semipermanent downhole pressure and temperature probes. The digital data are transmitted to surface with a four-conductor shielded wireline cable which suspends the probes from the wellhead.

Cavern development

All of TransGas' caverns have been developed in the Prairie Evaporite bedded-salt deposit; Fig. 1 [78174 bytes] shows the Elk Point salt basin, Fig. 2 [102536 bytes] the locations.

This formation is 3,000-5,000 ft below the surface and can contain natural gas at pressures between 2,000 and 4,000 psig. The formation thickness, in usable areas, is 300-500 ft.

A single cavern would typically be designed for a spatial volume of 900,000 bbl or a useable gas volume of 0.8 bcf at withdrawal rates of 40-90 MMcfd. Table 1 [40807 bytes] presents cavern statistics; Figs. 3 [30144 bytes] and 4 [29803 bytes], typical TransGas caverns.

Measuring storage-cavern gas volumes presents unique challenges. For many years TransGas has attempted to provide a second method to verify storage inventory.

Although scientific means (that is, an ideal gas equation) are available, the data required (cavern gas temperature) to use the science have not been readily available. Hence, the results are estimates at best and may not offer significantly more accuracy than meters located at the surface.

Since the cavern working pressures can be double or triple the normally encountered transmission-line pressures, specialized equipment tends to be expensive and difficult to maintain, particularly in Saskatchewan where temperatures drop to -40° C.

TransGas' practice has been to locate its meters in the lower pressure (500-1,000 psig) piping at storage sites. This avoids the excessive costs of metering equipment rated for higher pressure (1,000-4,000 psig) piping. It presents, however, other undesirable situations, as will be discussed.

Even the best surface measurement will result in metering differences and error accumulations in storage caverns after several withdrawal-injection cycles. In time, an adjustment to the perpetual inventory (as determined from surface measurement) will be unavoidable.

Only in rare cases will the plus and minus metering errors cancel out each other. Compressor pulsations can cause metering differences as high as 20% or more.

The methods used to verify surface pipeline's line pack and receipt volumes and eliminate deficiencies and error accumulations are unavailable in storage-cavern operations because measured pressure and temperature values are not usually available to determine the changes in cavern inventory.

The difference between measured withdrawal and measured injection is assumed to be remaining in the cavern. Therefore, if remaining usable gas volume is based solely on surface measurement, significant uncertainty remains about the amount of gas in inventory at any given time.

TransGas must also find an effective balance for maintaining its system and using its staff. The company's gas network is spread out over 140,000 sq miles, and maintenance requires extensive driving.

Use of such advanced technology as probes continually to record cavern status reduces requirements for a physical cavern check and thereby reduces system operating expenses.

New approach

TransGas uses a more reliable method for determining storage-cavern inventory.

Because the cavern spatial volume is known (through sonars, gas-production volumes, or brine removed), an accurate pressure and temperature reading of the gas inside the cavern plus current cavern conditions and gas composition will yield an accurate, precise theoretical volume calculation from the ideal gas equation of state.

An accurate temperature reading is critical to this calculation. Although bottomhole pressures can easily be calculated from surface data, such is not the case for the actual gas temperature inside the cavern. Long periods of stabilization are usually unavailable in order to allow use of formation-temperature data.

A pressure and temperature probe, permanently installed inside the cavern, will provide the required data. Dual probes are required to verify the accuracy of each instrument set and to determine when recalibration is required (two independent sets of sensing elements reduce error).

Periodic measurement of the temperature with wire line is a costly option and offers only limited data. Permanently installed probes offer far more data, and other uses are possible which can justify the additional cost.

TransGas deemed this technology the best approach for operating and monitoring its caverns cost effectively and efficiently. The cost of changing piping and installing other more-accurate measurement devices at the surface would be approximately C$50,000 higher than for installing the probes and would still yield the errors.

The conceptual investigation involved an analysis of project requirements and evaluations of various equipment. This followed the earlier evaluation of deficiencies of the trial cavern-wellhead metering equipment and material.

TransGas selected McAllister Petroleum Services Ltd., Calgary. TransGas' cavern operating conditions (gas composition + pressure [2,279 to 3,652 psig]/temperature [50 to 140° F.]) are ideal for the type of pressure and temperature probes selected. Accuracy limits are acceptable, and the cavern gas environment is not corrosive.

(An accompanying box presents probe specifics.)

This project began with the installation of the first test probe at Melville, Sask., in 1991. Probes are installed at one-third cavern height in order to measure average cavern conditions.

A controlled gas-withdrawal test at Melville in 1992³ verified the original cavern spatial volume from 1964 completion data. Therefore, a program was put together to install probes in remaining TransGas caverns.

In 1993, five caverns were fitted with probe instrument sets, and during 1996, four more caverns were fitted. The program will be completed this year.

There are now five caverns near Regina in which probes may not be installed because of abnormal downhole conditions.

Justification

These probes and associated pressure and temperature signals are required for inventory verification and cavern-health monitoring.

Inventory verification. The pressure and temperature data obtained from the probes, along with the known spatial volume, facilitate a more accurate calculation of the quantity of gas in the cavern.

This calculation is accomplished with the ideal gas equation of state and is performed at the same time each month. The month-to-month variance in the calculated volume determines the net change in cavern inventory (box: "Inventory verification."

The probe data for pressure and temperature are fed into the local station's remote terminal unit (RTU) and brought into the supervisory control and data acquisition (scada) host in TransGas' centralized gas control center in Regina. There the cavern gas volumes are calculated and monitored.

Fig. 5 [37372 bytes] presents the cavern probe scada interface which connects TransGas' scada system and the McAllister cavern probes. This brings the information back to the scada host and is made available via daily Ascii transfers to the corporate local area network (LAN).

Two significant benefits of this system are the accuracy of determining gas-cavern inventory and more-precise accounting of gas volumes. TransGas' cavern space is used by SaskEnergy Inc., the provincial distribution company, and is now marketed to other gas companies. Hence, it is essential that TransGas know the state of its natural-gas inventory.

Cavern-health monitoring. Regular monitoring and trending of cavern-volume data provides a history of the health of the cavern. Scada retrieves the pressure and temperature data and calculates and stores cavern inventories.

Cavern probes and surface flow equipment complement each other and provide more-accurate information on the structural integrity of the caverns.

Continually monitoring pressure and temperature data can reveal anything abnormal inside the cavern, particularly during shut-in.

TransGas has defined the maximum pressure decline allowable inside a cavern during production. Proper monitoring from gas control enables operators to avoid such future structural problems as roof falls, cracks in salt, and gas and brine leaks.

If probe data get knocked out, it can signal a roof or sidewall collapse or faulty instruments that may require removal for servicing. The faster this information becomes available, the quicker the response to protect a valuable asset.

For a single cavern only, a known amount of gas injection or production volume, and the starting and final pressures and temperatures, a calculation can determine if the cavern spatial volume has changed as a result of creep or other problems. Such changes have occurred in some Regina North and South caverns.

TransGas has set minimum operating pressure guidelines for all caverns so that actual pressures remain above these values to prevent cavern structural damage. Verification is possible by continual monitoring of the minimum cavern pressure conditions.

Finally, monitoring internal cavern conditions can save on sending manpower to verify surface conditions.

Another benefit is that having pressure and temperature data, and therefore the cavern volume, available to gas control continually provides a more-accurate picture of the inventory in the system. Gas-storage planning is improved and system optimization streamlined.

It should be noted that to the present, all pressure and temperature data from the probes have been collected at each site with an IBM-compatible personal computer. The first cavern-probe scada interface installation was commissioned in August 1996 and is successfully acquiring data from six probes on 15-min intervals.

Benefits

Successful completion of this project provides the following benefits:

• The pressure and temperature data coupled with regular cavern exercising (controlled production or injection testing) indicate cavern stability and its suitability for continued standard operations or restricted use, if a particular problem seems to be developing.

• Instantaneous pressure and temperature data trended during periods of inactivity provide conclusive evidence of cavern leakage.

• A monthly comparison of calculated volumes (based on the pressure and temperature data) to surface-measured volumes yields a monthly adjustment to metered volumes and avoids accounting inventory adjustments.

• Year end cavern inventory audits are more accurate and general inventories are available promptly and continually.

• Pulsation, a concern for surface measurement, does not affect cavern-inventory determinations.

It is important to note that the pressure and temperature probes are not intended to replace surface measurement already present. Surface measurement is still required to provide daily measurement of production and injection flows and to back up the cavern probe information.

Probe data cannot now be used for custody transfer because probes' accuracy has not been verified by the Canadian governmental agency Measurement Canada.

More cavern cycling may be required. This may restrict our ability to shut-in caverns and conduct individual cavern flow tests to determine P/ZT vs. Q plot for determining the structural integrity. (The present cavern piping does not allow for individual cavern flow measurement without shutting in the other caverns).

Having cavern probes in place gives TransGas more flexibility in its cavern operation.

Limits, hazards

To make the information more "user friendly," specialized software applications are required to provide health and integrity analysis (computer modeling).

Also, more operating and maintenance experience is necessary to be able to verify what is actually happening inside the cavern.

Access to cavern probes (that is, bent pipe, tubing production, etc.) requires a wire line service company to pull and run probes. A probe could become lost or unretrievable.

Probes also hinder quick re-entry to caverns for logging because they must be removed, then reinstalled. This activity adds 2 days to a work schedule which can result in additional costs of approximately $20,000 (Can.).

In addition, cavern roof or sidewall salt failures can damage the probe. Since 1991, however, this has not happened.

The significance of temperature variations becomes unreliable if a probe is submerged in fluid (brine) as a result of cavern leakage. Determining a changing fluid level is impossible. But fluid would have to rise to the one third of the cavern level for this to occur.

The frequency of maintenance or repair is unclear; dual instrument reading is necessary for drift identification. Only time will indicate the frequency for pullout.

Another limit relates to probe life-span, said to be 15 years. TransGas is hoping for at least a 5-8 year removal cycle to accommodate other cavern logging and gauge recalibration (to keep operating costs down). The company has had three probe sets in continuous operation since 1993 with no signs of any problems developing.

Finally, operation problems experienced with the probes have so far been related to wire line fault, as is noted in the box on equipment.

References

1. Crossley, N. Graeme, "Gas Storage in Saskatchewan Bedded Salt," Solution Mining Research Institute, Houston, Oct. 18-21, 1992.

2. Crossley, N. Graeme, "Converting LPG caverns to natural-gas storage permits fast response to market," OGJ, Feb. 19, 1996, p. 39.

3. Wickenhauser, Pat L. "Storage Cavern Volume Determination Using Pressure/Temperature Probes," Canadian Gas Association, 1993 Gas Measurement School, Winnipeg, June 1-2, 1993.

The Author