THREE KEY ELEMENTS NECESSARY FOR SUCCESSFUL TESTING

Christine A. Ehlig-Economides, Peter Hegeman Schlumberger Oilfield Services Houston, Gavin Clark Schlumberger Oilfield Services Aberdeen

Real-time surface readout during data acquisition, downhole shut-in, and appropriate pressure gauges are three key elements for successful well tests.

These elements are often overlooked in designing and implementing a successful well test.

This second in a series of three articles on well testing shows how these elements affected the testing of an example well. Also reviewed are the capabilities of several new testing tools and techniques.

This series of three articles began in OGJ on July 18.

FUNDAMENTAL COMPONENTS

At the well site, the main test objective is to acquire and validate data; that is, to verify that acquired data satisfy test objectives.

In the example well, the objectives of a conventional build-up test 1 were to determine well flow efficiency and average reservoir pressure. The well was shut in at the surface, and a low-resolution, wire line strain gauge was placed in the well bore just above the perforations.

Fig. 1 shows Homer and log-log diagnostic plots after 12 hr and 1, 4, and 8 days. Note that while the Homer plot' is an accepted analysis tool for determining permeability, skin factor, and average reservoir pressure (from which well productivity index and flow efficiency can be determined), it is not an effective diagnostic plot.

Specifically, the Horner plot should not be used to identify radial flow. Indeed, Fig. I shows that a straight line may not represent the radial-flow regime that signals a test of sufficient duration.

For example, in Fig. 1 all of the Homer plots exhibit a late-time, straight-line trend. If the Homer straight-fine were used to determine when to stop collecting data, the test might have been terminated after 12 hr. In that case the analysis yields a well flow efficiency of 2.2, and an extrapolated pressure (p*) of 3,109.2 psi.

Fortunately, when the log-log diagnostic plot is used to monitor well test progress, acquisition can continue until a flat trend in the derivative indicates radial flow in late-time data on the 8-day buildup.

From late-time data, the Horner plot yields a flow efficiency of 1.5 and a p* of 1,971.3 psi. A correct identification of radial flow on a log-log plot thus guaranteed that subsequent Horner analysis would provide the targeted results.

While this approach achieved the objective, the same well when tested 2 years later with a downhole shut-in valve and surface data readout, instead of a surface shut-in, obtained the response (Fig. 2) after only 10 hr.

In the comparative diagnostic plot (Fig. 2), the pressure change and derivative data are normalized (divided) by the production rate before shut-in. The flow regime identification (FRID) tool (described in the first article of this series) shows that radial-flow trends for the two tests are at the same level once the flow-rate magnitude has been accounted for. Consequently, the generalized Homer plot shows that late-time data for each test follows a line with the same slope.

The quantitative interpretation of the first and second tests yielded the same values with the exception of the extrapolated reservoir pressure, which varied because of a water injection program initiated after the first test. However, the second test with downhole shut-in required only a fraction (about 5%) of the elapsed time.

This example clearly illustrates the testing efficiencies gained from surface data readout, downhole shut-in, appropriate pressure gauges, and on site validation with log-log plots.

SURFACE DATA READOUT

Because the test was monitored through surface data readout and diagnosed with log-log plots as testing proceeded, the well site team avoided premature termination of the initial well test. Ultimately, this test showed a radial response, although well bore storage effects dominated the data for about 100 hr. Thus, real-time surface readout proved crucial for success.

Many options are available for surface monitoring during testing. In exploration wells, temporary completions can serve for disposing of produced fluids during testing. In this environment, a wire line in the drill stem test string is frequently prohibited.

Fig. 3a shows an inductive coupling device for retrieving data stored in downhole memory recorders located in the drill stem test string. Run on wire line, this device latches onto the test string and remains in place for real-time data transmission, as long as safety considerations permit.

Wire line telemetry is the most direct means for real-time data acquisition and display. A testing tool with downhole shut-in and a production logging tool, both of which can be run on wire line or coiled tubing equipped with wire tine, are shown in Figs. 3b and 3c. To perform properly, the testing tool must be latched to a landing nipple located near the tubing shoe.

Alternatively, although production logging tools are designed primarily for cased-hole logging of pressure, temperature, density, and flow velocity-vs.-depth, they can be positioned at any depth for stationary measurements.

The formation test is another testing technique that provides surface readout. The device is run on wire line or coiled tubing equipped with wire line. The recently introduced modular tool (Fig. 3d) can be configured at the well site for multiprobe formation testing, fluid sampling, and limited well testing procedures.' The tool can be used for characterizing reservoir parameters near the well bore, including horizontal, vertical, and spherical permeabilities.

Acquired formation pressures-vs.-depth enable reservoir pressure gradients to be determined, and, hence, fluid contact depths. Pressure profile anomalies may indicate barriers to vertical flow in the formation.

Wireless telemetry (Fig. 3e) now being field tested in many locations worldwide, offers a third way to access downhole data. In fact, it provides surface readout capability where it is most essential: in exploration well tests designed to characterize reservoir limits. Frequently, these tests are conducted in remote or hostile environments where safety precautions often prohibit wire line use. Wireless telemetry also is feasible in appraisal and development wells, an application that is expanding.

Because one main benefit of surface data readout is the opportunity to adjust the test sequence as necessary during data acquisition, improvements also have been made in downhole-tool control mechanisms. The remote implementation system (Fig. 4)4 provides complete flexibility for the job sequence.

The test valve allows downhole flow or shut-in, while the circulating valve allows communication with tubing for stimulation and safe killing of a well without formation damage. These valves operate when coded low-pressure mud pulses are sent, via the annulus, from the surface to the tool.

To further assist on site data analysis and decision making, operators are increasingly transmitting real-time data via satellite from the well site to office-based engineers. These experts assist well site personnel to adjust tests in progress and also help specify completion and stimulation procedures.

DOWNHOLE SHUT-IN

The example well clearly illustrates the time saved by downhole shut-in during the second test. Downhole shut-in could have eliminated 7 days from the first test. This translates into 7 days of additional production at about 1,700 b/d. Time savings is not the only benefit from downhole shut-in. Frequently when a well is shut in at the surface, well bore storage masks early time reservoir response. But with downhole shut-in, lengthy well bore storage transients are eliminated, often revealing trends that can quantify formation properties such as vertical permeability, damaged zone radius, damaged zone permeability layering, and dual-porosity heterogeneity.

Downhole shut-in is routinely available in drill stem testing tools, such as shown in Fig. 3a, and is also possible in production or injection wells equipped with an appropriate landing nipple near the tubing shoe (Fig. 3b).

With numerous wire line shut-in tools readily available in the marketplace, many equipped with surface data readout, it is rarely justified to conduct a build-up test without downhole shut-in.

In tubingless wells, however, downhole shut-in is only possible when testing hardware is lowered on the drill stem. Fortunately, an alternative to downhole shut-in is continuous measurement of both flow rate and pressure just above the formation. With suitable data processing,5 the response from combined flow rate and pressure measurements yields an analog for the log-log diagnostic plot of pressure change and its derivative. This strategy can offer the same quantitative analysis as lowering a shut-in valve to the same depths at which flow rates would be acquired.

Continuous sandface flow rate and pressure measurements can be obtained with a production logging tool (Fig. 3c). The advantage of this method is that flow-rate profiles are obtained during flowing or shut-in conditions.

The flow-rate profile permits direct evaluation of a partial completion penetration ratio (flowing interval divided by pay thickness). This, in turn, enables the total skin to be decomposed into skin damage and convergent flow skin components. Multiphase flow-rate profiles enable identification of water, oil, and gas entry depths.

When production from two or more formation zones is commingled, a layered reservoir test 6 can be conducted with the production logging tool. To do this, transient flow rate and pressure data are successively acquired above each zone and combined with flow-rate surveys derived at the end of each transient measurement. The test yields individual characterization of each formation.

This technique eliminates running successive drill stem tests and isolating each commingled interval with downhole packers.

A recent innovation in drill stem testing (Fig. 5) employs, in the test string, a Venturi flowmeter and gradiomanometer to continuously measure flow rate and density. This device offers downhole shut-in during pressure buildup and flow rate and density measurements while flowing the well. Produced fluid can, therefore, be identified during cleanup and testing. Further, the continuous flow rate and pressure data acquired during cleanup can be used to monitor skin reduction.

A more accurate flow rate prior to shut-in improves the accuracy of build-up transient interpretation. Also, for limit tests, interpretable drawdown data can minimize late-time distortions due to superposition effects that can complicate model diagnosis.

PRESSURE GAUGES

The final lesson to be derived from the example well tests concerns data quality.

The noise evident in the pressure derivative response for the first test (Fig. 1) acquired with a low-resolution gauge, is absent in the second test (Fig. 2), acquired with a high-resolution gauge.

In the two tests, the difference in quality of the transient response is striking. In fact in the first test, the signal noise during the late-time leaves room for doubt whether radial flow was encountered. In the second test, however, the minimal noise in the gauge response provides much greater confidence for identifying the flow regime and hence in calculating the final parameters.

Table 1 summarizes specific considerations for gauge selection. In general, the importance of gauge characteristics increases with reservoir-fluid mobility (rock permeability divided by fluid viscosity), and build-up tests typically require greater gauge resolution than do drawdown tests.

Greater gauge resolution as well as minimal gauge drift is important for long-term tests targeted to quantify distant reservoir heterogeneities that only appear after significant elapsed time. The same characteristics are essential for multiwell interference tests.

Short-term build-up tests 8 following a short period of production call for greater gauge resolution as well, particularly in situations with high-mobility fluid flow. Finally, formation tests of very short duration (a few minutes), require state-of-the-art pressure gauge technology.

TESTING STRATEGY

New technologies that support the four fundamental well test components are optimized by employing a sound well test design strategy. This strategy should include:

Staying abreast of modern testing techniques. Knowledge of available test tools increases the chances of obtaining maximum reservoir data for minimum cost. For example, installing a landing nipple in the tubing shoe (a minor expenditure) will enable downhole shut-in, thereby enhancing transient tests results during the life of a well.
Planning dynamic measurements to maximize interpretation value. For example in exploration and development wells, formation and open hole drill stem tests in isolated intervals provide pressure data that may be essential for understanding future production behavior, material balance, and interwell communication.
Also, to avoid interpretation complications from multiphase flow in the formation, tests designed primarily for reservoir characterization should be conducted before static pressure falls below reservoir fluid saturation pressure.
Weighing the value quantifying well and reservoir parameters required for designing well completion or stimulation. A transient test should be considered whenever the stimulation cost or the potential well productivity enhancement is significantly greater than the cost of quantifying the treatment design parameters with a well test.
In some cases, tests can be conducted alongside normal drilling, completion, or remedial operations. These tests require little or no extra time or equipment on the well site. In other cases, the cost and time required to obtain accurate values for horizontal or vertical permeability through transient testing are easily justified by their value.
Such data are used to define stimulation parameters such as optimal perforation interval geometry, perforation density and charge selection-vs.-depth, and (in low-to-moderate permeability formations) a target hydraulic fracture half-length. These data also indicate whether to consider drilling a horizontal well.
Further, permeability values, when considered together with skin, help determine whether well productivity (in vertical or horizontal wells) should be stimulated by matrix acidizing or hydraulic fracturing.
Insisting on well-defined test objectives. Properly designed tests begin with a set of objectives.
Tests are judged successful if specific objectives are achieved.
Using simulation to ensure maximum data at minimum cost. Simulations with estimated reservoir parameters and heterogeneity geometries can indicate the necessary well test duration and required Pressure gauge specifications. Simulations also can demonstrate the need for downhole shut-in and facilitate, where applicable, selection of a suitable flow-metering device.
Computerized modern test design and interpretation programs enable transient data simulation for a wide variety of test types, including drill stem tests, injection and production tests, build-up and falloff tests, impulse tests, horizontal well tests, layered reservoir tests, and horizontal well tests.
These programs help understand transient test response behavior.
Selecting appropriate testing hardware and sensors. The available wide selection of testing hardware can adapt to many well geometries and to the operations of a given test. Choosing adequate sensors is essential to testing success. Table 2 summarizes recent Schlumberger testing technology.

ACKNOWLEDGMENTS

The authors thank Elf Aquitaine for the example data, and Schlumberger's management for supporting this three-part series.

REFERENCES

1. Joseph, J.A., Ehlig-Economides, C.A., and Kuchuk, F.J., "The Role of Downhole Flow and Pressure Measurements in Reservoir Testing," SPE paper 18379, 1988.

2. Horner, D.R., "Pressure Build-Up In Wells," Third World Petroleum Congress, The Hague, Sec. 11, 1951, pp. 503-23.

3. Zimmerman, T., et al., "Application of Emerging Wireline Formation Testing Technologies," OSEA paper 90105, 8th Offshore South East Asia Conference, Singapore, Dec. 4-7, 1990.

4. Vella, Marc, et al., "The Nuts and Bolts of Well Testing," Oilfield Review, Vol. 4, No. 2, April 1992, pp. 14-27.

5. Kuchuk, F.J., "Applications of Convolution and Deconvolution to Transient Well tests," SPE FE, December 1990.

6. Ehlig-Economides, C.A., "Testing and Interpretation in Layered Reservoirs," JPT, September 1987, pp. 1087-90.

7. Brown, P.A., and Ehlig-Economides, C.A., "Use of Downhole Flowmetering Technique for Improved Drillstem Test Interpretation," SPE Annual Technical Conference and Exhibition, Washington D.C., 1992.

8. Ayoub, J.A., Bourdet, D.P., and Chauvel, Y.C., "Impulse Testing," SPE FE, September 1988, pp. 534-46.