MWD TOOL ACCURATELY MEASURES FOUR RESISTIVITIES

S.G. Mack, P.F. Rodney, M.S. Bittar
Sperry-Sun Drilling Services
Houston

A newly developed multiple depth of investigation measurement-while-drilling (MWD) resistivity, sensor that measures four depths of investigation produces logs with the look and function of those from a conventional wire line dual induction tool. The tool measures phase shift and amplitude attenuation resistivity from four independent receiver-antenna spacings and two electromagnetic frequencies.

Field test and modeling results show that phase shift resistivities have better bed resolution and accuracy than their equivalent spaced amplitude attenuation counterparts. Attempts to use only amplitude attenuation curves provided misleading invasion and true resistivity (Rt) results, as did a hybrid mix of amplitude attenuation and phase shift measurements. The addition of one deeper reading and two shallower reading sensors offers new information about formation invasion characteristics, Rt, and resistivity in the invaded zone (RXO in an MWD environment. Logging and modeling results show that invasion and Rt can be successfully measured or estimated over a wide range of logging conditions and formation exposure times.

EWR TOOL

An electromagnetic wave resistivity (EWR) tool consists of a steel drill collar specially adapted with antennas placed in cutouts machined into the outside surface of the collar. A transmitting antenna propagates an electromagnetic wave through the borehole and surrounding formation and is received at two antennas spaced at different distances from the source. Because the propagated wave arrives at the near receiver sooner than the far receiver, the travel time of the wave can be measured between the two receivers.

Because the propagating wave is at a fixed frequency, this time difference can be expressed as a phase difference.

The recorded phase difference between the receiver pair is a function of the formation conductivity. Formation conductivity can also be calculated in another manner by a measurement of the amplitude attenuation of the waveform between the receiver pair. The resulting amplitude attenuation and phase shift resistivities differ with respect to depth of investigation, vertical resolution, dielectric sensitivity, and accuracy.

BACKGROUND

Since the commercial introduction of propagating electromagnetic wave resistivity tools in 1983, the measurement while drilling industry evolved from having correlation-only resistivity measurements with short normal devices to wire-line-replacement quality propagating wave devices. 1 The MWD advantage of taking measurements before fluid invasion, combined with the good vertical resolution of these sensors, gave high quality resistivity information in a variety of logging environments.

These advantages notwithstanding, the propagating wave sensors were originally used in a manner which provided only one depth of investigation, with any degree of reliability, to an industry which was accustomed to looking at three different curves provided by a wire line dual induction log.

A high degree of confidence that Rt was being measured occurred only when the tool was located in a bottom hole assembly close enough to the bit to log the formation soon after penetration. Logging circumstances which did not meet these requirements, such as measurement after drilling (MAD), subjected the log interpreter to the possibilities of relying on a resistivity value compromised by invasion.

Logs produced by the first propagating wave sensors were most often derived only from phase differences, even though the ability to produce amplitude attenuation measurements existed (e.g., with the EWR since it was introduced). The phase-derived measurement had good response characteristics over a wide range of logging conditions, whereas the amplitude attenuation resistivity could not be relied upon to provide Rt in most pay zones. The shortcomings of amplitude attenuation measurements are poor bed resolution, loss of accuracy in resistivities over 20 ohm-m, and high sensitivity to changes in formation dielectric constant.

Despite an earlier recognition of the intrinsic response differences between amplitude attenuation and phase difference sensors because of the recognition of a need for a second depth of investigation, logs based on amplitude attenuation processing of electromagnetic wave data were extolled as a complement to phase shift resistivity much later in 1988. 2 The additional amplitude attenuation log is typically labeled as a deeper investigating curve, and the phase difference resistivity is thought of as a shallower curve. The value of the amplitude attenuation measurement in even moderate resistivities has been the center of heated debate, with some researchers seriously questioning whether the fog reads deeper under all logging conditions. 3

TOOL DESIGN

The EWR Phase-4 tool measures the phase difference and amplitude attenuation of a propagated electromagnetic wave between two receivers for each spacing (Fig. 1). The formation conductivity is computed as a function of phase difference and amplitude attenuation through algorithms developed from experimental tank tests with varying brine concentrations and finite element modeling. 4 5

The tool has the capability of producing both phase difference and amplitude attenuation resistivity at each of the four spacings, providing a total of eight curves. The tool differs from the conventional EWR by providing two measurements which are shallower, and one measurement which is deeper.

The phase difference combination was superior to measurements of amplitude attenuation only, and interpretations based both on phase difference and amplitude attenuation resistivities often produced misleading results about Rt and the invaded zone. After comparing all of the data available from the sensor, however initial field tests showed that using the phase difference resistivities provided a good value for Rt and more information about the invaded zone.

In water zones, the amplitude attenuation usually reads too deep for accurate assessment of invasion parameters while drilling. In pay zones, the log analyst can have problems with the largely dissimilar bed resolutions of the amplitude attenuation measurements at each spacing and the intrinsic accuracy problems at higher resistivities.

The need for relying on amplitude attenuation processing has been eliminated in this tool design by the longer spacing, deeper reading antenna-receiver combination. The two additional shallow spacings and one deep spacing provide phase difference resistivities for Rt and invasion characterization without resistivity or thin bed limitations.

The four phase difference resistivities are termed X-shallow (EWXP,), shallow (EWSP), medium (EWMP), and deep (EWDP), in the order of increasing depth of investigation. There are four amplitude attenuation resistivities measured at corresponding depths of investigation.

DEPTH OF INVESTIGATION

Defining the depth of investigation of wave propagation sensors is somewhat arbitrary because the response cannot be approximated with conventional geometric factor theory used for wire line induction devices. Instead, each combination of concentric resistivity zones around the borehole has unique effects which vary with formation and borehole resistivity changes.

Although the development of a wire-line-like pseudo geometric factor theory to approximate invasion depths for wave propagation tools is not applicable, a depth of investigation using a single, homogeneous medium can be calculated using tool response theory. 6 Fig. 2 shows the radial depths of investigation for a formation having an Rt of 25 ohm-m. The investigation diameters in this example for the EWXP, EWSP, EWMP, and EWDP are 25 in., 40 in., 55 in., and 80 in., respectively.

APPLICATIONS

Fig. 3 is a log example from a Gulf of Mexico oil sand logged with the EWR Phase-4 sensor. The four phase difference resistivity curves are plotted in Track 2, the conventional EWR-5 phase difference and amplitude attenuation resistivities are plotted in Track 3, and the wire line dual induction log is plotted in Track 4. In this example, the Phase-4 tool provides information about invasion which cannot be obtained with a conventional propagating wave tool or a wire line dual induction measurement.

This sand is divided into a top and bottom zone separated by a shale barrier at 8,260 ft. An oil/water contact is visible on all three logs at 8,330 ft, and a short transition zone between 8,320 ft and 8,330 ft indicates that this rock has good permeability. The lazy character of the amplitude attenuation, or "deep curve," is in contrast to the phase difference curves of both the EWR Phase-4 and conventional EWR, especially in the top lobe of the sand.

In this case, the term "deeper reading" is a misleading label for the amplitude attenuation because bed resolution problems actually make the curve appear to read shallower than the phase difference resistivity. In the lower part of the sand, invasion is not obvious with the conventional EWR because little curve separation exists between the phase shift (shallow) and amplitude attenuation (deep) measurements. The dual induction log shows separation which indicates invasion, but a tornado chart correction for diameter of invasion cannot be used because the appropriate ratios between the shallow, medium, and deep curves plot points which fall off the chart. Invasion is obvious from the separation of the EWR Phase-4 curves, and it was quantitatively computed as 14 in. in diameter. Because the medium phase difference (EWMP) and deep phase difference (EWDP) curves read the same resistivity, the interpreter is assured that these measurements are unaffected by invasion. This example shows the improvement in detecting movable hydrocarbons using the EWR Phase-4 tool over the conventional EWR, especially in the top sand at 8,250 ft, where differences in bed resolution between the phase difference and amplitude attenuation resistivity show RXO to be greater than Rt, an incorrect interpretation.

HEAVY OIL ZONES

Fig. 4 is a log from the Endicott field in Alaska, where log interpretation is necessary to distinguish heavy oil/tar zones from movable hydrocarbons. The operator had enough confidence in the MWD formation evaluation suite to use it as a replacement for wire line measurements; therefore, a dual induction log comparison was not available.

In this higher resistivity range, the conventional dual depth of investigation phase difference and amplitude attenuation resistivity combination cannot be used because of limitations in obtaining meaningful amplitude attenuation measurements. Instead, the operator compared an MWD EWR phase difference resistivity with a measurement after drilling (MAD) run acquired during a wiper trip 30-40 hr after bit penetration. The differences in resistivity between the MWD and MAD runs are used as an indication of movable hydrocarbons. 7

The conventional EWR phase difference resistivity curve (SEWR) is plotted in Track 3, alone, with the MAD run (SEWR-WP). The phase difference resistivities from the Phase-4 log are plotted in Track 2. From the separation of the MWD and MAD curves with the conventional EWR and the separation of the four curves with the Phase-4 log, the contact between the movable oil and the heavy oil/tar is interpreted as approximately 13,115 ft. The other high resistivity zones are saturated with immovable heavy oil/tar. The additional information provided by the Phase-4 sensor will be sufficient to determine tar from movable hydrocarbons, without the added expense of a MAD run.

FILTRATE INVASION

Fig. 5 is a comparison of Phase-4 MWD, Phase-4 MAD, and a wire line dual induction log in a water sand from the Gulf of Mexico. Each run has a different formation exposure time and different invasion profile. As expected, the invasion diameter increases with increasing formation exposure time.

The rate of invasion, however, is not expected from conventional wisdom of mud filtrate invasion and mud cake buildup.

The MWD run acquired 1 hr after drilling (Track 2) indicates there was no immediate spurt loss because of dynamic pressure changes at the bit, and the MAD run in Track 3 shows a diameter of invasion of only 23 in. after 8 days of formation exposure. After 12 days of formation exposure, the wire line dual induction log in Track 4 shows significant invasion, which is computed as 63 in. using the appropriate standard tornado chart.

Rapid invasion has been suggested to occur immediately as a consequence of spurt loss, followed by a buildup of internal and external mud cake which restricts further invasion. 8 In this case, it appears that invasion was minimal immediately after drilling and actually increased during the time it should have been retarded because of mud cake formation. The drilling mud had a fluid loss of 4.2 ml and varied less than 1 ppg during the drilling of the well.

The addition of the X-shallow curve (EWXP) to the Phase-4 tool was a modification to the original prototype, intended to provide information equivalent to a wire line microresistivity tool. Although this is not always possible, enough invasion has occurred between the MWD run and the MAD run for the shallowest phase curve to read RXO. In this example, invasion is so deep that the wire fine shallow focused log also reads RXO and matches a calculated value from Archie's equation.

Rt is measured during the Phase-4 MWD run, but both the MAD run and the wire line dual induction log require a small correction to obtain Rt. If the well plan did not originally call for MWD logs, the Phase-4 resistivity tool could still be run days later without loss of information if hole conditions deteriorated and interfered with wire line operations. The multiple depth of investigation features provide enough information, even in deeply invaded zones, to estimate Rt by applying invasion corrections.

REFERENCES

Rodney, P.F., et al.,"The Electromagnetic Wave Resistivity MWD Tool," SPE paper No. 12167, October 1983.
Clark, B., Luling, M.G., Jundt, J., Ross, M., and Best, D., "A Dual Depth Resistivity Measurement for FEWD," presented at the SPWLA 29th Annual Logging Symposium, 1988.
Shen, L.C., "Investigation of Coil Type MWD Resistivity Sensor," presented at the SPWLA 32nd Annual Logging Symposium, June 1991.
Rodney, P.F., Mack, S.G., Bittar, M.S., and Bartel, R.P., "An MWD Multiple Depth Of Investigation Electromagnetic Wave Resistivity Sensor," presented at the SPWLA 31st Annual Logging Symposium, 1991.
Bittar, M.S., Rodney, P.F., Mack, S.G., and Bartel, R.P., "A True Multiple Depth of Investigation Electromagnetic Wave Resistivity Sensor: Theory, Experiment, and Prototype Field Test Results," SPE paper No. 22705, October 1991.
Rodney, P.F., and Bartel, R., "Design of a Propagating Wave Resistivity Sensor in Order to Minimize the Influence of Borehole Fluids On the Sensor Response," SPE paper No. 18114, 1988.
Cunningham, A.B., Bullion, D., and Opstad, E.A., "Use of MWD Formation Evaluation in the Endicott Reservoir, North Slope, Alaska, USA," SPWLA paper C, presented at the 13th European Formation Evaluation Symposium, Budapest, Hungary, Oct. 24-27, 1990.
Allen, D., et al., "Invasion Revisited," Oilfield Review, July 1991.