LWD induction tools

Commercial resistivity measurements made while drilling first became available in the late 1970s. Because the drilling environment is much more adverse than the wireline logging environment, a simple short normal tool mounted behind the drill bit was used as the first LWD resistivity tool. However, short normal tools were only able to provide information for basic interpretation, such as correlation of geological markers and estimation of gross water saturation, because of their shallow depth of investigation and relatively poor vertical resolution. Being DC electrode devices, normal tools are also limited to conductive mud environments.

To expand the LWD resistivity market to oil-based mud (OBM) environments, induction-type propagation measurements were introduced in the early 1980s. The first commercial device was the electromagnetic wave resistivity (EWR) tool from NL Information Services (later merged with Sperry-Sun). Shortly after this, Schlumberger introduced the compensated dual resistivity (CDR) tool. All LWD propagation tools are run with a gamma ray tool for lithology estimation and correlation. Log data are transmitted uphole in real time using mud-pulse telemetry. Downhole memory and batteries allow raw and processed data to be stored for later retrieval.

Principles of the propagation measurement

Because the CDR is a simple tool that measures both attenuation and phase shift, it is used to demonstrate the basic concepts of propagation measurements. Conventional induction measurements are made with mutually balanced arrays of transmitter and receiver coils operating in the kilohertz frequency range and that phase-lock measurement electronics. Because it was difficult to engineer this type of measurement on a steel drill collar using the technology of the early 1980s, a higher-frequency propagation measurement was considered to be more practical for LWD. A frequency of 2 MHz was chosen because it was the lowest frequency at which accurate propagation measurements could be made on a drill collar at that time.

The CDR tool broadcasts a 2-MHz electromagnetic wave. A propagation measurement is made by taking the difference between the phases (phase shift) and amplitudes (attenuation) of the voltages recorded at the two receivers. Attenuation increases as a function of increasing conductivity, while the wavelength decreases as conductivity increases. Thus, the two measurements are proportional to formation conductivity and can be used to generate resistivity logs. Note that the CDR tool has two transmitters. The phase-shift and attenuation measurements generated by the upper transmitter between the two receivers, and by the lower transmitter between the two receivers, are averaged to symmetrize the tool response. This averaging is known as borehole compensation because it also reduces the effect of borehole rugosity. The transmitter-to-receiver spacings for the CDR tool are 25 and 31 in.

The phase-shift and attenuation measurements are transformed to two independent resistivities, which for the CDR tool are known as RPS (phase shift, shallow) and RAD (attenuation, deep). Because at 2 MHz dielectric effect can be significant at high resistivity levels, a dielectric correction is performed before the raw data are converted to apparent resistivity. Most service companies have developed their own proprietary algorithms to perform dielectric correction. Joint inversion for both resistivity and dielectric constant is also possible with today’s multiarray propagation tools. The dielectric-corrected phase shift and attenuation are converted to resistivity using a table look-up algorithm based on computed tool response in homogeneous isotropic media of known resistivity,R_t.

The phase-shift and attenuation measurements are both relatively insensitive to borehole size and mud resistivity. Borehole correction is only necessary in conductive holes with large washouts when the R_t/R_m contrast is greater than 100 to 1.

Invasion is usually quite shallow at the time of drilling when most LWD logs are run. However, LWD logs may also be recorded each time that the drillstring is pulled to replace the drill bit. At these later times, invasion can be much deeper. The two resistivity values, RPS and RAD, provide two independent depths of investigation as an indication of invasion.

The reason that two depths of investigation can be obtained from a single measurement is made clearer by examining the behavior of the electromagnetic field. The surfaces of constant phase are spheres because the wave travels with the same speed in all directions. The surfaces of constant amplitude are toroids because the wave is stronger in the radial direction than in the vertical direction, which is a normal characteristic of vertical magnetic dipole antennas. If we compare the phase and amplitude contour lines passing through the receivers, the amplitude extends to a significantly deeper region of the formation than the phase.

Depth of investigation can also be studied by modeling tool response in invaded formations. Fig. 1 shows CDR radial response for a case with R xo > R t , and Fig. 2 shows the radial response for R_xo < R_t. In both figures, RPS and RAD are plotted as a function of increasing invasion radius. In Fig. 1, RPS reads consistently closer to R_xo, indicating that RPS is the shallower of the two measurements. In Fig. 2, RPS is again consistently shallower than RAD. In this case, the RPS curve extends below the value of R_xo between radii of 30 and 50 in. because of wave reflection at the invasion front. In general, the depth of investigation of RPS is 10 to 20 in. shallower than that of RAD.

• 

  Fig. 1 – CDR radial response for R_xo < R_t.

• 

  Fig. 2 – CDR radial response for R_xo < R_t.

The depth of investigation and vertical resolution of 2-MHz tools is influenced to a significant degree by skin effect. 2D tool response can be characterized using Born response functions, which are geometrical factors that take skin effect into account. They show the amount of tool signal coming from a specific volume of the formation. The depths of investigation of both the phase-shift and attenuation measurements are shallower at lower formation-resistivity levels (higher conductivities) because the electromagnetic signal is more attenuated because of skin effect. In general, 2-MHz tools have a shallower depth of investigation and higher vertical resolution in low-resistivity formations, and a deeper depth of investigation and lower vertical resolution in high-resistivity formations.

Vertical resolution is characterized in more detail in Fig. 3, which compares CDR and Phasor induction logs. In this low-resistivity formation, the vertical resolution of CDR log is slightly sharper than that of the Phasor log. (The SFL log indicates that shallow invasion has taken place at wireline time.) However, the vertical resolution of 2-MHz logs deteriorates as the formation resistivity level increases. Because conventional induction logs always undergo vertical processing, while 2-MHz logs are seldom processed because their depth sampling is irregular, care must be taken when comparing wireline induction and 2-Mhz logs in resistive formations. Fig. 4 shows the variation in CDR vertical resolution as a function of resistivity level.

• 

  Fig. 3 – Comparison of CDR and Phasor induction wireline logs in a Gulf of Mexico well at low resistivity. The “permeability flag” is a function of separation between the phase and attenuation resistivities.

• 

  Fig. 4 – Vertical resolution of CDR measurements as a function of formation resistivity compared with Phasor and enhanced Phasor induction. The highly nonlinear response of propagation measurements makes resolution-matching and shoulder-effect correction difficult.

Multiarray propagation tools

During the 1990s, all major service companies developed multiarray versions of 2-MHz tools. Schlumberger introduced the array resistivity compensated tool (ARC5) 46 as a replacement for the CDR tool and to accommodate the increasing number of slimholes being drilled. The ARC5 tool makes five independent phase-shift and attenuation measurements. The number of measurements was deliberately chosen to be the same as that of wireline array-induction tools to allow the sharing of interpretation methods for analyzing complex invasion profiles and estimating R_t.

The ARC5 antenna configuration has five transmitters and two receivers. The phase shift and attenuation of the signal broadcast by each transmitter is measured between the two receivers for a total of five raw phase shifts and five raw attenuations. Because the transmitters are not arranged symmetrically above and below the receivers, conventional borehole compensation cannot be performed. Instead, the ARC5 relies on linear combinations of three transmitters to provide "mixed borehole compensation." This process results in five calibrated phase-shift and attenuation-resistivity logs which are characterized by the antenna spacings: 10, 16, 22, 28, and 34 in. The 28-in. spacing yields a log identical to a CDR tool.

Vertical resolution is related to the 6-in. receiver spacing that is common to all the measurements, but the depth of investigation increases as the transmitter spacing increases. The result is five phase-shift resistivity logs with different depths of investigation, and five deeper attenuation-resistivity logs, also with different depths of investigation. The phase-shift logs are matched in vertical resolution in conductive beds, but not in resistive beds. Similarly, the attenuation logs are better matched in conductive beds than resistivity beds. This difference in apparent vertical resolution is shown in Fig. 3.

The ARC5 makes a set of measurements at 2 MHz. Larger versions of the tool (6 3/4 and 8 1/4 in.) make measurements at both 2 MHz and 400 kHz. The 400-kHz measurement provides higher signal level in conductive formations (less than 1 ohm•m) and is less sensitive to borehole conditions, particularly in formations where R t < 1 ohm•m and with OBM. It also has a deeper depth of investigation than the 2-MHz measurement in conductive formations. However, 400-kHz measurements have less sensitivity to R t in resistive formations than 2-MHz measurements.

Fig. 5 compares ARC5 phase shift and wireline AIT logs. The vertical resolution of both sets of logs is similar, with the deeper AIT 90-in. curve giving a slightly higher value for R_t in the resistive beds. Fig. 6 shows a comparison with the ARC and the ARI dual laterolog in very nonconductive mud. In the bottom track, the ARC logs are inverted for invasion parameters.

• 

  Fig. 5 – Comparison of ARC5 phase shift and wireline AIT logs at 2-ft vertical resolution.

• 

  Fig. 6 – ARC logs compared with wireline ARI dual laterologs logs. In the bottom track, the ARC logs are inverted for invasion parameters.

In addition to the tools described here, the following LWD resistivity services are also available. References describing early versions of current tools are listed because old logs are sometimes used to interpret existing reservoirs. In 1991, Sperry-Sun introduced a version of the EWR tool with three phase-shift measurements and three attenuation measurements. In 1993, it introduced the EWR-Phase 4 tool with four phase-shift measurements and four attenuation measurements, giving a total of eight apparent-resistivity curves. The three shortest measurements are made at 2 MHz, while the longest measurement is made at 1 MHz. Both versions of the EWR are run without borehole compensation.

In 1989, Teleco introduced the 2-MHz Dual Propagation Resistivity (DPR) tool. This tool measured the phase shift and attenuation at receivers located 27 and 35 in. from a single transmitter (borehole compensation was not used). Teleco was taken over by Baker Hughes, and in 1993 the DPR tool was replaced by the multiple propagation-resistivity (MPR) tool. The MPR configuration consists of upper and lower long- and short-spaced transmitters surrounding a central receiver pair. Antenna spacings range from 23 to 35 in. The two receivers measure the phase shift and attenuation of 2-MHz and 400-kHz signals broadcast by each transmitter. Borehole compensation is performed by averaging the measurements from the symmetrically opposed long and short transmitter pairs. This yields a total of eight logs (long-spaced and short-spaced phase shift and attenuation at 2 MHz and 400 kHz). Resistivity is calculated and displayed either as "apparent" or "borehole-corrected" (hole size and mud-resistivity corrected). Further processing of the logs is available in the MPRteq processing. This processing corrects for a variety of environmental effects. Fig. 7 compares the processed logs with unprocessed logs.

• 

  Fig. 7 – Comparison of unprocessed MPR logs and MPRteq processed logs (courtesy of Baker Inteq).

In 1993, Halliburton introduced the 2-MHz compensated wave-resistivity (CWR) tool. After Halliburton’s acquisition of Sperry-Sun, the existing Halliburton business was spun off as PathFinder. The CWR tool makes a set of shallow and deep phase-shift and attenuation measurements with borehole compensation. The transmitter to receiver spacing is approximately 40 in. for the deep mode and 20 in. for the shallow mode.

Geosteering with LWD propagation tools

Because 2-MHz resistivity measurements are made behind the bit and can be sent uphole in real time, they are often used to steer the drilling of horizontal wells. Before drilling a horizontal well, potential hydrocarbon-bearing zones are first identified using vertical exploration wells. Then the horizontal well is drilled toward a target bed, with marker beds used to maintain the wellbore trajectory. 2-MHz resistivity logs recorded behind the bit are compared to logs from the exploration wells to identify the marker beds. Computer modeling of predicted resistivity-tool response at different well deviation angles is used to modify the well path. This process is called geosteering.

When comparing resistivity logs in a horizontal well to logs from a vertical exploration well in the same zone, the resistivity value often differs in shales and in laminated formations. This difference is caused by anisotropy (the variation of resistivity with direction). In addition to particle-size anisotropy, formations consisting of a series of thin beds of different lithology (such as sequences of sand and shales) also behave anisotropically if a logging tool is significantly longer than the bed thickness. In vertical wells, resistivity tools (conventional induction, 2-MHz and laterologs) read the effective horizontal resistivity, R_h, which can be calculated from the volume average of the layer conductivities (inverse resistivities),

....................(1)

where conductivities are expressed in mS/m. V_sand and V_shale are the bulk volume fractions (percentages) distributed throughout the layered region (layers are all assumed to be approximately uniform in thickness). The effective vertical resistivity, R_v, can be calculated in a similar manner from the volume average of the layer resistivities,

....................(2)

In deviated wells, the apparent resistivity R_a in anisotropic media can be calculated using the approximation

....................(3)

where α is the angle between the tool axis and vertical,

....................(4)

and

....................(5)

For α = 90° (horizontal wells), R_a = R. For α = 0° (vertical wells), R_a = R_h. Thus, the vertical resistivity cannot be detected at all by conventional resistivity logging tools in vertical wells.

Fig. 8 shows a modeled tool response illustrating differences caused by anisotropy between CDR logs in a vertical well (0° dip) and in a highly deviated well (80° dip). At 0° dip, the CDR log reads R_h. At 80°, the two CDR curves increase in the direction of R_v in the anisotropic bed, with the phase-shift resistivity reading higher than the attenuation resistivity.

• 

  Fig. 8 – CDR response at different dips as the tool logs and isotropic bed above an anistropic bed.

Induction and 2-MHz tools both generate azimuthally polarized electric fields that induce current loops that are tilted with respect to the bedding anisotropy. These tilted current loops sense a weighted average of R_v and R_h that depends on dip angle. Extensive modeling and analysis of 2-MHz tool response has demonstrated that radiation effects control the phase-shift measurement more strongly than the attenuation measurement. Thus, separation between 2-MHz phase-shift and attenuation logs provides a good indication of anisotropy (in the absence of invasion and shoulder-bed effect). There is sufficient sensitivity to invert for R_h and R_v only at considerable distances from bed boundaries. Also, note the polarization horn that occurs near the bed boundary at 80°. Polarization horns are a common occurrence at high dip angles, and are an indication in geosteering that the well has crossed into a target bed.

Nomenclature

Rh	=	resistivity in the horizontal direction (ohm•m)
Rv	=	resistivity in the vertical direction (ohm•m)
R	=	resistivity (ohm•m)

Noteworthy papers in OnePetro

Use this section to list papers in OnePetro that a reader who wants to learn more should definitely read

External links

Use this section to provide links to relevant material on websites other than PetroWiki and OnePetro

See also

Logging while drilling (LWD)

NMR logging while drilling

Resistivity and spontaneous (SP) logging

Induction logging

PEH:Resistivity_and_SP_Logging