Accelerometer: Definition, MWD Surveying, and Seismic Use

An accelerometer is a transducer that measures the acceleration of a body or the acceleration component of gravity acting along a defined axis. In the upstream oil and gas industry, accelerometers serve two distinct but equally critical roles: first, as the primary inclination sensor inside measurement-while-drilling (MWD) and logging-while-drilling (LWD) bottomhole assemblies (BHA), where a triaxial set resolves the gravity vector to compute borehole inclination; and second, as seismic receivers that capture ground motion or pressure waves during surface seismic, vertical seismic profiling (VSP), and ocean-bottom seismic (OBS) surveys. Modern accelerometers span a range of physical principles, from micro-electromechanical systems (MEMS) etched onto silicon wafers to precision servo force-balance designs capable of measuring changes in gravitational acceleration to sub-microgal levels. Understanding how each accelerometer type is selected, calibrated, and interpreted is fundamental to wellbore survey accuracy, seismic data quality, and safe execution of directional drilling and horizontal drilling programs worldwide.

How Accelerometers Work: Physical Principles

At the most fundamental level, an accelerometer measures force per unit mass. A proof mass suspended by compliant elements deflects under applied acceleration; that deflection is converted to an electrical signal by a capacitive, piezoelectric, or electromagnetic transducer. The four principal designs used in oilfield applications each trade different characteristics.

MEMS accelerometers are microfabricated from silicon using photolithographic etching. Interdigitated capacitor fingers attached to the proof mass change capacitance differentially as the mass moves, producing an output voltage proportional to acceleration. MEMS devices are inherently small (die size typically 3 mm x 3 mm), low power, and tolerant of high shock (rated to 2,000 g in some devices). Their main limitation is relatively high noise density, typically 50 to 300 micrograms per square root hertz (ug/rtHz), compared to sub-1-ug/rtHz for servo designs. Modern oilfield-grade MEMS accelerometers are calibrated over the full operating temperature range (-40 to +175 degrees C) to characterize temperature-dependent bias and scale factor drift, with calibration coefficients stored in onboard EEPROM and applied in real time by the tool's signal processor.

Servo (force-balance) accelerometers use closed-loop electrostatic or electromagnetic feedback to hold the proof mass at a null position. The feedback current required to maintain the null is proportional to applied acceleration, giving these devices exceptional linearity (typically better than 50 parts per million of full scale), very low bias instability (less than 1 ug at room temperature), and noise floors below 1 ug/rtHz. They are the technology of choice for strapdown gyrocompassing, fiscal metering inertial reference systems, and any application demanding the highest accuracy. Their principal drawbacks are higher cost, susceptibility to damage from severe shock if the feedback loop saturates, and somewhat larger physical size.

Piezoelectric accelerometers exploit the piezoelectric effect in quartz or ceramic: mechanical stress applied to the crystal lattice generates a surface charge proportional to acceleration. These devices are the standard choice for high-frequency seismic and vibration measurements (response to 10 kHz and beyond) and for detecting the shock signature of perforating guns or bit bounce. They have excellent high-frequency response but do not measure DC (zero-frequency) acceleration reliably, making them unsuitable for inclination measurement.

Quartz flexure accelerometers use a quartz pendulum whose angular deflection is detected by capacitive pick-off and rebalanced by electrostatic torquers. They combine the temperature stability of quartz (very low thermal expansion coefficient) with closed-loop linearity. Oilfield quartz flex units are rated to 175 degrees C continuous, making them suitable for deep high-temperature wells where MEMS temperature coefficients become large.

Accelerometers in MWD and LWD Survey Tools

Modern MWD survey sensors consist of a matched triaxial accelerometer set and a matched triaxial fluxgate magnetometer set, rigidly mounted in orthogonal orientations inside a non-magnetic drill collar. The three accelerometer outputs (Gx, Gy, Gz) measure the components of the local gravitational field vector resolved along the tool body axes. From these, inclination (the angle of the borehole axis from vertical) is computed as:

INC = atan2( sqrt(Gx^2 + Gy^2), Gz )

where Gz is the axial component (along the tool centerline) and Gx, Gy are the lateral components. This calculation is independent of azimuth and works regardless of borehole orientation. The magnetometer outputs (Bx, By, Bz) are used in combination with the accelerometer data to compute toolface and azimuth: azimuth is derived from the horizontal projection of the Earth's magnetic field vector, corrected for declination and dip using an International Geomagnetic Reference Field (IGRF) model.

The combined accelerometer-magnetometer sensor package must be housed inside a non-magnetic drill collar (typically an alloy of stainless steel 18Cr-5Mn or monel) of sufficient length to isolate the magnetometers from the magnetic permeability of the adjacent steel BHA components. Typical non-magnetic collar lengths range from 9 m (30 ft) to 15 m (50 ft), depending on the magnetic properties of adjacent components. The accelerometers themselves are not affected by magnetic interference but must be precisely aligned perpendicular to each other; any misalignment between axes introduces a systematic inclination or toolface error that is characterized during factory calibration and carried as an error coefficient in the tool's error model.

Key accelerometer performance specifications for MWD survey service are defined by individual tool manufacturers and validated against ISCWSA SPE 67616 error model parameters. Typical values for a high-specification MWD survey tool include: bias stability less than 0.05 mg (milligravity), scale factor error less than 300 parts per million, misalignment error less than 0.05 milliradians, and g-squared sensitivity (cross-axis coupling) less than 50 micrograms per g squared. Temperature compensation is applied using polynomial correction curves derived during multi-temperature calibration in a precision centrifuge. In high-temperature environments (above 150 degrees C), quartz flexure or temperature-compensated MEMS designs are preferred over standard industrial MEMS.

Gyroscopic MWD and the Accelerometer's Extended Role

In environments where magnetic interference prevents reliable magnetic azimuth measurement, such as within casing strings, in areas of strong magnetic anomalies, or in proximity to other wellbores in congested multi-well pads, gyroscopic MWD tools substitute MEMS rate gyroscopes for the fluxgate magnetometers. In these systems, the accelerometers retain their critical role as the inclination sensor, while the gyroscopes measure rotation rates about the tool axes to track azimuth changes by integration. The quality of the gyro MWD survey depends on the drift stability of the gyros and the bias stability of the accelerometers in equal measure, since any uncompensated accelerometer bias translates directly into a toolface error that corrupts the azimuth computation.

In continuous gyro survey tools conveyed on wireline, the accelerometer package also functions as the primary sensor for depth correlation: comparison of the gravitational field components at each depth station allows the tool to detect and compensate for cable stretch and sheave slip, improving depth accuracy of the resulting survey relative to a simple cable depth encoder. This is particularly important in deviated wellbores where cable tension variations are large.