Accelerometer Calibration Testing: Principles, Methods, and Standards

Capacitive MEMS Accelerometers

Capacitive MEMS sensors dominate consumer electronics, inertial navigation, and tilt measurement. Calibration addresses zero-g offset, scale factor accuracy, temperature drift, cross-axis sensitivity, and noise floor, often across multiple axes and over extended temperature ranges.

Servo Force-Balance Accelerometers

Servo accelerometers offer very high sensitivity and resolution for seismic monitoring, geotechnical instrumentation, and platform stabilization. Their closed-loop architecture requires careful static and low-frequency calibration, including gravity-based and centrifuge methods.

Reference Standard Transducers

Back-to-back reference standard accelerometers are calibrated at the highest accuracy level and then used as transfer standards for routine comparison calibration of working sensors. Their periodic recalibration, typically at intervals not exceeding one year, anchors the traceability chain of the entire laboratory.

Testing Items

Depending on the sensor technology, accuracy class, and end application, accelerometer calibration testing may include some or all of the following items:

• Reference sensitivity, commonly determined at 160 Hz and a specified acceleration such as 10 m/s² or 100 m/s² in accordance with ISO 16063-21
• Frequency response (amplitude and phase) across the operating band, for example from 0.5 Hz to 10 kHz or higher
• Amplitude linearity over the dynamic range, from the noise floor to the maximum rated acceleration
• Transverse sensitivity ratio per ISO 16063-31, typically required to be below 5 percent for precision sensors
• Mounted resonance frequency and damping characteristics
• Bias voltage, output impedance, capacitance, and insulation resistance for IEPE and charge-type sensors
• Zero offset, hysteresis, and long-term bias stability for DC-responsive MEMS and servo sensors
• Temperature response over the rated range, for example −55 °C to +150 °C or up to +250 °C and beyond for high-temperature piezoelectric designs
• Thermal transient response, base strain sensitivity, and magnetic field sensitivity
• Shock survivability and residual sensitivity shift after overload exposure
• Noise floor, resolution, and self-test functionality where applicable

Testing Methods

Primary Calibration by Laser Interferometry

Described in ISO 16063-11, primary calibration uses laser interferometry to measure the displacement of the shaker table directly, deriving acceleration from optical fringe data using fringe-counting, minimum-point, or sine-approximation methods. Because the acceleration is realized from length and time standards, no reference transducer is involved. Achievable uncertainties are typically 0.1 to 0.5 percent, and this method is primarily used by national metrology institutes and top-tier laboratories to calibrate reference standards.

Reciprocity Calibration

ISO 16063-12 specifies an absolute primary method based on the electromechanical reciprocity of certain exciters. It is less common in routine service but historically important for establishing primary vibration standards at specific frequencies.

Comparison Calibration

Comparison calibration per ISO 16063-21 is the workhorse method for working accelerometers. The sensor under test is mounted back-to-back with a calibrated reference transducer on an electrodynamic shaker, and both devices experience the same motion. The ratio of outputs yields the sensitivity of the test sensor. Typical uncertainties of 1 to 3 percent are sufficient for most industrial purposes, and frequency sweeps can cover the full audio range in a single automated run.

Static Gravity Calibration

For sensors with DC response, the local gravitational acceleration provides a convenient and highly stable reference. Two-point (flip) or six-position static tests rotate the sensor between +1 g and −1 g orientations to determine zero offset and scale factor with uncertainties often below 1 percent, without requiring any excitation hardware.

Centrifuge Calibration

Centrifuge calibration generates sustained, accurate acceleration levels beyond 1 g by rotating the sensor at a precisely measured radius and angular velocity. It is used for high-range sensors and for separating zero offset from scale factor at levels from a few g up to tens of thousands of g.

Shock Calibration

Shock calibration addresses sensors used for impact and pyroshock measurement. ISO 16063-13 defines primary shock calibration using laser interferometry on a shock exciter, while ISO 16063-22 defines comparison shock calibration against a reference transducer. Excitation sources include drop-ball shock machines and Hopkinson bars, covering acceleration ranges from a few hundred to tens of thousands of m/s² with typical uncertainties of a few percent.

Environmental and Supplementary Testing

Temperature response testing combines an electrodynamic shaker with a climate chamber to measure sensitivity variation across the rated temperature range. Thermal transient, humidity, strain, and magnetic sensitivity tests complete the characterization required by demanding aerospace and energy-sector specifications.

Applications

Calibrated accelerometers are a prerequisite wherever acceleration data influences safety, quality, or legal compliance. Typical application areas include:

• Automotive crash testing and safety system validation, where sensor accuracy directly affects injury criteria
• Aerospace flight test, structural qualification, and environmental vibration testing programs
• Vibration test laboratories performing product qualification and reliability screening
• Machinery condition monitoring and fault diagnosis of rotating equipment
• Structural health monitoring of bridges, buildings, and wind turbines
• Seismic, geotechnical, and railway dynamic measurement systems
• Consumer electronics and inertial navigation module verification
• Maintenance of internal reference standards in calibration and test laboratories

FAQ

How often should an accelerometer be calibrated?

Reference standard transducers should be recalibrated at intervals not exceeding one year as recommended by ISO 16063-21. Working sensors are typically recalibrated every 12 to 24 months, and immediately after any suspected overload, drop, exposure to temperatures beyond specification, or repair. Instruments used in legally regulated or safety-critical testing may require shorter intervals defined by contract or standard.

What is the difference between calibration and adjustment?

Calibration only measures and documents the relationship between output and input, including uncertainty; it does not alter the sensor. Adjustment corrects the sensor or its amplifier to nominal values based on the calibration results. A calibration certificate without a subsequent verification measurement cannot confirm the effect of any adjustment.

What accuracy can be expected from calibration testing?

Primary laser interferometry calibration typically achieves 0.1 to 0.5 percent uncertainty in sensitivity. Routine comparison calibration usually delivers 1 to 3 percent, while shock and extreme-environment calibrations typically range from 3 to 6 percent depending on the level and frequency.

Why is transverse sensitivity testing important?

A real accelerometer responds slightly to motion perpendicular to its sensitive axis. In multi-directional vibration environments, this cross-axis response can contaminate measurements by several percent. Testing per ISO 16063-31 identifies the maximum transverse sensitivity ratio and its angular direction, allowing users to mount sensors optimally or correct the data.

Do MEMS accelerometers require calibration testing?

Yes. Although many MEMS devices are factory-trimmed, zero-g offset and scale factor drift over time and temperature. Periodic static gravity and thermal characterization, performed by accredited third-party laboratories, is essential for navigation, tilt, and instrumentation-grade applications.

What is delivered after calibration is completed?

A calibration certificate reporting the measured sensitivity, frequency response data, applied methods, reference standards used, environmental conditions, measurement uncertainty, and the traceability chain. Certificates issued under an ISO/IEC 17025 accreditation carry international recognition and typically include the accreditation symbol.