Definitions for frequently used inertial sensor terms
The average over a specified time of accelerometer output measured at specified operating conditions that has no correlation with input acceleration or rotation. Bias is a signed quantity usually expressed in units of acceleration.
In general, any error that includes two or more sources; specific meaning is generated by naming the sources. More in detail, it is the maximum deviation of the output data from a specified output function. Composite error is due to the composite effects of hysteresis, resolution, nonlinearity, nonrepeatability, and other uncertainties in the output data. Errors Sources are typically RSS'd to obtain composite error.
Unit of acceleration, equal to a standard value of gravity or an otherwise specified level.
Undesired perturbations in the accelerometer output signal which are generally uncorrelated with desired or anticipated input accelerations. Noise in the accelerometer output is of two types: intrinsic and seismic.
Intrinsic noise is generated within the accelerometer and represents the limiting factor in making measurements. Intrinsic noise is random in nature and is characterized by a noise power spectral density (PSD) curve.
Seismic noise is a true input acceleration (usually unanticipated by the user). It results from noise sources in the local environment (running motors, seismic shocks, etc.) and their transmission to the accelerometer through the mounting structure which supports the accelerometer.
The closeness of agreement among measurements of the same variable, repeated under the same conditions, especially when changes in conditions occur or when operation is interrupted between the measurements
The ratio of the change in output (in volts or amperes) to a unit change of the input (in units of acceleration); thus given in mA/g or V/g.
The ratio of a change in response to a change in an undesirable or secondary input (as the scale factor variation to a unit of power supply voltage change).
The process by which performance characteristics are measured over the operating temperature range and mathematically reduced to obtain a constant and coefficients for a multi-order model. The model is normalized to one temperature (typically 20°C). The application of the model to raw data is referred to as 'correction'.
The sensitivity of a given performance characteristic (typically scale factor, bias, or axis misalignment) to operating temperature. Specified as worst case value over the full operating temperature range. Expressed as the change of the characteristic per degree of temperature change; a signed quantity, typically in ppm/°C for scale factor, mg/°C for bias, and mrad/°C for axis misalignment.
Vibration Rectification Coefficient (VRC)
A calculated number obtained by dividing the bias shift that occurs during vibration (rms)2 value of the applied input vibration. Usually expressed in mg/g2 units.
Vibration Rectification Error(VRE)
The apparent shift in accelerometer bias when subjected to sinusoidal or random vibration due to the accelerometers even order non-linearity. Usually expressed in mg or g units with a specified vibration applied
The specified maximum input rate (°/sec) over which full performance will be provided.
Full Range Output
The nominal voltage output for the specified full range input. The actual voltage output will be a complex compilation of several parameters including scale factor, bias, linearity and temperature characteristics.
The ratio of the change in output (in volts or amperes) to a unit change of the input (in units of angular rates); thus given in mA/(°/s) or V/(°/s).
The average over a specified time of gyro output measured at specified operating conditions that has no correlation with input rotation or acceleration. Bias is typically expressed in degrees per
Bias Thermal Sensitivity
The bounds within which the Bias may vary as the temperature varies across the operating temperature range, included in the "Over Operating Environments" specification.
The random variation in bias as computed over specified finite sample time and averaging time intervals (e.g. over 100 s or 1 day).
Self-generated electrical noise is defined by its power-spectral-density (PSD) and given in volts-squared per Hz over a given range of frequency. This defines the distribution of noise power in the output of the instrument, on a per-Hz basis.
MTBF: Mean Time Between Failures
The predicted elapsed time between inherent failures of a system during operation
The gyroscope operates from DC (zero frequency) to some upper frequency (typically defined by the frequency in which the frequency response function magnitude is down by 3 dB).
The maximum difference, at a fixed temperature, between bias readings during a full range excursion of input rates - first measured at zero input rate after having just returned from negative full range; and second, that bias reading measured at zero input rate after having just returned from positive full range.
The time required for the instrument to produce a usable rate output after power application
Vibration - Operating
The specified limit of random vibration in g-RMS, usually across the flat 20 Hz - 2,000 Hz input spectrum range, which the device will withstand while operating within its performance specifications or dedicated performance range.
Vibration - Survival
The specified limit of random vibration (usually expressed in g-RMS over a defined frequency bandwidth) which the device will survive for a limited period of time without damage.
The limit of shock which the device will withstand without damage.