The accelerometer-based shock-pulse method extracts bearing impacts from acceleration signals using band-pass filtering and envelope demodulation [S1].
The shock pulse approach gives rapid bearing-condition assessment, but it cannot by itself establish the overall machine condition, so vibration and machine-context measurements remain necessary [S1].
Measurement physics and operating distinction
When two pieces of metal in motion contact each other, the initial impact produces a shock or pressure wave whose amplitude is relative to the impact velocity [S4]. A shock pulse typically occurs around a center frequency of 36 kHz, and its amplitude is relative to the velocity of the impact [S4].
An accelerometer measures acceleration over its usable frequency range, but raw acceleration does not isolate bearing impacts by itself. The acceleration-based shock pulse method therefore applies a band-pass filter, performs envelope demodulation, quantifies pulse strength, and normalizes the result against bearing and operating conditions [S1]. Filter bandwidth is a design tradeoff: a narrower band improves rejection of noise, while a wider band retains more of the available pulse information [S1].
Shock and vibration are separate phases of an impact. The shock phase propagates as an ultrasonic pressure wave, while the rebound phase produces structural vibration controlled by mass, stiffness, geometry, and damping [S4]. One published implementation places the shock center near 36 kHz, and the Shock Pulse Method measures shock pulses using filtering combined with a specially characterized accelerometer [S4].
Decision criteria for transducer selection
Use a dedicated shock pulse transducer when the objective is fast screening of bearing impacts, lubrication condition, and recurring bearing damage. The Shock Pulse Method filters out the vibration phase of a collision because its magnitude is structure- and material-dependent [S4].
Use an accelerometer when the objective includes general machine vibration, fault-spectrum analysis, looseness, misalignment, rotor-related vibration, or integration with a broader condition-monitoring system. The acceleration-based SPM study found that extracting impacts from acceleration can widen the sensor's function, but it explicitly retains the limitation that shock pulse information does not represent overall machine health [S1].
Condition indicators and interpretation

Traditional SPM uses two principal readings.
SPM HD converts the same measurement concept into digital outputs. The instrument presents condition using green, yellow, and red bands and retains spectrum and time-signal views for further analysis [S3].
The amplitude of a shock pulse is relative to the velocity of the impact [S4]. SPM HD uses RPM-based sampling frequency and is particularly well suited to measurement on low-speed applications [S3]. In practical use, repeat the same operating point and normalization inputs when comparing trends, because a changed load, lubricant supply, alignment condition, or shaft speed can change the measured severity [S3].
Speed, installation, and signal constraints
SPM HD is specified for a speed range of 0.1 to 20,000 RPM and uses RPM-based sampling to support low-speed measurement [S3]. SPM HD uses sophisticated mathematical algorithms to make the measuring technique less sensitive to interference from other signals [S3]. These features are relevant where bearing impacts repeat irregularly or operating speed changes.
The sensor must still be mounted where the structural path can carry the impact signal. In contrast, acceleration transducers can be mounted for broader vibration analysis, but their usable low-frequency performance, sensitivity, resonance behavior, and noise behavior determine the quality of the final spectrum. Piezoelectric acceleration sensors use a proof mass and piezoelectric crystal and are particularly suited to high-frequency mechanical measurements [S6].
For electrically connected systems, piezoelectric accelerometer signal architecture affects the transmitter, cable, and input arrangement. That integration decision is separate from choosing between a dedicated shock pulse transducer and an accelerometer, but it affects installation cost, noise immunity, and field-service requirements.
Applicability, limitations, and failure modes

Dedicated SPM is best suited to large populations of rotating assets where operators need a quick, repeatable indication of whether a bearing needs lubrication, closer inspection, or continued monitoring.
SPM is not a replacement for general vibration analysis. It cannot monitor the overall state of the machine by itself [S1]. Misalignment, imbalance, structural resonance, looseness, and other rotating-machine faults can affect vibration readings without producing a clear bearing-specific impact pattern. A suitable condition-monitoring system therefore combines bearing-focused measurements with machine-level diagnostics where both are required.
The main failure modes are installation and interpretation errors: poor directional coupling, excessive path distance, unresolved impacts from an unsuitable sensor bandwidth, changing RPM, inconsistent shaft-diameter input, and confusing random external impacts with recurring bearing defects. A broadband accelerometer can also expose these limitations more directly because its output contains the full response within its calibrated range, while a dedicated SPM sensor provides greater selectivity for the impact band [S3][S4].
Recommended selection and implementation
Select a dedicated shock pulse transducer for routine bearing screening, lubrication checks, and trend-based damage alerts when high point counts and rapid interpretation are priorities. Specify the sensor's calibrated resonance, supported RPM range, normalization inputs, output definitions, and alert scheme before comparing suppliers. For a representative SPM HD platform, the documented figures are approximately 32 kHz resonance, 0.1 to 20,000 RPM, 24-bit A/D conversion, HDm in decibels, and HDc referenced to 200 shocks per second [S3].
Select an accelerometer when spectrum analysis, general machine vibration, or broader diagnostics are required. If its signal will be processed as acceleration-based SPM, specify the band-pass bandwidth, envelope-demodulation method, pulse-quantification algorithm, normalization method, and anti-noise treatment. The underlying research identifies these as the core processing issues, rather than treating the accelerometer output as a ready-made shock severity value [S1].
For critical machines, use both approaches rather than forcing a false choice. Keep a broadband accelerometer for machine-level and spectral diagnosis, and use shock pulse instrumentation for repeatable bearing-impact screening. Validate mounting location, RPM input, shaft diameter, alarm limits, and repeatability against recorded baselines before relying on automatic alerts [S1][S3].
Track two verifiable signals after commissioning: the normalized bearing-severity trend and the rate of repeat impacts. A rising carpet or HDc level points toward lubrication or operating-condition changes, while a rising max or HDm level above the background supports investigation for localized damage [S3][S4].