Enhanced Loose Particles (eLP)

Listen’s new enhanced Loose Particle algorithm offers accurate transient distortion measurements, even in the presence of background noise. This algorithm is highly accurate, as well as easy to configure and set limits. Results are easily correlated to audibility by listening to the loose particles in the recorded waveform.

This algorithm uses the same time envelope analysis as our pioneering (2004) Loose Particles algorithm, but rather than filtering and counting pulses, additional analysis is applied to measure the prominence, or impulsiveness, of the detected artifacts. The algorithm calculates the prominence of each loose particle event in the waveform by comparing the magnitude of a peak relative to the surrounding minimums, which reflects the impulsiveness of the artifact more accurately than the absolute magnitude of the peaks.  A user-defined prominence threshold determines the level at which the peak will be counted, and a user-defined count of events over the time window determines the pass/fail threshold.

Objective results obtained by this method are easily correlated to subjective analysis as you can listen to the recorded waveform with the fundamental removed, hearing only the loose particles. This enables the prominence threshold to easily be set based on defect audibility.

The correlation between audible transient distortion and the various steps in Listen’s enhanced Loose Particle algorithm

This analysis method is inherently reliable in a factory environment as external background noise events typically only occur once or twice during a measurement, whereas many loose particle transients will occur during the same timeframe. The event count is user-determined, and is set according to the background noise in the measurement environment. Prominence threshold and loose particle count are the only parameters that the user needs to define, so limit setting is simple, well correlated to audibility, and can be configured to give reliable results even when background noise is present.

Benefits

• Random transient distortion is differentiated from periodic harmonic distortion for rapid and efficient troubleshooting of production line defects
• Since it relies on a cumulative event count rather than a single event triggering a fail, false positives due to background noise are minimized.
• Since the measured metric is the number of loose particles greater than a certain threshold in a given timeframe, limit setting is simple as it is not frequency dependent

Applications

• End-of-line test of speakers, headphones and similar devices
• Buzz, Squeak and Rattle (BSR) measurements in cars
• Rattling components in audio devices (buttons, etc.)

eLP Methodology

Schematic diagram of Listen’s enhanced Loose Particle algorithm for transient distortion measurement

This schematic explains the analysis used in the algorithm. The measurement begins with our unique stepped sine stimulus STWEEP™ which has a smooth transition from one frequency step to the next. This avoids any transients introduced by changes in stimulus frequency that can sometimes occur when a continuous sine sweep is used. Since this is the same stimulus we generally use for other end-of-line tests, this measurement can be made alongside others with no increase in test time.

After playing the stimulus through the device under test, the resulting waveform is captured, and the artifacts are extracted from the waveform. This extracted Loose Particle waveform can be played through SoundCheck on a reference loudspeaker or headphone to listen to the recorded artifacts and enable correlation to the measured results.

Next, an envelope analysis is applied and we plot a time domain measurement of the energy in the recorded waveform against time. Transient defects are revealed as an anomalous burst of energy.

The Prominence of each peak is calculated, and a Prominence Threshold, the level above which an event will be counted, established.

Finally, the number of events over the measurement duration is counted. This results in a numerical output upon which a pass/fail limit can be set.

The 4 graphs below show the outputs of the four stages in the algorithm. They are all on the same horizontal time axis so that the relationship between them is clear.