Anywhere a human comes in contact with machinery — automobiles, planes, trucks, construction equipment, home appliances, tools — noise is an issue. It’s an issue for designers, engineers and regulators who are charged with producing appliances and machinery that is quiet, or at the very least, produces a sound that evokes a positive user response. The time-tested method for suppressing noise in a product has been a trial and error approach to “engineer in” the sound/noise attenuation measures to satisfy a noise level objective. This approach has been time consuming, expensive, and provides marginal results. Now, however, an acoustics engineer can quickly “design out” the noise in a product before it goes into production.

SenSound’s HELS-based (Helmholtz Equations Least Squares method) NAH (Nearfield Acoustic Holography) allows engineers to produce 3D visualizations of all sound values (pressure, intensity and velocity) of an object in a single measurement with very high spatial resolution, the broadest frequency range (3.4Hz – +6,500Hz) and verifiable accuracy. For design and acoustical engineers, this tool allows them to design out the noise in a product with precision.

Fig. 1.

Traditional NVH (Noise, Vibration ands Harshness) approaches measure signals through multi-channel inputs and outputs and analyze them using filtering, modulation, convolution, and correlation to gain insight into the problem. Alternatively, NVH analysis has used intensity scans to map sources. While these approaches have been the best methods available to date, they are usually time consuming and measurement position dependent. In addition, intensity probe results are valid at the measurement points only, and sources can be missed or misdiagnosed. And though Planar acoustic holography took a step in the right direction, it still suffers from three things:

Constraints in microphone spacing and positioning.

Multiple set ups when broad frequency ranges are of interest.

The inability to account for reflections and the limits of plane-to-plane reconstructions (because 3D surfaces cannot be reconstructed).

The SenSound HELS-based approach is very different in that it reconstructs in order to visualize the entire 3D acoustic field of a source object including pressure, intensity and velocity. This comprehensive approach provides the ability to:

Pinpoint sources and transmission paths.

Identify normal surface vibration response in order to extract the vibration modes that are responsible for sound radiation.

Distinguish between structure borne and airborne noise.

Identify the contribution of different sources relative to a given location.

Fig. 2.

The results are valid in the entire 3D space and surface and are much less measurement-position dependent.

To demonstrate the effectiveness of the software and how it can be integrated into the appliance product design process, SenSound conducted a test using both its SenAH (stationary) and SenNS (non-stationary) acoustical holography software options. Rather than diagnose a specific problem, the team sought to demonstrate the spatial resolution, accuracy and comprehensive sound data that can be realized from the test all around an object. In this instance, the test was performed on all sides of a juicer using two patches, one covering the upper half and the other covering the lower half. A single microphone was used as a stationary reference and each patch consisted of 30 microphones.

Fig. 3.

SenSound’s methodology calls for the use of conformal microphone arrays as opposed to the flat microphone arrays used to support Fourier methodology (planar technology). In this case, the microphone patches conform to the curvature of the surface of the juicer. Since SenSound’s technology allows the microphones to be placed very close to the source surface, the evanescent sound waves from the juicer are captured, something that is impossible with planar technology. As a result, HELS provides a complete analysis of what is taking place. In addition, HELS is valid in interior regions, making it suitable for analyzing sound coming into enclosures as well. As the accompanying figures illustrate, SenSound displays a narrow slice of information of a virtual ocean of data the software collects with every measurement.

In Fig. 1, the software displays the time history of acoustic pressure measured by one of the microphones starting at a point just before the juicer was switched on. The acoustic pressure increases during the run-up, and then drops down once the motor speed becomes steady. The acquired acoustic pressure was converted from time domain to frequency domain to analyze its spectral contents. Conditions under both “Run-up” and “Steady speed” have been identified for viewing.

Fig. 4.

The spectrogram in Fig. 2 displays the spectral information of the generated noise every 10 ms (milliseconds) at a point just before the juicer was switched on for a period of 4 sec. The graph clearly shows some distinct peaks corresponding to each time instant. The frequency of each peak increases with motor speed suggesting order related noise. The most prominent peak was found for the order corresponding to 227 Hz at steady motor speed. This graph also shows that most of the higher orders are loud during the latter half of the motor run-up and decay in amplitude as the motor approaches a steady speed. For steady speed, acoustic pressure was measured for both patches for 6 sec. and was time averaged.

The graph in Fig. 3 displays the acoustic pressure spectra for all microphones of the upper patch up to 1,000 Hz. Peaks were found at 227 Hz and its harmonics, that is, 454 Hz, 681 Hz, 906 Hz at higher orders. The motor of the juicer was spinning at 1,135 RPM or about 19 rotations per second, making 227 Hz the 12th order of the motor speed.

Fig. 5.

The software screenshot in Fig. 4 displays the distribution of acoustic pressure over the side face of the juicer at 227 Hz, clearly showing pressure hot spots. In this view, the user has the option of moving the cursor to the frequency peak of interest on the spectra and a corresponding cursor will indicate the maximum sound pressure for that peak on the object’s reconstructed image. The reverse is also an option. The pressure plot is overlaid on a digital photograph of the juicer.

The software screenshot labeled Fig. 5 shows another view of the 3D surface of the juicer. However, this screenshot displays the velocity distribution on the surface. The velocity is plotted on a log scale (Ref. 1E-8 m/s). SenSound has built in several software features to realize a thorough array of visualization options. For example, the user is able to compare two conditions simultaneously. It is possible to visualize an image of reconstructed velocity distribution side by side with intensity or pressure values. Every practical component that is measured to understand the noise phenomena is available for 3D viewing.

Fig. 6.

Fig. 6 is another screenshot of the software, but this one displays the intensity distribution on the juicer surface in 1/24th octave band centered around 229 Hz. The ability to view a pressure plot and its corresponding velocity and intensity values at a specific tonal frequency separates this technology from most other acoustical diagnostic tools. Noise sources are identified in an instant and the designer/engineer knows precisely where the problem exists. While this example looked only at the outer layer of the object, the process can be repeated with the “skin” off to identify the sources of excitation that are being transmitted through the skin at the points indicated by the analysis.

SenSound’s Windows-based software can be integrated with most 24-bit data acquisition platforms on the market. The software can also be provided as part of a turn-key system that is supported with Larson Davis’ DSS. For those who could use the capability on a short-term, non-permanent basis, the company provides both diagnostic services and rental system configurations.

SenSound offers the ability to visualize all noise values (pressure, intensity and velocity) at the surface of an object and in the surrounding field in a single measurement with high spatial resolution, the broadest frequency range, and verifiable accuracy. It is a methodology that provides more insightful information than traditional approaches, thereby leading to faster diagnosis of problems and more cost effective solutions at any stage of a product’s life cycle.