Best Practice Guide to Measurement of Acoustic Output Power (Radiation Force Balances)
Introduction | Radiation Force Balances | Measurement | Additional Information | Further Reading
Tank linings
A transducer under test should be operating into an acoustic 'free-field' - effectively is a reservoir of fluid of infinite extent so that acoustic reflections from interfaces do not perturb the main acoustic beam . Although physically unrealisable, these conditions may be approximated in practice through the use of suitable anechoic tank wall linings that absorb the majority of ultrasound incident upon them. The standards require that the overall echo-reduction should be better than -20 dB, which represents a combination of the echo from the front face and any that might return from the tank wall itself i.e. following a double-pass through the absorber. To realise this requirement, a range of materials have been used as ultrasonic tank wall linings, although published data on their properties is limited . A summary table is shown below.
| Material | Details of sample | Single pass transmission loss (dB) | Echo-reduction (dB) |
| Neoprene | Flat sample, thickness 3 mm. | 15 dB @ 1 MHz | - 11 dB over frequency range 1.5 to 4 MHz |
| Rubber car matting | Extreme thickness of 13 mm. Cones on base of thickness 3 mm. | 7 dB @ 1 MHz, 12 dB @ 3 MHz | <-30 dB |
| Plastic door mat | Extreme thickness of 13 mm. Foam base attached to lines of plastic spikes. | 10.5 dB @ 1.5 MHz. Very position dependent. | No data reported. |
| Carpet | Acrylic, hessian backed, overall thickness: 15 mm. | 20 dB @ 1 MHz. | < - 50 dB, not measurable. |
| High attenuation material | Parallel sided flat sample of thickness 14 mm. Internal pyramidal structure; polyurethane-based rubber. | 28 dB at 1 MHz, increasing linearly with frequency. | -40 dB at 1 MHz, increasing smoothly to -35 dB at 3 MHz |
On first look, carpet performs well, with very low echo-reduction and high transmission loss. However, ancillary problems arise when using carpet, in particular the large amount of air which remains entrained within the material after immersion and several hours' soaking. Also, fibrous material becomes relatively easily detached from the walls, and can, in extreme cases, nucleate cavitation activity.
It is the combination of the transmission loss and echo-reduction which govern the suitability of a material as a tank wall lining. In this way, Neoprene rubber, which has a very high transmission loss at 1 MHz, also has a relatively high reflection coefficient, due to the acoustical impedance mismatch at the front surface. The latter property would thus lead to an overestimate in the measured power at 1 MHz of 8%. The acoustic properties, particularly transmission loss, are very dependent on frequency so that commonly-available materials such as car or plastic matting which might be appropriate at 3 MHz cannot be used at 1 MHz.
In general, however, most materials perform better at higher frequencies. The data reported is for ambient conditions: at elevated temperatures, the acoustic properties can change significantly. The presence of coherent reflections from tank surfaces (usually an underperforming absorbing target or lining) can be investigated (and corrected for) by carrying out measurements of ultrasonic power as a function of distance between the transducer and the target - specifically, deriving the power value from measurements made at axial distances separated by a distance of λ/4.