Kohei Yamamoto

Expressions of Maekawa's chart for computation

Abstract Four types of numerical expression of Maekawas chart for the design of an acoustical barrier are presented. The maximum deviation of the calculated values given by these expressions from the readings of the chart is within 0·5 dB in the range of the Fresnel numbers −0·3 N .

A method for measuring oblique incidence absorption coefficient of absorptive panels by stretched pulse technique

Abstract This paper proposes a pulse method for estimating the acoustical performance of absorptive panels to reduce noise reflections when applied to road structures such as parallel walls and ceilings of double deck viaducts. The acoustical performance is specified on the basis of oblique incidence absorption coefficient. It is determined from the ratio of the reflected sound energy from the surface of a test panel to that reflected from a rigid floor. Considering its high power, flat spectrum and stability, an optimum Aoshimas time-stretched pulse (OATSP) is applied as a short duration test signal. The measurements were made on glass fiber with thickness of 50 and 100 mm. The absorption coefficients at oblique angles from 0 to 45° show good agreement with theoretical values based on the empirical model of Delany and Bazley. Ceramics and other materials with non-flat surfaces were also tested. For application, the averaged oblique incidence absorption coefficient, which is averaged over incident angles and weighted on the representative spectrum of A-weighted road traffic noise, is proposed as a single number index for the evaluation of noise control design at highways.

SCALE MODEL EXPERIMENTS FOR THE PREDICTION OF ROAD TRAFFIC NOISE AND THE DESIGN OF NOISE CONTROL FACILITIES

Abstract A method of scale model experiments for traffic noise prediction and design of traffic noise control facilities was developed. The scale model experiment technique may be useful to clarify complex factors, to verify theory and, furthermore, to formulate noise prediction. Two methods of application are separately discussed; by a simulation model experiment and an elementary model experiment. The former is an experiment simulating various conditions, such as noise source and surrounding environment. The latter is useful in revealing how specific factors are involved in noise propagation characteristics. In this report, noise sources developed and applied for scale model experiments and some results of investigations about traffic noise prediction are detailed.

The required sample size in measuring oblique incidence absorption coefficient Experimental study

Abstract It is an important issue to clarify the minimum required size of a specimen for the accurate measurement of oblique incidence absorption coefficient. Investigation is made on the basis of experimental studies in a laboratory, using Aoshimas time stretched pulse technique. Measurements are carried out for glass fiber with thickness of 50 and 100 mm as highly absorbing materials and for ceramic with a thickness of 52 mm as relatively low absorbing material. The tested surface areas range from 0.0625 to 42 m 2 for glass fiber and from 0.1563 to 39.168 m 2 for ceramics. The absorption coefficients are determined at incidence angles from 0 to 45° in increments of 15°. With consideration of an active surface bounded by ellipse orbit, the minimum required surface area is 16 m 2 with a ratio of length to width of about 1.25. This value is twice the area size of the active surface, which is calculated from the geometry of source and microphone, and the width of the temporal window.

A SIMPLE METHOD FOR THE CALCULATION OF NOISE ATTENUATION BY A FINITE LENGTH BARRIER

Abstract A practical expression for the estimation of sound attenuation by a barrier of finite length is presented. The sound attenuation is determined by taking account of sound contributions from rectangular sections of a plane barrier located between a point source and a receiver. The expression for the attenuation due to a finite length barrier is finally formulated by combining the attenuations due to semi-infinite plane barriers. Scale model experiments were carried out to check the validity of the expression and the results showed good agreement between them. The calculated values were also compared with the values estimated from the methods of ISO/DIS 9613-Part 2 and the Fresnel-Kirchhoff theory. It is shown that our expression is useful and applicable to the design of a finite length barrier when the sound source has a broad band spectrum and the overall sound pressure level is a matter of interest.