Transmission loss measurement method for car dash panel based on beamforming
Yang Yang,Chu Zhigang,Shen Linbang,Zhou Yanan,Duan Yunyang
DOI: https://doi.org/10.3969/j.issn.1002-6819.2014.05.006
2014-01-01
Abstract:Conventional methods for transmission loss measurement have drawbacks such as failing to identify poor sound insulation positions or time-consuming. In order to overcome these shortages, an innovative method was proposed, which is based on beamforming noise source identification technology and estimation of partial area sound power with beamforming scaled sound intensity. Its effectiveness was validated by comparing the transmission loss of a car dash panel before and after parts of dash panel were installed using the proposed method and the conventional sound intensity method. The experiment was conducted in a specialized laboratory consisting of a reverberation room and an anechoic room. In the reverberation room, an omnidirectional sound source (Brüel&Kj?r Type 4292) was excited by white noise emitted from PULSE front end and was amplified by a power amplifier (Brüel&Kj?r Type 2716), and a microphone (Brüel&Kj?r Type 4187) was utilized to measure sound pressure signals. In the anechoic room, different instruments were used for sound signal collection. For the test with the conventional method, a dual-microphone sound intensity probe (Brüel&Kj?r Type 3599) was used to scan and measure sound intensity from the surface of the car dash panel, while for the test with the beamforming method, a 36-channel sector wheel array with a diameter of 0.65 m (Brüel&Kj?r) was utilized to collect sound pressure signals. After signal processing based on the principles of the two methods, contour maps and transmission loss curves of the car dash panel were drawn. Results showed that compared with the conventional method, the proposed method could not only accurately locate the position where sound insulation was poor in the dash panel, but also evaluate the transmission loss of the dash panel at frequencies up to 2 000 Hz. Although the proposed method underestimated the transmission loss at frequencies more than 2 000 Hz because of influence of side lobes, the underestimated value could be used for assessing effects of different conditions and improvement technologies. Overall, the proposed method was an effective and fast measurement method of transmission loss. The test after parts of the dash panel were installed showed that the switch valve for inside cycle and outside cycle located in the air inlet of the air conditioner was the primary acoustic weak link because of its poor fit with the valve port. In order to fix this problem, sealing material was pasted on the valve body. The corresponding transmission loss curve acquired by the proposed method indicated modification dramatic improvement in the transmission loss of the car dash panel with a minimum increment of about 1.4 dB and a maximum one of about 6.2 dB. Furthermore, since acoustical material inside the car dash panel is an important factor affecting sound insulation performance, the original sound absorbing material composed of EVA (Ethylene Vinyl Acetate) rubber sheet and low-melting felt was replaced by a new one composed of EVA rubber sheet, low-melting felt and bi-component fiber. The corresponding transmission loss curve acquired by the proposed method revealed that transmission loss of the dash panel was significantly improved at frequencies less than 2 000 Hz, but unchanged at frequencies more than 2 000 Hz. In general, frequencies of the engine noise and the interior noise of a car are less than 2 000 Hz. Hence, the new sound absorbing material was superior to the original one in reducing interior noise of the car. By pasting sealing material on valve body and using a new sound absorption material, the transmission loss of the car dash panel increased by about 4.2 dB on average at frequencies of 500-5 000 Hz, suggesting that the both methods could greatly improve interior acoustical environment of cars.