SIMILARITY BETWEEN ACOUSTIC INDICATORS OF AN REAL HALL AND ITS MODEL/REALIOS SALĖS IR JOS MODELIO AKUSTINIŲ RODIKLIŲ PANAŠUMAS

Abstract

The similarity between acoustic indicators of an real hall and its model has been examined. A rectangular hall is 13.6 m long, 10.7 m wide and 7.0 m high. Its floor and ceiling are horizontal. The hall has plastered walls, parquet floor, and reinforced-concrete-slab ceiling. Thus, all surfaces of the hall are made of materials that reflect sound well. A hall model scaled 1:25 was made. The floor and ceiling of the model were made of fabric-based laminate and the walls were made of veneer 8 mm thick, lacquered three times. Therefore, the materials used to produce the model are similar to those of the real hall by their sound absorption properties. A 9 calibre sound pistol was used as a sound source for the investigations in the real hall. The sound signal was stored in the computer memory via a ½′ microphone, an amplifier and an analog-to-digital converter, then analysed by means of a acoustical signal analysis program developed by us. The signal was analysed within the frequency range of 50—5000 Hz. The main objective and subjective acoustical indicators of the hall were calculated using this program. A spark sound source was used for the experiments with the hall model. It was thrust through a hole in the floor in order to improve the radiation directivity diagram. The position of the sound source and a ¼′ microphone was the same in the real hall and its model. The signal was fed from the microphone to the amplifier, then to the analog-to-digital converter and recorded in the computer memory. The signal may be recorded via several different buffers allowing to record signals of varied length. The range of the frequencies investigated was from 1250 to 50000 Hz, the model scale being 1:25. The signal digitization frequency was 166.6 kHz and the digitization time was 6 mks. The decrease of the sound field of a non-filtered signal is of a similar nature during the first and the last 400 ms, i e during the early and the late periods of decrease. In the intermediate period, approximately from 500 to 3000 ms, the sound field decrease in the model exceeds the one in the real hall by only 1–3 dB. In the real hall, the sound field decrease is close to the straight line up to 2500 ms, while in the model—up to 1000 ms only, and the decrease is faster than in actual practice. The further field decrease has the character of a curve and the diffusive properties of the field are impaired. These results show that the sound field decrease in the real hall and in the model is quite similar. Investigations show that the sound field decrease in the real hall and in the model is almost analogous when the decrease is approximated every 10 dB from 0 to—30 dB. The reverberation time difference is 0.16–0.5 s and is lower than 10%. As the field decrease is approximated from − 5 to −35 dB, the reverberation time of the model exceeds that of the real hall by about 1 s, which makes up about 15%. The difference between the early reverberation time of the real hall and its model is only − 0.2 − 0.8 s even up to 500 Hz. This is mainly determined by the air sound absorption in the model at the ultrasound frequencies. As the sound field decrease is approximated from 0 to − 30 dB and from −5 to −35 dB, the difference between the reverberation time of the hall and its model in the frequency range up to 500 Hz is slight, only 0.2–0.9 s, which is less than 15%. The character of change in the sound absorption is analogous to that of the sound absorption coefficients. In the range up to 500 Hz, the sound absorption of the real hall and its model differs by 1–2 m2 only. As frequency increase, the difference reaches − 20 m2. For a non-filtered signal, the music sound clarity index C 80 is 5.6 dB for the real hall and 4.9 dB for the model.