The sounds of 115 Temple Bells are analysed and the frequencies of the fundamental and partial tones in their sounds are measured. The variations in the structure of sound spectrum concerning the frequencies distribution of partial tones with the lapse of periods are clarified. Following results are obtained. The intervals between the frequencies of adjacent partial tones are widest in the bells of Nara Period on the average and become narrower with the lapse of periods. In the bells of Nanpokucho Period they become narrowest, and after that there is the tendency to increase gradually. These variations with the lapse of periods are chiefly caused by the increase in thickness of "Koma-no-tsume"(Hoof-shaped lowest brim of the bell body) and the decrease in thickness of the upper portion of bell body as the periods go by. Considering these changes in the thickness of bell body, the empirical formulae clarifying the variations are derived.
The sounds of 115 Temple Bells are analysed and the sound pressure levels of the fundamental and partial tones are measured. The variations in the structure of sound spectrum concerning the distribution of the sound pressure levels of partial tones with the lapse of periods are clarified. Following results are obtained. The sound pressure level of fundamental tone in the striking sound of bells becomes higher on the average as the periods go by and reaches maximum in the bells of Nanpokucho Period. After that it decreases considerably. In the second partial tone the sound pressure level is higher especially in the bells of Nara and Edo Periods, and in the third partial tone it becomes remarkably higher since Edo Period. The number of partial tones constituting the striking sound of bells becomes smaller on the average as the periods pass by from Nara to Kamakura Period. After that it increases again. The variations with the lapse of periods above mentioned are chiefly caused by the increase in thickness of "Koma-no-tsume"(Hoof) and the decrease in thickness of the upper portion of bell body. Considering these changes in thickness, the empirical formulae clarifying the variations are derived.
As a primitive objective parameter of sound fields in rooms on subjective evaluations, the degree of interauralcross correlation (IACC) is being widely accepted as well as the listening level, the initial time delay gap between the direct sound and the first reflection and the reverberation time. The aim of this investigation is to determine contributions of the level and the IACC, and the independence between the both parameters on the subjective preference judgements. In oder to change the two parameters being kept other ones constants, sound fields were simulated by the aid of a computer with the Schroeder's reverberator. Results of the paired-comparison tests show that:(1) Preferred listening level depends upon the source music played, but the IACC. (2) Scale values of preference of sound fields decrease with increasing degree of the IACC, but they are independent of the listening level and the music. (3) The listening level and the IACC affect independently on the subjective preference, so that the scale value of preference of a given sound field can be obtained by adding the average scale values with respect to each parameter.
The detectability of clicks at onset or offset of brief bands of noise is measured as a function of the Sensatin Level(SL) of the signals ranging from 30 to 70 dBSL. The bandwidths are 0. 5, 1, 5, 10 and 24 critical bandwidths(CBW) cetered at 1950 Hz respectively, and each signal has a trapezoidal envelope. The results show that the critical rise or decay time (tc_1, tc_2) required to achieve clickless signal depends on the level and the bandwidth. As a band of noise is widened, the transient clicks are less audible, so that any click can not perceived for a broadband noise. The average data can well be approximated by the equation (I/I_0)・tc^<αj>_j=constant, where I and I_0 are instensity of the signal and its threshold respectively, α_j is a constant depending on the bandwidth, and tc_j is the critical rise (j=1) or the decay (j=2) time. This is the same equation as given for a sinusoidal signal. It is also shown that there exists temporal masking at transient parts of the signal. From above results, a mechanism of the perception of clicks is discussed, and it is suggested that only the upper or lower skirts of the transient excitation pattern at onset or offset of the bands of noise can contribute to the perception of clicks.
The hard hammer with force transducer, the weight of which is same as an actual piano hammer, is covered with a piece of felt cloth, named the test-elastic hammer in this paper. The spring constant of the test-elastic hammer is adjusted almost equal to the actual piano hammer, and at the same time the duration of contact between the hammer and the string so as to be almost equal to that between the actual piano hammer and the string. This paper describes that (1) the force-time curve is observed during the contact between the test-hard hammer and the piano string, and that (2) with the para-force-time curve during the contact between the test-elastic hammer and the piano string observed through the felt cloth, the supposed figure of the force-time curve is discussed during the contact between the actual piano hammer and the string.