We made a few experiments on a Japanese bell in order to clarify the relation between the character of striking rod and the timbre. In a previous paper, a model experiment was tried to investigate how the material, mass and speed of a striking block in place of the striking rod effected on the timbre of Japanese bell. From the results of model experiment, we tried some experiments using the striking rod in real use and got light on the relation of the size, mass and speed of the striking rod to the timbre. Moreover, we made some observations on the striking rod suitable for the sound of Japanese bell. The evaluation of the timbre was done in the same way as in the previous paper, namely, by the difference of sound spectra, which were given by the strengths of the fundamental tone, the first overtone and other tones. The sound spectra were obtained by means of a 1/3 octave analyzer. The results obtained from the above experiments are as follows: (1) The pattern of sound spectrum is varied by changing the position of the striking point. As the position is moved from the low place of bell to the high, the pattern is changed from "down type" and to "roof type" and "up type" as shown in Fig. 6. (2) Even though the striking speed of the striking rod is changed, the pattern explained in (1) is hardly changed, but the sound level is only varied. (3) When the ratio of the length of striking rod to the diameter of the bell (L/D) is ranged from 1. 5 to 1. 8, the pattern of sound spectrum becomes "roof type". On the contrary, this pattern is turned to "down type, when the ratio of L/D exceeds 3. 0. (4) When the load is applied on the striking rod within 2 kg and L/D is changed from 2. 1 to 3. 5, the pattern of sound spectrum is not varied. (5) The most important factor which effects on the sound spectrum does not consist in the ratio L/D, but in the weight of striking rod. If the weight of striking rod is smaller than a suitable weight the sound spectrum does not indicate "roof type", which seems to be agreeable. (6) There are the weight and striking speed of the striking rod suitable for magnifying the sound of bell.
Optimal transducer design of piezoelectiric ceramic transformers used for high voltage DC supply, such as DC sources of electrostatic electrography and horizontal deflection circuits of TV receivers are studied in this paper. A ceramic transducer has to be provided with suitable performance in practical use and high durability for a long time. Operating properties of the piezoelectric ceramic transducer in high power drive level is generally non-linear and detriorates with time. However, a ceramic transducer can be used with excellent durability and stability, when transducer is driven within the limits of the linear input level characteristics. The transducer in this paper is designed assuming an input level at a turning point of the linear relation to the non-linear relation with respect to input level characteristics. Such a relation between input resistance and input current is shown in Fig. 2. Resonant frequencies of ceramic transducers with different length ratios of l_g/l_d are calculated. The dimensionless frequency constant (α_1l_d) under conditions of no loading is shown in Fig. 4. For a loaded condition, expressed as a function of Q_<02> (loaded Q) and S (stray capacitance ratio, C_s/C_<02>), the dimensionless frequency constant at the maximum voltage step-up frequency of λ-type resonator (α_1l_d) is shown in Fig. 5. The length ratio of the λ-type resonator in practical use nearly equals √<S^D_<33>/S^E_<11>>, and α_1l_d nearly equals π. Consequently, the equivalent circuit representation shown in Fig. 8 and the relations in eq. (16) are practical and useful in determining the behavior of the ceramic transformer. The maximum voltage step-up ratio (γ_<Lm>) is calculated numerically from eq. (19) and Table 1 of physical constants of the ceramic transducer material as parameters of S and Q_<02>. Dependence of γ_<Lm> on Q_<02> seems to be divided into four portions as shown in Fig. 9. For the first portion, (a) range responding to high Q_<02>, γ_<Lm> can be expressed as eq. (21) which was derived by R. C. Rosen (1956). For the second portion, (b) range responding to medium Q_<02>, γ_<Lm> can be expressed as eq. (22). In the case of application to the horizontal deflection circuit of TV receivers, the values of Q_<02> usually exist in this (b) range. For other two portions, (c) and (d) ranges responding to low Q_<02> and very low Q_<02>, γ_<Lm> can be expressed as eq. (23), or eq. (24) respectively. Requirements of design factors for increasing the tendency of the voltage step-up ratios depend on the value Q_<02>. Design factors for obtaining the optimal voltage step-up ratio in each range of Q_<02>, are listed in Table 2. Other characteristics of the ceramic transformer depending on Q_<02> and S are studied. The field strength of the generator section, V_<out>/l_g is a important design factor depending on the maximum allowable strength (T_<gm>) of the the generator section as in eq. (32). General considerations for transducer design in the range of medium Q_<02> ((b) range) and an example of a 20 kV DC output transducer are described.
There are many interesting and practical problems in nonlinear effects of a sound field. In particular, progressive waves of finite-amplitude cannot be propagated without a change of waveform, which is a wellknown phenomenon and the subject of many reports. For a quantitative estimation of the waveform distortion, it is generally assumed that the wave at a source is purely sinusoidal or that the initial sinusoidal wave is radiated, and note only the progressive distortion by excluding other distortions mainly due to non-sinusoidal motion of the diaphragm of the source. In the present paper, we have relaxed that condition so as to investigate thoroughly the relation between the motion of the diaphragm and the received signal. This paper concerns the cancellation of second harmonic sound pressure of the progressive distortion. Theoretically, it is possible to cancel out the harmonic at some receiving point by adding a small signal to the driving current and determining the amplitude and phase of the small current appropriately. In the method, however, the assumption as an independent propagation of the fundamental and second harmonic waves is necessary, i. e. , this argument is valid only so long as the distortion is not so remarkable. Initially the theory is discussed in terms of finite-amplitude plane wave. Secondly, the analysis is extended to an arbitrary radiation field and examined by experimentation using a horn loudspeaker. Agreement between these results is satisfactory with respect to the driving current-form. Although this study is now at a basic stage in investigations on boundary problems in finite-amplitude sound fields, it can be applied to the construction of a no wave distortion field.