This paper is composed of two parts. The first part deals with the general theory of long-period seismographs having a device of the mechanical registration and the properties of horizontal pendulum of Zöllner's befilar type. As to be convenient for designing these instruments we investigated about the style of the suspended bodies as well as the tension of the fibre in the equilibruim state. The results thus obtained are graphically represented in Fig. 3, 4 and 5.
In the second part is described an electrical methol which makes to record mechanically the motion of long-period pendulum with a sufficient magnification and without any frictional resistance. For this purpose we adopted a system of series, i.e. horizontal pendulum (seismograph)→coil (generator)→amplifier→recording galvanometer. The seismograph is similarly constructed in size and figure as Galitzin's one. The horizontal pendulum, having a mass of 10kg hanged with piano wires of 0.7mm in diameter, bears a coil (20, 000Ω) and alminium plate. Both lie in the magnetic field and serve as the generator and damper respectively. In this seismograph it is not so difficult to get a period more than 40 sec. the pendulum continues to freely oscillate for a long while without any considerable decrease in amplitude if the effect of the damper be properly taken away.
The recording galvanometer was specially designed for this purpose. It is of a moving coil type and bears a recording pen with a straw (strengthened by bakelite suffusion) style in order to get registration on a smoked paper (See photopraphs of Fig. 16). The resistance of this galvanometer is 5, 600Ω, the proper period of it 0.4 sec., the damping ratio
v=4 in the case of open circuit. It causes a deflection of about 1cm. on the paper for a current of 0.1mA when the arm length is 16cm.
The current generated in the coil by earthquakes is in Galitzin's method lead to a long-period sensitive galvanometer and observed photographically, but here we magnified it by a voltage and current amplifier to get a current large enough to act the mechanically recording galvanometer of short-period mentioned above. The wiring of the amplifier is shown in Fig. 6. This amplifier consists of three stages of push-pull type directly coupled. Vacuum tubes B 228 and A 409 are used. The formers of earlier two stages are operated by 2 and 200 Volt storage battery and the latters of the last stage by rectified current obtained from 100 Volt A. C. line. We can obtain a sufficient stableness of this amplifier for the continuous routine observation and its adjustment and treatment are not so difficult. If the condensers used in this amplifier be omitted or changed properly for others of different capacity, we can get to some extent a desirable condition of amplification for the current generated by earthquakes according as its period varies. In a word, the magnification curve for stationary oscillations obtained by the present arrangement is very alike to that of Galitzin's instruments. But there are some inevitable defects because the recorded seismogram is magnified in very complex manner except when the earthquake motion takes place in regular oscillations, and even in this case a remarkable phase lag is observed for rapid oscillations.
In brief, our intension is to get the mechanically registered seismogram applied to the Galitzin's seismograph, because the photographic recording has various inconveniences for a routine work. Some examples of seismograms obtained by the present method are shown in Fig. 16. Of course there may be many things to be improved in our apparatus, it seems, however, to suggest that the registcring method of this kind may be useful for the routine observation of distant earthquakes by long-period seismographs, and also applicable to observations of other geophysical phenomena, such as terrestrial magnetism, earth current, wind, temperature and pressure of the air etc.
View full abstract