The effect of humidity on insulation resistance of some commercial fibrous insulating materials was investigated. To avoid experimental errors due to inconstancy of other conditions, the influence of temperature, applied voltage and etc. was also studied.
It is shown in this paper that, after the voltage is applied to the sample, the charging effect occurs and the temtperature rises and the distribution of absorbed moisture changes, owing to the leakage current through the material: therefore, the resistance at a given condition should be computed by obtaining the leakage current at the initial instant the e.m.f. is applied to the material. And it is stated that, generally the humidity has very profound influence on the insulation resistance of fibrous materials, and the relation between insulation resistance and relative humidity which is varied in a cyclic manner, represents a hysteresis loop due to the hygroscopic hysteresis property of the fibrous material, hence the resistance at a certain humidity should be exressed with the record of previous change of humidity; for the practical purpose, however, it is convenient to express the insulation resistance at a specified humidity with the average value of resistancelcorresponding to that humidity, obtained from the hysteresis curve of resistance against humidity. And it is recommended to express the relation between such average resistance and humidity by an empirical formula log₁₀R=a-bh/1-ch², where R is the average resisrance for relative humidity h, and a is a constant depending on the insulation resistance of the material itself at zero humidity, while b and c are constants expressing the effect of humidity on the resistance. For eleven materials, these constants are obtained by experimental results; and it is shown that they are very nearly equal for materials of vegitable origins, but they differ considerably for the other materials of animal and mineral origins or treated material. Finally it is stated that the effect of humidity on the insulation resistance of such materials may be determined by measuring average resistances corresponding to three different humidities and computing constants in the above formula.

View full abstract

Apart from the fact, that the dynamic characteristic of a discharge changes its shape according to the nature and the wave form of the electrical phenomenon, it is not inherently characteristic to the discharge itself, but is also affected by the nature of the entire circuit.
As an illustration, a carbon arc is lit from 60 cycle 300 volt altercating current source, and arc hysteresis curve is observed by means of a Braun tube oscillograph.
The mode of deformation of the curve for varied arc lengths shows us that the arc hysteresis may be considered to have two origins, one being the energy consumption for the ionization at and near the electrodes, and the other the electric and kinetic energy storage by the ions in the arc flame portion.
The arc hysteresis curves are reproduced, when the series impedance is a non inductive resistance, or an inductive reactance. The result for constant arc length, constant voltage (effective value) and constant arc current (effective value), but with the impedance gradually varying from noninductive to highly inductive, is the one most instructive.
It is well known, that a non inductive steadying resistance for an A. C. arc spoils the efficiency and an inductive steadying reactance lowers the power factor. In addition to this, the effect of the arc length and the nature of the series impedance should also be taken into consideration and they must be properly selected according to the purpose of the practical application of the arc discharge.

View full abstract

In this report the writer states the method of construction and some chara cters of the thereto converter which is used to measure milliampere of alternating current. This includes the followinz general heads.
I. The method of fixing of junction.
II. Thermo elements.
III. Heating wires.
IV. The effcet of the room temperature.
V. The relation between the sensibility of the thermo converter and the grade of vacuum.
VI. Some other experiments.
VII. The theory of the thermo converter.
VIII. The conclusion.

View full abstract

Here, the author calculates the penetration of flnx into a D. C. electromagnet under the assumption that the electromagnet is infinitely long solid cylinder.
At first, the assumes that the excitation occurs instantly, and compares his result with experimental result of Dr. Hopkinson. He ascertains that the both results approximately agree with each other.
In second case, he treats the penetration of flux into iron core when a constant emf is applied to the exciting circuit of the electromagnet.

View full abstract

The Noguchi Transformer is a special transformer designed by the auther for the purpose of producing audio-freqnency alternating current for Kohlrausch Bridges and for other similar purpose. This apparatus is very simple in construction and works without giving out any noise.
A short description about this transformer is given in the ournal, May 1921. This apparatus consists of a carbon filament incandescent lamp and a small transformer (5cm×4cm×3cm) in series. The core of the latter has a re-entrant for about 1/10th of the total length; in way of the re-entrant the sectional area of the core is made about I/4th of the remaining part.
Applying the ordinary A. C. town voltage (100 volts 50 or 60 cycles) to the Primary of this transformer, we obtaiin a narrow, steep and peaked voltrage from the secondary side; this is due to very high magnetic saturatition in the narrow portion of the core. This peaked voltage excites the telephone-receiver of a Kohlrausch-Bridge to give an audible sound having the natural frequency of the receiver. Thus we can obtain alternating current for Kohlrausch-Bridges in a simple manner without any noise which may disturb observations. With the use of this audic-frequency source the measurement may without difficulty have an accu acy of 1/10, 000.
In the present. Paper the auther dealt With the theory of production of the peaked voltage, and showed the process for designing the transformer to give the best result. The theoretical deduction is confirmed by oscillograms of the actual voltage and current given out by this transformer.

View full abstract

Since 1919 the establishment of an accurate standard wavemeter has been undertaken but at the least expense and labour. Among various methods, that suggested by the Bureau of Standards has been adopted at last. The method consists of calculating the inductance aud measuring the capacity of a standard wavemeter, and computing the wave lengths from these quantities combined. Thus the range of wave length from about 200 meters to about 22, 000 meters has been covered by the standard wavemeter.
The primary elements of the wavemeter consist of a single turn square coil and a Bureau of Standards type standard variable air condenser. At low frequeney the calculated inductance of the square coil with leads has been 7410 centimeters and the measured value 7412 centimeters which has been in close coincidence. But, as the variation of the inductance value of the square coil due to frequency has been expeeted to be rather great, it has been then calenlated at radio frequency and found to be 7080 centimeters at 106 cycles. The effect of self-capacity of the coil has been cousidered but found to be negligibly small. The inductance value 7080 centimeters at 106 cycles is used as one basis of calculation of wave lengths of the elemental circuit. The capacity of the standard condenser is considered to be constant at all frequencies and is used as one basis of calculation of wave lengths of the circuit. Thus the range of the wave lengths of the circuit has been from 230 meters to 438 meters.
In order to cover a wider range of wave lengths, a series of eight standard coils of simple forms with larger inductance than that of the squarecoil has been made. Combining the series of coils and the standard condensers, the required range of the standard wavemeter has been covered.
Knowing the fact that ample harmonics are drawn from the plate circuitof a triode generator, they have been used to the calibration of wavemeterand the harmonics as high as 200th. have been detected by means of acrystal detactor and a galvanometer.
The triode generator has been worked with a certain wave length and harmonics obtained step by step from the fundamental to the highest possible. When the wave lengths of harmonics get at higher order and come within the range of the elemental circuit, the wave lengths have been calculated by the elemental circuit. The fundamental wave length of the generator has been calculated by the product of harmonic numbers and the corresponding wave lengths given by the elemental circuit and the mean value obtained from which the wave length of each harmonic has been calculated. From these data, the curves of the standard wavemeter have been plotted. Thu s the shortest wave lengths have been carefully calibrated by the elemental circuit and the longer wave length calibrated upwards by turns from the standard of the shortest wave lengths.
The standard wavmeter thus obtained has been compared to the Navy standard which has been calibrate directly with a high frequency machine and the perfect coincidence of the both has been confirmed.

View full abstract