Vol. 13, No. 1 (1957)
In this paper a method for solving electric network is suggested. By this method, the solution of the network may be found conveniently without solving simultaneous equations.This paper consists of two parts. The first part deals with the method for solving the general linear network. The procedure of which is, firstly, to replace the e.m.fs. in the network by an equivalent source of currents; secondly, to eliminate the joints of the network one by one until the network transforms into an equivalent branch for finding the potential difference between the terminals of this branch; thirdly, to restore the equivalent branch to original network for finding the potentials of all joints of the network; finally, we can calculate the currents of the branches of the network easily. The second part deals with the special method for solving the linear network which contains only one e.m.f. The difference between this method and the former one is that the proceeding of restoring the equivalent branch to original network in this method may be omitted. In order to omit this proceeding, it is necessary to assume a certain amount of current source on every joint of the network before the transformation of the network. Therefore, when the network transforms into an equivalent branch, we may find the potentials of all joints of a network which contains only one e.m.f.
In the present work, a theory of multiple bunching including the effect of space-charge is developed through the use of space-charge wave concept. The results contain Feenberg's kinematic theory as a special case and reveal characteristics of multiple bunching system that could not be obtained from the kinematic approach.The theory will find applications in microwave electronic devices and some examples of practical significance are discussed.
The masses of nuclei from n to Ca41 have been derived from nuclear Q values exclusively, without recourse to mass spectrometric results, by an approximate least-squares adjustment. Main results are:(1) The comparison of the value of the mass difference 2D2-He4 derived from nuclear data with that from mass spectrometric data furnishes the most accurate experimental verification of the relation E=mc2. The accuracy of this verification is 1/6000.(2) Tables are given for the most probable mass values from nuclear data of the 79 nuclei from n to Ca41. Fundamental mass differences are n-H1=0.7824±0.001Mev, n+H1-D2=2.2255±0.0015Mev,2H1-D2=1.4431±0.0018Mev, 2D2-He4=23.840±0.004Mev. Mass values include n=1.008 9842(±17),H1=1.008 1440(±17),D2=2.014 7381(±29),He4=4.003 8732(±21),C12=12.003 807(±5),S32=31.982 188(±26),Ca40=39.975 204(±39) (3) The result of the present calculation demonstrates that the method used by the author, within the limitations of present-day nuclear experiments, has the same significance as an overall least-squares adjustment, attains same degree of accuracy, but is simpler in procedure and more convenient in the examination of the experimental results for systematic errors.(4) Comparison of nuclear results with recent mass spectrometric results shows that the masses of H1 and D2 are in agreement for the two systems, but definite discrepancy exists for C12 and much larger ones for S32 and Ca40.
1957, 34 (1): 58-68. doi: 10.7498/aps.13.58
The distribution function of crystal sizes on a cross-sectional surface can be obtained by direct observation. By using the well-known Scheil-салIтъIков method one can deduce the volume distribution function. Their method consists chiefly in solving a system of simultaneous linear algebraic equations, in which certain approximations are adopted without explicit criterions. The present paper proposes an analytical method of solution, in which operational calculus and Laplace transform are applied. The density function thus obtained is somewhat in variance with those of Scheil and others; namely, values obtained by the previous authors appear a little lowered as compared with ours. As the present method is rigourous and the numerical computations so far performed are within criterious in accuracy, the difference may well be attributed to the errors introduced by the approximations in the old methods. This fact is fully demonstrated by a concrete example in the text.
INTERNAL FRICTION PEAK ASSOCIATED WITH THE STRESSINDUCED DIFFUSION OF CARBON IN LOW-CARBON ALLOY MARTENSITE
1957, 34 (1): 69-77. doi: 10.7498/aps.13.69
Internal friction in hardened low-carbon nickel steel was measured with a torsion pendulum and an internal friction peak was observed around 155℃ with a frequency of vibration of about 2 cycles per second. The condition for the appearance of this internal friction peak is that the steel specimens contain martensite, alloying element and carbon. This internal friction peak has also been observed in chromium steel and chromium-nickel steel under suitable conditions. Systematic observations were made with steel specimens containing 29.7% Ni, and the height of internal friction peak was found to be proportional to the carbon content in the specimen. The experimental results mentioned above show that the new internal friction peak is associated with the stress-induced micro-diffusion of carbon in low-carbon alloy martensite.A preliminary model was suggested in which the carbon is assumed to be at the interstitial positions of 00(1/2) type. The presence of alloying atoms introduces an inhomogeneous distortion in the lattice, consequently the jumping probability of carbon atoms between two types of 00(1/2) interstial positions (Fe-C-Fe and B-C-Fe, where B represents an atom of alloying element) under conditions of thermal equilibrium is altered by the application of stress. Such a stress-induced movement of carbon atoms gives rise to internal friction. This model can explain qualitatively the observed experimental facts. A quantitative study on this subject is in progress.