Volume 26, Issue 1, 01 January 1955
Index of content:
26(1955); http://dx.doi.org/10.1063/1.1721856View Description Hide Description
The Čerenkov radiation emitted by a bunched electron beam passing along a dielectric material at a close distance is calculated for a flat beam near a plane surface, a flat beam between two plane surfaces and for a circularly cylindrical beam within a cylindrical hole.
Magnetic Materials for Digital‐Computer Components. I. A Theory of Flux Reversal in Polycrystalline Ferromagnetics26(1955); http://dx.doi.org/10.1063/1.1721867View Description Hide Description
It is proposed that magnetization reversal in polycrystalline ferro‐ and ferrimagneticmaterials is primarily due to the nucleation and growth of 180° Bloch walls. The origin of domains of reverse magnetization is discussed. The rate of growth of these domains is determined by a study of the elastic and frictional forces which retard the motion of their 180°‐Bloch‐wall boundaries. This theoretical model successfully explains the output‐voltage wave forms of polycrystallinematerials. A figure of merit for the magnetization reversal of magnetic cores is defined as the switching coefficient Sw =(Hm−H 0)τ, where τ is the time required to reverse the magnetization, Hm is the applied magnetic field, and H 0 is the threshold field value at which the average domain‐wall velocity is zero. Sw is composed of an eddy‐current contribution Sw e and a spin‐relaxation contribution Sw r . The value of Sw is derived in terms of various fundamental parameters of the material. It is shown that in ferrites and ultra‐thin metal tapes, Sw e «Sw r . Theoretical relationships expressing the contributions of spin‐relaxation, eddy‐current, and hysteresis effects to energy losses are derived on the basis of this model. A study of the factors which affect the magnetization‐reversal time of materials with square hysteresis loops indicates that a better figure of merit will result if the proper hysteresis shape is obtained by grain alignment rather than by the techniques currently employed in ferrites. A number of experiments are presented in support of this model.
26(1955); http://dx.doi.org/10.1063/1.1721855View Description Hide Description
Shadow pictures of turbulent flow regions give a particular kind of information about the three‐dimensional density field. If homogeneity and isotropy of the fluctuation field can be assumed, the statistical properties of the picture uniquely determine the correlation function or power spectrum of the three dimensional turbulent density field. Shadow pictures obtained in the wake of a supersonic projectile were analyzed in this manner and encouraging results were obtained. In order to check the validity of the method, the density fluctuations in a heated jet were measured both by the shadow method and by the more conventional hot‐wire technique.
Analysis of the Effect of Various Factors on Metal Transfer and Wear between Specimen Pairs of Same Metal and Same Shape. I. The Basic Scheme of Formulation of Metal Transfer and Wear26(1955); http://dx.doi.org/10.1063/1.1721857View Description Hide Description
The fundamental scheme of formulation of the amount of metal transfer and wear is based upon the theory suggested in the previous paper [J. Appl. Phys. 23, 1011–1019 (1953)]. The basic terms used in formulating metal transfer and wear are (1) the potential amount of metal transfer Mt *; (2) the mechanical factor of the chance of metal transfer φ m ; (3) the thermal factor of the chance of metal transfer ψθ; and (4) the adhesion factor of the chance of metal transfer X a . The expression for the amount of metal transfer Mt from one specimen to another is Mt=Mt *φ m ψθ, when ψθ is bigger than X a , and is Mt=Mt *φ m X a when X a is bigger than ψθ. The equation for the amount of wear Mw represented by the weight loss of a specimen is Mw=Mt *φ m (1‐ψθ), when ψθ is bigger than X a , and is Mw=Mt *φ m (1‐X a ), when X a is bigger than ψθ.
Analysis of the Effect of Various Factors on Metal Transfer and Wear between Specimen Pairs of Same Metal and Same Shape. II. Effect of the Surrounding Atmosphere26(1955); http://dx.doi.org/10.1063/1.1721858View Description Hide Description
The discussion in this paper is limited to the case where the surface layer is thinner than the average depth of roughening. The surrounding atmosphere influences the adhesion factor X a through its effect on adhesion that depends on such factors as the nature and thickness of the surface layer and also the degree of plastic roughening. The influence of the surface layer on the thermal factor ψθ is similar to its influence on X a . Because of the lack of the quantitative knowledge about its effect on adhesion and diffusion, the effect of surrounding atmosphere on metal transfer and wear is discussed qualitatively. The dependence upon such factors as the temperature, gas pressure, time of exposure, etc., of the effect of physically adsorbed and chemical‐compound surface layers on metal transfer and wear is also discussed in this paper.
26(1955); http://dx.doi.org/10.1063/1.1721859View Description Hide Description
This paper describes a structure, namely, a cross‐wound twin helix which overcomes the disadvantages of a conventional helix for high‐voltage traveling‐wave tubes. The disadvantages of a single helix suitable for high voltages are: (1) the impedance for electron interaction is reduced because of the energy content of the noninteracting space harmonics, and (2) the high impedance of some of the space harmonics can result in backward‐wave oscillation. In a structure consisting of two helices wound in opposite directions, the symmetry of the fields results in most of the electric energy being stored in the fundamental component and most of the magnetic energy in the space harmonics. This results in a higher impedance for the fundamental component and a reduced impedance for the space harmonics. Typical numbers for dimensions suitable for 10‐kv operation are an increase of a factor of 2 in the fundamental impedance and a reduction of a factor of about 20 in the −1 space harmonic for the twin helix as compared to the single helix.
26(1955); http://dx.doi.org/10.1063/1.1721860View Description Hide Description
Rates of evaporation of Ba and Sr from oxide cathodes in a practical tube structure were measured during life tests up to 20 000 hours. The effects on evaporation rates are shown for (a) impurity in the Ni support, (b) cathode temperature, and (c) space current. We conclude:
1. The product evaporating from commercial type cathodes under normal conditions is largely Ba metal; less than 5 percent is Sr; less than 2 percent is BaO; and less than 0.01 percent is SrO.
2. Chemical reducing agents in the support metal‡ are important in determining the rate of Ba evaporating during life.
3. A factor not closely controlled in the present experiments—anode and grid composition—affects the rate markedly. This is especially true for the evolution of Ba during exhaust.
4. Within the precision of these experiments, there is (a) no correlation between the rate of Ba evaporation and thermionic activity of individual cathodes, and (b) no effect of space current on the rate of Ba evaporation.
26(1955); http://dx.doi.org/10.1063/1.1721861View Description Hide Description
A detailed treatment of reactor kinetics is given. The kinetic equations are derived from a diffusion‐Fermi age model. An iterative technique and a matrix method are discussed as possibilities for obtaining solutions. The relationship between the solutions with and without delayed neutrons is pointed out. The concept of stability is discussed for the system which includes the delayed neutrons. The similarity of the treatment here to that used in other branches of mathematical physics (especially quantum mechanics) is noted.
26(1955); http://dx.doi.org/10.1063/1.1721862View Description Hide Description
Middleton and Sugar have shown that the scatter diagram of two random electrical quantities may be oscilloscopically displayed, and that a simple relation exists between the resulting pattern and the correlation coefficient, if a Gaussian distribution is assumed. The present paper shows that qualitative information concerning the correlation coefficient may be found from the scatter diagram, no matter what distribution is assumed.
Closed Expansion of the Convolution Integral (A Generalization of Servomechanism Error Coefficients)26(1955); http://dx.doi.org/10.1063/1.1721863View Description Hide Description
This paper presents a closed expansion of the convolution integral, which is useful in approximating the response of a linear time invariant system to an arbitrary forcing function. In a sense it is a generalization of the well known error coefficient expansion frequently discussed in connection with servomechanisms. This expansion is helpful when the response of the system cannot be evaluated exactly (i.e., the input function is specified graphically or input function cannot be transformed conveniently). This expansion differs from the error coefficient expansion in that it is valid for an arbitrary forcing function and is in a closed form accounting for the entire response of the system, rather than in an open form of Taylor's series that either neglects or simply bounds a portion of the total response.
26(1955); http://dx.doi.org/10.1063/1.1721864View Description Hide Description
Electrical polarization in thin, dielectric sheet material induced by handling or other means is not readily observed because the external field from a dipole charge layer is very weak compared with the field from net charges that are usually present. A method is described for measuringpolarization in the presence of net charges by placing the dielectric sheet in contact with a grounded metal backing plate and by placing a field meter at a short distance in front of the dielectric. The grounded metal backing plate has the effect of almost completely annulling the external field of any net charge by induction, and of doubling the external field of a dipole charge layer of polarization.
26(1955); http://dx.doi.org/10.1063/1.1721865View Description Hide Description
Extensive calculations have been performed with an electronic calculator to evaluate a problem in elasticity that simulates the effect of a cylindrical charge of high explosive detonated in intimate contact with a steel plate. The general method of calculation has been described in detail. Although elastic theory has been extrapolated into a regime where it is known not to apply, insight of a valuable general nature has been obtained on the nature of the negative component of the pulse.
26(1955); http://dx.doi.org/10.1063/1.1721866View Description Hide Description
Under certain conditions the usual gas‐discharge circuit was found to give rise to relaxation oscillations. The relaxation oscillations produce a series of very narrow pulses with random pulse repetition interval, which appear as noise. The amount of noise was found to depend upon the gas, the cathodeelectrode material, the power supply voltage, and the external circuit configuration. It seems that this mechanism may be responsible for most of the high level noise usually reported from gas discharges. The pulsed nature of the apparently continuous discharge appears to be a fundamental property of the low‐current, cold‐cathode arc which has not been considered previously.
26(1955); http://dx.doi.org/10.1063/1.1721868View Description Hide Description
The problem of armor penetration of thin plates is considered from a quasi‐dynamical approach. Equations are derived for the energy dissipation due to plastic deformation and for heating of the projectile target interface. Both the conical and the ogival head are considered in the application of the general equations.
26(1955); http://dx.doi.org/10.1063/1.1721869View Description Hide Description
For the study of high‐temperature gas dynamics, shock‐tube techniques have been developed earlier to produce shock waves strong enough to heat gases to high and accurately known enthalpy (for argon up to 18 000°K or 40 percent ionization at equilibrium). This paper reports a study of the visible radiation from the highly luminous argon following strong shock waves.
Preliminary spectrograms showed a strong continuum and that the prominent argon lines were broadened and shifted to the red. Correlation of the frequency shifts with theoretical treatments permitted an evaluation of the ion density in the gas. Development of a drum camera spectrograph (film speed 700 ft/sec) in which time effects could be resolved to about 1 μsec indicated that equilibrium ion density was reached rapidly, and provided a rough measurement of the rate of decline of ion density due to cooling.
Absolute photoelectric spectrophotometric measurements of the continuum radiation were made and correlated with theoretical expectations. Confirmation of the expected variation of continuum intensity with wavelength and temperature was obtained and an undetermined factor on the theoretical intensity was evaluated. Determinations of the cooling rate of the high‐temperature argon from time variation of the continuum intensity, the line shift, and the electrical conductivity (by others) are in agreement and show that continuum radiation was the dominant heat loss.
26(1955); http://dx.doi.org/10.1063/1.1721870View Description Hide Description
Shock tube techniques for the production of shock waves up to Mach number 20 have been developed and reported previously by Resler, Lin, and Kantrowitz, J. Appl. Phys. 23, 1390 (1952). These techniques can produce high temperature gas with accurately known enthalpy (e.g., in argon 25 percent ionization has been produced following an incident shock). Spectroscopic studies of high temperature argon produced this way by Petscheck, Rose, Glick, Kane, and Kantrowitz, ``Spectroscopic studies of highly ionized argon produced by shock waves,'' J. Appl. Phys. 26, 83 (1955), showed that equilibrium ionization can be reached in the time available in these experiments (of the order of 100 microseconds). This paper reports a study of the electrical conductivity of high temperature argon produced by shock waves.
At low degrees of ionization (less than 10−3 for argon), the diffusivity of electrons and thus the gas conductivity is determined by the cross section for electron‐atom collisions which has been measured by mobility and by scattering techniques. At high degrees of ionization (larger than 10−3 for argon) the diffusion of electrons is primarily limited by long range Coulomb interaction with positive ions and thus is independent of the chemical nature of the gas. Theoretical treatments of this case have been given by Chapman and Cowling, Cowling, and by Spitzer and Härm. At intermediate degrees of ionization additive effects of both of these resistivity mechanisms would be expected.
Preliminary measurements with electrodes indicated large surface resistances. These effects were avoided by the development of an electrodeless technique in which the moving ionized gas deflected a magnetic field. Resultant voltages induced in a search coil were related to the conductivity distribution in the gas following the shock wave. At temperatures greater than 8000–10 000°K (depending on the gas density) the gas conductivity quickly reached a maximum value (up to 80 mhos/cm). The maximum conductivity obtained at these high temperatures agreed within 10 percent with theoretical expectations. It also agreed well with measurement of electrical resistivity in the cesium discharge by F. L. Mohler, Bur. Standards J. Research 21, 873 (1938). At lower temperatures the oscillograms indicated that the conductivity was still rising at the end of the hot region. Under these conditions maximum conductivities reached were much lower than the theoretical values. The ionization rate obtained varied considerably with the gas density.
At the highest temperatures the conductivity declined quickly from the maximum value and the rate of decline could be correlated with the expected cooling due to recombination radiation. Indications of a high conductivity associated with luminous shock fronts were obtained.
26(1955); http://dx.doi.org/10.1063/1.1721844View Description Hide Description
The ``superstate'' in a magnetic ion source has been investigated experimentally and theoretically. The space‐charge density is approximately constant throughout the source. The radial plasma density does not differ significantly from a Gaussian distribution. The radial distribution of space potential is approximately parabolic. The theoretical results have been tested by probe measurements, and good agreement has been found.
26(1955); http://dx.doi.org/10.1063/1.1721845View Description Hide Description
Experimental results are presented of shock wave and contact front velocity measurements, in air, obtained in a 3 in.×3 in. wave interaction tube. A diaphragm pressure ratio range up to 10 000 was employed, while the distance was varied simultaneously from the origin to 142 in. beyond.
It is shown that when shock‐wave attenuation occurs, it consists of two portions; (a) a decrement due to formation, and (b) a further attenuation due to the distance traversed by the shock wave.
Concurrently with the attenuation phenomenon, the contact region spreads with time and its front boundary accelerates. The increase in velocity consists of two portions; (a) an increment due to formation, and (b) a further rise in velocity with the distance travelled by the contact front.
A satisfactory empirical relation is developed for the total shock‐wave attenuation.
A Rayleigh‐type incompressible ``pipe''‐flow analysis applied to the experimental results overestimates the attenuation for stronger shock waves.
26(1955); http://dx.doi.org/10.1063/1.1721846View Description Hide Description
Alloyed germaniump‐n junctions of 0.1 to 0.2 mm diameter or less have been alloyed upon crystals of roughly ten times greater linear dimensions such that when these units are suitably electrolytically etched, ring formations (both scattered rings and tracks of overlapping rings) occur on the crystal faces in the neighborhoods of the junctions. Best current evidence indicates that the observed effects are etch produced. An explanation of this is given in terms of the action of bubbles accompanying the etching procedure.