Magnetostriction Oscillator Research Articles

The Mechanism of Cavitation Erosion

Abstract After pointing out the interrelation between corrosion resistance, caustic embrittlement, fatigue, creep, and other phenomena of metals, the author cites the results of experiments on glass and quartz rods, subjected to high liquid pressures, as indicative of the action which probably occurs in cavitation erosion. To demonstrate the mechanism of cavitation on this basis, an investigation was conducted, in which tests were made on gray-cast-iron specimens in water, alcohol, glycerin, and paraffin oil, using a magnetostriction oscillator as the source of energy. The results of the tests which are reported bear out the theory. Materials and protective coatings which reduce cavitation erosion greatly, and the effect of the presence of hydrogen in the liquid are discussed.

ニッケル單結晶の彈性恒數

Single crystals of nickel, of a slender cylindrical rod form, were prepared by the method of slow solidification, and Young's moduli of elasticity of annealed crystal specimens having various orientations were measured by the method of magnetostrictive oscillation at room temperature. The relation between the reciprocal of the modulus, 1/E, and orientations(γ1, γ2, γ3) was found to be 1/E=0.8349-2×0.6739 (γ12γ22+γ22γ32+γ32γ12)×10-12cm2/dyne, where γ1, γ2 and γ3 are direction cosines of the rod axis of a crystal specimen referred to the tetragonal axes; Young's moduli in directions of the three principal crystallographic axes, [100], [110], and [111], are E[100]=1.198× 1012, E[110]=2.008×1012, and E[111]=2.592×1012 dynes/cm2, respectively. Then, principal elastic parameters, Sik, were determined by the use of the reliable measured data for the volume compressibility of polycrystalline nickel, which was 0.535×10-12cm2/dyne; results obtained are S11=0.835×10-12, S12=-0.328×10-12, and S44=0.978×10-12cm2/dyne. The rigidity moduli in directions of the three principal crystallographic axes, computed from the above-determind principal elastic parameters, are G[100]=1.022×1012, G[110]=0.605×1012, and G[111]=0.533×1012 dynes/cm2. Hence, the elastic anisotropies are E[111]/E[100]=2.16 and G[100]/G[111]=1.92, being nearly the same as those of iron crystals. Further, three principal elastic constants, Cik, were computed to be C11=2.44×1012, C12=1.58×1012, and C44=1.02×1012 dynes/cm2; so-called Cauchy's relation C12=C44 does not hold. The elastic moduli of quasi-isotropic polycrystalline nickel were calculated from the elastic constants of single crystals by theories of Voigt, Reuss, Huber and Schmid, and Bruggeman, and the calculated values were compared with the measured ones. It was found that theories of Voigt and, particularly, of Bruggeman gave the most accurate values, which were 2.07×1012 and 1.92×1012 dynes/cm2, respectively, for Young's modulus and 0.78×1012 and 0.72×1012 dynes/cm2, respectively, for the rigidity modulus. Further, Debye's characteristic temperature was calculated by the use of the computed values for Poisson's ratio, 0.315 and 0.329, which were obtained by the theories of Voigt and Bruggeman, and the results were 451 and 432 °K, respectively.

鐵コバルト合金の<i>ΔE</i>効果の動的測定

The ΔE-effect (change in Young's modulus of elasticity with magnetization) in the annealed state of iron-cobalt alloys has been measured at the room temperature by the method of magnetostrictive oscillation. The measured range of magnetic field is up to about 900 oersteds. It was found that the negative ΔE-effect occurred at weak fields or at low magnetizations in alloys of the composition range excluding 45_??_50 and 100 percent cobalt. The negative minimum value of ΔE/E (change of Young's modulus, relative to that in a non-magnetized state) is, at most, of the order of -0.1 percent in a low-nickel-content range, while it amounts to -2 percent in a high-nickel-content range of α solid solution. The maximum measured values of the ΔE-effect, (ΔE/E)max, are approximate to the saturation values, (ΔE/E)∞, excluding only that for pure cobalt, and they are all positive in a whole composition range of the alloys. The (ΔE/E)max vs. composition curve exhibits a very sharp and high peak at the composition of about 50 percent cobalt. The value at this composition is 22.10 percent, which is the highest value of the ΔE-effect at ordinary temperatures ever observed. The course of the curve in the α-solid-solution range agrees qualitatively well to the result of an approximate calculation. The results of the present dynamical measurement were compared with those of a statical measurement by Honda and Tanaka, and it was concluded that marked differences discerned between them were due to differences impurities of specimen materials used and in the state of stress in specimens at the time of measurement. Further, a secondary increase preceding to a saturation of Young's modulus of pure iron, and a maximum and minimum at weak fields of the modulus of pure cobalt, found previously by the present author, have been again confirmed by the present measurement.

強磁性ニッケル-銅合金のヤング彈性率とそれの磁化に伴ふ變化

Young's modulus of elasticity and its change with magnetization, that is, the ΔE-effect in the annealed state of ferromagnetic nickel-copper alloys, have been measured in detail at ordinary temperatures by the method of magnetostrictive oscillation, which was described in a preceding paper. The measured range of magnetic field for the ΔE-effect is up to about 900 oerstcds. The general course of the change of Young's modulus accompanied by an increase of magnetic field or intentisity of magnetization, is similar for alloys of any compositions here investigated. The modulus increases quickly at weak fields, but the rate of increase diminishes gradually with growing field, finally the change of the modulus attaining a saturation. As for the relation between the change of the modulus and the intensity of magnetization, the modulus increases more and more rapidly from the low to the high magnetization range, but once the rate of its increase diminishes slightly, and afterwards the modulus shows a rapid increase in the magnetization range adjacent to its saturation. The detailed observation of the ΔE-effect reveals, however, the following two particulars. In the first place, Young's modulus in alloys of an intermediate composition range (5_??_21 weight percent copper) decreases, that is, ΔE/E (the change of the modulus, relative to that in the non-magnetized state) becomes negative, though slightly, at first at very weak fields or in the very low magnetization range, and makes a minimum before it increases rapidly with further increasing field or magnetization. Secondly, in nickel and alloys of high nickel compositions (below 15.5 percent copper), the modulus increases somewhat rapidly again in the field range just preceding the saturation. The saturation value of the relative change of Young's modulus, (ΔE/E)∞, decreases with increasing copper content, but the course of decrease is by no means monotonous, making an inflexion in the composition range of 10_??_15 percent copper. An extrapolation of the (ΔE/E)∞, vs. composition curve shows that the ΔE-effect vani-shes at the composition of about 34 percent copper, and this composition coincides with that at which the ferro-magnetism disapears at ordinary temperatures. The curve calculated on the basis of theories agrees qualitatively well with the measured one. It is also shown that a quantitative difference recognized between (ΔE/E)∞ vs. composition curves of the two sets of alloys of different origins can be explained in terms of the difference in their initial susceptibilities. The rate of a rapid increase of Young's modulus at weak fields, the intensity of magnetic field at which its increase saturates, its temporary decrease in the very weak field or magnetization range, and the intensity of magnetic field at wihch its secondary somewhat rapid increase occures, all decrease with an increase of copper content. The relation between Young's modulus at the non-magnetized state and the composition shows a considerably complicated aspect. The modulus increase at first with an addition of copper to nickel, makes a maximum at about 10, a minimum at about 15, and again a maximum at about 20 percent copper, and then it decreases with further increasing copper. The existence of the minimum at about 10 percent copper has never been found.

鐵-コバルト及び鐵-二ッケル合金のヤング彈性率と密度の測定

Young's moduli of elasticity of annealed iron-cobalt and iron-nickel alloys have been measured by the method of magnetostrictive oscillation at ordinary temperatures. In iron-cobalt system, the modulus vs. composition curve shows two maxima at about 25_??_30 and 60 percent and a conspicuous minimum at about 50 percent cobalt in the α-solid-solution range, and moreover makes a minimum in the composition range where the α and γ solid solutions co-exist. The modulus, then, recovers rapidly in the γ-solid-solution range (80_??_95 percent cobalt), but in the narrow range of ε solid solution it seems to decrease. The curve for iron-nickel system shows a linear and rapid decrease in the α-solid-solution range and in the higher composition range where the α and γ solid solutions co-exist, and it makes α minimum at about 40 percent and a maximum at about 85 percent nickel in the γ-solid-solution range. Further, according to the results of measurements of the densities at the room temperature of these two binary alloys, the density vs. composition curves show a parabola-like increase in both composition ranges of α solid solution of iron-cobalt system and γ solid solution of iron-nickel system and moreover anomalous changes at their boundaries.

THE EFFECT OF SONIC VIBRATIONS ON PHAGE, PHAGE PRECURSOR, AND THE BACTERIAL SUBSTRATE

1. A nickel tube magneto-striction oscillator of 320 watts output producing sonic vibrations of 9,300 cycles per second frequency is described. Certain structural innovations contribute to operating efficiency and permit more convenient exposure of test materials than in earlier types. 2. The rate of phage inactivation by sonic waves proceeds logarithmically with time and serves as a satisfactory measure of energy output during operation of the generator. The curve for phage inactivation taking place in the presence of homologous staphylococci follows that for phage alone but soon reaches a plateau after which no further loss of activity is noted. In general higher concentrations of bacteria more effectively inhibit phage destruction than do lower concentrations. 3. Cells that have attained a resting state after a preliminary phase of rapid growth normally have the capacity of inducing a very rapid and marked increase in [phage] when added to phage. This effect has been attributed to the presence of intracellular phage precursor. The store of phage precursor in activated cells is destroyed by sonic waves in about 30 minutes. The number of cells (plate count) shows no reduction until after the precursor is entirely inactivated. 4. Attempts to extract phage precursor from activated staphylococci by exposing the cells to sonic vibrations were unsuccessful.

New Instruments in Physics

AN interesting account of the debt of modern physics to recent instruments comes from the Bell Telephone Laboratories in New York and is given by Karl K. Darrow (Review of Scientific Instruments, 12, 1-10; Jan.; 53-61, Feb. 1941). The instruments are classified into three types according as they are devoted primarily, to observation, to measurement or to achieving new conditions. In the first article consideration is given to those instruments in which substances can be subjected to new combinations of physical forces or in which one variable is pushed to extreme limits. Bridgman's scheme for achieving pressures of the order of 100,000 atmospheres, Beams's ultracentrifuge rotating at more than 20,000 turns per second, the piezo-electric or magnetostriction oscillator, the Dickel and Clusius use of convection and thermal diffusion to separate isotopes, the attainment of vacua of the order of 10˜8 or 10˜9 mm. mercury, by the aid of oil diffusion pumps and ‘getters', the attainment of temperatures of only a few hundredths of a degree absolute, and of high electric voltages, and magnetic field strengths are all dealt with. The second article is devoted to instruments of observation and instruments of measurement without enforcing too sharp a distinction to the two types. Both articles are well illustrated and give an excellent idea of the present limits reached in most branches of modern physics. The developments in that protean tool, the thermionic vacuum valve, and its resulting developments in acoustics are so great that another article would have been necessary to deal with them.

鐵,ニッケル及びコバルトの<i>ΔE</i>効果に就て

The ΔE-effect, i.e. the change of Young's modulus of elasticity with magnetization, in the annealed state of electrolytic iron, decarburized 0.1 per cent. carbon steel, electrolytic nickel and electrolytic cobalt has been measured in detail, in the magnetic field up to 900 Oe, by the method of magnetostrictive oscillation and measuring apparatus, which were described in a previously published paper. In every metal the course of change of Young's modulus by magnetization has been found to be not so simple as was hitherto known; the moduli of iron and nickel display both a secondary increase in the field range preceding the saturation, and that of cobalt shows a maximum and a minimum in the weak field or the low magnetization range. The ΔE-effect is positive for all of the metals examined. The saturation values of the change of Young's modulus relative to that in the non-magnetized state, (ΔE/E)∞ are 0.2822, 0.4851 and 12.60 per cent. for electrolytic iron, decarburized mild steel and electrolytic nickel, respectively, and for electrolytic cobalt the value of the relative change of the modulus at the highest magnetic field applied, (ΔE/E)max is 0.2208 per cent., the value of (ΔE/E)∞ obtained by the extrapolation being 0.464 per cent. The saturation value of ΔE-effect in decarburized 0.1 per cent. carbon steel coincides perfectly with the value expected from the experimental results on the ΔE-effect in carbon steels of varions carbon contents, described in the previous report, and this fact contributes to the further establishment of the proposition that the amount of ΔE-effect in carbon steels disminishes continuously with the increase of carbon content. The comparison of the experimental results of the ΔE-effect of both iron and nickel with the theories proposed by Akulov and Kondorsky, Brown, Takagi, and Becker and Doring indicates that, every theory agrees well with the experiment for the ΔE-effect of iron, while for that of nickel only two theories by Takagi and Becker-Doring give the calculated values coinciding with those measured. A theoretical consideration also shows that the difference between measured values of (ΔE/E)∞ of iron of to kinds is primarily due to that of their susceptibilities.

Specific Aggregation of Streptococcal Proteins Adsorbed on Oil Droplets.

Emulsions of oil droplets in solutions of certain streptococcal proteins have proven to be a means of transforming precipitative reactions into agglutinative reactions. Type-specific aggregation, by means of antistreptococcal antisera, of oil droplets emulsified in solutions of Lancefield's “M” substance are of particular interest. The solutions used have been solutions of the proteins extracted from Streptococcus pyogenes (Group A αhemolytic streptococci) at neutrality as well as the acid-soluble type-specific protein (Lancefield's “M”) from Streptococcus pyogenes., The antisera used were rabbit sera prepared by injecting rabbits with living cultures of Type 1 and Type 6 streptococci, and by injecting rabbits with a solution of the neutral extractable proteins. The experimental technic employed follows: Preparation of the antigen. 0.5 cc of olive oil was emulsified in 20 cc of distilled water. This emulsification was accomplished by putting the material in the magnetostriction oscillator of Chambers† and Gaines and treating with sound waves., To this emulsion was added the protein solution. This was prepared by dissolving 25 mg of protein in 5 cc of distilled water and carefully adjusting the pH to 7.4 (brom thymol blue indicator). The protein solution and emulsion were then shaken for 15 minutes in a Kahn-test shaker and allowed to stand overnight in the refrigerator. Examination of such b.n emulsion under the dark field microscope showed many minute droplets of oil homogeneously dispersed in water. Macroscopic Test. To 0.2 cc of antiserum dilution was added 0.2 cc of a 1:3 dilution of the antigen. After thorough shaking, the tubes were placed in the 56°C bath overnight. In the morning the tubes were centrif uged and read. In reading the test it must be borne in mind that agglutinated oil droplets will be found in a creamy layer at the surface of the liquid.

An Improved Magnetostriction Oscillator

An Improved Magnetostriction Oscillator

The propagation of supersonics in capillary tubes

The velocity and attenuation constants for the propagation of supersonic vibrations, of frequencies 40 to 115 kc./sec., in air contained in narrow tubes, of diameters 0.06 to 0.15 cm., are measured, the sources being monel needles in magnetostrictive oscillation. On comparison with Kirchhoff's theory, the results show that the absorption in the tubes is considerably greater in practice than in theory, though not markedly different from that observed at low frequencies.

An Improved Magnetostriction Oscillator

An Improved Magnetostriction Oscillator

鐵及び炭素鋼のヤング彈性率の磁化に伴ふ變化

The change of Young's modulus of elasticity by magnetization, ΔE/E, in annealed Armco iron and various carbon steels was measured by the method of magnetostrictive oscillation. Main results obtained are as follows:-Young's modulus of elasticity in these materials increases by magnetization, and the amount of increase is of the order of 10-3. The rate of increase is large in weak magnetic fields, and becomes gradually less as the field increases. For a given magnetic field, ΔE/E decreases continuously as the carbon content increases. Every ΔE/E vs. intensity of magnetization curve consists of three stages. Akulov and Kondorsky's theory was confirmed for the total change of elasticity in Armco iron. The initial susceptibility plays an important role in the change of Young's modulus of elasticity by magnetization in iron and carbon steels.

種々の鋼の彈性係數に就て

The modulus of elasticity of various kinds of high yield point steels and carbon steels was measured with an ordinary triod valve magnetostriction oscillator, and it was found that the addition of small amounts of special elements to steels causes hardly any change in the modulus of elasticity. The change in the modulus of elasticity of a carbon steel containing 0.1 percent. of carbon, due to cold working was also measured, the result of which shows that the cold working causes no appreciable change in the modulus of elasticity of the steel. The modulus of elasticity of armco iron and of carbon steels containing 0.27, 0.46, 0.55 and 0.81 percent. of carbon was measured at elevated temperatures up to 600° the result of which is satisfied by the following formula Et2=Et1e-a(t2-t1)/lm-t2 where tm is the melting points of steelss.

The Magnetostrictive Oscillation of Quartz Plates

Small quartz plates were vibrated by mechanical impact from magnetostrictive rods cut to the proper lengths. Sand scattered on the plates gathered at the nodal lines and produced figures similar to those on Chladni plates. The quartz plates were then vibrated in an electrical field produced by the same vacuum tube circuit which was used to vibrate the magnetostrictive rods. Electrical and mechanical vibrations gave exactly similar patterns on each crystal provided the frequencies were identical.

Magnetostrictive Alloys with Low Temperature Coefficients of Frequency

Thirty-four magnetic alloys of iron, nickel, chromium, and cobalt were prepared and studied in order to find compositions which have substantially zero temperature coefficient of frequency of longitudinal vibration. Rods of these alloys were made to be used as secondary frequency standards, to stabilize the frequency of magnetostriction oscillators. The same alloys would be valuable for tuning forks. It was found that the temperature coefficient of frequency is a function of composition, heat treatment, temperature, and magnetization. Seven compositions were found which gave temperature coefficients of the order of one cycle in a million per degree centigrade, when properly heat-treated and magnetized. Five of these showed large dynamic magnetostriction, and gave good frequency stabilization when used with a magnetostriction oscillator.

A Study of the Velocity of Sound in Air

This study of the velocity of sound in air involves the use of the magnetostriction oscillator in an experiment similar to those of G. W. Pierce and others, who used piezoelectric crystals as sources of sound. The wave system generated by the source is investigated by observing the reactions of the source when the sound waves are reflected back upon it from a moveable reflector. The range of frequencies used extends from 20,000 to 70,000. The irregularities of the wave system near the source are studied in some detail. It is found that these irregularities vary with the frequency, as well as with the diameters of both the source and the reflector, and that they vanish at relatively large distances from the source. The values of the velocityof sound then obtained show no variations with respect to the frequency.

A Magnetostriction Oscillator Producing Intense Audible Sound and Some Effects Obtained

Apparatus is developed capable of pushing the magnetostrictive oscillation of nickel to the limit set by its mechanical strength. The intense audible sound produced is introduced under water from beneath the surface, continuous operation of the vibrator being secured by spray cooling. The interiors of corks are charred to cinder by the sound. A fountain 6 cm high forms on the surface of water, and is proved to be due to pumping action of the reciprocating rod and not to radiation pressure. Erosion of metals and of glass by cavitation and resultant water hammer is obtained. Colloids of oil and of carbon in water are produced. Curious fatigue effects in glass are observed. Whitened appearance and erratic motions of air bubbles released in water in the presence of the sound are observed, and theory is developed to explain these and the ``atomization'' of liquids. Flattening of a drop of liquid against a vibrator is observed and attributed to sonic attraction and to the reaction of sonic radiation. Frogs, fish, larvae, and water fleas are killed by the sound. Bacteria are killed so quickly that partial sterilization of milk in continuous process is effected. Striations in a Kundt tube that are parallel to its axis are produced, and an explanatory theory developed.

Further experiments on magnetostrictive oscillators at radio-frequencies

Work on magnetostrictive oscillators at radio-frequency is continued from a previous paper. It is shown that the coil surrounding the bar can be in either branch of a simple tuned anode circuit. When the bar-coil is in the inductive branch the circuit may be operated as a series or parallel arrangement; in the latter case the direct plate current does not pass through the inductive branch of the fly-wheel circuit. The variation in either the anode or the grid current can be used to indicate resonance. Comparative experiments with corronil, nickel and glowray suggest that glowray is the most suitable of these materials for high-frequency oscillators. The experiments are carried on to the case of a glowray oscillator 1.9 mm. long which has a frequency of 1280 kc./sec. and weighs less than a fiftieth of a gramme.

The equivalent circuit of the magnetostriction oscillator

The Joule effect in magnetostriction is defined by the relation S = λB, where S is the stress produced by a small increment of induction B of an existing induction B0; and the converse effect by the relation H = κξ, where H is the magnetic field produced by a small strain ξ. Conservation of energy requires that κ = 4πλ. These equations are used in investigating the motion of a magnetostriction oscillator consisting of a closed circular ring inside a toroid, eddy currents and hysteresis being taken into account. It is shown that the oscillator can be represented by an equivalent electrical circuit comprising a pair of parallel impedances Zc, Zm, in series with an impedance Zl, where Zl represents leakage impedance and Zc impedance due to core flux in the absence of motion, while Zm is a resonant shunt to Zc and represents the effect of the motion. Expressions for the elements of the circuit in terms of the fundamental constants of the material are given. The circle-diagram of impedances is deduced, and the modifying effects of eddy currents and hysteresis are investigated. Some simple geometrical relations between the vectors in the diagram are derived. It is shown that the size of the circle diagram for solid material having a large value of λ depends mainly upon permeability and resistivity and only slightly upon λ, and that a high degree of resonance may be associated with poor magnetostriction quality; the circle-diagram may, in fact, lead to erroneous conclusions unless interpreted in the light of an adequate theory. An experimental investigation of the resonant radial vibrations of solid and laminated nickel rings verifies the theoretical deductions and leads to numerical values of λ and κ in agreement with values deduced from the work of Masumoto and Nara and of McKeehan and Cioffi. For nickel in the annealed state, λ=1.76×104 and κ = 22.1 × 104 at a point on the curve corresponding to H0 = 14.5 gauss.