摘要:

The stators of two identical wound-rotor induction motors are connected to a common supply to produce fields revolving in the same direction. The shafts are coupled mechanically and the rotor circuits are electrically connected together with the voltages initially phased so that no rotor current flows. By rotating one stator relative to the other through an angle β the rotor voltages are no longer in anti-phase, rotor currents flow and the unit produces torque.Boucherot adopted a similar scheme in the construction of a composite machine with two stators and a common squirrel-cage rotor. The mid-points of the rotor bars were connected together through a resistive element so that currents always flowed in the rotor circuits and zero torque could not be obtained.Expressions for torque are derived for the shunt-interconnected rotor developed by Boucherot and for the purely series-interconnected motor investigated by the author. It is shown that, for the former, torque cannot be controlled from zero without voltage variation, whereas in the latter arrangement zero torque is always obtained when the phases of the induced rotor voltages are in opposition.The properties and applications of the series interconnected motor circuit are discussed. By the insertion of impedance the torque/speed characteristics of the motors can be modified retaining the zero torque position. Possible applications include small hoists, cranes or lifts, and positioning, manipulating or tensioning drives for which precise control is required.

DOI:10.1049/pi-c.1961.0038

出版商:IEE    

年代:1961

数据来源: IET

摘要:

The frequency response of feedback relay amplifiers is analysed by the dual-input describing-function method. At low signal frequencies, the amplifier behaves like a saturating linear amplifier. This picture is modified at higher frequencies, owing to synchronous effects, and it is difficult to achieve satisfactory operation when the ratio of the input-signal frequency to the periodic excitation frequency is appreciable. To eliminate synchronous effects within the signal band, restrictions must be imposed on the forward linear-transfer function. The theory is compared with an experimental example.

DOI:10.1049/pi-c.1961.0039

出版商:IEE    

年代:1961

数据来源: IET

摘要:

Given a conductance matrixGof ordern, the problem of realizing it by a network withn+ 1 nodes can be split in two parts: to find the tree of the ports corresponding to the given matrixG; and to determine the actual resistances of the branches of the network.A procedure to solve the first (topological) problem is proposed, which is simply concerned with the signs of the elements of the given matrixG. The successful development of the procedure can be considered as a necessary and sufficient condition for the existence of a tree corresponding toG.

DOI:10.1049/pi-c.1961.0040

出版商:IEE    

年代:1961

数据来源: IET

摘要:

The behaviour of networks, whether lumped, distributed, active or passive, is analysed from a fundamental travelling-wave point of view. Generalized wave parameters are derived for a 2-port network from which image, conjugate and iterative parameters follow as special cases. The relationship of these to short- and open-circuit parameters are also given.A matrix organization of waves in a multi-port network is next presented, which distinguishes between waves internal and external to the network. Through special transformations, voltage- and powergain matrices are derived and their relationship to the conventional scattering matrix is shown.Finally, a visual representation of the wave trains set up in cascaded stages is given in the form of wave flow diagrams, reminiscent of Mason's feedback graphs.

DOI:10.1049/pi-c.1961.0041

出版商:IEE    

年代:1961

数据来源: IET

摘要:

A method of rendering feedback control systems amenable to treatment by the Wiener theory is applied to the case in which the controller operates on a sampled measurement. An explicit expression is obtained for the minimum attainable mean-square error for certain classes of system transfer functions and disturbance power spectra, and the form of the optimum controller is derived. The results show the inherent limitations in controllability imposed by the structure of the controlled system and by the sampling process.

DOI:10.1049/pi-c.1961.0042

出版商:IEE    

年代:1961

数据来源: IET

摘要:

The paper describes tests made to compare the magnetic stability of representative martensitic and precipitation-hardening (isotropic and anisotropic) permanent-magnet materials. The weakening of magnets with time after magnetization, when left as free from disturbance as possible, was measured over a period of about three years, both on unstabilized magnets and on magnets artificially stabilized by weakening them 1% or 5% by applying demagnetizing fields. Anisotropic alloys were found to be much more stable than isotropic alloys, the stability of Columax being outstanding. Alnico, in spite of its much higher coercive force, was not more stable than the martensitic steels.The effect on stability of the working-point of the permanent-magnet material was also examined and found to be small above the(BH)maxpoint on the demagnetization curve. Below the(BH)maxpoint, however, the stability of both isotropic and anisotropic materials was considerably reduced.Further tests were made on the effects of heating at various temperatures up to 220° C and of mechanical stress and impact. Stability in high-temperature treatment is only roughly related to stability in time tests. The results of the mechanical tests were inconclusive as the magnet assemblies available, designed primarily for long-term tests, were not suitable for subjection to severe mechanical treatment.

DOI:10.1049/pi-c.1961.0043

出版商:IEE    

年代:1961

数据来源: IET

摘要:

The paper deals with the topological synthesis of non-reciprocal resistance networks specified by the so-called driving-conductance matrix. Synthesis of non-reciprocal networks has already been treated by H. J. Carlin and Y. Oono, and their networks are specified by scattering matrices and consequentlyn-port impedance matrices. One of the merits of this paper, which differs from their paper, is in the determination of the connection and ideal transformers. Factors which determine the network are the connection and element resistances, and they therefore make possible the determination of the node-branch incidence matrix and resistances so that the driving-conductance matrix may equal the given matrix. When we denote the branch e.m.f. matrix byVand the branch-current matrix byI,I=GV, whereGis called the driving-conductance matrix, which is assumed to be given, and, in general,G′≠G. If a maximal set of solutions forGx= 0 andx′G= 0 is denoted byXas the matrix expression,Xis called the zero factor ofG. Provided that the zero factorXsatisfies the character of a node-branch incidence matrix, the connection can be fixed directly. If not, the connection is determined by the part ofXwhich satisfies the character of a node-branch incidence matrix, and the ideal transformers needed are determined by the other part. After the connection and the ideal transformers are determined, the branch resistances are determined by comparing the given matrix with the matrix calculated from the newly determined connection and the unknown branch resistances.

DOI:10.1049/pi-c.1961.0044

出版商:IEE    

年代:1961

数据来源: IET

摘要:

Approximate analytical expressions are derived for the impulse response of identical circuits in cascade. The results obtained from these approximate expressions, tested for an exactly solvable case, show remarkable accuracy. Cruder approximations to obtain a quick estimate of peak-to-peak ring in response to a step function, and the first zero of impulse response, are also given.

DOI:10.1049/pi-c.1961.0045

出版商:IEE    

年代:1961

数据来源: IET

摘要:

The theory of microwave propagation through an unbounded ferrite medium magnetized along the direction of microwave propagation is considered, and expressions are derived for the components of the microwave field. Boundary conditions are applied, and characteristic equations are derived for four circularly symmetric shapes, namely(a) Ferrite-filled round waveguide.(b) Ferrite rod at the centre of round waveguide.(c) Ferrite tube adjacent to the wall of round waveguide.(d) Ferrite tube in the centre of round waveguide.Some propagation constants have been computed for the last three shapes. The method is such that it may easily be extended to give the characteristic equations for any circularly symmetric shape of ferrite inside round waveguide in addition to the characteristic equations given in the paper. At 9370Mc/s some computed results of Faraday rotation have shown reasonable agreement with those measured on ferrite rods. The results of a specimen investigation of the variation of the propagation coefficient with change of ferrite dimension are included.

DOI:10.1049/pi-c.1961.0046

出版商:IEE    

年代:1961

数据来源: IET

摘要:

Faraday rotation of the plane of polarization of an electromagnetic wave, propagated through a medium in the direction of an independently applied steady magnetic field, has been interpreted in terms of gyromagnetic action and in certain circumstances as a consequence of Hall effect.The paper calls particular attention to the Hall-effect mechanism, which is shown to lead to different propagation coefficients for the right-handed and left-handed circularly-polarized components of a plane wave.The contention is advanced that a Hall effect can be expected to arise from a displacement current in the material medium and that, as a consequence, the Faraday rotation effect in dielectrics and semiconductors may, in some cases, be significantly modified at frequencies above the microwave part of the spectrum. Some elementary calculations are given in support of this argument.

DOI:10.1049/pi-c.1961.0047

出版商:IEE    

年代:1961

数据来源: IET