Thesis (Ph. D.)--Ohio State University, 2004. / Title from first page of PDF file. Document formatted into pages; contains xv, 129 p.; also includes graphics. Includes bibliographical references. Available online via OhioLINK's ETD Center.

Thesis (Ph.D.) -- University of Rhode Island, 2008. / Typescript. Includes bibliographical references (leaves 87-92).

Thesis (Ph. D.)--Université de Liège. / Includes bibliographical references (p. 243-263).

Thesis (Ph.D.)--Tennessee Technological University, 2006. / Bibliography: leaves 164-169.

Anomalous electrical behaviour may appear in the d.c. and a.c. characteristics of metals or semiconductors in which the current is determined by both voltage and temperature. Theoretical investigations have been carried out by assuming different models of heat flow and the conditions for the appearance of thermal breakdown and Negative Resistance (NR) have been obtained for both metals and semi-conductors. For purely longitudinal heat flow, NR is predicted for metals while for semi-conductors, the d.c. characteristic is of the "breakdown" type without NR. On the other hand, the radial heat flow model predicts NR for semi-conductors and the conductivity modulation due to the radial temperature distribution causes a concentration of current-density along the axis, giving rise to the "thermal pinch" effect. For metals, this model does not predict NR and the resistivity modulation confines the current-density within a small depth from the surface giving rise to the "thermal skin effect". NR is also predicted for the model considering longitudinal heat flow with surface heat loss, the d.c. λ-K thermal theory, in semi-conductors whereas for metals, the theory does not predict NR. For the applicability of the λ-K thermal theory, the specimen must be thin enough to ensure an isothermal cross-section. The a.c. impedance of the specimen with a small a.c. voltage superimposed on the d.c. bias has been obtained for the λ-K thermal theory. Theoretical analysis shows that a non-zero surface loss parameter λ is essential for attaining NR in semiconductors. Experiments were performed with metals and semiconductors in an attempt to check the d.c. and a.c. λ-K thermal theories. Comparison of the experiments with the theory shows that for semi-conductors, the λ-K thermal theory is valid for current-density J ≤20 amps.cm-² while for metals, it is valid for J ≤5 x 10⁴ amps.cm-². The measured a.c. characteristics at both low and high frequencies are interpreted on the basis of the a.c. λ-K thermal theory but over the intermediate frequency region, the theory offers no explanation for the "circular arc" locus of impedances observed experimentally for both metals and semi-conductors. The present investigation enables one to determine the character of heat flow from measurements of the electrical characteristics of the specimen and also to distinguish the thermal effects which may be present in other experiments (e.g. on the "magnetic pinch"). / Science, Faculty of / Physics and Astronomy, Department of / Graduate

No description available.

In a grounded direct current system, motor faults may occur which will not be indicated by ordinary protective devices, but which will affect motor operation or create unsafe conditions. All such faults are manifested by the fact that the current in one lead supplying the motor will be greater than that in the other. This current differential can be made to operate a direct current differential relay. About the two supplying leads, when such a fault occurs, will exist a magnetomotive force. This mmf will set up a flux in the core and armature of a relay which forms a magnetic circuit about the supplying conductors. Since the armature of the relay is balanced on a knife edge, the force of attraction between core and armature which is caused by the flux passing from one to the other will move the armature, and in so doing close a pair of contacts. A model relay was built which operates on this principle. Testing demonstrated that it would operate on a differential of approximately one and one-half amperes and that the relay would release when the differential was reduced to approximately four-tenths amperes. The test also showed that these differentials did not vary when the total currents involved changed. These results indicate that a relay built with a slightly different design, (so as to overcome the shortcomings of test model, notably the insecure armature mounting, and so as to provide certain other features, such as control of the current differential for which the relay will operate), would be responsive to current differentials of less than one-half ampere. Such a relay would be a practical device and would be capable of performing, in conjunction with a second, conventional, relay and a circuit breaker, an important control function. / M.S.

A method of measuring the negative-sequence current in an unbalanced three-phase circuit. A four-winding current transformer, in conjunction with a phase shift network and an alternating current ammeter, was used for obtaining a current proportional to, and in phase with, the negative-sequence current. After calibration, readings were taken of the three line currents in a three-phase three-wire circuit and of the current in the negative-sequence ammeter. The negative-sequence current. was calculated from the line currents and compared with the ammeter reading. The difference between the two values, in most cases, was less than five per cent of the ammeter reading. It was found that. the circuit had to be calibrated for each type of ammeter used. It was concluded that direct readings of negative-sequence currents can be made with accuracy consistent with commercial requirements. / Master of Science

Master of Science

The original circuit diagram was drawn previous to any reference work on the subject; however, in the construction of the system it was found necessary to add to and subtract from the original circuit. For instance, in order to isolate the oscillator from the modulator and to amplify the carrier voltage it was necessary to add buffer amplifier stages; since only one system was built and since the system worked on the tuned detector principle, no filters were deemed necessary. Other circuit changes of minor importance are; the use of individual plate voltage supplies to minimize feedback between stages; the use of regenerative detection; and a variation in the obtaining of the oscillator output. The system works satisfactorily, but is very critical; this is largely due to the fact that regenerative detection is used. Once the system is adjusted for quiet distortionless operation, it is found that an increase in modulating potential sufficient to over modulate the carrier introduces distortion. This is probably due to the fact that the tube is not operating on a parabolic portion of the grid voltage plate current characteristic curve during a large portion of each cycle, thus introducing third order, fourth order, fifth order, and nth order distortion terms.* * It is to be remembered that the plate current can be expressed as a power series of n terms of the grid voltage, i.e. --- [see equation] The number of terms depends upon the portion of the curve over which the tube is operating; thus, for "square-law" or second order modulation, which occurs over the parabolic portion of the characteristic curve, n = 2; for third order modulation, n = 3; etc. / M.S.