Showing papers in "Transactions of The American Institute of Electrical Engineers in 1932"

Equivalent Circuits-I

Abstract: Equivalent circuits have been found valuable in the solution of certain problems in that they simplify the analysis. Furthermore, equivalent circuits provide a means of replacing magnetic coupling with simple impedance links thereby permitting the use of the calculating board for solution. In Part I of this paper, the general equivalent mesh for the n-winding transformer is derived, and other less rigorous transformer circuits are discussed and compared. In Part II the mesh equivalent for the general network having (m + l) points of entry is derived and its application is explained. The generalizations of Parts I and II provide a background for the particular problem treated in Part III in which the study of system networks involving groups of parallel transmission lines is considered. The usually difficult problem of accurately determining ground fault currents in such a system is readily solved by applying the special equivalent circuits developed in Part III to a calculating board set-up of the system. These circuits are particularly valuable in the solution of any problem involving an analysis of the zero phase impedance diagram of a system. The methods and circuits developed in this paper are perfectly general and have many evident applications other than those treated.

Transmission Line Vibration Due to Sleet

Abstract: PHENOMENA of vibration in transmission lines due to the action of a transverse wind are encountered most frequently at moderate wind velocities (about 5 miles per hr.) and are characterized by rather high frequencies (5 to 15 cycles per sec.) and small amplitudes (up to a few cable diameters). This type of vibration is caused by the “Karman” vortices forming behind the wire and is well understood.

Geometrical Circuits of Electrical Networks

Abstract: If all the electrical properties are abstracted from a given electrical network, there remains a geometrical circuit, completely characterized by the sets of branches terminating at the various vertexes. In this paper, enumerations of geometrical circuits are given, classified according to two different parameters, the nullity (number of branches minus number of vertexes plus number of separate parts) and the rank (number of vertexes minus number of separate parts). There is also a partial list of geometrical circuits which are symmetrical with respect to all the branches and all the vertexes.

Three-Phase Multiple-Conductor Circuits

Abstract: During the past year the subject of power transmission by multiple-conductor circuits has received considerable attention. The formulas and estimating curves given in the paper may therefore be of interest at this time. Formulas are developed for the inductance and capacitance to neutral per phase and the approximate corona starting voltage of perfectly transposed, multiple-conductor, three-phase transmission lines having any number of conductors per phase. For certain special arrangements of the conductors, curves are given for the 60-cycle reactance, capacity susceptance, and corona starting voltage. These curves show the effect of variations, throughout their practical range, in conductor diameter, spacing between phases and between conductors of the same phase. Two, three, four and five conductors per phase are considered. A comparison is made of multiple- and single-conductor circuits with respect to charging current, voltage at no load, power which can be carried with the same voltage drop, and stability.

A High Sensitivity Power Factor Bridge

Abstract: This paper describes an a-c. bridge of high precision, used for power factor measurements in one of the dielectric investigations being carried on at the Johns Hopkins University. The bridge, a modified form of the Schering bridge, possesses several novel features and advantages. The detecting instrument is a moving coil a-c. galvanometer; its field excitation is supplied from a phase shifting transformer, which permits obtaining independent ratio and phase angle balances. A shielded transformer electrostatically isolates the galvanometer from the bridge circuits. This bridge is completely shielded and guarded and the analytical theory of the resulting mesh connection is given in the paper. The mathematical treatment is general and may be applied to any similar network. A full discussion is given of the effects upon the balance relation of various sources of error, for example, as failure to balance guard circuits properly and the presence of leakage resistance between measuring and guard circuits. Experimental verification of the equations is presented. Power factors of specimens of cable material have been measured ranging from 0.00007 to 0.16 with a maximum variation of ± 0.000005.

Performance Calculations on Induction Motors Practical Straightforward Means for Calculating Performance of Single-Phase or Polyphase Induction Motors

Abstract: Practical means for calculating the performance of both single-phase and polyphase induction motors from previously determined constants are developed in a straightforward manner. The methods are suitable for accurate calculations but can be simplified for approximate calculations. The methods given are complete and no auxiliary charts or sets of curves are required, nor need any diagram be constructed. The single-phase calculation method is based on the cross field theory presented by H. R. West. The polyphase calculation method is based on the equivalent circuit.

Characteristics of Surge Generators for Transformer Testing

Abstract: The requirements and limitations of surge generators for transformer testing can be summed up as follows: The voltage that can be obtained is, in the first place, largely determined by the capacity of the surge generator when the capacity of the transformer is great. If a wave is desired without superimposed oscillations, it is necessary to insert resistance within the generator. The front of the surge then depends upon the inductance of the generator and the capacity of the transformer. This is a matter inherent to all surge generator test circuits. Thus, in general, waves of very steep front cannot be obtained with large transformers. Further, the inserted resistance also limits the voltage by an amount dependent on the proportion of this resistance to the load resistance. For transformers of low inductance, the length of the wave obtained is largely determined by the capacity of the surge generator. This results in a requirement of large generator capacity if very long waves are to be produced. If the lead between the transformer and generator is appreciably long, other oscillations will be set up which are highly damped but are appreciable at the generator end. These oscillations occur on the front of the wave only, and will be prominent when measurements are taken at the surge generator.

Vibration of Overhead Transmission Lines

Abstract: STUDIES of vibration of overhead conductors have been prosecuted vigorously for several years in laboratory and fields. The importance of magnitude and distribution of stresses in multiplestrand conductors led to the development of a unique method for measuring stresses in vibrating conductors in the laboratory under conditions closely simulating field conditions, and numerous schemes for effecting better stress distribution at conductor supports have been investigated. These tests have been facilitated by the development of special Machinery. Field work has been conducted on special outdoor spans, and observations have been made also upon operating lines at widely separated locations.

Interconnection of Primary Lightning Arrester Ground and the Grounded Neutral of the Secondary Main

Abstract: This investigation of the relative merits of various possible transformer connections and protective devices was the result of statistical data secured, over a long period of years, by the Commonwealth Edison Company of Chicago regarding the number and types of distribution transformer failures during lightning storms.* It refers to standard, 60 cycle, overhead, power and lighting distribution systems having 4,000-volt, four-wire primaries and three-wire, 230-volt, three-phase power and 115-230-volt lighting secondary mains. Five spans of standard pole line construction were exposed to transient potentials of steep wave front, the latter being induced by the sudden charging of the well insulated artificial clould suspended over the line. The cloud was charged to potentials, up to 150 kv., by means of a surge generator. Measurements of potentials, ranging as high as 40 kv. in some instances which were induced upon various portions of the distribution system were made by means of ionized sphere gaps and the cathode-ray oscillograph. Some of the important conclusions reached as the result of this investigation are: 1. The value of an experimental wood pole distribution line with an insulated artificial cloud charged by means of a surge generator was definitely established for lightning protective investigations involving induced as well as direct stroke potentials. 2. The practicability and economy of studying, by means of such laboratory equipment, the operation of various transformer, lightning arrester and ground connections, when exposed to surges approximating those of lightning, were demonstrated. 3.

Effect of Transient Voltages on Power Transformer Design-IV Transition of Lightning Waves from One Circuit to Another Through Transformers

Abstract: Three preceding papers of the same general title have dealt with transient voltage stresses developed within high-voltage transformer windings. The most important conclusions arrived at by the author and presented in these papers and the discussions are: 1. Transformers of all conventional constructions undergo oscillations when subjected to lightning or switching surges. 2. The amplitudes of these oscillations may be dangerously high, depending on the amplitude and the shape of the applied voltage. Their frequencies range from a few thousand to a few hundred thousand cycles per second.1,2 3. A lightning wave of a given shape produces very different stresses in different transformers.1,2 4. In practical design, neither amplitude nor frequency of these oscillations can be controlled by arrangement or proportions of windings.1,2 5. A lightning wave chopped by flashover of line insulation can produce stresses in transformer windings equal to or even in excess of those produced by a long wave of the same amplitude and front.1,2 6. Unless means are taken to obtain uniform distribution under all lightning conditions it is entirely possible to design transformers that will pass A.I.E.E. test which are inadequate for service conditions. This follows because test voltages in neither magnitude nor distribution of stresses correspond to that produced by transient voltages in transformers of ordinary construction. This difference is particularly great in transformers with solidly grounded neutrals, where the potential test allows the insulation from high-voltage winding to low-voltage winding and ground to be reduced as the neutral is approached.1,2 7.

Lightning Protection for Distribution Transformers

Abstract: Usual methods of connecting arresters for the protection of distribution transformers are often inadequate. In service some transformers still flash over although the arrester, in itself, is capable of protecting the transformer with a large margin of safety. Surge current flowing to ground through the ground lead of the arrester gives rise to inductive drop and resistance drop voltages. These voltages added to that permitted by the arrester can be great enough to flash over the transformer bushings to secondary neutral. By interconnecting the ground lead of the arrester and the transformer secondary neutral, the voltage across the transformer insulation is limited to that permitted by the arrester alone. Tests have been made to determine the conditions of surge required to duplicate flashovers that occur in practise. Under such conditions measurements were made of the voltages appearing on a representative secondary circuit when bushing flashovers are permitted, and when flashover is prevented by interconnecting the arrester ground and the transformer secondary neutral.

Predetermination of the A-C. Characteristics of Dielectrics

Abstract: From the charge and discharge currents of any type of dielectric under continuous potential it is possible to predict accurately the loss, power factor and capacity at 60 cycles. This is done by an empirical determination of the equation for the relaxation function of the dielectric at a given temperature, followed by the application of von Schweidler's method. The method developed is available at any frequency, provided that the continuous potential charge and discharge currents may be measured over initial time intervals comparable with the alternating period. It is shown that agreement of the relaxation function with the t-nexpression is not general, but on the contrary, it is confined to dielectrics showing negligible or no irreversible conduction. The inadequacy of the t-n expression is proved on experimental and analytical reasons. Chief among them is the failure of the t-n expression to predict accurately the variation of dielectric loss with frequency. A convenient and sufficient expression for the relaxation function is shown to be a sum of three exponentials. Further, experiment and analysis prove that the method of three exponentials predicts accurately the a-c. behavior of a dielectric at 60 cycles. The usual forms of irreversible conduction encountered are defined and classified. The case where the initial constant current does not obey Ohm's law has been considered analytically in its contribution to the a-c. behavior. The manner in which the complete a-c. behavior of a dielectric is accurately predicted from suitable d-c.

Extinction of A-C. Arcs in Turbulent Gases

Abstract: A recent theory of a-c. arc extinction in gas-blast circuit interrupters is reviewed. It is predicted on the basis of this theory that the interrupting ability of an a-c. arc should increase very considerably with the degree of turbulence to which the arc gases are subjected, and also that the recovery of dielectric strength by a turbulent arc space during a period of zero current should be very rapid for the first few microseconds and then comparatively slow for a considerable time thereafter. Experimental results obtained with an arc in atmospheres of various common gases verify these predictions, and also reveal the importance of the gas medium. A curve has also been obtained showing recovery of dielectric strength with time during the current zero period by an arc in an expulsion fuse tube. This curve has the characteristic shape predicted by the turbulence theory, thus proving the validity of the theory as applied to the behavior of a practical gas-blast device.

Transient Oscillations of Mutually Coupled Windings

Abstract: IN THE PAST, the mathematical analysis of transient oscillations in transformer windings has been based upon a single winding having self and mutual inductances between its turns, capacitances along the stack and to ground, and, in one instance, resistances representing the losses. Such a circuit ignores the presence of secondary circuits, but strangely enough proves adequate to describe the internal high frequency transients of the winding under consideration, not only qualitatively but quantitatively as well. Cathode ray oscillograms, however, do show some difference in the characteristics of the oscillations, depending upon the terminal connections of the secondary circuit.

Design of Capacitor Motors for Balanced Operation

Abstract: THE PRESENT methods of calculating the full load performance of capacitor motors are quite long and tedious. A simple convenient method of calculating full load conditions is given in this paper, including the necessary theory and formulas.

Measurement of Stray Load Loss in Polyphase Induction Motors

Abstract: FOR LACK of a satisfactory method of measurement, no provision is made for stray load losses in the A.I.E.E. standard efficiency tests for induction motors. These losses may amount to from 2 to 5 per cent in squirrel cage motors from 1 to 5 hp. in size, decreasing gradually with increase in motor rating until they are less than 1 per cent for motors of several hundred horsepower. Development of an easy method of measuring stray load losses would not only rectify this situation, but also would permit the losses themselves to be studied more intensively and perhaps ultimately to be reduced greatly.

Single-Phase Short-Circuit Torque of a Synchronous Machine

Abstract: This paper extends the analysis of synchronous machines previously published by Doherty and Nickle to include the calculation of torque due to single-phase short circuit. The torque is expressed as the sum of odd and even harmonic series which are simply related to those previously derived for the armature current. The effect of amortisseur windings is also taken into account. As an illustration of the application of the equations, the short-circuit torque is calculated for a 100,000-kva. generator. The mathematical analysis is given in Appendixes as follows: A. Short-circuit armature current. B. Short-circuit torque. C. Decrement factors. D. Single-phase and three-phase reactances.

Torque-Angle Characteristics of Synchronous Machines Following System Disturbances

Abstract: A method of predetermining the torque-angle characteristic of a synchronous machine following a system disturbance is of importance in calculating the transient stability limit and in evaluating the effect on this limit of additional rotor circuits or of a change in excitation. In the first part of this paper equations are derived which are of sufficient generality to evaluate such factors. Two types of system disturbances are considered: 1. Switching in or out of a connecting line, 2. Occurrence of a balanced three-phase system fault. In the second part, an actual case of switching out a connecting line is calculated and the results compared with field tests taken on the New York Power and Light Corporation's system. No attempt has been made to draw any general conclusions as to the effect of amortisseurs or changes in excitation. A subsequent paper will present results which have been obtained from the application of the method presented in this paper.

Recent Developments in Arc Rupturing Devices

Abstract: THE EXTINCTION of an a-c. arc by the deion grid method has been very successfully applied during the past two years to the design of an extensive line of arc rupturing devices for oil circuit breakers of all voltage classes. Refinements have been made in the theory of operation of deion grids and in the construction of several types. Field operation and special tests have given much information regarding these devices.

The Theory of Oil-Blast Circuit Breakers

Abstract: DIRECT experimental proof of the oil blast theory of arc interruption has been assembled. A circuit breaker has been constructed having oil driven between its electrodes at a measurable velocity, and tests show that a correlation exists between the rate of recovery voltage rise and oil velocity. This correlation corresponds to an oil dielectric strength of 55 kv. per 1/10 in. of oil. To determine whether or not this correlation is accidental, relations between voltage and oil velocity and between current and oil velocity have been studied, but correlation does not appear. Impulse tests also have been made on clean and carbonized oils.