of the large power systems of the country on the basis of excessively high voltages across the reactor has created an impression in some quarters that the reactors themselves were the source of the high voltages. This is an erroneous impression. When reactors are added to a system they simply add inductance, thereby increasing the total amount which the system already has by virtue of the generators, transformers, transmission lines and other equipment. Obviously, no voltage can be developed by the reactor itself. This being the case, the question arises as to whether the additional inductance can increase the possibility or the magnitude of voltages. developed from outside sources such as lightning and switching disturbances, arcing grounds or other phenomena. Disturbances of this character are almost entirely absent from systems which operate with the neutral grounded and which distribute underground. Even where there is some overhead distribution, disturbances originating on this section of the system are quickly damped out when they enter the cables on account of the dielectric loss in the cables. This loss, which may be small under normal conditions of voltage frequency and temperature, increases very rapidly when any of these are increased until at the voltages and frequencies encountered in traveling waves, this loss would consume the energy in a very short distance. Resonant conditions on a system do not necessarily mean excessively high voltages. Resonance in a single circuit requires an impressed voltage of the same frequency as the natural period of the circuit. No such voltage can exist in a power system except through the discharge of a condenser through the medium of an arcing ground or other similar disturbance. When this occurs, the voltage cannot be increased above that to which the condenser was initially charged, namely, the line voltage. To produce any increase in voltage, the energy stored in the condenser must be discharged into another of much smaller capacity located in a circuit having the same natural period as the first circuit and mutually coupled to it. The condition under which two circuits will have the same natural period is that the product of the inductance by the capacity in one circuit is equal to the same product for the other circuit, or L,C, =L.C2. With these requirements in mind, and a knowledge of the actual magnitudes of the circuit constants on modern power systems, one would have to stretch the imagination beyond all rational limits to predict resonance. Resistance, shunting the reactor, has been proposed as a means of "relieving all apparatus in the circuit from excessive voltages due to resonance and impulses by absorbing the energy in 12R loss." The first result of adding such a shunting resistor would be to change the natural frequency of the circuit. This would be just as liable to throw the circuit into resonance with some other circuit as to throw it out of resonance. The relief, intended to be provided by the shunt ing resistor, is already supplied to the system by the cables, inasmuch as the dielectric loss in the cables constitutes the most effective and efficient form of shunt that could be devised. At normal voltage and frequency the loss is low, but it increases tremendously with increase in either factor. Thus it would appear that the only excuse for shunting a reactor with resistance is to protect the reactor itself. This is directly opposed to the wellrecognized principle that any protective device should be able to take care of itself without the aid of auxiliary apparatus.. Even this might be permissible if the addition of the resistance had no effect on other apparatus. A concentrated inductance acts as an impassable barrier to all high frequency disturbances. Thus a reactor forms an absolute stop against high frequency getting past and into the expensive equipment centered in the generating station. Shunting the reactor offers a by-pass for a portion of the energy so that disaster to the more expensive and important equipment may follow, especially in view of the almost insignificant energy loss which can be permitted in the resistance shunt. For these reasons the Westinghouse Company has striven to increase the strength of their reactors beyond the possibility of failure without relying on any auxiliary apparatus to protect the reactor at the expense of other more important and more costly equipment. C By Charles Le G. Fortescue It has been proposed as a remedy for preventing the breakdown of current limiting reactors under operating conditions to shunt the reactor with a resistance preferably of the type having a lower resistivity to high frequency currents. Such a device will undoubtedly reduce the voltage stress to which the reactor is subjected under operation, but on the other hand it will correspondingly increase the stress on the apparatus which the reactor is intended to protect. It may be taken for granted that for slow, periodic variation in current, shunting the reactor has little effect on either its own internal stresses or those in some other part of the circuit. The possibility of high voltages due to arcing grounds need not be considered, as with grounded systems conditions for producing them are not favorable. The most frequent source of high voltage stress is due to the sudden interruption of an arc during short circuit. In this case a high frequency oscillation may be set up in the system. The initial stress under this condition may, however, be higher with the shunted reactor than with the plain reactor. It is quite easy to insulate reactance coils to withstand these conditions without having recourse to shunting resistances, and it seems to me undesirable to add a protective device to protect another protective device, especially when this is done at the expense of the original protection. All the surges that have to be handled by this type of apparatus have large energy content and therefore the beneficial effect of damping is very slight, as the damage may be done long before the stress has decreased to a small value and the reduction of the effective reactance by shunting increases the initial value of the stress on the rest of the apparatus. Report of Oil Circuit Breakers, Accessories and Switchboard Equipment Subcommittee This Subcommittee has confined its activities, during the past year, almost exclusively to carrying forward the work started three years ago, of endeavoring to place the method of rating oil circuit breakers on a basis which would better meet the needs of the operating companies. It was felt at that time that the present standard method of the manufacturers did not suffice for the requirements of operating practice, but no general authoritative data were available by which to substantiate the belief. It was also felt that in view of the difficulty of testing oil circuit breakers of high interrupting capacity, a considerable amount of information relative to their capabilities might be obtained, which would be beneficial both to operating companies and to the manufacturers, if the former would send to the Subcommittee descriptions of any operations of their breakers under abnormal conditions, whether successful or unsuccessful. With the desire for information on the above two phases of the subject, it was decided to send out a request for data. In the course of the preparation of the request, committees of the *A.I.E.E., and of the Electric Power Club, who were also desirous of the same data, joined with us, so that eventually the request became a joint one of the three organizations, and was sent out during August of last year, under the auspices of the N.E.L.A. The request was divided into two parts, the first dealing with operating practices, experiences and opinions, for immediate answer, and the second dealing with the results of specific cases of operation of oil circuit breakers under abnormal conditions, to be answered as the cases occurred. It was sent out to a list of fifty-four representative companies in the United States and Canada, selected by the three organizations compiling it. Part I Replies to Part I were received from forty-six companies on the list, and in addition nine other representative companies sent in replies, making a total of fifty-five. These replies were carefully analyzed with the following results: Ques. A. What is your practice with regard to the number of times an oil circuit The A.I.E.E. Subcommittee on Oil Circuit Breakers and Switches of the Protective Devices Committee summarized the replies to this request from the standpoint of design limitations, in a paper presented by Mr. H. R. Woodrow before the A.I.E.E. Midwinter Convention during February, 1921, and appearing in the March issue of the A.I.E.E. Journal. classes of service, please state for each, and reasons. The replies to the above question fall under three headings-distribution, high tension cable and high tension overhead transmission oil circuit breakers, practice differing very materially among them. In considering the replies, it should be noted that when a breaker is reported to be closed a certain number of times, it means that the breaker is closed that number of times after it has opened automatically, and that if it continues to open after each closing operation it is then left open for inspection. The number of times of opening is therefore one more than the number of times of closing reported in the replies. Oil Circuit Breakers for Distribution Circuits Twenty-eight clear replies were received. The practice of ten of the most representative companies is to close three times immediately. Ten close three times at intervals from a fraction of a second, when using automatic reclosing breakers, to one minute, and four companies close twice at intervals of from one to three minutes. The practice of the remain ing four companies differs from any of the fore going. Oil Circuit Breakers for High Tension Cable Lines Of seventeen clear replies, thirteen companies operating in the largest cities of the East and Middle West follow the practice of not reclosing a breaker after it has opened once, until tests and inspections have been made. Two of the remaining companies close once, with a two-minute interval, and the other two companies close twice with one and two-minute intervals, respectively. Oil Circuit Breakers for High Tension Overhead Transmission Lines The replies show lack of uniformity as to practice, the main agreements being as follows: ardized, and which is to open twice and then to be in condition to be closed and carry the rated current until it is practicable to inspect the breaker and make any adjustments which may be found necessary, does not, for distribution service, give as large a number of times of opening, or as short an interval between openings, as is required by operating practice, a point which should be taken into consideration when selecting breakers for such service. For high tension cables the manufacturers' duty cycle provides for one more opening and for a shorter interval than is usually required in practice. For high tension overhead transmission, the manufacturers' duty cycle is in general agreement as to the number of times a breaker is closed, but does not give a sufficiently short time interval to agree with operating practice. Ques. B. What would you consider satisfactory operation of an oil circuit breaker? For instance, would you consider that a breaker should be capable of performing a certain duty cycle and then be in condition to be closed for nonautomatic operation, or, if not, what would you consider satisfactory operation? This question was included for the purpose of getting the opinions of the operating companies as to what they would consider satisfactory operation of a breaker, as it was realized that opinions differed widely in that respect. Unfortunately the question was worded in such a way that one of the main points at issue, namely the condition of the breaker at the end of its duty cycle, escaped the attention of many of those making replies, and attempts were made to define what a suitable duty cycle should be, leaving out, however, in many cases, the class of service to which the duty cycle pertained. However, if practice is an indication of what is desired of an oil circuit breaker for different service conditions, the opinions are fairly well indicated in the replies to Question A. Some replies specified a number of openings with certain intervals between and that the breaker should then be in condition to be closed for automatic operation, stating that they would not care to reclose if the breaker were good for non-automatic operation only. It should be borne in mind, however, that to say that a breaker is in condition for one more automatic operation, carries with it the obligation of describing what the condition of the breaker will be after it has opened. This condition may be anything from destruction to capability for one or more further automatic operations, or the intermediate condition adopted by the manufacturers. There must be a stopping place somewhere, and that adopted by the manufacturers, which is to "be in condition to be closed and carry the rated current until it is practicable to inspect the breaker and make any adjustments which may be found necessary," seems reasonable, except that it should go further and say whether the breaker will be capable of breaking normal current. The latter is the intent of the present manufacturers' definition. It would seem, however, that there might be some question as to whether the former might not be more acceptable to the operating companies in view of the fact that if the breaker were incapable of interrupting normal current, a shutdown of an entire station might be necessary to take the breaker out of service. In order to obtain the consensus of opinion of the operating companies, a special question is being sent to those companies who replied to our request for oil circuit breaker data, the results of which question we hope to present at the annual convention. Those companies which desire that after a certain duty cycle a breaker shall be in condition for automatic operation can obtain such service by specifying a breaker which will be good for one more opening than they wish it to perform. Ques. C. Would you prefer to have an oil circuit breaker rated at its ultimate interrupting capacity, or at some percentage of that value? If the latter, would you consider 50% reasonable, which corresponds to a factor of safety of two? Thirty-one companies favored an ultimate rating, eighteen favored a factor of safety of two, and five companies thought the question immaterial. One member writes in part that the ultimate rating should include a factor of safety, determined by the manufacturers and necessary to them in order that they may guarantee the rating, this factor of safety to cover inequalities in material, manufacturing processes, etc., but not that which a purchaser might desire to take into account due to operating changes, poor maintenance and conditions entirely foreign to the manufacturer. Conclusions Regarding Question C The ultimate rating is the present standard of the manufacturers, and it would seem that their choice has been confirmed by the operating companies. It is to be hoped that the manufacturers will clearly bring out this method of rating in their publications and in their advice to their customers for the benefit of the smaller companies who may not have an engineering staff to give consideration to the factors involved in the selection of an oil circuit breaker for a given set of conditions. Ques. D. What is your practice as to the frequency of oil circuit breaker inspections, and how do you make them? The replies indicate the frequency of inspections to be as follows: oil circuit breaker at once if it shows signs of severe stress, and a number of companies make more frequent inspections of breakers used for hard service than for the remainder of their breakers. The nature of the periodic inspections is generally to remove the oil, test it and filter it if necessary. to examine, adjust and repair all contacts and operating mechanisms, and to examine the condition of bushings, tanks, etc. Ques. E. Have you had fires on oil circuit breakers or caused by oil circuit breakers in operation or placed in operation during the last two years? If so, please describe fully, giving the type and size of breaker and the year purchased. The replies received to this question are not in all cases of a sufficiently definite nature to draw the proper conclusions. Some fires have been reported, but apparently there were few cases where any real damage occurred to the surrounding objects. The experience indicates, however, that care should be exercised in securing circuit breakers of a proper rating for the system on which they are to be used, remembering that these systems grow rapidly. The installation of the switches should be made so that the hazard either to operators or from possibility of fire will be removed as far as possible. Ques. F. Have you observed after heavy short circuits which have occurred during the last two years, any evidence that the contacts of the oil circuit breakers have been forced apart due to the mechanical strains set up by magnetic fields, or that the breaker showed any signs of distress due to the thermal effects of heavy current? If so, please describe the circuit breaker fully, giving type, size and year purchased, describing also the effects observed, and giving best information as to the amount of current which caused the trouble. Only six companies reported such troubles, and forty-two replied that they had not observed any. The answers, however, may not give a true indication of the frequency with which such troubles have occurred, inasmuch as the cause of the trouble is often obliterated at the time an oil circuit breaker opens. It is quite possible also that breakers which have given trouble may not have been inspected with the above subject in mind. The results of the experiments conducted by the New York Edison Co. and described by Mr. Philip Torchio in his paper appearing in the February Journal of the A. I. E. E., show very conclusively that such troubles are quite possible on those types of construction where the tendency of the magnetic field is to force the contacts apart instead of together, and as a consequence, the manufacturers have modified their designs in some instances to guard against such trouble. Ques. G. Have you experienced any trouble dur ing the past two years on oil circuit breakers due to heating under normal operating conditions? If so, please describe the breaker, the manner of installation, and the trouble. The replies indicate that there has been little trouble on modern breakers. Some heating trouble has been experienced when the breakers have been installed in poorly ventilated compartments, but an improvement in ventilation has remedied the trouble. Overheating has occurred in several instances, however, on subcell disconnecting switches. Ques. H. Is it your practice to use a factor of safety in selecting an oil circuit breaker to provide against a decrease of current carrying capacity with age? If so, what factor do you use? Most companies do not consider that there is any permanent decrease of current carrying capacity with age, as the contacts can be readily replaced, and therefore do not provide for it. It is general practice, however, to purchase a breaker larger than immediately needed to provide for growth and any temporary decrease of current carrying capacity. Practice ranges from 25 per cent oversize to 100 per cent. Frequently the size next larger than actually required is taken. Ques. I. Have you experienced any trouble due to potential stresses produced under normal operation or during surges, or any trouble from static sparks? If so, please describe breaker and experience. There has been comparatively little trouble from the above source, although there have been several cases where static sparks have ignited the explosive vapor over the oil and have blown up the circuit breaker. Several cases of corona have been reported between the terminals of high voltage breakers and the inside of bushing type current transformers around them, which have resulted in disintegrating the insulation on the inside of the transformers. There have also been a number of failures of bushings under surges and some failures of insulation in the tanks. Ques. J. Have you any evidence as to the effect of low power factor on the interrupting capacity of oil circuit breakers? (By power factor being meant the ratio of resistance to impedance of the entire circuit which determines the amount of short circuit current.) If so, please describe. There is practically no experience reported except in the case of some tests conducted by the Pennsylvania Water & Power Company, in which a breaker interrupted a short circuit of a certain magnitude at 88 per cent p.f. and at 16 per cent p.f. with very little, if any, difference. However, from old test data of another company, it would appear that low power factor does impose a heavier duty upon a breaker, as has been the generally accepted belief. Ques. K. Describe any recent tests you may have made on oil circuit breakers in connection with interrupting capacity, their object, the kind of equipment used for the test, and give any suggestions as to methods or equipment which might be used to advantage on future tests. No recent tests are reported except those made by the New York Edison Company during March, 1919, some of which are described in Mr. Torchio's paper previously mentioned. The Pennsylvania Water & Power Co., in conjunction with the Consolidated Gas, Electric Light & Power Company of Baltimore and several manufacturing companies, is conducting a series of tests, the results of which, it is hoped, will be available to the interested committees and will throw considerable light on the subject at hand. Ques. L. Have you changed the speed of mechanical operation of any of your oil circuit breakers, giving a faster parting of contacts? If so, with what results? Nine companies report having speeded up the travel of several makes and sizes of breakers, very materially improving their operation. These improvements were in the nature of prolonging the life of contacts, reducing oil throwing, and reducing distortion of tanks. The increases in speed have been given in two cases as 100 per cent. No data are given in the remaining cases. It was stated by the representatives of the Power Club present at the January meeting of the Electrical Apparatus Committee that the benefits to be derived from such speeding up would generally apply only to old breakers or to those on which the normal operation had been slowed down by long operating rods, many bell cranks or stiff operating mechanisms. The representatives claimed that modern breakers are so designed and constructed that little increase of interrupting capacity is to be expected by an increase in the speed of contact travel. Ques. M. Have you had any experience which throws light upon the question as to whether it is better to confine the gases in an oil circuit breaker tank upon opening under short circuit, or whether it is better to vent the gases to the atmosphere under such conditions? (a) When the breaker opened once; (b) when the breaker opened more than once and the interval between. The recommendations received, probably based on a limited experience, were as follows: 24-Vent 2-Confine The main points brought out are that the accumulation of gas above the oil renders the second opening a hazardous one if it occurs soon after the first and that, therefore, it is well to vent the tanks. The difficulty in venting is to provide free passage for the gases, but to prevent the oil from escaping. In the tests being conducted by the Pennsylvania Water & Power Company, a ventilating scheme designed to carry off part of the gas directly from the arc zone to the outside was tried with apparently fair results. Two large companies report the desirability of preventing the gases given off by oil circuit breakers from accumulating in their compartments and also the desirability of eliminating or covering up any auxiliary switches, the arcing of which might ignite the explosive mixture. One of the companies reported having suffered from such an occurrence, and has since installed vents to the outside of the compartments. The other company uses fans which operate automatically on each opening of the breaker and can also be operated independently by the operator. Ques. N. There is at present no standardization The replies are as follows: In all cases where oil is interchanged, or one va riety of oil used in breakers of all manufacturers, no trouble is reported from the substitution. See also answer to Question P. The reasons given in all cases are to reduce the kinds and amount of oil carried in stock, for convenience, and because when emergencies arise frequently only one kind of oil is available. standardization of oil among the leading manufacThe Subcommittee is endeavoring to effect a turers so that the present disadvantages may be eliminated. Ques. O. What is your practice regarding periodically filtering or changing or testing the oil in circuit breakers? The replies as to the intervals of periodic inspection were as follows: 1-1 month 5-3 months 1-4 months 9-6 months 16-yearly 1-2 years 18-Have no routine inspection. It is the general practice to test the oil, and filter or change it if necessary when the breakers are regularly inspected, or when they have opened short circuits and it is suspected that the oil has been damaged. A number of companies make a practice of testing the oil at more frequent intervals than those at which the periodic inspections are made, and then filtering or changing if found necessary. Ques. P. What has been your experience with regard to the operation of outdoor oil circuit breakers in cold climates? Please describe the breakers in question, the range of temperature under |