The protection of persons and property against electrical dangers has in recent years become an important problem. In the beginning of the art of distributing electrical energy to consumers voltages were low, direct current was used, and in the case of each system the points of consumption were confined to a comparatively restricted territory. Then the dangers to persons from electrical circuits were practically negligible, but the dangers to property were not because insulation and methods of construction were undeveloped. At the present time, however, far different conditions exist. Electrical systems have been enormously extended. The development of alternating current machinery and methods of transmitting electrical energy over long distances has been chiefly instrumental in bringing about the new conditions. With the increase in the voltages at which electrical systems are operated, dangers to persons and property havemultiplied. It is true that improvements in methods of construction and quality of insulation have kept pace with the growth of electrical systems, or vice versa, but at the same time many dangers arise from electrical circuits which require special means and devices to render them harmless. One of the most important sources of danger is that due to the close proximity between highvoltage and low-voltage circuits. The actual danger in this case arises from the entrance of voltage and current from the high-voltage circuit upon the low-voltage circuit through faults in insulation, contacts between wires and in other ways. The great importance of this source of danger is due to the fact that large numbers of people arc continually coming in contact with apparatus and appliances connected to these low-voltage circuits. It is the purpose of this paper to discuss briefly the principal causes of dangers just mentioned, the desirability of connecting lowvoltage secondary circuits to water pipes to avert these dangers, and also the effect on the pipe systems of such earthings.

Dangers from Electrical Systems.

One of the most common forms of lowvoltage circuits are those fed from highvoltage alternating current circuits through step-down transformers. In this case the danger arises from the entrance of current and voltage from the high-voltage circuit either through leaks in the transformer insulation or contacts between wires. Take a transformer connected to a 2,200 volt distributing system which transforms energy for lighting a group of dwelling houses by means of a 3-wire 110-220 volt circuit. If during a storm a high-voltage line wire breaks and falls on a low-voltage wire, or any one of the possible accidents occur which may have a similar result, a condition may arise such as that illustrated in Fig. 1. That is, one side of the low-voltage circuit is, in effect, connected by a wire to one side of high-voltage circuit. If both circuits are insulated from ground at all points, an electrostatic voltmeter. or some other voltmeter which takes but slight current to operate it will show at V a difference of potential approaching 1,100 volts between low-voltage circuit and ground, the actual voltage indicated varying with a number of factors. It is not likely, however, that either circuit will be thoroughly insulated at all points. There may be faults in the insulation, and if there are the voltmeter may read as high as 2,200 volts. This would be the case if the opposite wire of the high-voltage line were down and touching ground somewhere, or were even in contact with a tree. In either case if the low-voltage circuit were in any way accidentally grounded, the flow of current to ground might be considerable. In the first case the current flow that might occur would be due to the electrostatic capacity between the high voltage line and ground, the line acting as one plate of a condenser and the earth as the other. The alternating voltage of the line itself would tend to charge and discharge this condenser through the fault “A” in Fig. 1, in series with whatever accidental ground happened to be on the lowvoltage circuit. This might be the body of a person in contact with lighting fixtures. The value of this condenser current can be shown to be about 10 milliamperes per mile of line. In the second case the current would depend upon the sum of the resistance between low-voltage circuit and ground and between the opposite wire of the high-voltage circuit and ground. Similar conditions may arise with any low-voltage circuit if it comes in contact with a high-voltage circuit.

Abstract of paper read at Convention of Central States Section of American Water Works Association, at Columbus, O., 1915.

Protection Against the Miximum Degree of Danger.

It is obvious that when there do exist such high voltages between low-voltage circuits and ground, there is great danger to persons handling lamp sockets and other appliances; moreover, there is an extreme liklihood of fire following a failure of insulation of the lowvoltage circuit because of current flowing from the high-voltage circuit to ground and heating inflammable substances. The condition described, however, represents the maximum degree of danger which can arise, and the one which is likely to be met with the most infrequently. From the dangers just described a great deal of protection can be obtained by connecting the low-voltage circuit to earth as in Fig. 2. For the best degree of protection it is necessary that the resistance to flow of current from the lowvoltage circuit through the earth wire “B” and earth connection “F” be very low. That is, a current must easily pass from the lowvoltage circuit into the earth in case of contact with a high-voltage line. If this is the case, and the resistance to flow of current is low enough, a dangerous voltage cannot exist between low-voltage circuit and ground. For example, instead of the voltmeter at “V”, Fig., 2, reading between 1,100 and 2,200 volts, as it did under the condition shown in Fig. 1, it will now read a value which may be found by multiplying the current flow in amperes in wire “B” by the resistance in ohms to flow of current into the earth at “E”. If the resistance at “E” is low the voltage measured at “V” will be low even though a large current is flowing. The lower this voltage is the better, and for a high degree of safety it should in no case exceed 150 volts. The current which will flow in wire “B” depends upon the current carrying capacity of the protective devices on the line from which the current comes. This may be small or large, and the maximum allowable resistance in the earth connection at “E” will depend upon the maximum value of this current. The maximum current which a line will carry permanently is limited by the rated current of the fuses or automatic circuit breakers which protect it from over load. Hence, the maximum allowable resistance in the earth connection can be expressed by the following formula: R=150/I A, where “R” is the resistance in the earth connection, 150 is the allowable voltage between low-voltage circuit and ground, and I A is the current required to operate the fuses or circuit breakers which open the circuit. If this current is 100 amperes, “R” will be 1.5 ohm; 200 amperes, 0.75 ohm; and so on. One hundred-fifty volts, however, is the maximum allowable voltage rise between low-voltage circuit and ground.


Earth Connections.

The next question is that of obtaining a low value of resistance for the earth connection “E”. There are several methods by means of which electric circuits.can be earthed; namely, by utilizing water pipe systems, driven pipes or buried plates. Buried strips of metal of any desired length may also be used under special conditions, and there arc on the market many patented devices for making earth connections. In this list, however, water pipe systems easily come first in point of desirability. In the first place, on account of their great extent, they offer but little resistance to flow of current away from them into the earth, the resistance of water pipe connections being found to be but a fraction of an ohm in most cases. Wherfe these low resistances are found the pipe joints for a considerable distance from the earth connection must give as good metallic contact as will ordinarily be found with lead or screw joints. In the second place, water pipes are easily accessible at service pipes or at other places, and, in the third place, it can be shown that with certain precautions all possibility of damage to the pipes, or injury to employees of the water company can be easily avoided. It may be stated as a well known fact that if pipes are driven in the earth at a distance from each other and connected together by a wire the combined resistance to flow of current away from the several pipes into the earth will be much less than that of a single pipe. If the resistance of a single driven pipe earth connection in a certain locality were as low as 15 ohms, to obtain a resistance comparable with that of a water pipe system in the same place would therefore require 50 or 60 pipes driven in the earth at 10 to 20 feet distant from each other and electrically connected together. This is plainly impracticable. It must be granted, of course, that by the use of driven pipes or buried plates, a certain degree of protection can be obtained; in the case of small transformers and lines of limited capacity, even an ample degree of protection, but as the kilowatt capacity of the lines and transformers increases the resistances required of the earth connection must decrease. The most obvious solution of the problem is to use water pipe systems which, in most cities, cover approximately the same areas covered by the electrical systems. There are, of course, localities not reached by water systems and in such places it is necessary to resort to driven pipes and other means of making connection to earth.

Effect on Water Pipes of Using Them as Earth Connections.

It has been shown above that it is very desirable to connect low-voltage secondaries to water pipes. It must also be shown, however, that the use of the water pipes for making earth connections is not in any appreciable degree a disadvantage to the pipe-owning company. In the past it has been stated that trouble for the company would arise from three sources: (1). electrolysis by stray currents from the earthed circuits. (2), danger to employees of the water company while working on service or other pipes to which low-voltage circuits were earthed; and (3) complications from allowing a second public service company the use of the pipes. The subject of electrolysis in this case may be considered under two heads: (a) electrolysis by alternating current, and (b) electrolysis by direct current. (a). Electrolysis by alternating current—Referring to Fig. 3, “A” represents an extensive low-voltage alternating current circuit which is earthed at a number of points to service pipes, these service pipes being all connected to the same water system. It is readily apparent that under normal conditions of operation, or with insulation everywhere in good condition, the opportunity for current flow in the pipes is very slight. With unbalanced load on the system current would flow in the middle wire, setting up differences of potential between the points “E”, “F” and “G”. This would cause current to flow in the pipes, but the voltages between “E”, “F” and “G” being necessarily small, the currents would be small, since there is more or less resistance in the pipes. On the other hand, there can be no interchange of current between circuits “A” and “B”. In order for this to occur, the flow would have to take place from “A” through earth wires “A”, “B” and “C” in parallel, over the water pipe to wire “D” and circuit “B” and back to “A” on the high-voltage line. Two layers of transformer insulation prevent this, so such an interchange of currents is impossible unless there are faults in the insulation of both transformers. If this should occur, fuses 1, 2. 3, or 4, would be likely to go out and isolate the low-voltage circuit. Or. if the fuses did not go out, the current would be limited to their rated capacity. With a low-voltage circuit connected to the pipes at a single point or as in “B”, Fig. 3, there is obviosuly no opportunity for any but the very slightest flow of current into the pipes over wire “d”, because there is no place for it to go. The slight current that would flow would be due to electrostatic capacity effects and would hardly be measurable by any ordinary means. With a condition existing such as shown in Fig. 1, however, considerable current might flow for a greater or less time. If the leak should occur to circiut “A”, for example, the current would pass from one side of the highvoltage line through the fault to circiut “A” and to the pipes over earth wires “a”, “b” and “c”. whence it would flow through the earth back to the opposite side of the high-voltage line. Nevertheless, granting that under normal conditions of operation slight alternating currents may flow, and that in the event of failure of insulation large currents may flow, it has been shown by experiments recently conducted at the Bureau of Standards, that the damage by electrolysis from such currents is practically neglible. These experiments, made in both iron and lead, in moist clay, show that with alternating current at commercial frequencies the amount of metal dissolved is but a fraction of one per cent, of that dissolved by direct current, the quantities of electricity passed being the same. From the stray alternating currents which may be found in pipes from earthing low-voltage alternating current circuits to them no perceptible damage through electrolysis need be feared. The only possibility of damage is that due to heating if the current flow is heavy, but the likelihood of such currents flowing is so small as to be of no importance. Unless the pipes were very small the currents required to damage them would be thousands of amperes, values beyond the range of possibility in any ordinary case. (b). Electrolysis by direct current may arise, or be increased in two ways. In the first place, where the middle wire of a low-voltage direct current system is earthed to a pipe system at a number of points, the pipe forms a conductor in parallel with the middle wire, in which case when current flows in the middle wire current will also flow in the pipes. If the resistance of the pipes is comparable in magnitude with that of the middle wire these currents will also be comparable, being shared in inverse proportion to the resistances. This is the same condition as was shown in the previous paragraphs to exist in the case of low-voltage alternating current circuits, but with the alternating currents the damage is negligible, whereas in the case of the direct currents, the damage may be serious, because of the marked corrosive effects of direct current. Earthing direct current systems to water pipes at a number of points is therefore undesirable, and such earthing should be confined to but one point on the system, and that at the power house. Earthing the middle wire at more than one point is unnecessary, and even when earth connections arc made, say, with one to the water system at the power house and others to driven pipes or plates, current flow may be considerable and lead to undesirable results. It is to be recommended, therefore, that the middle wire of 3-wire direct current systems be earthed at the power house only, and all other earth connections to the middle wire omitted. An exception to this can be made in the case of connections to service boxes because in such cases the resistance to ground will in general be high. There can, however, be no serious objections to earthing to pipes, machine frames, conduit, or other metallic objects near or enclosing direct current circuits, and it is desirable to earth them to promote personal safety and safety to property. In this case current would flow only while faults existed in the insulation and since this flow would be temporary, lasting only until warning could be given of the fault and repairs be made, the resulting damage would be negligible. In the second place, electrolysis may be increased or transferred from one locality to another, if a low-voltage circuit of any kind is earthed at different points to two metallicly separate pipe systems. For instance in cities where electric railways with ground returns are in operation, it may be found that one pipe system is at a considerable difference of potential with respect to another. If a low-voltage circuit is earthed to both of them, current will flow from one pipe system to the other, with damage of overheating the wires and also of increasing electrolysis on the other system. Moreover, if a low-voltage circuit is earthed at one point to a pipe system that is at a potential difference of several volts against earth, and at another point to the steel-work of a building,, the steel-work of the building takes the potential against ground of the pipe, and danger of electrolysis in the building ensues. On the other hand, if different low-voltage circuits fed from the same high-voltage distribution circuits are earthed to different pipe systems no current can flow from one pipe system to the other, because, as shown in Fig. 4, the insulation between transformer windings prevents the flow of current, and current flow would necessarliy have to take place over the high-voltage line in order to get from one pipe system to the other. This might occur, of course, if there were faults in the insulation of both transformers, but this is a contingency that can be neglected. Hence this general rule can be laid down. In every case where it is necessary and allowable to earth a lowvoltage circuit at more than one point, the earthing should be done to the same pipe system, and in no case should separate pipe systems be used.

FIG. 2FIG. 3

Insulating Joints.

In connecting a low-voltage circuit to a pipe system at more than one point, it is important that there be no insulating joints in the pipes that may come between the points of connection. If insulating joints are present the low-voltage circuit will act as a shunt to the joint and heavy current flow over the wires may follow if stray currents from street railways are on the pipes. This presents a dangerous condition because of the possibility of fire, and in the case of direct current systems, of increased electrolysis, and should be avoided. Moreover, if there are insulating joints near a point where an earth connection is made to a service pipe, the joints may so restrict the useful pipe surface in contact with the soil so as to cause the resistance to current flow from the low-voltage circuit into the earth in case of an accident to insulation to be so great as seriously to impair the usefulness of the earth connection. For this to be the case it is, of course, necessary to have insulating joints on both sides of the point where the earth connection is made. With an insulating joint on one side of an earth connection and lead or screw joints on the other for several hundred feet, the earth connection may be considered good. With insulating joints nearby on both sides, however, as would be the case with a cement joint pipe line, the earth connection would not be very effective. Therefore, in making earth connections using water pipe it is very important to know the character of the joints in the pipe.

Picking Up Stray Currents by Low-Voltage Circuits.

If a low-voltage circuit is earthed to a lead jointed pipe system at several points, and’ there is current flow in the pipe from electric railway circuits, the current picked up by the low-voltage circuit is not likely to be of importance.

Danger to Employees of the Water System.

The question of danger to employees of the water company has been many times discussed. In general, it may be stated that danger is likely to arise in only one case and that is illustrated in Fig. 5. This shows a low-voltage circuit earthed at one point to a service pipe, a failure of transformer insulation having occurred at “B”. If the service pipe is disconnected at “A”, the person making the disconnection is liable to an electric shock. The severity of the shock depends upon a number of factors which have previously been discussed, and it has been shown that conditions may easily arise under which the severity of the shock would be great enough to cause death. Therefore, with low-voltage circuits earthed to service pipes at a single point precautions are necessary to prevent such accidents to employees of the water company when working on them. This can be readily provided for by requiring the electric company to disconnect earth wires for service pipes when work is to be done on the pipes and reconnect them when the work is finished. This is a reasonable requirement, is sufficient, and has been in force in several places for years with satisfaction to both parties. With multiple earth connections to water pipes, however, as shown in Fig. 3, even the precaution mentioned above is hardly necessary. Unless all of the service pipes to which earth wires have been attached are disconnected there is little danger. A person may work on one of the service pipes in perfect safety as far as electric shock is concerned if the other service pipes are undisturbed. Moreover, it is an advantage to the consumers using the circuit to have more than one connection to the water pipe, for if one of the wires becomes disconnected the others still maintain the circuit in a condition of safety. Multiple connection of low-voltage, alternating current circuits to water pipes is therefore to be recommended and encouraged. It should be emphasized again at this point, however, that these multiple connections should be confined to a single water system, for reasons previosuly set forth in this paper. Moreover, when multiple earth connection to water pipes are made, the resistance of the pipe line between the two farthest removed points of connection should be measured in order to make certain that no insulating joints are present. If there are no insulating joints the resistance will in general be found to be but a small fraction of an ohm. If there are insulating joints it may be as much as an ohm or even several ohms.


Mechanical Construction of Water Pipe Earth Connections.

Where connection is made to a cast iron pipe line with bell and spigot joints the most satisfactory method is probably that prescribed by the National Fire Protection Association. Thi* connection is made by drilling a hole in the bell, tapping it and screwing in a brass plug to which the earth wire is soldered. This joint can be marie with a reasonable amount of labor and will be permanent, especially if the surface of the plug and pipe in the immediate vicinity is heavily coated with pitch or something else to prevent corrosion. Where connection is made to service pipes which can be drained of water while work is being done on them the earth wire can be wrapped several times around the pipe and soldered to it. If the service pipe cannot be drained conveniently the next best thing is a clamp to go around the pipe, with a lug to which the earth wire can be soldered. There are a number of amalgams on the market with which the pipe can be treated before putting on the clamp, making a better electrical connection between pipe and clamp than would otherwise be attainable. Many of the clamps sold for this purpose, however, seem of rather flimsy construction. To obtain good results a clamp of sturdy construction is necessary. Otherwise it is difficult to Ret as good a contact with the pipe as is in most cases desirable. The point at which connections to service pipes are made must be considered. This will in most cases be determined by local conditions. In general, however, it may be stated that this point must be chosen so that there is the least likelihood of the pipe being disconnected between the earth connection and the water main.

Connecting to Gas Pipes.

Connection directly to gas pipes should in every case be prohibited. On the other hand, the connection of low-voltage alternating current secondaries to water pipes should be made compulsory.


1. Many of the dangers which arise from electrical circuits may be eliminated by earthing these circuits. 2. Of the means of earthing electrical circuits water pipes are by far the most desirable. 3. There is no danger of electrolysis by such stray alternating currents as may result from earthing low-voltage alternating current circuits. 4. Direct current circuits should be earthed at but one point, preferably at the station. 5. Where conditions permit two or more ground connections should be made on each secondary system. 6. In case of any individual lowvoltage circuit in which a number of earth connections to pipes are made the earthing should be confined to one pipe system. Measurements of the resistance of the pipe line between the farthest removed points of connection should be made to make certain there arc no insulating joints in the part of the pipe line which may cause the electric circuit to carry heavy stray currents from railways. 7. In making cither single or multiple earth connections to water systems unless it is positively known that there are no insulating joints nearby the adequacy of the earth connection should be determined by measurement of its resistance. Earth connections should not be made to cement joint lines, or other lines in which there are many joints of high resistance. 8. Connection to pipe lines may be made by screwing a brass plug into the bell in the case of cast iron pipes or by clamps or soldering to service pipes. Soldering is preferable to clamps where it is possible to do it. 9. Earth connections to scrvice pipes should be made at the point where there is the least likelihood of the pipe being disconnected between the connection and the main. If the water meter comes between connection and main a jumper of the same size as the ground wire should be put around it. 10. Earth connections to gas pipes should be prohibited chiefly on account of the chance of explosions or fires, and also because the almost invariable presence of water pipes makes such connections unnecessary. 11. Because of the great advantage to the public and the slight disadvantage, if any, to the public service corporations resulting from earthing to water pipes such earthing of secondary systems should be made compulsory.


No posts to display