Modernizing Electrical Equipment in Water Plants
Connection Between Source of Power and Motors — Emergency Equipment—Elements of Cost—Motors—Simplicity and Flexibility
THE interest manifested by superintendents in the use of electricity as a motive power in water works plants, especially since the increase in the cost of coal as a fuel, makes the following article particularly timely and the facts as given will, no doubt, prove very instructive.
A water plant should be designed to give the most reliable and adequate service at the lowest possible total cost to operate and maintain. Stated, in other words, the most important factors in determining the equipment of a water plant are: 1, reliability; 2, adequacy; 3, economy; 4, simplicity of operation; 5, flexibility of location; 6, ease of control.
I shall endeavor to keep these factors in mind in discussing the application of electrical machinery in the water plant.
Adequate Lines Between Source of Power and Motors
There has been a rapid increase in number of water plants driven by electric motors, but in spite of this we find many plants which hesitate to change to motor drive because they fear that the service may prove unreliable. Electrical equipment as now made and installed has been developed to a point of very high reliability, so the continuous operation of the motor drive depends largely on the power source and the line. Since the source of electric power supply is different for each pumping plant, it is rather difficult to make a general statement regarding the reliability of the supply. All modern electric power plants are provided with reserve equipment so that there is practically no condition which may cause a complete shut-down of the plant for an appreciable time. It remains therefore to insure adequate lines between the source of power and the motors driving the pumps. Again local conditions determine what are adequate lines. If the pumping plant in question is one of several supplying the same system, and an interruption is not of great moment, a single line may be adequate; on the other hand when an interruption cannot be afforded, the expense of duplicate lines would be warranted. Whether one line or more should be used, the motors should always be properly protected from lightning, overload, and any other abnormal condition. It is possible to install separate lines preferably from independent feeders and provide the pumping station with relays, which, in case of trouble, will automatically switch the motor from one line and connect it with another. Of course this plan involves additional expense, and in many cases will not be warranted, but it indicates what can be done if necessary. Lightning arrestors should always be provided for the line, and the motor be equipped with relays which will disconnect it before it burns out from overload, or in case the voltage drops too low. Not only should these devices be provided but a check should be made from time to time to insure that the devices are in a good condition to function. If the proper size fuse is used, or if the circuit breaker is properly installed and set, it should function only when an abnormal condition exists. If the circuit breaker opens several times, there is a tendency for the operator to blame the circuit breaker instead of taking its warning to indicate that the motor, or some other part of the system, is not functioning normally.
Steam, Oil or Gasoline Emergency Equipment
Where it is not feasible to install complete duplicate lines, or such a plan is not desirable for some reason, it is possible to insure against interruption of pumping service by installing emergency equipment independently operated by steam, oil or gasoline. Quite often this takes the form of a prime mover arranged to be coupled to the same pump which is driven by the motor, in case of interruption of power to the motor. Since this prime mover is for emergency use only, and will operate only occasionally, it need not be of an efficient or even highly efficient type. After all the best measure of efficiency of a pumping plant is the average cost of pumping a thousand gallons including not only the operating costs but the overhead as well.
Even with the stand-by equipment it will be found that the fixed charges on this auxiliary equipment when added to the fixed charges on the electrical equipment and the operating charges will frequently produce a total cost for electric pumping below what can be obtained by any other method.
Owing to the lower efficiency of the centrifugal pump, the combined overall efficiency of the motor driven centrifugal pumps of the larger sizes is usually somewhat less than that of the triple expansion steam driven pumping unit, but this is counterbalanced by the decreased floor space, and the greater cost of the steam unit with correspondingly greater interest, depreciation and maintenance costs.
Elements of Cost of Electric Power
In discussing the costs of electricity for pumping, no figures can be given as there arc so many local factors to be considered. A brief analysis of the elements entering into the costs of producing and distributing electric power may not be amiss and may help us to realize some of the things which affect these costs:
Electric power costs are made of three principal elements as follows: (a) The consumer cost which includes the cost of having the customer on the books, reading his meter, collecting the bill, etc., all of which is nearly independent of the load or amount of energy used by the consumer.
(b) A demand cost which is determined by the maximum demand which the consumer can put on the system at one time. There is no satisfactory method of producing electric power during times of light demand and storing it for use during the peak; it must be produced when and as needed, consequently if consumer A demands 50 KW. of power when the load is already 200 KW., the power plant must have enough capacity to furnish a total of 250 KW. Since consumer A has by his demand, necessitated an additional capacity in the plant of 50 KW, the interest and depreciation on this additional capacity is a legitimate part of the cost of his load. This cost is called the demand cost, and has an important bearing on the cost of power for industrial use. It includes a considerable part of the salaries of general office force and general office expense, taxes, insurance, etc.
Consumer’s Load a Factor
Another factor which has a bearing on this cost is the power factor of the consumer’s load. This matter of power factor is little understood except by those having technical training in electrical engineering. I will not undertake a technical discussion of it, but will endeavor to give you some conception of its effect. With direct currents, we have no such a thing as a power factor, and the power is the product of the current by the voltage. With alternating currents the voltage and current must be multiplied by a factor not greater than unity which is called the power factor. This means that with a low power factor the current for a given power is larger than is required for the same power where the power factor is high. Some motors operate at a lower power factor than others, which means that for a given power consumption they require more current. Since no additional power is produced by this current it is called a wattless current. Unfortunately, the rating of generators, transformers and transmission lines, is governed more by current than by power, and greater capacity is required because of this wattless component of current where the power factor is low. An analogy may be found in the case of a pump in which part of the water from the discharge is by-passed and allowed to return to the suction. This water is doing no useful work, and yet it is using part of the capacity of the pump. Unfortunately, the elimination of the wattless current in an electric circuit is not so simple as closing the bypass of the pump for the low power factor is an inherent characteristic of certain types of motors. The effect is illustrated by the fact that a 100 horsepower motor operating at 60 per cent, power factor will use up as much of the capacity of the generator, lines and transformers as a 166 horsepower load at unity power factor. The demand cost of the low power factor motor therefore is greater than for a motor putting out the same power at a higher power factor.
There is no standard method of carrying the higher cost of low power factor power into the rate, but in some cases it is included directly, and in others indirectly. Since it does involve a definite increase in cost, it is reasonable to include it in the rate. A high power factor motor is always desirable from the standpoint of the power company. In my discussion of motors suitable for operating water plants, mention will be made of the power factors at which they operate.
Third Element Energy Cost
(c) The third element in the cost of electricity is the energy cost, which depends on the Kw-hr of energy used. It covers cost of fuel, labor, repairs and other items which vary with the load.
For a water plant or other large user of electricity, the first item of consumer cost is not a relatively large factor. The item of demand cost is most important however. For the usual light and power plant a curve showing the load throughout the twenty-four hours would show one or more peaks, the highest one usually being soon after sundown. The demand at this time determines the capacity generating units which must be provided. If the consumer referred to as having a 50 KW. demand must operate his motor during this peak load on the power plant, it is necessary to have the 50 KW. additional capacity for him, but if he can so adjust the time of operating his motor that he does not call on the plant during the peak, but only at such times as the load is otherwise light, then no additional capacity is needed to take this load, and a correspondingly lower rate should be made. If the water system is not provided with storage capacity but must maintain the water pressure by direct pumping, then the power plant must have the additional capacity to operate the pumps during the peak load and the cost will include the full demand cost for the capacity required. On the other hand if the water system has a standpipe which will allow the pumps to be shut down during the electric plant’s peak, and the electric plant can be assured that the pumps will not make the demand during the peak, then this saving to the electric company should be reflected in the rate. Another reason why water storage capacity is desirable where the pumps are electrically driven is that it makes less necessary the construction of duplicate lines and other precautions to insure against short interruptions to electric service.
Motor Larger Than Pumping Load Unnecessary
There is often a tendency in water plants to install a motor considerably larger than the pumping load requires, perhaps on the theory that it may be needed sometimes and that it is better to have too much capacity than too little. While those responsible for the operation of the water plant may not realize it, they are usually making themselves pay more for their electric energy than they would with a motor no larger than needed for operating the pumps. There are too many systems of rates to discuss here, but practically all of them undertake to make a charge which will cover the three costs noted above. The usual plan is to use some type of step rate, wherein the first block of so many kw.-hrs. per horsepower of motor capacity carries a certain rate, the next block a lower rate, and so on. Note that this is based on the power of the motor. It is true that a forty horsepower load carried by an 80 horsepower motor consumes only slightly more power than the same load will require when carried by a 40 horsepower motor. (This slight additional power is due to the higher losses in the larger motor), but the possible demand in the electric plant is greater, more capacity must be available and since it costs the power company to provide this additional capacity, a greater charge must be made. Also the 80 horsepower motor, if of the induction type, will operate on a lower power factor when half loaded than will the 40 horsepower motor fully loaded, with the bad effects of low power factor.
(Continued on page 402)
(Continued from page 400)
Types of Motors
There are several types of motors which are suitable for use in water pumping and I will discuss their characteristics briefly before taking up the application of each type to the pumps.
Direct current motors are of interest to few water plants, for the power supply now is almost universally alternating current. For direct current there are two types of motors which might be used: the shunt motor and the compound— the shunt motor has the advantage of being very nearly constant in its speed from light load to full load. The speed can be changed and when the rheostat is set for a given speed, will maintain that speed nearly constant with varying load. The compound motor is not so nearly constant in speed with changing load, and the speed will decrease with load. It can be varied as needed. This motor also has a larger starting torque, and altogether is more satisfactory for pump drive than the shunt motor. The single phase alternating current motors are little used for pumping and will not be discussed here. Among the two and three phase alternating current motors there are several types which should be considered. There is the squirrel cage induction motor w’ith which many of you are familiar. This is a rugged type of motor easily started either by hand or a remote control system. Its operation requires little skill, and its principle requirements are that it be not overloaded for any appreciable time and that its bearings be kept oiled and free from grit. There are no moving rings or brushes to require attention. This motor is practically a constant speed motor. Though its speed decreases slightly with the load, the speed can not be varied. Its starting torque is not large, and if used where even a moderately large starting torque is required, it draws a heavy current from the line. Its efficiency is high but its power factor is lower than is desirable especially when lightly loaded, and this means larger transformers, lines, generators, etc., with a higher cost for the power used.
Wound Rotor Induction Motor
The wound rotor or slip ring induction motor also has the disadvantages of being a low power factor motor and it is usually slightly less efficient than the squirrel cage type. This motor may have a starting torque equal to, or even exceeding its full load torque, and it has the further advantage that it can be operated at a speed below normal operating speed. Its cost is somewhat greater than that of the squirrel cage motor and, while it is also a rugged type, it has brushes which must receive occasional attention. It can be started from a point near at hand or from a distance. Its power factor is about the same as that of the squirrel cage motor, but when operated below normal speed its efficiency is low. To illustrate, if the full speed efficiency of the motor with a certain load is say 90 per cent., when the speed is reduced to 70 per cent, of full speed, the efficiency is reduced to about 70 per cent, of 90 or 63 per cent.
Brush Shifting Motors
One of the newer types of motors which is of interest in pumping plants is the variable speed, brush shifting motor. This motor is not a constant speed motor, but by shifting the brushes, the speed of the motor can be changed. With the brushes in a given position, the speed of the motor will decrease with increasing load which is a desirable characteristic for pumping. This motor has a high starting torque, and, except at low speeds, a better power factor than the woundrotor slip ring type induction motor. Unlike the wound rotor induction motor, the speed of which can be varied only by steps, it permits an infinitely large number of speed changes and at a better efficiency except when operated only slightly below full speed. Compared with the induction motor, it is somewhat more complicated to operate, requires more skilled attention; cannot be operated so well from a distance, and is higher in first cost. These disadvantages must be set against the better efficiency, the higher power factor, and the greater flexibility of speed control.
All types of induction motors have the disadvantage of causing a low power factor, which is undesirable for reasons already mentioned. The most satisfactory method of obviating this low power factor is by using synchronous motors. The synchronous motor is an absolutely constant speed machine. and its principle advantage is that it can be operated at a high power factor; in some cases so high a power factor as even to counteract the low power factor of induction motors operating at the same time. Though the more the low power factor of other motors is to be corrected, the larger the synchronous motor must be. The synchronous motor must have a direct current exciter operated with it and requires some attention from time to time. It may be started by hand switch, or by remote control, but its greatest disadvantage is the comparatively low starting torque, which will not usually exceed 50 per cent, of full load torque, from start to a point just below full speed, though absolutely full speed once reached, a torque several times rated full load torque may be supplied. Its efficiency is high but due to the need of an exciter and its more complicated operation, the synchronous motor is seldom used except in the larger sizes and in plants where it can receive a reasonable amount of attention. Its speed is absolutely constant and it will either operate at synchronous speed or not at all.
A type of synchronous motor has recently been developed which should find its ready application in pumping plants where large pumps are used and an operator is in attendance. This motor has been called the supersynchronous motor and consists of a synchronous motor arranged so that both armature and fields can revolve, the fields being on the shaft which is connected to the load. The starting switch is closed in the usual way and the armature allowed to rotate backward freely until it reaches normal speed. The motor is equipped with a braking device which is slowly applied to the armature after it reaches full speed. As the armature is caused to slow down, the field rotor with its load gradually picks up speed so that the relative speed between fields and armature remains normal. By the time the armature is brought to a stop the field rotor is carrying the load at normal speed and it has been capable of exerting several times full load torque in accelerating the load. After having brought the load to full speed, the motor operates as any other synchronous motor and with similar characteristics, namely: constant speed and high power factor. Obviously, it must be started by an attendant, and not from a distance. It is somewhat higher in first cost than the synchronous motor.
Of the several types of motors which have been discussed, the most suitable for a given installation will depend on the characteristics of the pump to be driven and the conditions under which it is to operate. The principal types of pumps are: 1, reciprocating; 2, rotary; 3, centrifugal.
From the standpoint of the requirements of the driving motor, the several kinds of reciprocating pumps may be placed in one class.
In starting a pump of this kind, the positive displacement of the piston or plunger required full load torque and in addition the static friction which must be overcome at starting is high due to sliding pistons and water tight stuffing boxes. Depending on the tightness of fit of pistons and plungers, this starting torque may vary from 25 per cent, above normal operating torque to as high as 150 per cent, above normal. When otherwise feasible the high starting torque requirements may be reduced by opening a by-pass allowing the water to pass freely from discharge to suction thus reducing the head and with it the starting torque required.
Except for slip, the reciprocating pump discharges a definite quantity of water at each stroke, and the output is independent of head. At a given speed the load on the motor varies with the head. If the speed is increased, the quantity of water will increase and that part of the head which is due to pipe friction will increase approximately as the square of the speed. With a direct current supply, the compound wound motor is preferable for driving reciprocating pumps because of its high starting torque. If the pump can be bypassed on starting or if it is of a very small size the squirrel cage induction motor should prove satisfactory. The latter type may be operated automatically or from a distance.
In most cases the wound-rotor induction motor is more satisfactory as it will develop the necessary torque for bringing the pump to speed without drawing excessive current from the line. It may be started automatically or from a distance if desired.
On account of high starting torque required, the synchronous motor is seldom suitable but in exceptional cases it may be used when the pump is by-passed.
If the low power factor of the induction motor is an item to be considered, and when it can be started by an operator, the supersynchronous motor is probably the most satisfactory. It is especially well suited for large reciprocating puntping plants.
For variable speed operation the only alternating current motors available are the wound rotor and the brush shifting motor. The former is preferable if the speed is to be reduced only a small amount, or on few occasions, otherwise the brush shifting motor is much the better because of its higher efficiency and power factor.
The rotary pump is similar to the reciprocating pump in that, except for slip, each revolution of the pump handles a definite amount of water. At a given pump speed the power requirements vary with the head as in the case of the reciprocating pump. It differs from the reciprocating pump in that it requires only about 50 per cent, of normal running torque for starting. When pumping against a fixed head the power required varies with the speed. It must be borne in mind however, that the part of the head due to friction will increase as the square of the amount of water discharged and consequently as the square of the speed. The motor requirements are practically the same as for the reciprocating pump, though the by-pass is not usually required owing to the lower friction at starting.
The centrifugal pump is most important from the standpoint of use in water plants, and it seems that its power requirements are least understood. There are two fundamental facts about a centrifugal pump and these combine to form a third. 1. The head varies as the square of the speed. 2. The quantity varies directly as the speed. 3. The power varies as the cube of the speed.
The starting torque of the centrifugal pump is low and increases to normal at normal speed. If, however, the discharge is kept closed, the water will churn and the torque for normal speed will not rise above about 60 per cent, of full load torque.
Fig. 1 shows some typical curves for a centrifugal pump driven at constant speed. The normal operating condition exists when the pump delivered 375 g.p.m. against a total head of 325 ft., and the power required is 50 horsepower. One point worthy of note is that if the head decreases to, say, 195 feet the pump will deliver 575 g.p.m. and the horsepower requirements increase to 62. In other words the horsepower requirements increase rapidly with decrease of head. This fact has been the cause of some trouble with motors in the past, especially when driving pumps working against a head, which was subject to variations, as when pumping into a standpipe where there was considerable variation in height of water. If a motor is used which will give the proper power at normal discharge and head, it will be overloaded when operated against the decreased head.
(Continued on page 416)
(Continued from page 403)
By properly shaping the impellor blades, the pump can be so designed that the discharge will increase less rapidly with decreasing head, and this type of pump should be used where the head may vary appreciably. The power curve will not continue to rise with discharge but will finally begin to flatten and then even decrease with decreasing head. This is shown by Curve B in Fig. 1. The normal operating condition should be such that the pump operates at perhaps 10 per cent, or 15 per cent, below the maximum power. For a decreasing head, the power will not exceed a 10 per cent, or 15 per cent, overload on the motor, and the motor should be able to carry this for a time. It is not always possible to care for all conditions with this pump however, and there are other methods of remedying the difficulty. Among these methods are:
- Install a motor large enough to carry the maximum load, and have it operate at less than full load at other times. This will prevent the motor from overheating, but it will not usually be the best solution of the problem. The larger motor will cost more so the interest and depreciation will be greater; its losses will be slightly greater; it will take more current and at a lower power factor, and the rate for the power actually used will be greater because it is based on rating of the motor.
- Install a motor of the proper size for normal operation and throttle the discharge to keep the head and the discharge practically constant by adding friction head to a static head. This method is simple but it is not efficient, and should be used only when the condition of low head is infrequent.
- Install a variable speed motor and when the static head is low, reduce the speed, and with it, the output until the power requirements are within the limits of the load that the motor can carry.
For the centrifugal pump with constant head, a squirrelcage induction motor up to 50 horsepower may be used. The starting torque requirements are low, and constant speed is satisfactory, though the power factor is low. It is simple to operate and easily started from a distance. If a direct current motor is to be used, it should be a compound wound type: For the large size induction motors the wound rotor is required. In the larger sizes, however, the low power factor may create too great a demand on the power plant, a condition which can be met only by the use of a synchronous or supersynchronous motor. Where the pump is to be started with discharge valve closed the synchronous motor may have sufficient torque but for any installation where this is not the case, the supersynchronous type will develop all the torque needed. In case of appreciable increase in load from decreased head, the load on the squirrel cage induction motor or either of the two types of synchronous motor can only be kept within normal limits by throttling the discharge. This is not an efficient method. If the head decreases 19 per cent, and it is necessary to throttle the discharge and raise the head until the discharge is back to normal the useful output is reduced to 81 per cent, and the efficiency is only 81 per cent, of its former value. On the other hand since the head varies as the square of the speed, a speed reduction to 90 per cent, will accomplish the same result. With the wound rotor motor this speed reduction is obtained with a reduction of efficiency of 10 per cent. With the brush shifting motor it is obtained with very slight reduction in efficiency, perhaps less than 5 per cent. It will thus be noted that there is a great saving in cost of operation by using a variable speed motor if pumping against lower than normal head is to be a frequent or prolonged occurrence.
Simplicity of Operation
Simplicity of operation depends somewhat on the type of motor used, but an operator with sufficient skill to operate other sources of pump drive should be able to operate any type of motor.
Flexibility of Location
No other type of pump drive adapts itself so readily to use with small and widely scattered motors as for pumping scattered wells. This has proved an especially valuable feature for cities built on two levels, as partly on a bluff and partly below it. To place the standpipe high enough to supply all the city would necessitate pumping against a head not needed for the lower part of the city besides putting a greater strain on the pipes. In many places this problem has been solved by having a standpipe for the lower part of the city and another for the higher part and keeping the higher supply up with motor driven pumps automatically started and stopped by the height of water in the standpipe.
Perhaps the two most notable advances in electrical equipment suited for use in water plants have been the development of the brush shifting motor for variable speed and the supersynchronous motor which provides high starting torque with high power factor.
(From a paper read before the Sixth Short Course for Water Works and Filter Plant Operators at Waco, Tex.)