Motor Testing – Saturation Tests

Saturation Tests

In order to ascertain the characteristics of the magnetic circuit, a test known as a saturation test is made. The characteristic curve may be obtained by either the generator saturation method or the motor satura tion method.

To obtain a saturation curve by the generator saturation method, the machine is driven at normal speed as a generator. The brushes of de machines should always be set on neutral, and the machine run at its normal rated speed. In taking the saturation curve on polyphase ac generators, a reading of the voltage across each phase must be taken at normal field current to see if all phases are properly balanced. If they do not balance, they must be made to do so. With synchronous converters, careful readings must be taken of the de voltage as well as the ac voltage between all phases, with the field excitation giving a normal voltage. The phase voltages must also be closely balanced.

The usual method of taking the generator saturation curve is to hold the speed constant, and then increase the field current step by step until the full field-excitation voltage has been reached, taking simultaneous readings at each step of armature voltage, field voltage, and field amperage. About twelve points are sufficient. Four readings should be taken from zero field to 90% of the normal rated voltage of the machine.

Five points should be taken between 90% and 100% of the normal rated voltage, taking one of the points at the normal rated voltage of the machine. Three points should be taken between 110% normal rated voltage to the full field current on the machine.

After reaching the maximum value of the field current, and without opening the field, reduce the current gradually in four or five steps, and again take readings to determine the value of residual magnetism at various points along the curve. Special care must be taken to ensure accurate readings at and above normal voltage, for in ac generators this is the portion of the curve used for calculating the regulation load. On separately excited high-voltage de generators, the saturation curve should be carried only to 25% above the normal rated voltage.

When it is inconvenient or impossible to drive the machine as a generator, a motor-saturation curve may be made. For this method, the machine is operated as a free-running motor. The driving power must be furnished from a variable-voltage circuit. A certain voltage is impressed upon the armature and the motor field increased or decreased (in the case of de machines) to give normal speed, and a record made of the armature voltage, armature amperage, field amperage, field voltage, and speed. The starting voltage should be at least 50% lower than the normal rated voltage of the apparatus. The applied voltage of the armature should be increased by steps to 25% above the normal value, and the field increased correspondingly to keep the speed constant, the same readings being recorded at the various steps as before. Readings should also be taken at three or four points as the impressed voltage and field current are lowered to approximately the values at the beginning of the test.

Care must be taken in testing de apparatus, because unstable electrical conditions may develop and excessive speeds result. The circuit breaker in the armature circuit of the motor driving the machine must be accessible to the tester reading the speed.

On ac apparatus, the machine is run as a motor and the impressed voltage varied as already described. The speed is independent of the motor field in this case, and instead of regulating the motor field for speed, it should be regulated to give minimum input current at each voltage. Readings should be taken of each voltage impressed, the armature amperage, field amperage, and field voltage. With induction motors, it is only necessary to impress variable voltages at a constant frequency and record the readings of the impressed armature voltage, armature amperage, and speed. A curve should be plotted, using the armature voltage as ordinates and the field amperage as abscissae. Fig. 11 shows a typical saturation curve.

Core Loss

There are two methods commonly used to measure the core losses of rotating de machines and ac synchronous apparatus. They are the running-light and belted methods.

For de machines, the following conditions must be met in order to give satisfactory results: The brushes must be shifted on the commutator to the mechanical neutral point. The driving power must be supplied from a variable-voltage source that is not subject to sudden fluctuation. Readings must not be taken when the rotating parts are accelerating or decelerating.

In the running-light method, this test is made by running the machine free as a motor. It is made on most de generators and motors which are given a running test, and occasionally on ac synchronous apparatus. When running any de machine as a motor, the following procedure should be followed: Set the brushes on mechanical neutral. Be sure that the water box (or other variable resistance) for starting is plugged across the main switch, that the main switch is open, and that the blade of the water box is nearly all the way up. (See Fig. 12.) Plug the power to the armature, bring the supply voltage up, close the breaker, and lower the blade of the water rheostat slowly, watching the reading on the line ammeter. If the machine does not start to rotate at a value of current below full load, turn the power off and investigate the cause. When the machine is running at rated voltage, close the line switch and raise the blade of the water box again. If the machine runs too fast at normal voltage, turn off the power. The connections should be carefully checked to see that the field is wired properly. It may be that the field is connected directly across the main switch. If such is the case, the field current will fall rapidly as the starting resistance is cut out, and the motor will speed up. To test for incorrect wiring in the field, observe the field voltage during starting, because it will drop if the field is incor rectly connected.

When a running-light test is made on generators, the observations must be made with a full-load field flux. The potential applied to the armature must be equal to the normal rated voltage of the generator, increased by theIR drop under full load. TheIR drop is taken as 4% of the normal voltage for both motors and generators. With this voltage impressed, the field current is varied until normal speed is obtained, when careful readings must be made of the armature current, armature voltage, field current, field voltage, and speed.

If the machine under test is a de motor, the voltage applied to the armature should be equal to the normal rated voltage of the motor, less the IR drop under full load. The field current is then adjusted to give normal speed, and electrical readings taken as previously outlined for de generators.

The power supplied to machines running free will equal that absorbed in bearing, friction, brush friction, windage, and core loss, when the armature l²R losses have been subtracted.

The record of these tests must clearly show whether the running-light current consists of the armature current plus the shunt-field current or whether it is the armature current alone . To obtain the running-light core-loss test, the machine should operate as a shunt motor.

For series-wound motors, the field should be separately excited and extreme care should be taken to see that the motor does not lose its field excitation.

To obtain running-light core loss in ac synchronous motors, they should be operated at the proper frequency and rated voltage. For the best results, both frequency and voltage must have a steady value.

With normal voltage on the armature, the de field should then be varied until minimum armature current ,is obtained. Readings should then be taken of the current and voltage of all the phases. At minimum input current, unity power factor is obtained and, therefore, the power to drive such machines will be the volt-ampere input. Wattmeters may be used, in addition, to check the volt-ampere readings. This mea surement includes friction and windage losses, together with open-cir cuit core losses, plus the I²R loss in the armature. If the value of the core loss need not be separated from the other losses, the test is useful for checking load efficiency.

On motor-generator sets, unless otherwise specified in the instructions, the following running-light tests should be made: With the de brushes down, running at the rated speed from the de end with the line voltage 4% high as for any generator, and with no field on the synchronous motor, record the readings of the line voltage and current, field voltage and current, and speed. Continuing to hold the line voltage and speed constant, bring the field excitation up on the synchronous motor until the normal voltage on the armature is obtained. Take and record the readings of the synchronous motor armature voltage, field voltage and amperes, together with another set of input readings on the generator as before. The difference between these two sets of input readings will be the core loss of the synchronous motor.

By means of the belted core-loss method, the core loss can be separated from the bearing friction, brush friction, and windage. A small de motor is used to drive the machine under test as a generator at its rated speed. (See Fig. 13.) A belt drive between these machines is most commonly used, but whenever great accuracy or a high speed is necessary, a direct drive by means of a coupling is often used.

The driving motor for this test should be such that good commutation is obtained with a fixed setting of the brushes for all the loads required by the core-loss test. With the maximum voltage on the machine under test, it should not carry more than 100% of its normal rated capacity.

A good rule to follow is to select a motor that has a rated capacity of approximately 10% of the rated output of the machine under test. When the brush setting to give the best possible commutation at all loads has been obtained, the brushes should be left in a position that will give the motor a slightly falling speed characteristic. With the brushes in this position, the voltage applied at the armature will have to be increased as the load is applied in order to maintain constant speed. The commutator surface should be in first-class condition and should have the brushes closely fitted to it.

The belt should be of minimum width and weight to carry the load without slipping. When testing motor-generator sets, synchronous con verters, and other machines that do not require belts in practice, the tension of the belt must be kept as low as practical so that the bearing friction is not increased by the belt pull. Endless belts should always be used in preference to laced belts. The diameter of the pulleys should be selected so that the driving motor will run at or near its normal rate of speed when the motor under test is running at its normal speed.

The driving motor should have its field separately excited from a constant source, and other wiring so arranged that readings may be taken of the armature current, armature voltage, field current, and speed. The test wires should be firmly attached to adjacent brushes, which should be insulated from the brush holders so that the true armature voltage may be obtained.

The machine under test should be wired as a separately excited generator with provision for reading the armature voltage, field voltage, field current, and speed. The tests should be carried out as follows: The field of the driving motor should be adjusted to about normal value and held constant, and the speed regulated by varying the voltage applied to the armature terminals. Careful readings should be made to make sure that no belt slipping occurs, by taking simultaneous readings of the speed of both the driving motor and the machine under test: (a) with no field on the machine under test, (b) with normal field excitation. The two readings of speed should be identical. The machine should be run a length of time sufficient to allow the friction to become constant. This will be the case when the input to the driving motor remains constant when the machine under test is driven without any field excitation.

Throughout the entire test, readings must be taken at absolutely constant speed when the rotating parts are neither accelerating nor decelerating. The readings should be taken as follows: (a) take the input of the driving motor, with no field on the machine under test and with all brushes down on the commutator; (b) take the input, with all brushes raised and with no field on the test machine. The difference these two readings is the brush friction. Then, starting with zero field on the machine under test, and with the brushes raised from the commutator, observations of input to the driving motor should be recorded at various values of field up to that which will give 125% of normal voltage; and at least half of the readings should be taken between 90% and 110% of normal voltage.

The friction reading with zero excitation on the machine under test should be repeated at least three times during the test namely, at the beginning, again near the midpoint of the curve, and finally at the end of the test.

The driving motor should then be disconnected and a running-light reading taken as follows: Without changing the brush lift, hold the same field current that was held during the core-loss test, and take a reading of the input to the motor to give the same speed that was read on the driving motor at the beginning of the test. The armature voltage should be lower than for any other reading taken during the core-loss test.

To check the results of the core loss as the test proceeds, the power input to the driving motor required by the core loss at a given excitation should be plotted against the voltage generated by the armature.

Correct the motor input at various field strengths by deducting the/2R loss in the armature of the driving motor, and subtracting the power input to the driving motor with zero field on the machine under test. The core losses left correspond to the various field strengths. By subtracting the running-light input to the driving motor from the input at zero field on the machine under test, the bearing friction and windage losses of the machine are obtained.

On series motors, core-loss tests should be taken at several different speeds covering the range of the speed curve. The method used is identical with that previously described.

Synchronous ac machines generally have loss measurements as previously outlined, on open circuit, and also with the armature of the machine under test short-circuited. In the latter case, the increase in power supplied by the driving motor over that required by the friction loss is plotted as ordinates against the armature amperage as abscissae, or the open-circuit armature voltage due to a given excitation. A curve is obtained similar in character to the open-circuit core-loss curve. Such test is commonly known as the short-circuited core loss. In making this test, careful measurements must be made of the resistance of the short-circuited armature circuit, including all leads. Observations should be made starting at 200% normal current value and reducing by steps to zero current. Several thermometers should be placed on the stator winding and readings taken of the temperature after each step on the core-loss curve.

HEAT-RUN TESTS

Heat runs are taken primarily to determine the amount of temperature rise on the different parts of the machine while running under a specified load. This rise in temperature is measured by the rise in resistance of the current -carrying part, by means of a thermometer, or by both. The results obtained by the rise in resistance, as a general rule, are used only as a check on the results obtained by reading the thermometers placed on the different parts.

The average temperature of the winding is obtained from the resis tance measurements by using the following equation:

where,

T is the hot temperature in degrees C,

R is the hot resistance,

r is the cold resistance at temperature t,

t is the cold temperature of the winding in degrees C .

Erample: Assume that the cold resistance is 0.50 ohm at 25° C and the hot resistance is 0.61 ohms. Then:

The temperatures measured by detectors or by the change in resis tance are generally higher than thermometer measurements and are closer to the true hottest-spot temperature in the machine. For this reason, the Standards permit higher observable temperature rises when measured in this manner.

Special copper leaf brushes are commonly used for measuring the voltage of the fields of alternators and synchronous motors by the volt-ampere method. The resistance by the volt-ampere method must be made to check within 1% of the Wheatstone-bridge resistance mea surement before the machine is started the first time. Three readings of volt-amperes should be taken at 20, 25, and 30% normal field voltage, with thermometers on two poles to be read at each of the foregoing voltage readings.

Self-excited de machines must be checked, as described previously, immediately after the voltage is brought up the first time. As the heat run progresses, the temperature rise of the field should be calculated and recorded at the same time the thermometer readings are taken. All meters to be used on a heat run should be checked against other meters before the run is started. The same meters must be used throughout any given test.

Thermometers should be carefully examined for a broken mercury column before being used and should not be inverted or placed on a machine so that the bulb is at a higher level than the other end. Before starting a heat run, thermometers should be placed on the stationary accessible parts of the machine. Each thermometer should be attached with the bulb in contact with the part of which the temperature is required, and should have the bulb covered with an amount of putty sufficient to secure it to the machine and to shield it from the surrounding air . Extreme care must be exercised regarding the amount of putty used, as too much putty is as bad as too little. Just enough should be used to do the work required. There should be no restriction of the natural windage of the machine, or no radiation from the coil of which the temperature is being measured.

Thermometers which are to register the temperature of the air ducts should be placed so that the bulbs cannot make contact with the iron laminations while the machine is running. Thermometers which are liable to be shaken off by continued action of windage or vibration should be securely fastened to the machine. When placing thermome ters on field coils, care should be taken to see that they are not placed on the fiber strips protecting the outside terminals. These fiber strips run from one terminal to the other and from a nonconducting wall between the coil and its outside insulation, and thus do not represent the true temperature of the coil. Coils above the horizontal center line of the machine should be used as the top of the machine. They are usually somewhat hotter than the bottom coils. On small machines, two thermometers will be sufficient on the coils, but larger machines should have at least four.

One thermometer will be sufficient on the frame of small machines, but two should be used on large units. At least two thermometers should be used on the laminations and ducts of small machines, and four or more used on the larger machines.

Any large machine requiring considerable floor space should have the room temperature taken at four or more nearby points. These ther mometers should be placed about 6 feet ( 1.829m) from the base and at the elevation of the center line of the machine. These thermometers should be placed two together; one set should be placed in an oil cup designed for this purpose, and the other set should hang in the air to record the air temperature. All of the room-temperature thermometers, whether they are placed in air or oil, should be so located that they are not affected by the windage and radiation of the machine under test.

The machine should be shielded from currents of air coming from adjacent pulleys, belts, and other machines. A very slight current of air will cause great discrepancies in the heating results. Consequently, a suitable canvas screen should be used to protect the machine under test when necessary. Great care must be used, however, to see that the screen does not interfere with the natural ventilation of the machine under test. Care should always be taken to see that sufficient floor space is left between machines to allow free circulation of air. Under ordinary conditions, a distance of 6 feet (1.829m) is considered to be sufficient.

Immediately after the heat run is started, various checks should be made. On all heat runs where meters are connected in both the output and input circuits, especially motor-generator sets and converters, the power-output readings plus the losses of the apparatus under test must equal the power-input readings. On all synchronous machines, the power factor specified in the instructions should be held and checked against the phase characteristics on motors and against saturation on generators. In such cases, the field current held during the heat run should be greater than the unity-power-factor field current.

SUMMARY

Electric motors are tested to make sure that they have been con structed properly, that they meet the manufacturer’s guarantee, and that they can perform their assigned duty with safety under normal condi tions of operation. The tests can be applied to ac machines as well as to de machines.

Prior to making a test, it is important that the necessary instruments and accessories are available, in addition to the proper source of power. Instruments and accessories used most commonly are tachometers, frequency indicators, meggers, megohmmeters, ohmmeters, volt meters, ammeters, shunts, graphic recording instruments, wattmeters, temperature meters, potential transformers, current transformers, etc.

Careful inspection of the machine should be made prior to any test to see that each point connected to the field, armature, commutator, brushes, brush rigging, terminal block, etc., conforms with the manufacturer’s specifications. The number of terminals on each coil indicates the form of field-winding-series, shunt, or compound.

Various resistance tests are made on commutating machines. These tests are classified as armature resistance, location of defects, field winding resistance, insulation resistance, and dielectric strength.

The availability of a motor for a particular duty is determined almost entirely by two factors-variation of torque with load and variation of speed with load. The more common motor characteristics that can be obtained by testing are:

1. Efficiency, torque, and speed as a function of current,

2. Saturation curves,

3. Speed-torque curves (series, shunt, cumulative, and differential compound).

Heat-run tests are made primarily to determine the amount of temperature rise on the various parts of the machine while running under a specified load. The average temperature of the winding is obtained from the resistance measurements by the equation: