Handbook of Small Electric Motors Episode 11 pdf

Reduced-voltage idle heat run: The motor operates without load at voltagesreduced by 10-V increments below rated voltage until the windings reach ther-mal equilibrium for 60 min without

CHAPTER 9 TESTING

con-In order to obtain accurate test results, the maximum motor torque should equal

at least 30 percent of the dynamometer rating Voltage drops during the test willdrastically affect results and require heavier wiring than specified by the NationalElectrical Code A good rule of thumb would increase wiring in conduit by six wiregauges and bench wiring by four wire gauges Voltage control devices such as powerstats must also minimize voltage drops, either by voltage regulation or by selecting arating of at least four times the maximum motor current The power supply fre-quency should not vary more than 0.5 percent from the rated value In addition,polyphase systems should not exceed 0.5 percent voltage unbalance between phases

9.1.1 Acceleration Test

This test determines the acceleration characteristics of a motor for a given amount

of inertia and load torque The inertia test stand shown in Fig 9.5 includes a set ofinertia wheels that will provide a range of inertias from 50 to 4000 lb⋅in2in 50-lb⋅in2

increments and an analog tachometer to monitor speed In order to minimize lations, the setup uses low-backlash couplings between the motor, the inertia wheelshaft, and the dynamometer An oscilloscope monitors the voltage from the analogtachometer and provides voltage and time data to a PC through the general-purposeinterface bus (GPIB) connection At this point, the PC calculates the speeds based

oscil-on the tachometer voltage coscil-onstant and plots the acceleratioscil-on curve (see Fig 9.6)

FIGURE 9.1 Typical motor test bench diagram.

FIGURE 9.2 Typical test station.

FIGURE 9.3 Speed-torque curve and summary sheet generated by a computerized test station.

FIGURE 9.4 Speed-torque, efficiency, current, watts, and power factor curves generated by a puterized test station.

com-FIGURE 9.5 Inertia test stand.

FIGURE 9.6 Acceleration plot.

AC motors and open-loop dc motors could use the bench shown in Fig 9.1 toacquire and document additional data such as torque, efficiency, current, and watts.However, closed-loop motors require a different approach, as outlined in the dcmotor test section on speed profiles.

9.1.2 Good Test Practices

1 Review test requirements for consistency with UL and other requirements to

avoid wasted test time Obtain clarification as needed

2 Measure the resistance of the motor windings using a precision multimeter or

bridge and record ambient temperature Ensure that resistance meets windingspecifications prior to any tests

3 Perform high-potential tests per UL requirements prior to energizing the

motor In the absence of other information, a good rule of thumb would apply avoltage equal to (1000 plus twice the rated voltage) times 1.2 for at least 1 s

4 Select stable capacitors within at least 1 percent of specified values and check

for drift rather than relying on old measurements marked on the capacitor.Always discharge the capacitors prior to checking the value to avoid damage tothe meter Switch between all selections on capacitor decade boxes when bleed-ing charge

5 Confirm that the dynamometer has appropriate cooling operating.

6 Use brass-tipped setscrews in the coupling and check tightness periodically to

avoid bad data

7 Select appropriate ranges for equipment that does not autorange.

8 Minimize coupling backlash and torsional deflection.

9 Monitor display to confirm at least the initial printout.

10 If the motor has a ground wire, connect it to the workstation ground.

11 Recheck motor connections against specifications prior to test.

12 Adjust speed-torque test time and minimum speed to minimize motor

oscilla-tion below breakdown

9.2 AC MOTOR THERMAL TESTS

Thermal tests for ac motors determine the thermal protector trip and reset atures under locked and running condition, as well as the leveling temperatures forrunning conditions See Fig 9.7 for test setup These tests usually run at nameplate-rated voltage, frequency, and current, but UL may require tests at different voltages,

Running overload: A dynamometer holds the motor current per the nameplaterating until the motor windings attain thermal equilibrium for 60 min or until theprotector trips The dynamometer controller then increases the torque to provide acurrent increment per Table 9.1, and maintains this current until the winding tem-perature reaches equilibrium This sequence repeats until the thermal protector

opens or the motor stalls (Note: For running-overload tests, UL requires

perform-ing the first test at nameplate current If the no-load current of the motor at ratedvoltage equals or exceeds the nameplate current, the first running-overload testoccurs at idle If the thermal protector opens at the nameplate current—above theno-load current—then the motor test must also occur under no-load condition atthe nominal test voltage If the protector opens at idle, then the idle test repeats for

TABLE 9.1 Nominal Voltages and Currents for Running-Overload Test

Nameplate current Current increment Nameplate voltage Nominal voltage

voltages reduced by 10-V increments until the winding temperature stabilizes out opening the protector See Fig 9.8 and Table 9.1 for these test requirements.)

with-Idle heat run: The motor operates without load at nameplate voltage until thewindings reach thermal equilibrium for 60 min

Reduced-voltage idle heat run: The motor operates without load at voltagesreduced by 10-V increments below rated voltage until the windings reach ther-mal equilibrium for 60 min without tripping the thermal protector

FIGURE 9.8 Logic diagram for idle thermal test.

9.2.1 Test Conditions

UL 2111 allows the following maximum temperatures for Class B insulation withthe various thermal tests:

Locked rotor 225°C in first hour

200°C in the second hourFull-load heat run 165°C

Running overload 165°C if protector opens

175°C if protector does not openIEC 34-1 requires a maximum temperature rise of 85°C above room temperature

as measured by thermocouple or 90°C as measured by resistance For Class F lation, the maximum temperatures increase to 110 and 115°C, respectively

insu-For full-load, running-overload, and idle heat runs, UL requires a temperaturerise variation at temperature equilibrium of no more than ±1.0°C for 2 h

Correct size and type of coupling between motor and dynamometer

Motor test stand and means of securing stand to stationary mount

Type J thermocouples and accompanying chart equipment

Printer

Computer hardware and software

Locking bar for locked-rotor test (optional)

Powerstat

Fan for quick cooling of the test motor

9.2.3 Test Setup

Motor Setup Requirements

1 Mount the motor in a test stand with the protector at the six o’clock down

posi-tion with the motor connected to the dynamometer

2 Couple the motor to a dynamometer with a thermally insulating coupling For

locked-rotor tests, either apply enough torque with a dynamometer to overcomethe motor starting torque or use a thermally insulated lock bar to stall the motor

3 Align and secure test stand to dynamometer.

4 Connect thermocouples from the motor to chart equipment with electrical

isola-tion

5 Connect motor to capacitor or relay (if specified) and to power source per

out-line

6 Confirm that the dynamometer cooling and dissipation rating meets the test

needs before operating the dynamometer

For locked-rotor tests or tests on intermittent-duty motors, let the test run withthe motor off for a short time to clearly document a room temperature start Othertests may start without delay All locked-rotor tests, regardless of thermal rating,must start with winding temperatures within 5°C of room temperature For otherthermal tests, only intermittent-duty motors must begin test with windings within

FIGURE 9.9 Typical computerized dc motor test bench diagram.

9.3.1 Voltage Constant Test

The K etest checks the voltage constant in volts per thousand revolutions per minute(V/krpm) for a backdriven dc test motor Any motor capable of maintaining anexact speed under a varying load can serve as a backdrive motor For brush dcmotors, measure the dc voltage generated by the test motor with a multimeter (gen-erally in both rotations) For brushless dc motors, acquire the peak-to-peak voltage

with an oscilloscope and divide by twice the drive speed in krpm to obtain the K e

9.3.2 Terminal Resistance Test

While multimeters can accurately measure brushless dc motor resistance, brush dcmotors experience variations in brush contact drop as well as resistance changesbased on the relative position of the brush to the commutator bars Brush dc motorstherefore require averaging several locked-rotor measurements to provide a stablereading, or, preferably, a dynamic measurement, while backdriven at low revolutionsper minute For either test, attach the motor terminals to a dc power supply and setthe current limit high enough to reduce contact drop fluctuations and low enough tominimize heating during the test In the absence of a specification, use a current limit

of 25 percent of the rated motor current Acquire the voltage necessary to drive the

current through the motor for calculation of resistance (R =V/I) For dynamic

mea-surements, reverse either the polarity of the voltage or the rotation of the drivemotor and average the two measurements to remove the counter-emf contribution.Backdriven speeds of 30 to 100 rpm work reasonably well and will generally providebetter repeatability than a locked test

FIGURE 9.10 Typical computerized dc motor test bench.

9.3.3 Speed-Torque Test

The speed-torque test provides a curve of the speed and the test motor torque (seeFig 9.11) Couple the shaft of the test motor to the dynamometer and attach themotor terminals to a programmable power supply Set the supply voltage limit at therated motor voltage and the current limit to the rated peak current Acquire voltageand current from the programmable power supply and speed and torque from thedynamometer controller

By using the dynamometer controller to take the motor from idle to a mined maximum torque and back to idle speed at a controlled rate, inertial effectswill nearly disappear simply by interpolating for the same speed points and averag-ing the two sets of data Points from the resulting curves will then compare very wellwith static measurements made at the same performance point In most cases, a dcmotor curve will not include the locked point, since a test at stall will often riskdemagnetizing the motor and/or damaging the commutator

predeter-9.3.4 Demagnetization Test

The demagnetization test determines the amount of current the test motor can draw

before reducing the K e by 5 percent Couple the shaft of the test motor to adynamometer and attach the motor terminals to a dc power supply Use an oscillo-scope to monitor a current probe or current shunt on the positive output of the dcsupply Set the supply voltage limit to the rated motor voltage and the current limitwell beyond the calculated demagnetization point Using the oscilloscope to deter-mine the current, quickly apply torque until reaching the desired current Remove

all torque immediately and repeat two more times Recheck the K eof the test motor

FIGURE 9.11 DC motor speed-torque curve.

after allowing the motor to fully return to room temperature, generally after about

30 to 60 min A reduction in K egreater than 5 percent means the test motor hasdemagnetized Otherwise, repeat the test at a higher current (typically in 5 percentincrements)

9.3.5 Thermal Resistance Test

The thermal resistance test determines the temperature (°C) rise per watt loss of thetest motor Place the test motor into a stand and lock the shaft For brush dc motors,route wires from the commutator and under the bearings to the outside for thewinding resistance measurement, to avoid errors introduced by the brushes and thecontact drop Measure the cold winding resistance with the multimeter and recordthe ambient temperature Attach a thermocouple to the shell to monitor the tem-perature rise Attach the motor terminals to a programmable power supply Slowlyincrease the voltage until reaching the rated current Hold the rated current for 1 hafter the shell temperature levels Quickly detach the motor terminals and take thehot winding resistance with the multimeter Record the ambient temperature.Repeat the test for the rated running condition, preferably with a different motor forbrush dc tests to avoid the possible effects of a burned commutator The thermalresistance constant equals approximately:

Rth=Use a winding constant for copper of 0.00393

where Rth=thermal resistance

Rhot=hot winding resistance

Rcold=cold winding resistance

ambcold=cold ambient temperature

ambhot=hot ambient temperature

voltagehot=hot winding voltage

currenthot=hot winding current

9.3.6 Safe Operating Area Curve Test

The safe operating area curve (SOAC) determines the boundaries of safe operation.Use the running thermal resistance constant to estimate the winding temperatureand maintain safe operating temperatures (usually below 85°C plus ambient) Placethe test motor into a stand and couple the test motor to the dynamometer Attach athermocouple to the shell to monitor the temperature rise Attach the motor termi-nals to a programmable power supply Set the supply voltage limit at the rated motorvoltage, and the current limit to the rated peak current Start the current at the ratedcontinuous motor current with the appropriate torque on the dynamometer Adjustthe voltage on the power supply to obtain the desired shell temperature when level.Acquire the speed and torque from the dynamometer controller, and the voltageand current from the power supply Adjust the torque as needed to keep the wind-ing temperature at 85°C plus ambient Repeat the test at the next desired level.Acquire data at different speeds and torques to plot the SOAC The winding tem-perature rise equals approximately:

(Rhot/Rcold)−1+(ambhot−ambcold)



winding const×voltagehot×currenthot

Tempwinding=Rth (watts lost)

where torque is in ounce-inches Record the cold winding resistance at the beginning

of the test, the hot resistance at the end of the test, and the ambient temperature foreach measurement

9.3.7 Holding Tests

The holding torque test determines the current, speed, and temperature at a fied torque Place the test motor into a stand and couple the test motor to thedynamometer Attach a thermocouple to the shell to monitor the temperature rise.Attach the motor terminals to a programmable power supply Set the voltage on thepower supply to the rated motor voltage, the current to the rated continuous cur-rent, and the dynamometer to the desired torque Run at the desired torque until theshell temperature levels Acquire the speed and torque from the dynamometer con-troller, and the voltage and current from the power supply Record the cold windingresistance at the beginning of the test, the hot resistance at the end of the test, andthe ambient temperature for each measurement

speci-The holding speed test determines the current, torque, and temperature at a ified speed The holding current test determines the speed, torque, and temperature

spec-at a specified current The holding temperspec-ature test determines the speed, torque,and current at a specified temperature The PC monitors the specific parameter andcontrols the test to hold it constant, similar to the holding torque test The applica-tion determines which test will provide the most relevant data, and the PC automat-ically controls the test and acquires the data

The PC then prints data sheets for all tests, prints summary sheets for SOAC andholding tests, and graphs for SOAC, speed-torque, and holding tests For brushless dctesting, the servo drive takes the place of the programmable power supply The volt-age divided by the current provides an estimate of resistance During the SOAC testthe PC uses the servo drive to control speed and the dynamometer controller to con-trol torque to obtain the desired temperature

9.3.8 Torque Ripple*

The output torque of a dc motor at low speeds appears constant, but closer

exami-nation reveals a cyclic component called torque ripple, as illustrated in Fig 9.12 This

torque ripple results from the switching action of the commutator, from the ture reluctance torque and sometimes from the bearings

arma-Torque ripple usually constitutes a very small percentage of the rated outputtorque and proves negligible for most uses However, torque ripple may become crit-ical in some applications, thereby requiring a means of measurement

The apparatus illustrated in Fig 9.13 can accurately measure torque ripple, aslong as the moment of inertia of the measuring device remains much smaller thanthe motor moment of inertia (otherwise inertia filtering invalidates the ripple mea-surement)

voltage×current−speed×torque



1351.7

FIGURE 9.12 Torque ripple test.

FIGURE 9.13 Torque ripple test setup.

The percent peak-to-peak ripple torque equals:

T r=

9.4 MOTOR SPECTRAL ANALYSIS

This section provides examples of the value of sound, vibration, and current spectralanalysis for electric motors Since spectral analysis encompasses a tremendous range

of diverse technical areas, requiring many books to adequately cover, this sectionprovides primarily hands-on snapshots of a few uses for motors The Vibration Insti-tute in Willowbrook, Illinois, offers a wide range of literature and training for thosebeginning in this field

Figure 9.14 illustrates a typical time signal, with time on the horizontal axis andamplitude (usually in volts) on the vertical axis An oscilloscope commonly providesthis type of display After a fast Fourier transform (FFT), the horizontal axis changes

to frequency, as shown in Fig 9.15 However, the basic analysis techniques remainthe same, regardless of whether the electrical signal derives from sound, vibration,current, flux, surface finish, roundness, or any other parameter containing data of aperiodic nature For variable-speed motors, the waterfall display in Fig 9.16 simpli-fies the tracking of suspect frequencies with speed Fixed frequency problems such

as resonances and critical speeds stand out clearly as compared to speed-relatedproblems

peak-to-peak torque ripple (100)



average output torque

FIGURE 9.14 Time history amplitude graph (Courtesy of Pemtech, Inc.)

FIGURE 9.15 Spectral amplitude graph (Courtesy of Pemtech, Inc.)

FIGURE 9.16 Real-time waterfall display (Courtesy of Pemtech, Inc.)

9.4.1 Ball Bearing Analysis

The spectrum in Fig 9.17 illustrates a motor with predominantly inner race and balldefect frequencies in the sound spectra As shown in Fig 9.18, subsequent micro-scopic examination revealed false brinells on the side of the inner race closest to therotor Since the preload spring forces the balls toward the rotor, the inboard orien-tation suggests that the damage occurred following assembly However, this type ofmark results from extended periods of vibration on a nonoperating bearing, such asmight occur during shipment, but not during assembly If at all possible, motorsshould ship with horizontal shafts to avoid exciting the axial rotor-spring resonanceand to take advantage of the typically higher radial stiffness

Qualification of motor packaging on shaker and oscillatory tables will preventfalse brinelling from inadequate packaging; Figs 9.19 and 9.20 are examples of thisequipment Shock testing will further qualify packaging to prevent housing distor-tion or true brinells in bearings Swept sine wave tests will reveal resonances in prod-ucts or packaging that shipping vibration could excite Sine dwell tests at theresonance points will then ensure that normal transportation vibration would notdamage the product at these frequencies Spectral analysis of motors before andafter testing will reveal damage by changes in bearing frequencies or frequenciesassociated with other motor components that produce periodic vibration

Motors with true brinells tend to produce sound spectra with strong inner raceand ball defect frequencies and lesser outer race defect frequencies However, the

FIGURE 9.17 Log power spectrum of bearing showing inner race and ball defects.

nature of the defect will vary the contribution of each component and often willcause modulations that produce families of sidebands mingled with harmonics.These sidebands can greatly complicate bearing spectral analysis, particularly whencombined with large numbers of harmonics and sidebands from gears or other com-ponents The simple sound spectrum in Fig 9.21 includes mostly inner race harmon-ics generated by a severe true brinell as shown in Fig 9.22.

FIGURE 9.18 False brinell on inner race of bearing.

FIGURE 9.19 Shaker table.

Bearing contamination generally creates considerable white noise, combinedwith mixed bearing frequencies For example, the spectra in Fig 9.23 consist mainly

of white noise with some inner race and ball defect harmonics This bearing noiseresulted from contamination denting (see Fig 9.24) caused by very fine paper dustcontamination shown on the bearing housing in Fig 9.25 Harder particles such asgraphite brush dust will produce larger denting until the bearing mills the particlesize down At this point, ball indentations will produce large numbers of transfermarks on the raceways Because of the smaller race curvature lending less support at

FIGURE 9.20 Oscillatory table.

FIGURE 9.21 Sound spectrum with inner race harmonics, ∆x= 130.875 Hz.

the point of contact, the inner race usually suffers more damage than the outer race.Consequently, the ball and inner race defect frequencies tend to dominate the spec-tra and will often produce ball sidebands next to the ball harmonics, but separated

by the difference between the ball and inner race frequencies This type of spectrum(see Fig 9.26) almost always indicates hard particle contamination

FIGURE 9.22 True brinell.

FIGURE 9.23 Spectra consisting mainly of white noise with some inner race and ball defect

harmonics: (a) inner race,∆x=131.500 Hz, and (b) outer race,∆x= 77.500 Hz.

(a)

(b)

FIGURE 9.24 Denting caused by contamination.

FIGURE 9.25 Paper dust contamination shown on the bearing housing.

Although the bearing defect frequency calculations in Fig 9.27 appear forward, the presence of hundreds of intermingled harmonics, sidebands with unre-lated carriers, and difference frequencies can greatly complicate analysis Inaddition, field returns often generate bearing frequencies at variance with new bear-ings because of the effects of wear and slippage The spectra in Fig 9.28 furtherdemonstrate unexpected harmonics and sidebands resulting from unusual defects.The ball bearings in an optical encoder had 15 out of 18 cage ball pockets rubbing onthe shield, producing a mixture of cage harmonics and sidebands as well as a 15-times cage harmonics and sidebands A change in internal shaft and housing shoul-ders solved the problem Various unusual defects can take hours to diagnose,depending on the complexity of the product involved.

straight-FIGURE 9.26 Spectra indicating hard particle contamination: (a) inner race ball pass,∆x= 170.375

Hz, (b) double ball spin,∆x=141.750 Hz, (c) double ball spin,∆x=142.000 Hz, and (d) double ball

spin and inner race ball pass differential, ∆x= 28.375 Hz.

(a)

(b)

(c)

(d)

9.4.2 Magnetic Noise

Even though ball bearing defects cause the most concern in noise issues, magneticnoise remains the dominant contributor to overall motor sound For ac motors, vary-ing the voltage provides the easiest method of separating magnetic and mechanicalnoise Since magnetic sound varies roughly as the square of the voltage, cutting the

FIGURE 9.27 Bearing defect frequency calculations.