So your motor is already installed — what tests can be performed in-field to determine its condition?
Today, many processes require continuous operation of motors. Failure — because of the expense of production loss — is not an option. Luckily, proper maintenance, monitoring, and testing can extend motor life and eliminate unplanned downtime.
One reason for measuring a motor’s operating current is to determine how much power it’s producing. Suppose a motor’s full-load current is 100 A and it is drawing 75; the motor is producing threequarters of its nameplate horsepower, right? This is a false assumption because as the load on a motor increases, its power factor improves. This phenomenon is very common in modern motors. As a result, measuring only current will not give an accurate indication of the motor load. Approximate load may be determined by comparing measured current versus the motor performance data.
No-load currents vary widely from motor to motor because of differences in efficiency, power factors, and design; uncoupled, these motor currents can range from 30% to 50% of full-load current. As long as measured values of all three current phases are near the motor’s typical no-load current value, within 1% of each other, and with balanced voltage, current consumption is satisfactory. No-load current can also be determined from the motor’s performance data sheets.
Measuring current is done with a portable clamp-on current transformer with an appropriate voltage insulation level. If the equipment has a panel-mounted amp meter or ammeter, all the better. The current signature can be used to identify motor problems such as broken rotor bars or loose laminations. These faults can be detected by analyzing current with a spectrum analyzer or through a computer program. These faults are very rare, and the procedures followed to perform the tests depend on the equipment being used.
Voltage measurements on operating equipment should be made as close to the motor as possible, so that the voltage drop of the motor’s feeder doesn’t influence results. For most industrial installations the motor starter is close enough to the motor so that measurements are satisfactory. With appropriately rated equipment, the phase-tophase voltage of all three phases should be measured. Each measured value should be 6 10% of the motor’s nameplate voltage, and all measured values should be within 1% of each other. Next, the voltages in the three phases to ground should be measured; ideally these voltages should be equal to each other and equal to the phase to phase voltage divided by = 3.
If phase-to-ground voltages are not equal, it implies that ground fault problems exist that must be immediately corrected. If this unbalance is significant and not corrected, the motor’s insulation system will be severely overstressed, resulting in significant reduction of motor life.
The measurement of a running motor’s wattage is the most accurate measurement of a motor’s produced work.
Approximate values of efficiency and power factor can be obtained from charts. Today’s high quality, easy-to-use instruments make taking power readings relatively simple.
Non-contact infrared pyrometers make abnormal hot spots, as well as bearing, air flow, and cooling problems, stick out like a sore thumb.
Standard motors are designed to operate with a maximum total winding temperature rise of 80°C (by resistance) for NEMA class B or 105°C (by resistance) for NEMA class F above 40°C ambient. (The motor’s insulation class can be found on the motor’s nameplate.) If the surface of a motor approaches these temperatures, it may indicate a winding or cooling problem; the motor winding temperatures should be measured or the motor should be taken out of service as soon as possible so that the problem can be fixed.
Anti-friction or ball bearings should not be operated above 130°C. Elevated temperatures degrade their greases and may begin to anneal rolling elements. If exact bearing temperature measurements can’t be made, approximate measurements will suffice. If the bearing box or shaft next to a bearing approaches 100°C, the bearing is probably near 130°C and is in danger of sustaining thermal damage. The motor should be shut down immediately and the problem rectified. Sleeve or babbitted bearings have a temperature limitation of about 110°C. The bearing material begins to get soft at about 130 to 170°C, depending on whether the bearing is made from tin or leadbased materials. If there are no thermometers or temperature detectors embedded in the bearing, an indication of the bearing’s temperature can be obtained by scanning the bearing box and shafting next to the bearing, and removing sight plugs to inspect oil rings and bearing metal.
Continue on page 2
Bearing temperatures anywhere near 90°C should be monitored continuously to determine if they are increasing. If they do increase, the motor in question should be shut down for repairs. If temperatures are constant, they should be monitored at six to twelve hour intervals to make certain they remain constant.
Routine vibration measurements taken at regular intervals across the entire equipment train can spot problems well in advance of failure. Measurement frequency depends on the importance of the equipment. The most critical equipment should incorporate permanent vibration monitors. Routine vibration measurements should include high-frequency spectrum analysis on anti-friction bearings. Such spectral analyses indicate bearing failure well in advance of serious problems. Foundation and equipment bases should be checked regularly for movement or looseness. Testing is performed by placing a seismic vibration probe at locations of interest. Any movement greater than 25% of the equipment’s normal vibration indicates looseness and should be investigated.
Also known as the Megger test, high-voltage ohmmeter testing is performed with the motor out of service. It is one of the most common tests performed on electrical equipment because insulation system failure is a very common electrical problem. Motor insulation may deteriorate because of contamination, mechanical movement, cracking, mechanical impact, exposure to solvents, and many other factors.
To test for loss of insulation integrity, a voltage is placed across the insulation — at motor leads to frame. This voltage is usually dc and larger than the normal operating voltage, but not large enough to damage the insulation. This voltage stresses insulation so that if a weak area exists, it will become evident.
Motor insulation resistance is very temperature sensitive. Note: The higher the insulation resistance, the better. The equation for calculating insulation resistance at 40°C states that:
Rc = Kt × Rt
where Rc = insulation resistance (in Mv) corrected to 40°C Kt = insulation resistance temperature coefficient at temperature t Rt = measured insulation resistance (in MV ) at temperature t
IEEE Standard 43 recommends a minimum winding resistance per the following calculation:
Rm = kV + 1
where Rm = recommended minimum insulation resistance (in MV) at 40°C of the entire machine winding kV = rated machine terminal to terminal potential, in RMS kilovolts
Winding polarization index (PI)
This test is also performed with the motor out of service, and indicates the general condition of the winding insulation system. It determines the amount of dirt or moisture in windings, how much deterioration of insulation has occurred, whether the insulation is suitable for hi-potting — high-voltage testing of large motors — and the general suitability of a motor for continued operation.
The motor insulation resistance in question is usually tested at 500 or 1,000 V for 10 minutes. Readings are taken routinely to insure that the value of insulation resistance increases with time and that there is no insulation deterioration. The 10-minute insulation value is then divided by the one-minute insulation value to obtain the polarization index.
A low PI index may result from a dirty or wet insulation system. The number should never fall under two for a motor; anything below that indicates that the machine will not provide reliable, long-term service. Servicing the motor may raise the PI index above two.
Performed with the motor out of service, measuring winding resistance reveals open or shorted sections of winding. There are three standard methods used to conduct this test.
Continue on page 2
1. Measure phases A to B, then B to C, and finally C to A with an ohmmeter. It is important to use a highly accurate instrument that has a high resolution at low resistance values for this test to be of any value. The three values should be within 1% of each other; if variation between the measurements is more than 1%, use Method Two.
2. Measure the reactance of the winding by repeating the above test — except use an adjustable source of ac and a precision voltmeter and ammeter instead of an ohmmeter.• Connect the ammeter in series with the variable ac source and two phases of the motor. • Connect the voltmeter across the two phases being tested. • Increase the voltage to a convenient value that allows no more than 25% of full-load current to flow in the winding. • Record the readings and then calculate the impedance using Ohm’s Law (E=ZI). Repeat the test on the other two phases using the same value of applied voltage. If the difference in calculated impedance for each phase is more than 5%, there is probably a winding problem.
3. Use a surge comparison tester. The surge comparison test probes a motor’s coils for shorts or opens by comparing one set of coils with the motor’s other coils. This test is very definitive and can reveal subtle faults in a motor’s winding. Because the test is very equipment specific, follow the instructions provided with the test equipment.
Before performing any tests on a motor, you should be familiar with the testing equipment being used, have the proper personal protective equipment, and know the risks involved in testing motors. Know how to safely work with energized rotating equipment.
Though most people know they’re facial tissues, the name Kleenex has stuck. Megger Group Ltd. of Dover, England enjoys a similar fame. The name Megger dates back to 1889, when the company introduced the first portable insulation tester. Now the brand name is so well known that maintenance professionals often use it as a verb to refer to any wire insulation testing.
Looking out for the little guy
Single and polyphase fractional horsepower motors outnumber integral-horsepower motors in industry, so it’s important to know of special servicing and testing they require. Many big-motor tests apply; comparing currents and voltages to nameplate values, measuring surface temperatures, and monitoring insulation resistance can also help determine the condition of fractional horsepower motors. Still, standard and traditional testing for new small-motor applications is sometimes insufficient. To find out when this is the case, the authors asked Joe DePrisco and Brent Lahey of Adaptek Systems Inc. some size-specific questions; following are the small-motor tips they provided.
Are small motors typically tested in-field?
Joe: Unless there are suspected problems with the drive system, typical motors are not regularly field-tested in the end application. Where possible in higher volume production and critical applications, small motors may be best tested in the final configuration in production to establish how all variables will affect the motor such as gear trains, linear actuators, variable loads, and part tolerance variation issues. In these cases, production testing all small motors is highly desirable if, in conjunction, a database of critical performance values is utilized for design feedback, in-field checks, and performance improvement.
Brent: Field testing of small motors is also limited by instrumentation available. One common test performed with a portable digital multimeter, or DMM, checks winding resistance. A quick measurement of each winding resistance can be compared with the rated nameplate resistance to determine a bad or failed winding in the motor. DMM temperature adapters allow the testing of surface temperature under operation, which could identify problems with failing windings or bearings.
When do small motors require special testing?
Joe: Because small motors are finding their way into many emerging applications, testing them is increasingly tailored to uncover design issues or evaluate performance consistency. The application and a strong understanding of the end-use help define what tests are really needed. As an example: In some end-use applications a simple, lowcost motor may only need resistance or speed testing. But if this same motor was used and failed in some critical automotive applications, it could cause significant liabilities, magnified by the volume produced, safety issues, and high replacement costs. In these cases, more extensive testing is easily justified.
Another example: Some small motor applications now include very low or high rpm units. Final configuration output speed can be rated in fractions of rpm to over 100,000 rpm. Obviously, vibration testing for the second scenario may be critical, and not even considered for the first.
Brent: Demanding applications, new technology, extended life requirements, environment, and motor control electronics often dictate customized testing beyond the industry basics to ensure each motor’s performance or life. Tests for response time, high-speed waveform, or high or low temperature can be integrated with the more basic tests for a total picture of the motor’s capabilities.
More specialized testing is probably in the works. Increasingly exotic failure modes might spur the reinvention of motor testing as nano-motors (built molecule by molecule) develop. Needless to say these microscopic motors will push the testing limits of new technology motors even further.
Does the physical size of small motors ever require special testing setups?
Brent: Smaller motors typically have tighter mechanical and electrical tolerances. It’s not uncommon for a small motor to have a winding resistance of less than one 1 Ω. By manufacturing standards, this would require the capability to repeatedly measure under 0.1 Ω, with an instrument accuracy of 0.01 Ω. This starts getting into specialized test equipment.
For more information, circle 223 on the Reader Service Card or visit www.adapteksys.com.