Tom Bishop, P.E.
EASA Technical Support Specialist
Have you ever had to deal with a rewound motor that had high no-load amps? That is almost a rhetorical question as most of us have experienced this situation. The focus of this article will be on steps to take before rewinding in order to avoid the condition of high amps after the rewind.
Steps that should be performed on every AC stator rewind:
- Inspect the stator bore and rotor outside diameter for evidence of machining or damage.
- Record the original winding data exactly as found.
- Test the stator core before winding removal.
- Verify the winding data.
- Test the stator core after winding removal and cleaning. Applying these five steps will help avoid the vast majority of situations where a rewound motor will exhibit high no-load current. If these steps were not all followed and a motor has high no-load current, if possible, perform any steps above that were omitted.
Most of the time the cause of high no-load current will then be identified. The five steps do not address all of the factors that can cause high no-load current; however, our focus will be on these key preventive steps to avoid high no-load amps after a rewind.
Preventative steps to follow
The steps have been outlined in the order in which they are normally most effectively performed. Let’s go through the steps and the reasons for them in detail.
1. Inspect the stator bore and rotor outside diameter for evidence of machining or damage. The purpose of this inspection is to determine if any machine work has been performed that would have increased the air gap. An increase in air gap will increase the no-load current. Look for evidence of machine work on both the stator core bore and the rotor outside diameter. Either or both could have been machined. Also inspect for a stator-rotor core rub that resulted in smearing of laminations. That could short laminations together and increase losses and heating.
If it is suspected that the motor draws high no-load current and the motor is disassembled, check the length of the stator and rotor cores. Although rare, it is possible that the rotor and stator cores are of markedly different lengths, possibly due to an error at time of manufacture. Another possibility to check for is axial misalignment of the stator and rotor cores. If the stator extends beyond the rotor, the rotor beyond the stator, or one is longer than the other, the effect is as though the air gap was increased. That will cause the no-load current to increase.
2. Record the original winding data exactly as found. Take exact original winding data. Most of us are very careful about getting the connection, turns, pitch (span), and wire size correct. The jumpers and coil grouping may also be critically important. Incorrect jumpers can result in noise or reduced torque; and incorrect grouping can lead to circulating currents, noise, heating and reduced torque. The circulating currents will increase the no-load current. Be particularly vigilant with two-winding motors. The number of circuits or jumpers or coil grouping (or any combination of these) may be very unusual.
Another factor that has become more common is the use of unequal turns in a winding. Manufacturers are “tweaking” winding designs to obtain the highest efficiency. Make sure to count the turns for at least one complete group of coils. Rewinding with fewer turns than the original will increase the magnetic flux densities and the no-load current also will increase. Increased flux in a motor with flux densities near magnetic saturation tend to increase the no load current with increased applied voltage at a faster rate than designs with lower flux densities. For this reason, a slight over-voltage on a high flux design can result in an exponential increase in no-load current. Check no-load current on these designs at rated voltage. Slow-speed motors, e.g., 8 poles or more, also tend to have higher no-load currents (sometimes nearly full-load current) and are also sensitive to over-voltage.
3. Test the stator core before winding removal. Prior to the coil removal process, i.e., before cutting off a winding extension or placing the stator in the burnout oven, Avoiding High No-load Amps On Rewound Motors perform an initial core test. The core can, and should, be tested with the faulty winding in it. Core testing can be performed with a commercial core tester or by the loop test (see Tech Note 17 and “Proper Use Of The Core Tester,” CURRENTS, May 2003). Hot spots that may be identified could indicate that the core is so badly damaged that it should be replaced. It is certainly better to determine that condition before the stator is rewound. Test values that should be recorded include the watts per pound (kilogram) of loss, hot spot temperature, ambient temperature, and duration of the test.
4. Verify the winding data. The winding data can be compared to similar motors in the EASA winding database. If the data, air gap density and current density (circular mils per amp or A/mm2) are in line with data on file, that is increased assurance that the original data is correct. If the data is significantly different from that on file for similar motors, it should be verified by other means gap flux densities greater than 65,000 lines per square inch (10,000 gauss or 1 Tesla) may result in excessive or unexpectedly high no-load current (see “A Closer Look At The No-load Current,” CURRENTS, May 2001). Although a high current density will not affect no-load amps, it can overheat the windings. Typical lower limits are 330 CMA (circular mils per amp) for open motors and 450 CMA such as EASA technical support or the motor manufacturer. In addition to data you may have on file for comparison purposes, the EASA Motor Rewind Data CD has information on over 350,000 windings. Further assurance that the winding should not result in excessive no-load current can be attained by calculating the actual air gap, tooth and back iron (core) magnetic flux densities, and the current density.
If you have the EASA AC Motor Verification and Redesign program (See Figure 1), it is a good practice to check every three-phase winding. Air for totally enclosed motors. The metric equivalent limits are 4.70 A/ mm2 (amps per mm squared) and 3.44 A/mm2, respectively. Note that the metric units are inversely proportional to the CMA values, therefore the metric limits are maximum values.
5. Test the stator core after winding removal and cleaning. The core test should be repeated after the winding has been removed and the core cleaned and prepared for coil insertion (See Figure 2). A key to verifying the core condition is to test it in the ready-to-be-wound condition. That assures that any cleaning or preparation tasks did not affect the core loss. The core loss test values after winding removal should be comparable to the initial test of Step 3 above. The core should be repaired if the watts per pound (kilogram) increase more than 20% or hot spot temperature rise increases.
What can happen if we omit one or all of these steps? We will use two motors to illustrate the effect of an increase in air gap: an error in the number of turns, and an increase in core loss. The example motors are a 4-pole, 200 hp and a 2-pole, 15 hp.
Example 1: Increased air gap
For this example we will consider the effect of increasing the air gap of each motor by .010” (0.25 mm). The 200 hp (150 kW) motor had an original no-load current of 55 amps with an air gap of .028” (0.71mm). The increase in air gap to .038” (0.97 mm) resulted in a no-load current of 65 amps. The 36% increase in air gap caused an 18% increase in no-load current.
Now let’s look at the effect on the 15 hp (11 kW) motor. The no-load current with the original air gap was 11 amps and with the air gap increased from .020” (0.51 mm) to .030” (0.76 mm) the no-load current increased to 13.5 amps. The 50% increase in air gap resulted in 23% increase in no-load amps. A note of caution: The percentage increase in no-load current for both example motors is about half that of the increase in air gap. Do not try to apply this ratio as a rule-of-thumb to all motors; every design is different.
Example 2: Turns count error
Manufacturers are making increasing use of windings with unequal turns per coil. Our 200 hp motor originally had equal numbers of 5-turn and 6-turn coils, for an average of 5.5 turns. The unequal turns were missed and the motor was rewound with 5 turns in all coils. That means the new winding had 9% less turns than the original winding. The no-load current with the original winding was 55 amps and with the new winding the no-load current is 74 amps. The 9% turns reduction has resulted in a 35% increase in no-load amps.
The 15 hp motor would also be affected by a turns count error. In this case we will consider a winding that originally had 11 turns per coil, and was rewound with 10 turns per coil. The no-load current with the original winding was 11 amps and with the new winding it is 16 amps. The 10% reduction in turns has resulted in a 45% increase in no-load current. Two-pole motor designs generally use lower flux densities than lower speed (4- or more pole) designs. In this example the no-load current changed significantly. However, in many cases the lower flux density may “mask” the effect of increased flux because the relatively low flux density does not significantly increase no-load current if effective turns are reduced modestly.
The effect of incorrect turns will affect each motor design differently. As with the caution about no-load current and air gap, there are no “rules of thumb” to suggest the percentage change in no-load current associated with a percentage change in winding turns. We have used turns as a winding design change factor; however it is the change in flux densities that actually causes the increased current. A mistake in the pitch or connection can be as harmful, or worse, as an error in turns.
Note: Changes in wire size have almost no effect on no-load current. For example, increasing the wire size (area) will not cause the motor to draw more current.
Example 3: Increased core loss
Comparison of the initial core loss test results to the pre-winding test can reveal degradation associated with the winding removal process. Good practice dictates that the core will be repaired so as to correct any degradation. The factor that cannot be detected is the condition of the core “as-received” compared to “asmanufactured.” For economic reasons, manufacturers do not perform core tests on new motor stators. Here we will consider the case of a core that has had the watts per pound (kilogram) core loss double due to a prior fault, e.g., a ground fault that has damaged laminations.
The core loss of the 200 hp motor was originally 2.29 watts per pound (5.05 watts/kg) and increased to 4.58 watts per pound (10.10 watts/kg). The original 55 ampere no-load current increased to 127 amps with the doubled core loss wattage. That is, the 100% increase in core loss resulted in a 131% increase in no-load current.
The 15 hp motor had an original core loss of 1.61 watts per pound (3.55 watts/kg) that increased to 3.22 watts per pound (7.10 watts/kg). The no-load current in this case increased from 11 amps to 21 amps, a 91% increase.
The doubling of core loss watts may appear to be an extreme case. However, note that the watts per pound (watts/kg) values for both the original and as-found cores of both motors were within the range of generally acceptable core test values (0-6 watts/lb (0-13 w/kg)).
Prevention is the key
What do the examples tell us? In general, we do not want to do anything to the motor that will markedly increase the air gap, decrease the turns or increase the core loss. Of these three factors, the turns and core loss change had the greatest impact. The turns change represents an increase in magnetic flux. Therefore any change that increases flux can increase no-load current. The best solution to the problem of high no-load current is prevention. Follow the five steps and most, but rest assured not all, of your high no-load current problems with AC motors will be avoided.
ANSI/EASA AR100
More information on this topic can be found in ANSI/EASA AR100
Print