Chuck Yung
EASA Technical Support Specialist
A common observation about 2-pole machines fitted with sleeve bearings is the inherent weak magnetic centering force. The classic symptom is chronic axial movement: a 2-pole rotor drifting “to and fro” from the established magnetic center position. This article addresses the numerous causes of this phenomenon, colloquially referred to as “hunting.” Although the focus is on 2-pole motors, much of this information applies to sleeve bearing motors of any speed rating. Identifying the cause of a problem is good, but solutions are a lot more useful, so I’ve included those as well.
We can use magnets to describe how a motor works. Opposite poles attract; like poles repel. The magnetic field rotating within the stator turns the rotor, and magnetic force affects the axial position of the rotor relative to the stator core.
The basics required to explain axial “hunting” are:
- Opposites attract; like poles repel
- Inverse relationship between distance and magnetic force
- Gravity
- Rotor bar skew
- Aerodynamic forces
- Various combinations of the above forces
Attraction/repulsion
Consider an electric motor as a collection of magnets: a north and south pole constitute each pole-pair. Opposite poles attract. The more magnets (of equal strength), the stronger the attracting force. Axial centering forces are proportional to the number of poles. Two-pole machines—with only 1-polepair—have weak axial centering forces. The more pole-pairs, the stronger the forces acting to hold the rotor in its axial position.
The components that influence magnetic centering force are:
- Position of stator core ends relative to the rotor core ends
- Air gap between rotor and stator
- Level of the shaft
- Skew, if any, of rotor bars
- Ventducts, if present
- Line voltage
- No-load current
- Endring extension of the rotor, beyond the rotor core The axial centering force can be determined by: F=(K x E x I x [Ef + Df]) / L
Where: K = a constant, 0.02
E= line-to-line voltage of stator
I= No-load line current
L= Stack length (inches)
Ef = Sum of core end forces
Df = Sum of individual stator-rotor vent force factors
If the shaft is not level, gravity acting on the rotor produces an axial force towards the low end. If the machine has no vent ducts, then Df would be zero, while for a machine with a large number of vent ducts Df
Some applications hamper our ability to use limited endfloat couplings, as stipulated in the third column of Table 1 (that’s the part of MG1 many motor users miss.) Axial movement during operation can be far more than a nuisance. Expensive bearings and possibly couplings may be damaged, production interrupted, and vibration-monitoring equipment might give nuisance alarms.
The relative axial position of the rotor and stator is often mistakenly attributed to physical centering of the rotor within the stator core. In fact, the magnetic centering effect is on the center of mass of the laminated core. In most cases, the two—center of core length and center of mass— coincide. But the exceptions can be challenging.
Vent ducts
Vent ducts in the stator and rotor must be symmetrical with respect to the midpoint of the stacked core. When practical, manufacturers align the stator and rotor vent ducts to maximize the effectiveness of airflow required to cool the machine. For machines with higher peripheral speeds, a siren effect sometimes results from the close proximity passing of vent duct supports in the stator and rotor. Objectionable noise forces the designer to offset the vent ducts of the stator and rotor.
The rotor will seek the center of mass of the laminated stator core. If a laminated stator core has vent ducts and those ducts are not symmetrically spaced, the rotor will shift towards the end with the most iron.
There are two obvious ways this might occur. There are designs where the center of the iron is not the same as the midpoint of the stator core length. Irregular vent duct spacing, perhaps due to unique ventilation requirements, may be the reason (Figure 1). There are cases where an OEM designs a non-symmetrical motor, and someone later reverses the stator to change the mounting (e.g. F1 to F2). Such a dissymmetry might also be caused by an error of stacking during manufacture, or an improperly done restack of the stator (or rotor) core. In either case, the magnetic center position when loaded is liable to move from the unloaded position.
In the second example, it might be necessary to restack the stator core. An alternative is to machine false vent ducts in the rotor to correspond to the stator vent ducts, as described in EASA Tech Note 15.
Level
For 2-pole machines, with their weak magnetic centering forces, accurate leveling of the shaft is important. With most equipment installed decades ago, perhaps by contractors unfamiliar with the need for extremely accurate leveling, and foundations settling over time, it is not uncommon for a motor shaft to be off level. A 2-pole motor coupled to a pump can, upon shutdown, coast for upwards of an hour. If the motor is not level, significant sleeve bearing damage is possible during coast-down.
Tapered air gap
The air gap is the physical distance between the stator and rotor. If the air gap differs from one end to the other (tapered rotor or tapered stator bore), the rotor will be pulled towards the end with the smaller air gap (See Figure 2). The explanation is simple: There is a square relationship between magnetic force (F) and the inverse of the distance (d) between the parts.
F = (1/d)2
Figure 2. The result of tapered air-gap is the same, whether it is the stator bore or the rotor that is tapered.
If the air gap is 20% closer at one end of the machine, the magnetic attracting force is 144% greater (1.2 x 1.2 = 1.44) at the end with the closer air gap. The rotor will be pulled axially towards the closer air gap, moving the shaft off the expected magnetic center. There are crane motor designs, with tapered stator bore and rotor, which take advantage of this fact to release the brake.
A tapered rotor can result from machine tool wear during the manufacturing process or an offset tailstock on a lathe. For the service center, correction may be as simple as a very light skim cut of the rotor surface. Consultation with the OEM is strongly recommended to determine the maximum allowable air gap before machining the rotor OD. Tool speed must be controlled to avoid surface shorting of the rotor laminations.
Two recommendations from manufacturers are:
- Sharp tool at 400 – 450 feet per minute
- #4 radius tool at 850 feet per minute
A tapered stator bore is more difficult to correct. Machining the stator bore is fraught with the risk of a grounded winding. To rewind and restack a stator core is costly and time-consuming.
Mounting a rotating grinder on a tool post minimizes lamination smear that could increase core losses and reduces the chance of a lamination being pushed into a coil.
When the air gap is uniform on one end but offset on the other, the effect is similar to a tapered air gap. Although the axial displacement force is not as strong as with the tapered air gap, it can still be problematic.
It has long been known that the rotor should be concentric to the stator within 10% of the average air gap. For 2-pole machines, other factors— including reduced shaft stiffness—a tolerance of 5% can reduce problems with noise and vibration.
Rotor cage skew
Rotor cage skew is frequently used to smooth out the torque curve and/or reduce electrical noise (Figure 3). Torque developed in the rotor is perpendicular to the rotor bars. When the rotor bars are skewed, the result is an axial component to the torque. The greater the skew, the greater the axial portion of torque results from the skew and the more likely that the rotor will be displaced axially when the machine is loaded. The OEM skew should never be changed without discussing the situation with the manufacturer or a qualified engineer. Elimination of rotor bar skew can cause significant changes in the torque curve, as well as electrical noise.
Aerodynamic force
Aerodynamic force is especially strong in 2-pole machines, which often have directional fans located inboard of both bearing housings. Steep fan blade pitch not only increases the airflow but also the axial aerodynamic force.
When the aerodynamic forces resulting from opposing fans are not equal, the stronger fan may pull the rotor off magnetic center. The relative position of fan and baffle are factors, as is the distance from fan to bracket, and the duct friction in each end. WP-I and WP-II enclosures are sometimes designed with nonsymmetrical ventilation passages. Blocked openings can upset the balance of airflow.
As the rotor is displaced in the axial direction, the force required to restore the magnetic center increases. The rate at which the required force increases decreases with the distance from magnetic center. The increase in axial restoring force can be approximated by raising the increase in axial offset distance to the 0.75 power. For example, if the rotor is offset 0.04” vs. 0.08” the force increases by about the 20.75 power, or 1.7 times. If the offset were a factor of 4 instead, the increased force would be roughly 2.8 times as great (40.75 = 2.8).
Thrust-limiting auxiliary bearing
When the only negatives resulting from axial float are cyclical vibration and the actual movement, there is a mechanical solution. The solution is to install— on the opposite drive end—a ball bearing in an oversized housing, to prevent axial movement of the rotor.
The ODE bracket should be machined with a flat mounting surface perpendicular to the axis of rotation and a locating fit machined before the machine is assembled. During the test run, a magnetic center is marked, usually by scribing the drive end shaft where it exits the labyrinth seal. After running the machine, the shaft is positioned with the rotor on magnetic center and dimensions taken so that the ball bearing housing and stub shaft can be constructed and installed. The housing should be designed with enough depth to permit shimming of the captive bearing to hold the shaft on magnetic center.
Unbalanced voltage
Line voltage and current affect axial centering force. It may come as a surprise to know that unbalanced voltage affects the rotating magnetic stator field. Uneven magnetic force, combined with any of the above-listed factors, can cause axial oscillation of the rotor.
Causes and solutions for axial hunting of 2-pole rotors are summed up in Table 2. When all else fails, the endplay-limiting ball bearing can be relied on to stop axial movement, as well as dual magnetic centers.
There are two additional issues with magnetic centering of a rotor in a sleeve bearing motor. First, the magnetizing current (no-load amps); and second is the load current. If a magnetic centering challenge seems to be load related, we can focus on the load-related factors:
• Rotor skew; since torque is developed perpendicular to the rotor bars, and is proportional to rotor current, rotor skew effect is noticed with increasing load.
• Endring component; like skew, the current carried by the endrings varies with load.
Both of these contribute to the brief axial thrusting that may occur during starting, when the rotor current is several times greater than full-load current.
Dual magnetic centers
Tapered air gap, a tapered rotor, non-symmetrical stacking of the stator or rotor; these can result in dual magnetic centers. If the magnetic center changes from the no-load test in the service center to the customer’s loaded condition, those are prime candidates.
Unbalanced voltage might explain why the customer’s no-load shaft position differs from the results you had on the test panel. It might just be a case of a shaft that is severely out of level.
Print