Interior Permanent Magnet Motors Power Traction Motor Applications

Robust, magnets sparing designs provide quasi-constant power over a broad speed range.

For decades, surface permanent-magnet (SPM) motors dominated the market for permanent magnet motors. In recent years, however, the emerging hybrid vehicle market, and, to some extent rising rare-earth magnet costs, have boosted demand for interior permanent magnet (IPM) motors. With advantages like near-constant power over a broad speed range and a magnet-retaining design, IPM motors provide a good solution for applications like traction motors and machine tools. Let's take a closer look.

At its simplest, a rotary motor consists of a rotor and a stator, one of which is stationary while the other moves to perform work. Although a variety of permutations exist, for purposes of this article, we’ll be talking about designs with an outside stationary rotor and an inside cylindrical rotor coupled to the output shaft. The stator features a collection of copper coils that are energized in succession to generate magnetic flux. The stator flux interacts with the rotor, which is mounted inside the stator in such a way that it can turn to align itself with the fields generated by the coils. This alignment force generates torque, turning the rotor so that it moves the shaft and, hence, the load.

The turning force applied to the rotor can arise from multiple sources. In the case of permanent magnet motors, the primary source of torque arises from the interaction between the stator flux and the north/south poles of the magnets. As the name suggests, an SPM motor features slabs of magnet material applied to the rotor surface with adhesive and, in some cases, mechanical strapping or housings. The rotor of an IPM motor consists of a stack of metal laminations stamped with slots so that the stack forms a cavity parallel to the axis of the motor. The magnets fit into these slots, creating accurate alignment without need for special tooling and retains the magnets even in the face of high shock, vibration, and centrifugal force (see figure 1).

IPM motors produce torque based on two different mechanisms. The first is permanent-magnet torque which is generated by the flux linkage between the PM rotor field and the electro-magnetic field of the stator. It is the same torque produced by SPM motors. IPM designs produce a second force known as reluctance torque. The shape and location of the slots in the rotor laminations are designed to channel magnetic flux so that even if the slots were left as air gaps, the rotor would experience a force to align the magnetic flux lines with those generated by the stator coils. Because those coils are energized in sequence to create a rotating series of alternating north-south magnetic poles, the rotor will follow that progression, generating reluctance torque and causing it to continually turn.

Because IPM motor designs augment permanent magnet torque with reluctance torque, the magnets used in the motors can be thinner. That's important in the current market. Although the prices of rare-earth oxides (REOs) have dropped significantly from their peak in August 2011, rare-earth magnets still represent a significant cost source in permanent-magnet motor design, so IPM designs can provide a cost savings.

Winding designs
Two common variants of IPM motors are distributed-winding designs and concentrated-winding designs. A distributed winding design has multiple coils per magnet pole (24 slots to 4 poles, for example) whereas a concentrated winding has only a few (6 slots to 4 poles, for example; see figure 2). Distributed winding designs lend themselves to producing reluctance torque whereas concentrated windings do not. "Strictly speaking, it's an interior magnet machine but it turns out that the magnetics of the motor are designed intentionally to be more like a surface permanent magnet machine, so for that reason you don't get very much reluctance torque out of them," says David Fulton, director of advanced engineering for Remy International.

As a result of the reluctance torque, distributed winding IPM motors can produce more torque than the concentrated counterparts. Because distributed windings have longer end turns, they suffer from higher losses in the copper and can be more difficult to manufacture with automated winding techniques. They can also be more vulnerable to shorts. “There are areas where those phases overlap in space and were not for some insulation they would touch," says Fulton. "In concentrated-winding IPM motors and even switched-reluctance motors, those phases don't touch or overlap at all so the opportunity for that failure mode is not there in the concentrated wind types.”

Concentrated designs lend themselves automated manufacturing but they must be manufactured to tighter tolerances. Not surprisingly, they also exhibit more torque ripple than their distributed counterparts.

Concentrated windings are finding increased applications in hybrid vehicles with transverse engines. In a transverse engine, both motor and transmission need to fit between the front wheels of the vehicle. Concentrated-winding IPM's are shorter than their distributed counterparts, taking them a good fit for these designs.

Making trade-offs
Current sweet spot applications for IPM motors include traction motors for all variety of vehicles and high-speed machine tools such as CNC spindles.

Perhaps the biggest advantage of IPM designs, one that gives them an edge in vehicle applications like traction motors, is the high-speed performance. The power versus speed curve for SPM motors is roughly hyperbolic, rising to a region of quasi-constant power over a narrow speed range, then falling off.

IPM motors, in contrast, provide a much broader region of more or less consistent torque. Using a technique called field weakening, designers can apply current to modify performance. As speed rises, the permanent magnets and motor generate highers voltage. At very high speeds, the back EMF of the motor times the speed can exceed the voltage of the battery, which limits drive current, and torque. Field weakening essentially involves tuning the magnetic field of the stator to partially oppose the effect of the permanent magnets. The process involves a control scheme known as direct (D) and quadrature (Q) axis current control. The D-axis runs through the center of the rotor pole while the Q-axis lies between two adjacent rotor poles in the center. “By breaking the stator vector into two vectors, and applying one current to the quadrature axis and one to the direct axis, they control the current phase angle between them, which allows much wider constant power control," explains motor designer and IEEE fellow Jim Hendershot, co-author of Permanent Magnet Brushless Motors and Generator Design. 

For vehicle applications, the technique provides big benefits compared to SPM motors. “The IPM configuration allows more control over the magnetization of the magnetic circuit," says Hendershot. 

That's not to say that field weakening isn't possible with SPM designs, as well, but because of the size of the air gap, the technique requires far higher currents. “Because of the current limit of the inverter on thermal limitations of the motor, you can't field weaken it enough to produce torque at high speeds," says Fulton.

At low speeds, SPM motor and an IPM motor of the same size can generally produce about the same amount torque, or the SPM design may even produce a bit more up until they reach the corner point RPM. At speeds higher than the corner-point RPM, torque from SPM designs drops rapidly. “If both of them have a base of 3000 RPM, the SPM motor will probably have zero torque at 5000 RPM whereas the IPM could continue on to 10,000 or 12,000 RPM," says Fulton. The behavior makes IPM motors a good fit for traction motor applications, which tend to demand high torque over a broad speed range. “With IPM designs you get the best of both worlds—you can get very good acceleration at low speeds and then run at very high speeds while at almost the same power level.”

SPM designs can be more forgiving as far as manufacturing tolerances go. IPM motors require the smallest possible air gap between rotor and stator to maximize the reluctance torque. Tight tolerances can increase manufacturing cost, however. Budget constrained projects trying to minimize manufacturing costs may choose an SPM motor, although any cost-benefit must be balanced against increased magnet content. In context, though, a wider air gap may be a big benefit.

“It allows you to use larger tolerances, which can be a nice thing to have if you're doing a direct drive motor off an engine," says Fulton. "Those machines typically rely on the transmission bearings so you have a very large number of tolerances and they stack up. You also tend to have a lot of vibration off the engine that can cause the rotor to create [excess friction] between the rotor and the stator.”

The designs are different when it comes to thermal management, as well. In SPM designs, a significant amount of heat is generated in the magnets, which can cause demagnetization. Adding a small amount dysprosium to the magnet material can significantly enhance heat tolerance, but dysprosium is currently expensive.

IPM machines with distributed windings typically do not generate a significant amount of heat in the rotor—roughly 90% of the motor losses tend to occur in the stator, which can be easily cooled by heat sinking, oil cooling, or watercooling.

Given the wide range of designs available, it can be dangerous to make generalizations. That said, in an apples-to-apples comparison—same size, same power consumption—an SPM motor would deliver higher torque density. There are trade-offs, though. “Surface permanent magnet motors have the potential to have the highest torque density but you pay more because of the additional permanent magnet material," says Fulton. "You can get almost the same torque density with an IPM machine at a lower cost because you don't have to use permanent magnets get all your torque."

SPM designs also exhibit higher back EMF than IPM motors. This comes into play at a systems level—the inverter in the system can only tolerate a certain amount of back EMF before it requires switches and IGBTs with higher voltage ratings—and higher cost.

For high-speed applications like traction motors and machine tool spindles, an IPM motor can be a good fit. “You use less magnet material because you get some of the torque from the saliency of the rotor plus you have the superior control of the magnetization of the magnetic circuit by D and Q axis control of the current, which allows you to operate very efficiently over very wide speed range," says Hendershot. "The fact that the rotor is robust is a bonus.".