Oct 04,2026
Permanent Magnet Synchronous Motors: PMSM, FOC & Selection
Learn how PMSMs work, including SPMSM and IPMSM topology, d-q torque, FOC, MTPA, field weakening, sensorless control and selection.
1. PMSM Fundamentals: What “Synchronous” Means
The mechanical synchronous speed is determined by electrical frequency f_e and total number of poles P:
N_s=120f_e/P
Equivalently, electrical angular speed and mechanical angular speed are related by:
ω_e=pω_m
Where p=P/2 is the number of pole pairs. In a PMSM running in synchronism, rotor speed follows the commanded rotating-field speed. Unlike an induction motor, it does not require steady-state slip to induce rotor current. The practical system still experiences speed error during transients, limited controller bandwidth, compliance, and feedback/estimation error—but the magnetic operating principle is synchronous.
| Motor family | Rotor excitation | Steady-state speed relationship | Typical control requirement |
|---|---|---|---|
| Induction motor | Rotor current induced by slip | Rotor operates below synchronous field speed in motoring mode | V/Hz, vector control, or DTC. |
| PMSM | Permanent magnets | Rotor locks to electrical synchronous speed when controlled correctly | Electronic commutation with rotor position/angle information. |
| BLDC motor | Permanent magnets | Synchronous electronic commutation | Often trapezoidal/6-step, sometimes sinusoidal FOC. |
| Synchronous reluctance motor | Rotor saliency, no magnets | Synchronous under inverter control | Vector control with saliency-based torque production. |
2. PMSM Construction and the Role of Each Component
A typical PMSM contains a laminated stator core with three-phase windings, a rotor with permanent magnets, bearings, housing, shaft, encoder/resolver or sensor interface where applicable, and a three-phase inverter-driven control system. The motor may be integrated into a servo package, traction system, compressor, pump, spindle, robotics axis, or industrial variable-speed drive.
| Component | Function | Key engineering consideration |
|---|---|---|
| Stator laminations | Guide alternating magnetic flux | Core loss rises with frequency and flux density. |
| Three-phase windings | Produce the rotating stator magnetic field | Resistance, inductance, insulation, and thermal class determine performance. |
| Rotor magnets | Provide rotor flux without rotor excitation current | Material, temperature capability, retention, and demagnetization margin matter. |
| Rotor core | Carries magnetic circuit and mechanical loads | IPMSM geometry creates saliency; high-speed strength is critical. |
| Bearings and shaft | Support torque transmission and rotation | Electrical bearing current, load, lubrication, and speed limits require review. |
| Position feedback | Provides electrical rotor angle for commutation | Encoder, resolver, Hall sensors, or sensorless estimator must match duty. |
| Inverter | Applies controlled sinusoidal PWM currents | DC-bus voltage, current rating, PWM, EMC, filters, and protection are system-critical. |
3. Permanent Magnet Materials
Permanent magnets determine rotor flux, torque density, thermal behavior, cost exposure, and fault tolerance. Common material families include neodymium-iron-boron (NdFeB), samarium-cobalt (SmCo), and ferrite. A motor designer chooses magnet grade and geometry from the required operating temperature, demagnetization margin, corrosion environment, rotor speed, torque density target, availability, and lifecycle cost.
| Magnet family | Relative strength / energy density | Thermal and corrosion direction | Typical design use |
|---|---|---|---|
| NdFeB | Very high | Requires careful temperature and corrosion design; grade-dependent | Compact high-torque motors, servo axes, traction, compressors. |
| SmCo | High | Generally strong high-temperature stability, but costly/brittle | High-temperature or specialized aerospace/industrial service. |
| Ferrite | Lower | Cost-effective and corrosion resistant; larger magnet volume needed | Cost-sensitive drives and designs where size can increase. |
| Bonded magnets | Moderate, geometry-flexible | Application-dependent | Complex shapes and selected compact assemblies. |
A permanent magnet is not immune to thermal or magnetic stress. Excessive temperature, strong opposing d-axis current, fault current, inadequate rotor cooling, or poor material selection can reduce magnet flux or cause irreversible demagnetization. The motor’s allowed current, temperature, and field-weakening range must therefore be taken from the manufacturer’s operating envelope, not inferred from generic PMSM formulas.
4. SPMSM versus IPMSM
An SPMSM mounts magnets on the outside of the rotor core. It often has a relatively uniform magnetic air gap and small difference between d-axis and q-axis inductance. Its torque is predominantly magnet torque, its modeling is comparatively simple, and its rotor can be compact. Magnet retention becomes particularly important at high speed.
An IPMSM embeds magnets inside the rotor. This produces magnetic saliency: d-axis and q-axis inductances differ. Saliency adds reluctance torque to magnet torque, improves opportunities for maximum-torque-per-ampere control, and usually supports stronger field-weakening behavior. The embedded rotor can also provide mechanical protection to magnets at high speed, though the rotor geometry is more complex.
| Attribute | SPMSM | IPMSM |
|---|---|---|
| Magnet location | On rotor surface | Embedded in rotor laminations. |
| Saliency | Usually low; L_d≈ L_q | Meaningful; typically L_q>L_d. |
| Dominant torque source | Magnet torque | Magnet torque plus reluctance torque. |
| MTPA benefit | Limited relative benefit | Often important for current/efficiency optimization. |
| Field weakening | Possible but can be more constrained | Often favorable due to saliency and rotor topology. |
| High-speed magnet retention | Requires robust sleeve/bonding/retention design | Magnets mechanically protected within rotor. |
| Rotor design complexity | Lower | Higher. |
| Typical direction of use | Compact servos, low/medium-speed direct drive | Traction, high-performance drives, broad constant-power speed ranges. |
Texas Instruments notes that IPMSM saliency produces reluctance torque because of the difference between d-axis and q-axis inductance, and that MTPA selection of the current vector improves efficiency for high-saliency IPMSMs [2].
5. The d–q Axis PMSM Model
PMSM control is usually described in a rotating direct–quadrature (d–q) reference frame aligned to rotor flux. The d-axis is aligned with a rotor magnetic pole; the q-axis is 90 electrical degrees away. In this frame, the three sinusoidal phase currents become two controlled current components.
A standard voltage model is:
v_d=R_si_d+L_ddi_d/dt-ω_eL_qi_q
v_q=R_si_q+L_qdi_q/dt+ω_e(L_di_d+\lambda_m)
Where R_s is stator resistance; L_d and L_q are d- and q-axis inductances; i_d and i_q are d- and q-axis currents; ω_e is electrical angular speed; and \lambda_m is permanent-magnet flux linkage.
The general electromagnetic torque equation is:
T_e=3/2p\left[\lambda_m i_q+(L_d-L_q)i_di_q\right]
The first term is magnet torque. The second is reluctance torquecaused by saliency. In an SPMSM where L_d≈ L_q, reluctance torque is small and torque is predominantly controlled through i_q. In an IPMSM where L_q>L_d, a negative i_d combined with positive i_q can create positive reluctance torque.
Imperix presents the rotating-reference-frame PMSM voltage equations and a torque expression based on d/q flux and current components, explaining that d-axis and q-axis current components can be controlled independently in FOC [3].
| Quantity | Physical meaning | Control role |
|---|---|---|
| i_d | Direct-axis current, aligned with magnet flux | Flux adjustment, MTPA, and negative d-axis field weakening. |
| i_q | Quadrature-axis current | Primary torque-producing current. |
| L_d, L_q | Axis inductances | Quantify magnetic saliency and current response. |
| \lambda_m | Permanent-magnet flux linkage | Sets magnet torque/back-EMF contribution. |
| ω_e | Electrical speed | Couples voltage demand to speed. |
| p | Pole pairs | Converts electrical model to mechanical torque/speed. |
6. Worked SPMSM and IPMSM Torque Example
Consider a 4-pole PMSM with two pole pairs p=2, magnet flux linkage \lambda_m=0.12 Wb, and q-axis current i_q=50 A.
For an SPMSM with i_d≈0, torque is approximately:
T_SPM=3/2×2×0.12×50=18.0\ N·m
Now consider an IPMSM with L_d=0.5 mH, L_q=1.0 mH, i_d=-20 A, and i_q=50 A:
T_IPM=3/2×2\left[0.12×50+(0.0005-0.001)(-20)(50)\right]
T_IPM=19.5\ N·m
The additional 1.5 N·m in this illustrative IPMSM case comes from reluctance torque. At 3,000 rpm, 19.5 N·m corresponds to approximately 6.13 kW mechanical output:
P=Tω_m=19.5×2π×3000/60≈6.13\ kW
If a constant-power field-weakening region supported the same 6.13 kW at 6,000 rpm, torque would fall to about 9.75 N·m. This calculation illustrates the torque/speed trade-off; it does not define an approved overspeed range for any real motor.
| Example condition | i_d | i_q | Calculated torque | Interpretation |
|---|---|---|---|---|
| SPMSM reference | 0 A | 50 A | 18.0 N·m | Predominantly magnet torque. |
| IPMSM MTPA-style illustration | −20 A | 50 A | 19.5 N·m | Magnet torque plus reluctance torque. |
| IPMSM at 3,000 rpm | −20 A | 50 A | 19.5 N·m | About 6.13 kW shaft power. |
| Constant-power illustration at 6,000 rpm | Control-dependent | Control-dependent | 9.75 N·m | Torque approximately halves as speed doubles. |
7. Field-Oriented Control (FOC)
Field-oriented control (FOC)is the standard high-performance PMSM control approach. It measures phase current, obtains or estimates rotor electrical angle, transforms currents into d–q coordinates, regulates i_d and i_q with current controllers, transforms voltage commands back into three-phase quantities, and applies PWM through the inverter.
The basic objective is to align the d–q control frame with rotor flux. Imperix explains that rotor position must be known for this alignment and can be measured with an encoder or estimated with a sensorless technique [3]. TI likewise describes FOC as using three-phase current measurement, Clarke and Park transformations, d–q current references, current regulators, inverse transformations, and space-vector PWM to drive the inverter [2].
| FOC block | Function | Key design consideration |
|---|---|---|
| Current sensing | Measures phase current | Sensor offset, bandwidth, synchronization, and fault robustness matter. |
| Clarke transform | Converts three-phase currents to stationary α\beta frame | Requires correct current reconstruction. |
| Park transform | Rotates α\beta to d–q frame | Accuracy depends on electrical rotor angle. |
| i_d controller | Regulates flux/field-weakening component | Saturation and magnet protection require limits. |
| i_q controller | Regulates torque component | Current limit governs torque capability. |
| Speed/torque loop | Generates torque request | Must match load inertia and dynamic requirements. |
| Inverse transforms and PWM | Produce inverter gate commands | DC-bus voltage, PWM frequency, dead time, and EMC influence performance. |
8. Sensored and Sensorless Rotor Position
A PMSM cannot create controlled torque without correct electrical rotor angle. The system may use a resolver, incremental encoder, absolute encoder, Hall sensors, or sensorless estimation. The choice is determined by required speed range, zero-speed torque, dynamic accuracy, safety needs, environment, cost, and fault tolerance.
At high and medium speed, sensorless estimators can use back-EMF or model/observer information effectively. At standstill and very low speed, back-EMF is small, so sensorless control may require rotor alignment, an open-loop start sequence, high-frequency signal injection, or another specialized method. There is no universal sensorless algorithm that is optimal for every PMSM and load.
| Position method | Strengths | Limitations | Typical direction |
|---|---|---|---|
| Resolver | Robust in harsh, high-temperature environments | Cost and signal-interface complexity | Traction, aerospace, industrial servo. |
| Incremental encoder | High resolution and mature ecosystem | Requires homing/index logic; environmental protection | Servo axes, automation. |
| Absolute encoder | Knows position after power-up | Higher cost/protocol integration | Robotics, complex positioning. |
| Hall sensors | Simple commutation reference | Lower angular precision | Cost-sensitive drives and start support. |
| Sensorless observer | Eliminates feedback device and cable | Low-speed/start performance can be challenging | Pumps, fans, compressors, selected traction/industrial drives. |
| High-frequency injection | Enables low-speed sensorless saliency estimation | More control complexity, acoustic/EMC considerations | High-performance IPMSM systems. |
9. Maximum Torque per Ampere (MTPA)
Below base speed, an IPMSM controller often applies maximum torque per ampere (MTPA)control. MTPA selects i_d and i_q references to generate the requested torque with minimum stator-current magnitude. Because current creates copper loss and consumes inverter capacity, MTPA improves utilization and efficiency in saliency-rich machines.
For an SPMSM, the common idealized choice is i_d≈0 below base speed because reluctance torque is limited. For an IPMSM, negative i_d can generate useful reluctance torque, so the MTPA trajectory usually uses a nonzero negative d-axis current. TI states that MTPA is necessary to maximize the efficiency of high-saliency IPMSMs and explains that proper d/q current-vector selection combines magnet and reluctance torque efficiently [2].
| Operating objective | SPMSM tendency | IPMSM tendency |
|---|---|---|
| Maximum torque below base speed | i_d≈0, increase i_q within current limit | Follow MTPA curve with negative i_d and suitable i_q. |
| Minimum copper loss for requested torque | Primarily optimize i_q | Optimize current-vector angle, exploiting saliency. |
| Low acoustic torque ripple | Sinusoidal current/accurate angle control | Same, plus saliency/saturation model quality. |
| High load transient | Fast i_q regulation | Fast vector control with MTPA/current-limit coordination. |
10. Field Weakening and Constant-Power Operation
At higher speed, PMSM back-EMF rises. Eventually the inverter reaches its available output-voltage limit. To move above base speed, the controller applies negative d-axis current to oppose part of the magnet flux, reducing effective air-gap flux and therefore voltage demand. This is field weakening.
MathWorks describes field weakening as increasing motor speed above rating at the expense of reduced torque, using negative d-axis current to counter permanent-magnet air-gap flux. It describes MTPA below base speed, current and voltage constraints, and maximum-torque-per-volt operation in deep field weakening [4].
The operating boundaries can be represented conceptually by a current-limit circle and voltage-limit ellipse:
i_d²+i_q²\le I_max²
v_d²+v_q²\le V_max²
| Region | Primary control objective | Torque/speed behavior | Main constraint |
|---|---|---|---|
| Below base speed | MTPA or i_d≈0 control | Approximate constant torque | Inverter/motor current and thermal limit. |
| At base speed | Rated flux/voltage reached | Rated torque and power reference | DC-bus and modulation voltage limit. |
| Field weakening | Negative i_d reduces effective flux | Torque falls as speed rises; power may be approximately constant | Current, voltage, magnet, and mechanical limits. |
| Deep field weakening | MTPV/current-voltage constrained optimization | Maximum feasible speed/torque envelope | Voltage ellipse, current circle, magnet safety margin. |
Field weakening is not permission to overspeed a motor arbitrarily. Rotor mechanical stress, bearing speed, magnet retention, windage loss, acoustic behavior, cooling, gearbox/coupling limits, back-EMF fault voltage, and inverter voltage all define the approved maximum speed.
11. Thermal Design and Magnet Protection
PMSMs avoid rotor I²R loss from a rotor winding or cage current, but they still generate stator copper loss, core loss, magnet eddy-current loss, rotor-sleeve loss, bearing loss, inverter loss, and mechanical windage/friction loss. High speed raises iron and windage losses; high torque raises copper loss; field weakening creates negative d-axis current and can increase total current stress; poor cooling can reduce magnet strength.
| Thermal factor | PMSM-specific concern | Engineering response |
|---|---|---|
| Magnet temperature | Excess heat can reduce flux and demagnetization margin | Use magnet temperature sensing/modeling and derate per motor data. |
| Stator copper loss | High i_q and i_d raise I²R loss | Apply current limits, MTPA, and suitable cooling. |
| Core loss | Increases with frequency/flux waveform | Validate high-speed efficiency and lamination design. |
| Magnet eddy-current loss | PWM harmonics and spatial harmonics can heat magnets | Use segmentation, rotor design, filtering, and validated operating map. |
| Rotor retention | High speed increases centrifugal stress | Observe maximum rpm and rotor mechanical design limitations. |
| Low-speed cooling | Shaft fan may provide inadequate airflow | Use forced cooling or approved continuous-torque derating. |
| Field-weakening current | Strong negative i_d can stress magnets | Enforce manufacturer current/temperature/voltage limits. |
12. Inverter, PWM, EMC, and Cable Integration
PMSMs require an electronic inverter. Its PWM output supplies controlled sinusoidal current components but also introduces fast voltage edges, common-mode voltage, and electromagnetic interference. Motor cable length, shielding, grounding, output filters, and bearing protection must therefore be designed with the inverter and motor as a single system.
The same core VFD integration issues found in high-performance induction drives apply to PMSMs: reflected-wave voltage, insulation stress, common-mode bearing current, cable charging current, and sensor-signal noise. High-resolution feedback systems are especially sensitive to grounding and routing quality.
| Integration issue | What can go wrong | Design direction |
|---|---|---|
| PWM voltage edges | Winding insulation stress and reflected-wave overvoltage | Follow approved cable length; specify dV/dt or sine filtering where required. |
| Common-mode voltage | Shaft voltage and bearing electrical discharge | Apply grounding, shaft grounding, insulated bearing, or common-mode measures when system design requires. |
| Motor cable shielding | EMC and feedback disturbance | Use drive-compatible shielded cable and correct shield termination. |
| Encoder/resolver cable | Noise corrupts rotor angle | Route separately from motor power; use appropriate shield/ground method. |
| PWM frequency | Trade-off among current ripple, switching loss, audible noise, and EMC | Select within inverter/motor thermal and acoustic limits. |
| DC-bus voltage | Limits base speed and field-weakening range | Match inverter supply, motor back-EMF, and target speed envelope. |
13. PMSM Selection Workflow
PMSM selection starts with the complete torque–speed envelope, not rated kW alone. A motor delivering high torque at low speed, constant torque to base speed, and field-weakening above base speed may be ideal for traction or servo duty. A simpler SPMSM may be best for a compressor or direct-drive appliance. The motor, inverter, feedback, cooling, gear ratio, and process must be specified together.
| Step | Define this requirement | Selection impact |
|---|---|---|
| 1 | Continuous and peak torque vs. speed | Determines current, rotor topology, cooling, and inverter overload capacity. |
| 2 | Base speed and maximum overspeed | Determines pole count, back-EMF, DC-bus voltage, and field-weakening needs. |
| 3 | Load inertia and acceleration time | Determines peak torque and regenerative energy. |
| 4 | Duty cycle and ambient temperature | Determines thermal sizing and magnet grade/cooling. |
| 5 | Required precision at low/zero speed | Determines encoder/resolver versus sensorless control. |
| 6 | Size, mass, and torque-density target | Influences SPMSM/IPMSM and magnet choice. |
| 7 | Efficiency map and operating-hours profile | Determines lifecycle energy performance. |
| 8 | Motor cable length and installation environment | Determines insulation, filters, grounding, and EMC design. |
| 9 | Fault/safety conditions | Determines brake, safe torque off, overspeed, and back-EMF protection. |
| 10 | Service and supply-chain requirements | Determines magnet material, component availability, and maintainability. |
14. Applications
Microchip lists PMSM applications including air-conditioner and refrigerator compressors, direct-drive washing machines, automotive electric power steering, machine tools, traction control, and data storage [1]. More generally, PMSMs are attractive wherever a compact electronically controlled motor delivers value through energy efficiency, dynamic response, or precise speed/torque control.
| Application | PMSM value proposition | Typical topology/control direction |
|---|---|---|
| Servo axes and robotics | High torque density and fast dynamic torque response | SPMSM or IPMSM with encoder/resolver FOC. |
| Electric traction | Wide speed range, high power density, field weakening | IPMSM with MTPA, field weakening, robust feedback/sensorless control. |
| HVAC and refrigeration compressors | Efficiency and variable-speed capacity | PMSM with sensorless FOC. |
| Pumps and fans | Efficiency across a variable operating profile | PMSM with inverter speed control. |
| Machine-tool spindles | High speed, precision, field weakening | PMSM with high-resolution feedback and careful mechanical design. |
| Direct-drive appliances | Reduced mechanical transmission and quiet operation | SPMSM/BLDC-derived sinusoidal control. |
| Precision medical/automation equipment | Accurate, smooth speed/torque behavior | PMSM servo with encoder/resolver. |
| Renewable-energy systems | High-efficiency electromechanical conversion | Application-specific synchronous PM machines and power electronics. |
15. Maintenance and Troubleshooting
PMSMs eliminate brushes and mechanical commutators, reducing one category of maintenance. The principal reliability focus shifts to bearings, cooling, rotor feedback, cable/connector integrity, inverter health, insulation, grounding, and parameter management. A PMSM failure investigation should consider the motor and inverter together.
| Symptom | Possible causes | First engineering checks |
|---|---|---|
| Failure to start or rough startup | Incorrect rotor angle, sensorless alignment issue, phase order, current limit | Verify encoder/resolver polarity, phase sequence, startup parameters, and current sensing. |
| Low torque | Incorrect motor parameters, i_q limit, angle error, inverter undervoltage | Check d/q current data, DC bus, feedback angle, and parameter set. |
| High current with limited output | Bad FOC tuning, saturation, incorrect L_d/L_q/flux data, mechanical overload | Validate motor identification, load, current sensors, and control limits. |
| Overspeed or field-weakening fault | Voltage limit, incorrect negative i_d control, speed feedback error | Check DC bus, speed loop, field-weakening settings, and mechanical rpm limit. |
| Overtemperature | Excess torque, low-speed cooling, high ambient, magnet/core/stator loss | Review duty cycle, cooling, temperature sensors, and current waveform. |
| Bearing noise or fluting | Mechanical load or common-mode electrical current | Inspect grounding, cable shielding, bearing protection, and vibration spectrum. |
| Position feedback fault | Damaged cable, resolver/encoder failure, EMI, connector issue | Inspect shields, routing, supply, signal integrity, and mechanical coupling. |
| Intermittent inverter trips | Loose terminals, cable fault, insulation degradation, EMI | Review fault log, insulation, grounding, cable routing, and bus conditions. |
Conclusion: PMSM Performance Comes from the Complete Motor-Control System
A permanent magnet synchronous motor combines rotor magnet flux with electronically controlled stator current to deliver compact, efficient, synchronous motion. SPMSMs emphasize simple magnet-torque operation; IPMSMs add rotor saliency, reluctance torque, MTPA optimization, and often broader field-weakening capability. In both cases, the motor’s real performance is inseparable from the inverter, feedback system, control algorithms, cooling, cable, and driven load.
The most important engineering practices are to match the PMSM topology to the required torque–speed envelope, use the d–q model and FOC appropriately, respect current and voltage limits, protect magnets from excessive thermal and demagnetizing stress, evaluate field weakening as a constrained operating region, and design PWM/EMC/grounding as a system. When these requirements are met, PMSMs provide a high-performance foundation for modern motion, traction, appliance, compressor, and industrial-drive applications.
References
Editorial note: This article is intended for engineering education and preliminary product selection. Final magnet grade, motor parameters, inverter current/voltage ratings, maximum speed, field-weakening current, thermal limits, cable/filter selection, feedback architecture, functional safety, and commissioning settings must be confirmed using the exact motor and drive manufacturer documentation and validated by qualified personnel in the intended application.
Related News
LATEST
INFORMATION
Get the latest product information of the company
NAVIGATION
PRODUCTS
CONTACT US
Telephone: +86 13305761511
Email: info@cntecho.com
Add: RM. 602, BUILDING B, NO.1551 SHUANGSHUI RD., LUQIAO DIST., TAIZHOU CITY, ZHEJIANG, CHINA
Copyright © 2026 TECHO ELECTRICAL & MECHANICAL (TAIZHOU) CO., LTD. All Rights Reserved.