From commutation logic to sensorless startup, explained through a design-and-debug workflow
Editorial note: This is an independently rewritten educational article. It uses general engineering knowledge and a new narrative structure; it is not a translation or a substitute for permission to reuse any protected source text.
A brushless motor is easiest to understand as a closed loop: the controller creates a changing magnetic field, the rotor responds, and feedback tells the controller when to change the next phase. Once that loop is clear, the three wires, pole count, Hall signals, and back EMF stop looking like separate topics.
The goal here is practical: give UAV builders a mental model for selecting a motor, interpreting a datasheet, and diagnosing a rough startup.
1. Begin with the commutation loop
An ESC does not simply send power to a motor. It chooses which stator phases conduct, controls current, and advances the magnetic field so the rotor can keep following it. In a common six-step drive, two phases are actively driven while the third is left floating for part of the cycle.
A useful way to visualize the sequence is:
· Command: the flight controller requests torque or speed.
· Drive: the ESC energizes a selected phase pair.
· Motion: the rotor magnets align with the resulting stator field.
· Decision: the ESC estimates or measures the new rotor position.
· Advance: the ESC commutates to the next phase pair.
Design insight: Torque depends on the timing relationship between stator current and rotor position. A motor can have sufficient electrical power and still perform poorly when commutation is mistimed.
2. Read the motor as two magnetic assemblies
The stator is the stationary laminated core carrying the windings. The rotor carries permanent magnets and rotates around, or inside, the stator depending on the motor geometry. The mechanical arrangement is simple; the electrical timing is where most of the control problem lives.
The three external phase wires are normally the accessible ends of a three-phase winding. In a star-connected winding, the three coil groups meet at an internal neutral point that is usually not exposed. A delta connection joins the coil groups differently and changes the electrical behavior seen by the ESC.
Do not infer motor performance from wire count alone. KV rating, winding resistance, pole count, magnet strength, cooling, and propeller load all influence the final operating point.
|
Parameter |
What it changes |
Why it matters |
|
Pole pairs |
Electrical cycles per mechanical revolution |
Affects the relationship between commutation frequency and shaft speed |
|
Winding connection |
Phase impedance and current behavior |
Changes how the motor responds to voltage and load |
|
Winding resistance |
Copper loss and heat at a given current |
Helps explain efficiency and temperature rise |
|
Magnet and air gap |
Available torque and back-EMF amplitude |
Influences startup margin and sensorless detection |
3. Choose how rotor position will be known
There are two broad feedback strategies. A sensored system reads rotor position directly. A sensorless system infers position from the voltage generated by the motor while it is turning.
|
Method |
Strengths |
Limitations |
Good fit |
|
Hall feedback |
Reliable at low speed; predictable startup |
Extra wiring and sensor alignment; temperature and vibration can matter |
Gimbals, actuators, frequent start-stop motion |
|
Back-EMF sensing |
Simple motor assembly; fewer wires; economical at scale |
Weak signal near zero speed; startup needs a special routine |
Propulsion systems that spend most of their time rotating |
Back EMF is not an extra component inside the motor. It is the voltage induced in a non-driven phase as the rotor magnets move past its windings. The ESC can look for a characteristic crossing point and use it as a timing reference. This is why a sensorless motor may behave well in flight but hesitate, buzz, or twitch before it has built enough speed.
A practical sensorless startup sequence usually includes rotor alignment, a controlled acceleration ramp, and a handoff to back-EMF tracking. The exact implementation varies by ESC firmware, so startup problems should be separated from high-speed commutation problems during testing.
4. Turn the theory into a bench workflow
Before changing firmware or buying a larger motor, isolate the failure mode. Remove the propeller, secure the airframe, and change one variable at a time.
1. Check phase-to-phase resistance and look for a clear outlier.
2. Verify battery voltage, connector condition, and ESC current capability.
3. Confirm the motor pole-pair count and ESC configuration.
4. Test startup without a propeller, then repeat at a low controlled throttle.
5. Add the intended propeller only after startup and commutation are stable.
6. Record current, RPM, and temperature instead of judging performance by sound alone.
|
Symptom |
Likely area |
First useful check |
|
Motor clicks or jitters |
Startup alignment, phase order, feedback |
Remove the propeller and test a slow acceleration ramp |
|
Runs but heats quickly |
Over-prop, timing, resistance, cooling |
Measure current with the intended propeller and battery |
|
Smooth at speed but poor at launch |
Sensorless low-speed margin |
Try a longer alignment or gentler acceleration |
|
Vibration across the range |
Mechanical balance or pairing |
Inspect shaft, propeller, mount, and phase resistance |
5. A compact decision rule
For a typical multirotor propulsion unit, a sensorless ESC is attractive because the motor is expected to spin continuously and the wiring stays simple. For a mechanism that must hold position, start under load, or reverse frequently, direct position feedback is usually worth the added complexity.
Bottom line: Select the control method from the motion profile, not from the parts list. The best motor is the one whose torque, speed, thermal margin, and feedback method match the propeller and the aircraft mission.
Independent rewrite note: Technical facts and formulas may be shared across engineering references, but the wording, examples, sequence, and presentation in this article were newly composed for this document. For publication, keep records of your own experiments and obtain permission for any third-party images, diagrams, or quoted passages.
0 comments