Why do my motors get hot?

The symptom

You land after a normal flight, put a finger on a motor bell, and take it off again quickly. Maybe all four are like that. Maybe one is worse than the others.

The framing that makes the diagnosis fall out: a motor turns electrical energy into thrust, and heat is the fraction it failed to convert. Every degree of temperature rise is watts you paid for and did not get airborne.

There are only two places those watts go. Copper loss — current heating the winding resistance — scales with the square of current, so 40% more current makes roughly twice the heat. Iron loss — hysteresis and eddy currents in the stator laminations — rises with RPM, hysteresis roughly in proportion to it and eddy currents with its square, which is why iron loss turns punishing on a high-KV motor spun hard, even lightly loaded. Everything below asks which of the two you have inflated, and why.

The rule of thumb, honestly: if you cannot hold the bell for a few seconds, you have a problem. Warm is fine. Uncomfortable is a warning. Painful means you are approaching the temperatures at which magnet adhesive softens and the magnets can begin to lose flux — and a motor that has lost magnet strength draws more current for the same thrust from then on. One hot flight is unlikely to ruin it; a season of them will.

Check this in 60 seconds

1. Which motors are hot? All four means a build-level problem — propping, tune, or load. One means a mechanical or ESC fault on that arm, and one motor behaving differently from the other three is a diagnosis of its own. This question alone halves the search space.
2. Spin each motor by hand, props off. It should coast freely and quietly. Grinding, notchiness or side-to-side play is a bad bearing. A bell that binds at one point in the rotation is a bent shaft or a packed gap.
3. Pull the motor screws and compare their thread length against the depth of the mounting hole. A screw that is too long bottoms out into the windings. More below — it is missed constantly.
4. What throttle do you hover at? Above roughly half stick, the aircraft is working hard just to stay airborne, and no tune will make it run cool.

The ranked causes

1. The D-term is cooking them

The most common cause by a wide margin, and counter-intuitive because the aircraft flies fine.

D is a differentiator, and differentiators amplify noise. Every bit of gyro noise your filters let through gets multiplied by D and sent to the motors as a command. The motors obey — hundreds of times a second, at amplitudes you cannot see as motion, only hear as a whine. Those oscillating currents do no useful work at all. They go straight into copper loss, and since copper loss goes as current squared, a small oscillation on top of hover current makes disproportionate heat.

Hot motors after a gentle hover is the single most reliable sign that D is too high, or that filtering has collapsed. The aircraft was asked to do nothing, and burned energy anyway.

Fix it in order: drop D by 10–15% and re-fly; verify the RPM filter is working (bidirectional DShot enabled, motor poles correct — it fails silently); only then touch the low-pass filters, remembering that every hertz of filtering is delay added to the loop. If the motors are hot and the aircraft buzzes, read why does my quadcopter shake — same fault, seen from the other side.

2. You are over-propped or overloaded

If the aircraft hovers at high throttle, the motors sit near their thermal limit before the tune is even involved. This is arithmetic, not a settings problem. Every version of it is "too much load for that motor at that voltage":

• Too much prop. More diameter or pitch means more torque per rev, more current, quadratic heat. A prop one size too large runs a motor hot at any tune.
• Too much KV for the cell count. The motor spins fast, and that is the iron-loss term above.
• Too much aircraft. Payload creeps. A build that hovered at 40% throttle on its first flight hovers at 60% once you hang a camera, a gimbal and a bigger pack on it, and thermally it is now a different aircraft.

Aim to hover below half throttle, with real headroom for manoeuvring. Heavy platforms have little margin here, which is why 5 kg builds burn motors and 250 g racers do not. The fix is hardware: less prop, lower KV, more motor, or less aircraft. Nothing in the CLI will save you — and the same excess current heating the motors is sagging your pack as well.

3. Mechanical drag

Anything that makes the bell harder to turn becomes heat, because the FC just commands more current to hold the RPM it wanted. In descending order of likelihood:

• Motor screws that are too long. The screw bottoms out against the windings inside the bell. Sometimes it only adds drag; sometimes it slices the enamel, shorts a winding, and that motor burns current for nothing. It is the most-missed cause here, and it gets created every time somebody swaps a frame or uses "the screws that came in the bag." Add a washer if a screw is marginal.
• A dying bearing. Grit, water, or a hard landing. Spin it by hand and listen.
• A bent shaft, usually after a crash — the bell wobbles and rubs.
• A prop rubbing an arm, a wire, or a gimbal. Look for shiny marks and scored plastic.
• Debris in the gap between bell and stator. A steel splinter on a magnet drags every revolution.

The test that resolves all of these at once: swap the suspect motor onto a different arm. If the heat follows the motor, it is mechanical. If it stays on the arm, read on.

4. The ESC

An ESC that is not commutating cleanly wastes energy on both sides of the wire. A desync — the ESC losing track of rotor position — heats the ESC and the motor, and on a high-inertia motor it is often audible as a stutter under fast throttle changes.

On a hot arm with a healthy motor, check that the ESC firmware and motor protocol match what the FC is sending, that all four ESCs run the same firmware and settings, that the timing setting suits the motor, and that the solder joints are clean. A high-resistance connection is a heater, and it heats one arm only.

5. Airflow and ambient

Last, because it is rarely the whole story. A motor that runs acceptably at 10 °C sits near its limit at 35 °C in direct sun, and a long hover in still air is thermally the worst thing you can ask of one — no forward speed across the bell. If yours are hot only in those conditions, you may not have a fault. You have physics — and no thermal margin, which is what you draw on when something else goes wrong.

How to know you actually fixed it

Not "it feels cooler." Your hand is a bad thermometer, and you badly want your last change to have worked.

• Fly the same profile. Same duration, throttle, weather and pack. A cooler motor after a shorter flight tells you nothing.
• Check current, not just temperature. If the fix was real, hover current went down. Heat is the lagging indicator of something the log shows you immediately.
• The bar is that you can hold the bell, and that all four are equally warm. If you still cannot hold one for a few seconds, what you changed was not the problem — or not the only one.

Why this is worth simulating first

Two of the five causes are settled before the aircraft leaves the bench, and both are cheap in a simulator and expensive in a field.

Over-propping is arithmetic. A physics model with winding resistance, iron losses and battery sag tells you what a given motor, prop, cell count and take-off weight actually hover at — before you buy the props. D-term oscillation is control-loop behaviour, and the real firmware lets you wind D up until the motor outputs ring and watch the current climb, without committing real motors to the experiment.

The causes you cannot simulate are the ones you fix with a screwdriver: a long screw in the windings, a bearing full of grit. Do those on the bench.