By the Baoluo Motor Engineering Team · Published 2026-10-22
A wind farm called us about a yaw motor that kept throwing a winding over-temperature alarm. The first instinct on the other end of the phone was to send the motor back as a factory defect and demand a replacement. We told them to hold off. Before anyone ships a motor back, somebody has to measure it, and somebody has to read the control log. Both of those led somewhere the customer was not expecting: the motor was not the root cause. The wind farm's own yaw mechanism had seized, stalled the motor, and burned the winding from the outside in.
This article walks through that failure, the measurements that pointed the right way, and why the answer was sitting in the control data all along. If you run DC compound wound motors on a wind turbine yaw system, the same sequence plays out more often than anyone likes to admit.
Rule of thumb: a hot winding is almost never the beginning of the story. It is the last link in a chain. Find the link before it — the stall, the jam, the over-current — and you stop shipping good motors back to the factory for nothing.
1. The call: a yaw motor flagged for overheating
The machine was an XQD-4-21, a 24 V, 4 kW DC compound wound motor built for a wind turbine yaw drive. The yaw drive is the system that turns the whole nacelle on top of the tower so the rotor faces the wind. Its job is short and brutal: turn a little, hold, turn back a little, hold, thousands of times across the life of the turbine. That stop-start, forward-reverse duty is exactly the kind of work a compound wound DC motor is chosen for, because it delivers the high starting torque of a series winding with the speed stability of a shunt winding.
The alarm came first as a winding temperature warning, then as a trip. The turbine's protection system refused to let the yaw drive run until someone looked at it. The operating company's first reading was that the motor windings had failed, and they wanted it swapped under warranty.
We asked for two things before agreeing to anything: a resistance measurement of the windings, and the yaw control system's log of the days leading up to the alarm. Those two requests turned the whole diagnosis around.
2. The first measurement told the real story
Resistance first, because it is cheap and it cannot lie about a copper circuit. The customer measured the rotor (armature) winding and the stator field winding with a precision micro-ohmmeter. Here is what came back.
| Winding | Failed motor | Known-good motor | Ratio |
|---|---|---|---|
| Rotor (armature) | 71.226 mΩ | 38.371 mΩ | 1.86 × |
| Stator field | 6.8 mΩ | 7.0 mΩ | 0.97 × (normal) |
Two numbers jump out. The armature measured 71.226 mΩ against 38.371 mΩ on a healthy motor — that is 1.86 times the normal value. The field winding, at 6.8 versus 7.0 mΩ, was effectively unchanged.
The armature reading is the giveaway. In a DC motor the armature winding is made of several parallel branches. A reading of roughly double the healthy value is the classic signature of one of those parallel branches going open — a broken or cold-soldered connection in the armature circuit. The current that used to flow down two branches now squeezes down one, the single branch carries double the current, and the heat it generates goes up by the square of the current. That is how a winding that is not "burned out" as a whole still trips an over-temperature alarm.
The healthy field winding is equally important. It says the stator did not short, did not ground, and did not soak up moisture. The fault is pinned to the rotor circuit: the winding itself, the riser solder joints, or the commutator contact surface.
3. What the numbers ruled out
Before pointing at any one cause, we ran the reading through a short list of suspects. Each one predicts a different resistance behaviour, and the measurement eliminates most of them on the spot.
| Suspect fault | What resistance should do | Does it fit this case? |
|---|---|---|
| Turn-to-turn short | Resistance drops | No — measured value went up |
| Moisture ingress or full burnout | Erratic, open, or grounded | No — value is stable, not open or grounded |
| One parallel branch open | Resistance roughly doubles | Yes — 1.86 × fits (slightly under 2 is residual contact resistance at the break) |
| Bad measurement (brush contact) | Readings drift high | Unlikely — the field reading came out normal on the same meter |
So the winding had a real open branch, not a vague "it got hot." But that only tells us what broke. It does not yet tell us why. A solder joint can open on its own as a manufacturing defect, or it can be torn open after the motor was cooked by a stall. To separate those two, we needed the timeline — and that is where the control data comes in.
4. Why a stalled motor burns its own winding
To read that timeline, it helps to know what a stall does to a DC motor. A running motor generates a back EMF — a voltage that rises with speed and pushes back against the supply. In a DC motor the back EMF is E = k · n · Φ: it scales with speed n and field flux Φ. The armature current is the difference between supply voltage and back EMF, divided by the armature resistance:
I = (U − E) / R
Now jam the rotor so it cannot turn. Speed drops to zero, the back EMF vanishes with it, and the current is limited only by the resistance of the armature circuit: I = U / R. That locked-rotor current is several times the rated current. And because resistive heating follows the square of the current (P = I²R), a current that doubles produces four times the heat, a current that triples produces nine times. The copper, the insulation, and the solder joints all take that heat in the worst possible place — inside a nacelle, on top of a tower, with poor cooling and nowhere for the heat to go.
That is the mechanism that connects a mechanical jam to a burned winding. The stall is not a motor fault. It is the motor doing its job — trying to turn against something that will not move — until the heat breaks the weakest part of the armature circuit. On this design that weakest part is usually a commutator riser solder joint, which fatigues and opens under repeated thermal cycling.
5. The control log closed the case
This is the step that changed the conclusion. We asked the customer to pull the yaw control system's logged data for the days before the alarm. Two things showed up in that log, sitting right next to each other:
- A locked-rotor event. The log showed a stretch of time where the yaw motor was commanded to move but the rotor was not turning — a stall.
- An over-current event. In that same window the armature current spiked far above its normal running value, exactly as the formula above predicts.
The sequence in the log was unambiguous: mechanical resistance stopped the rotor, the stall collapsed the back EMF, the current shot up, and the winding overheated. Only after that heating cycle did the armature branch let go, which is when the resistance jumped to 1.86 times normal and the over-temperature alarm locked the drive out.
That ordering settles the earlier question. The open branch was not a manufacturing defect waiting to happen. It was the result of the stall. The motor failed because the mechanism it was bolted to could not move, not because the motor was built wrong. Sending it back for a warranty swap would have put a brand-new motor onto the same jammed yaw drive, and that one would have failed the same way a month later.
6. The mechanical fault behind the stall
Once the control log proved a stall, the search moved from the electrical side to the mechanical side of the yaw drive. A yaw system stalls for one of a short list of reasons, all of them mechanical, and all of them worth checking before you blame the motor:
| Mechanical fault | How it stalls the motor |
|---|---|
| Yaw ring / slewing bearing misalignment | Pinion and ring teeth bind instead of meshing, locking the drive |
| Worn or dry slewing bearing | Friction climbs until the drive cannot overcome it |
| Yaw reducer gear jam | A chipped tooth or foreign object locks the reduction stage |
| Foreign object in the gear ring | Debris caught in the teeth acts as a hard stop |
| Grease thickening in cold weather | Lubricant resistance rises sharply, slowing or seizing the drive |
| Yaw brake not fully releasing | The brake drags against a motor told to turn |
Which specific one of these this particular turbine had is a matter for the on-site teardown. The point of the article is not the exact part number of the stuck bearing — it is the order of operations. The customer proved, from their own data, that a mechanical fault caused the stall, the stall caused the over-current, and the over-current caused the winding failure. That is a very different conclusion from "the motor was defective," and it changes who pays, what gets fixed, and whether the replacement motor will survive.
7. What this means for your yaw drive
If you operate wind turbines or maintain any machine with a DC drive motor, this case gives you a three-step habit that costs almost nothing and prevents almost everything:
- Measure before you blame. A micro-ohmmeter reading of the armature and field windings takes minutes and tells you whether the copper circuit is open, shorted, or intact. A roughly doubled armature resistance points straight at an open parallel branch.
- Pull the control log before you ship anything back. Look for a stall flag and an over-current spike sitting in the same time window. If they are there, the motor was a victim, not a culprit.
- Check the mechanical side at the same time. A yaw ring, slewing bearing, reducer, brake, or a lump of cold grease can all produce the same stall signature. Fix the mechanism, and the next motor does not repeat the failure.
This is the same discipline we apply to every motor that comes back to us, whether it is a yaw motor or a DC motor failure of any other kind. The failure is rarely in the copper first. It is almost always in the load, the alignment, or the control, and the copper is simply the part that could not take it any longer. Getting this right matters twice: it saves a good motor from a false warranty claim, and it keeps a bad mechanical problem from eating motor after motor.
FAQ
Why did the armature resistance read 1.86 times normal?
The armature winding runs as parallel branches. When one branch opens, the healthy motor's two parallel paths collapse to one, and the total resistance roughly doubles. The value came in slightly under exactly 2 because a broken or cold-soldered joint still has a little residual contact resistance at the break point.
How does a mechanical stall overheat a DC motor?
A turning motor produces a back EMF that limits armature current. A stalled motor produces none, so the current is set only by the armature resistance (I = U / R), which means several times the rated current. Heating scales with current squared, so the winding temperature climbs fast and the weakest solder joint fails first.
Was the motor a factory defect?
No. The control log showed a stall and an over-current spike before the branch opened. The winding failure was the result of the stall, not a manufacturing fault. Replacing the motor without fixing the jammed mechanism would have failed again.
What should I check if my yaw motor overheats?
Measure the armature and field resistance first, then pull the yaw control log for a stall flag and an over-current spike, and inspect the mechanical drive — slewing bearing, gear ring, reducer, brake, and grease condition — at the same time.
Why is a compound wound DC motor used for yaw drives?
A compound wound motor combines a series winding and a shunt winding. The series part gives the high starting torque a yaw drive needs to break the nacelle loose against wind load, and the shunt part keeps the speed stable when the load drops, so the motor does not run away. That combination suits the stop-start, forward-reverse duty of a yaw system. See our yaw motor application note for the full picture.
How do I keep this from happening to my next motor?
Fix the mechanism, not just the motor. Address the jam, the misalignment, or the brake drag, and keep the drive's protection settings set so a stall trips the drive before the winding cooks. For a plain-language guide to what else kills brushed motors, see our notes on carbon brush life and brushed DC motor quality.
Sources
- IEC 60034-1, Rotating electrical machines — Part 1: Rating and performance — thermal classes, temperature rise limits, and duty types (S2 short-time duty) for motor winding insulation.
- IEC 61400-1, Wind turbines — Part 1: Design requirements — yaw system, protection system, and braking system design requirements for wind turbines.
- Back EMF and armature current relationships (E = k·n·Φ, I = (U − E)/R) — standard DC machine theory; see ScienceDirect topic Armature Resistance.
- Resistive heating P = I²R — Joule heating, standard electrical engineering.
- Winding resistance measurements in this case (71.226 mΩ / 38.371 mΩ armature, 6.8 / 7.0 mΩ field) — customer's precision micro-ohmmeter readings, anonymized.
- XQD-4-21 rating (24 V, 4 kW, DC compound wound, wind turbine yaw duty) — Baoluo Motor product data, enterprise knowledge base.
Need a yaw motor that survives the duty, or a second opinion on a motor that came back hot? Tell us the application and we will help you size it and read the failure: send us your project details.
