Aircon compressor windings: the motor in the refrigerant
Half of a compressor is a pump and the other half is an electric motor. The motor half usually settles what happens to the machine, because it is the half anyone can put a reading on. A replacement recommendation should rest on that reading.
By Team Snowflake | Updated 7 Aug 2026
What a compressor motor is made of
A compressor motor is copper wire wound into coils and packed into a steel core. Those coils are the windings. Current through them produces the turning force, and the pump sits on the same shaft doing as it is told.
Nothing separates one turn of that wire from the next except a film cured onto the copper. The wire arrives at the winding machine already coated, and nothing is wrapped around it afterwards. Two thicknesses of that film are all that stand between neighbouring turns.
The same film keeps the windings clear of the metal around them. The core they sit in is steel, that steel is fixed to the body of the compressor, and the body is earthed through the unit. A coil pressed into a metal slot with a coating between the two is the whole arrangement.
Everything on this page follows from those two facts. The insulation is thin, and nothing else shares that job with it. Almost every electrical death of a compressor is that film letting go at one point along its length.
Two windings, and three wires out through the shell
Most compressors in local flats run on single-phase supply, and those carry two windings rather than one. The run winding does the work once the motor is turning. The start winding, helped by a capacitor, breaks it away from standstill.
Three connections leave the shell to serve those two, with one shared between them. How those connections cross steel without letting anything out is its own topic. What matters here is that both ends of the wire are reachable from outside. That is the reason any of this can be read at all.
Why the motor is sealed inside the refrigerant
Putting the motor inside solves a sealing problem. A motor mounted outside would drive the pump through a shaft crossing the wall of the shell, and a shaft crossing a wall has to pass through something that wears. Welding the whole machine shut removes that joint along with the leak it eventually becomes.
The arrangement also feeds the windings their cooling. Gas returning from the room enters the shell first and washes over the motor before it reaches the pump. Heat coming off the wire is picked up there and leaves with the stream.
That ties the state of the motor to the state of the circuit. What cools the wire is the same refrigerant that cools the room. Less of it arriving means less heat carried away, and the motor then runs hot for reasons sitting nowhere near the motor.
So a low charge is not only a cooling complaint. A system short of gas returns warmer vapour to the compressor, and warmer vapour cannot pull the same heat out of the windings. A circuit quietly undercharged for a long stretch has been holding its motor above its design temperature for exactly as long.
Restriction reaches the same place by another road. Anything narrowing the path back to the compressor reduces what arrives at the windings. A choked outdoor coil works differently again, by raising the temperature the machine has to discharge into.
The coating has to survive what surrounds it
Sitting in the stream means sitting in the chemistry. Wire for a sealed motor is not ordinary wire. It is qualified against refrigerant and oil under a published magnet-wire standard, because the coating has to keep insulating while soaked in both for the working life of the machine.
The oil is often the harsher of the two. Testing work behind that standard notes that the lubricants used in current compressors can be more aggressive on wire coatings than the refrigerants they travel with. A circuit given a product it was never matched to becomes a motor question, not only a chemistry one. That is part of why counterfeit refrigerant is worth refusing on price alone.
What wears the insulation down
Heat is the ordinary one. Every run puts current through the wire, current makes heat, and insulation ages faster the hotter it is held. Nothing dramatic has to happen for this to matter. A motor kept warm season after season arrives there on its own.
Moisture is the one worth understanding, because it never attacks the coating directly. Water left inside a sealed circuit finds the refrigerant and the oil to work on, and acid is what comes out of that reaction. Refrigeration guidance ties those acids to deterioration of motor winding insulation. That is why a system gets dried before it is charged, and then given a drier to hold it dry.
The drier is a wear part rather than a fitting, and what a filter drier does is written up on its own page. The point here is narrower. Its condition and the condition of the wire are one story told at two places on the same circuit.
Starting is where the wire takes its heaviest load. A motor at standstill takes a great deal more current than the same motor turning, and anything making the break-away harder stretches that moment out. Tired starting components belong in this group. So does a restart attempted before pressures have settled, which is what compressor lockout holds a unit back to prevent.
| What the wire is exposed to | What puts it there | What an owner can act on |
|---|---|---|
| What the wire is exposed toHeat with nowhere to go | What puts it thereA charge sitting below the figure the equipment was sized for | What an owner can act onAsking why gas was needed, rather than accepting another top-up |
| What the wire is exposed toAcid formed from water, refrigerant and oil | What puts it thereA circuit opened and closed again without being dried out | What an owner can act onAsking what was done to dry the system after any repair |
| What the wire is exposed toHeavy current on every attempt to start | What puts it thereWeak starting components, or restarts against unsettled pressure | What an owner can act onRecording how often the unit stops and tries again |
| What the wire is exposed toA product the coating was never matched to | What puts it thereRefrigerant or oil of unknown origin put in at some point | What an owner can act onKeeping the paperwork from every gas visit the unit has had |
| What the wire is exposed toWarm return vapour on a long run | What puts it thereA restriction quietly starving the line that feeds the shell | What an owner can act onReporting rooms that cool slower than they used to |
Nothing about this shows from the room
Insulation ageing announces nothing. The film does not thin evenly and then give a date. It holds, and holds, and then fails at the one place where conditions happened to be worst.
What an owner can see is the history that put it there. A unit topped up more than once, a circuit opened several times, a machine that stops and restarts more than it used to. None of those are winding faults. All of them are the conditions a winding fault grows in.
The electrical outcomes a technician separates
Every electrical verdict on a compressor lands on one of three findings. The copper path is broken, two things that should stay apart are touching, or a winding is touching the body. Each says something different about the machine.
An open winding means the path along the copper is broken somewhere. Nothing flows, so nothing turns. The outdoor unit stays silent while the rest of the system behaves perfectly normally, and the indoor fan carries on as though nothing has happened.
A short means the film gave way between two conductors meant to stay separate. Current then takes a shorter route than the one the motor was built around. A machine in this state often still runs. It pulls more than it should, makes heat where it should not, and delivers less cooling than it did.
A winding touching the body is the third, and it ends the discussion fastest. The steel shell is earthed through the unit, so a winding in contact with it sends current to earth. That is the exact condition earth-leakage protection exists to catch, and these usually announce themselves by tripping the ELCB the moment the unit is asked to run.
What separates the three in the field is what the machine does while power is still on it. Silent outdoors with everything else alive points one way. Running badly points another. Cutting the supply the instant it starts points at the third and rarely at anything else.
| What has failed | What the machine does | What it means for the compressor |
|---|---|---|
| What has failedThe copper path along a winding is broken | What the machine doesOutdoor unit silent, the rest of the system behaving | What it means for the compressorFinished, once a tripped protector has been ruled out |
| What has failedFilm gone between turns of the same winding | What the machine doesStill runs, pulls more, cools less than it did | What it means for the compressorFailing, and the reading will keep moving |
| What has failedFilm gone between the two separate windings | What the machine doesStruggles to break away, or starts and then stops | What it means for the compressorFinished as a motor, whatever the pump is like |
| What has failedA winding is resting against the earthed body | What the machine doesProtection cuts in as soon as it is asked to run | What it means for the compressorFinished, and not something to keep retrying |
| What has failedNothing yet, the internal protector has opened | What the machine doesSilent while hot, alive again once it has cooled | What it means for the compressorNothing yet, but something made it that hot |
An open reading is not always an open winding
A compressor with a protector built into it reads as an open circuit while that protector is out. The device sits inside the shell among the windings and opens when the motor gets too hot. Until the motor cools, a meter sees what a broken wire would show it.
That is the most expensive mistake available on this component. A hot compressor tested on the spot can be condemned for a fault it does not have, and the heavy parts inside hold heat long after the shell feels cool to the hand. Letting it come down and reading it again settles the question.
A protector that has opened is still worth something as information. It says the motor reached a temperature it was built to retreat from. The pattern of those retreats is the useful part, not the single event.
Why the windings can be measured when the rest cannot
Almost everything else about a sealed compressor is inferred. Pressures, currents and temperatures describe how the machine is behaving, and a technician reasons backwards from them toward what must be happening inside. The windings are different, because the wire has ends and those ends come out through the shell.
Two separate things get read there. One is the resistance along the windings, which says whether the copper path is intact and whether the two windings still stand in the relationship they should. The other is the resistance between a winding and the steel body, which says how well the film is still holding current away from earth.
That second figure is what people usually mean when they say the compressor was tested. It measures the coating rather than the copper, and it is the only direct evidence anyone gets about the condition of the insulation. Everything else on a sealed part is argument from behaviour.
One figure on one day is not a verdict, and the compressor manufacturers say so themselves. Copeland's published position is that a single poor reading should not condemn a hermetic compressor. Contaminated oil and contaminated refrigerant both move the number.
Temperature moves it as well. A compressor standing cold with liquid refrigerant lying in the shell reads lower than the same compressor warm and running, because what surrounds the windings has changed. Two honest technicians can measure one machine hours apart and disagree without either being wrong.
None of that makes the measurement worthless. It makes it evidence rather than a verdict, and the distinction is worth holding once a large number is attached to the answer.
What to ask before agreeing to a replacement
A verbal summary is not a reading. The figures either got written down or they did not, and asking for them is a reasonable request that a competent supplier answers without friction.
- The figures themselves rather than a description of them, and what each was read across
- Which state the compressor was in at that moment: standing cold, warm, or under load
- Whether it was read again once the compressor had come down to room temperature
- What got ruled out first, since starting components produce the same complaint for far less money
- Whether anything was found in the oil, because that widens the job past the part
Where the answer goes from here
A confirmed winding fault does not by itself decide what happens to the unit. Whether a compressor is worth replacing at all gets weighed against the age of the system and the state of everything around it. Repair vs replace is where that judgement is set out in full.
An electrical ending also leaves something behind that a mechanical one does not. What that residue does to the pipework, and how the next compressor inherits it, belongs to compressor burnout. This page stops at the wire.
A motor still holding but no longer strong shows itself well before any of this. The signs of a compressor failing describe what that looks like from inside the room, which is the earlier and cheaper place to catch it.
Ready to get started?
Tell us what’s going on. Symptoms, setup, photos, anything we should know. We’ll assess and come back with the right next step.