A replacement compressor with three unmarked terminals and no wiring diagram is a common enough job that you shouldn’t need to guess at it, and guessing is exactly how sound compressors get condemned for faults they don’t have. Three resistance readings and one line of arithmetic tell you which terminal is common, which is start, and which is run, and confirm the windings are sound — all before line voltage ever touches the machine. By the end of this article you’ll be able to run that identification with confidence and know exactly what to do if the numbers don’t add up.
Why the two windings read differently
A single-phase induction motor needs two windings displaced in the stator to produce a rotating field: a run winding (main winding) and a start winding (auxiliary winding). They are not built the same, and that difference is exactly what makes terminal identification possible with nothing more than a multimeter.
The run winding carries current for as long as the compressor runs, so it is wound with heavier gauge wire and more turns — lower resistance, typically a few ohms on a domestic or light commercial compressor. The start winding is only ever in circuit for the second or two it takes the rotor to reach speed, so it is wound with finer wire — higher resistance, commonly two to four times the run winding’s value on the same compressor, sometimes more. That resistance gap is not incidental. It is the entire reason a meter on an unmarked compressor can tell you, from the numbers alone, which terminal is which.
The three-reading method
With the compressor isolated, locked off, and proved dead per the OHS Act 85 of 1993 and SANS 10142-1, measure resistance across all three terminal pairs. Work through the logic in this fixed order every time:
- The highest reading is between S and R. Current between these two terminals passes through the common point and both windings in series — start resistance plus run resistance, which always reads higher than either winding alone.
- The terminal absent from that highest reading is C. If the highest reading was between terminals A and B, terminal D is common.
- Of the two remaining readings — C to each other terminal — the lower one is C-R (the heavier-wound run winding) and the higher one is C-S (the finer-wound start winding).
- Confirm it: C-S + C-R must equal S-R, within a few percent. This is the check that the three readings are internally consistent, not a marginal connection or a dying meter battery.
The arithmetic check, worked
| Measurement | Reading | What it tells you |
|---|---|---|
| Terminal A to Terminal B | 13.2 Ω | Highest of the three — this is S to R |
| Terminal A to Terminal D | 9.8 Ω | Middle reading |
| Terminal B to Terminal D | 3.4 Ω | Lowest reading |
Terminal D appears in both the middle and lowest readings, and is absent from the highest (A-B) — so D is C. Between D’s two readings, 3.4 Ω is lower, so D-B is C-R, meaning B is R. That leaves A as S, with D-A = 9.8 Ω as C-S.
Check: C-S + C-R = 9.8 + 3.4 = 13.2 Ω, matching S-R (A-B) exactly. The arithmetic closes, so the identification stands.
A second worked example, from an actual replacement job with terminals labelled only 1, 2 and 3:
- Terminal 1 to Terminal 2: 11.6 Ω
- Terminal 1 to Terminal 3: 8.4 Ω
- Terminal 2 to Terminal 3: 3.2 Ω
The highest reading, 11.6 Ω, is between 1 and 2 — that’s S to R. Terminal 3 doesn’t appear in that pair, so 3 is C. Of the two readings involving terminal 3: 3-1 = 8.4 Ω (higher, so C-S, meaning 1 is S) and 3-2 = 3.2 Ω (lower, so C-R, meaning 2 is R). Check: C-S + C-R = 8.4 + 3.2 = 11.6 Ω, matching the measured S-R exactly. The identification is confirmed by arithmetic, not assumed from terminal layout.
When it doesn’t add up
A gap of more than a few percent between C-S + C-R and the measured S-R — say 11.9 Ω calculated against 13.2 Ω measured — means one of the three readings is unreliable. Before you trust any part of the identification, check for the usual causes: a corroded or oxidised terminal pin giving a falsely high contact resistance, meter leads not fully seated, the meter left on the wrong range, or a weak meter battery affecting the ohms function specifically. Clean the terminals, reseat the leads, and re-measure all three pairs from scratch rather than patching the one reading that looks wrong — a single bad contact can quietly shift more than one of the three numbers.
If the readings still won’t close after a careful re-measurement, do not proceed to wire it up on a best guess. Compare against the manufacturer’s service data if it’s available, since absolute winding values vary by frame size and horsepower even though the S-R = C-S + C-R relationship itself does not. Treat a compressor that won’t confirm cleanly as a winding fault under investigation, not as a terminal-identification problem to be solved by trial and error at line voltage.
Proving the windings, not just the terminals
Identifying C, S and R correctly tells you nothing about whether the compressor is actually sound, so finish the job with two more checks before reconnecting. Insulation resistance, each terminal to the compressor shell, at 500 V DC for 230 V equipment: reject anything below 1 MΩ, the absolute floor, and expect a healthy compressor to read well above 100 MΩ. Continuity on each winding should show a finite, low resistance matching the values already measured — OL (no continuity) means open, most likely a burnt joint or a burnout, which is covered in depth in our guide to insulation resistance testing on compressors.
A shorted turn is the hardest fault to catch with a standard meter at this stage: two turns shorted out of a hundred loses only a couple of percent of total resistance, easily lost in meter resolution and lead resistance on a reading of a few ohms. The winding can still show continuity, a plausible resistance, and an arithmetic check that closes. What a shorted turn actually does is raise running current and localised heating once the compressor is running — confirmed operationally, with a clamp meter against nameplate FLA, rather than by a cold resistance test alone.
Motor type affects torque, not the identification method
RSIR, CSIR, PSC and CSR are all single-phase induction designs solving the same problem with different hardware, and none of that changes how you identify C, S and R — the resistance method above works identically regardless of motor type. What differs is starting torque and the capacitor arrangement each design needs to run correctly, which is exactly where a correctly identified but incorrectly wired compressor causes trouble: fit the wrong capacitor to a correctly identified HERM terminal and you get the hard-starting, high-current symptoms covered in our guide to testing a capacitor on an air conditioner. Read the motor type and capacitor values off the nameplate or service data, never assumed from compressor size or refrigerant alone.
Final Thoughts
More sound compressors are condemned from a guess at the terminal markings than from any actual winding fault. Three resistance readings, one line of arithmetic, and two follow-up checks on insulation resistance and continuity are what stop the guessing — and the same discipline of testing to a number rather than eyeballing a connection runs through every chapter of the complete guide to electrical troubleshooting and fault-finding. Isolate, measure, check the arithmetic, and only then reach for line voltage.
+ There are no comments
Add yours