Call Us
+86-18072289720
- All
- Product Name
- Product Keyword
- Product Model
- Product Summary
- Product Description
- Multi Field Search
Views: 0 Author: Site Editor Publish Time: 2026-04-06 Origin: Site
Most of what circulates online under "how to check a compressor" was written for a domestic refrigerator. It tells you to shake the start relay, meter a run capacitor, and add three resistance readings across a Common, Start and Run terminal. If you take that procedure to a cold store, a blast freezer or a fish carrier and try it on a T-Series semi-hermetic reciprocating compressor, you will stand in front of a machine with no start relay, no run capacitor and no Common-Start-Run terminals at all, and you will conclude that the compressor is broken because your test does not fit it.
We build both ends of this problem: the compressors, and the service valves and spare parts that make them testable and repairable. This guide is the diagnostic order we expect a plant engineer to follow on a three-phase semi-hermetic reciprocating machine or an industrial screw unit, and where the readings should land.
One framing point before the tools come out. On a commercial refrigeration plant the compressor is usually the messenger, not the culprit. A machine that has been running at a high compression ratio, swallowing liquid or fighting a blocked filter is being killed by something outside itself. Checking the compressor properly therefore means recording what the system asked it to do, not only whether its motor conducts.
The Common-Start-Run resistance method applies to split-phase appliance motors. A three-phase semi-hermetic has three identical windings, so equal readings are normal and prove nothing about drive type.
Isolate the supply and read the nameplate before you test anything. The allowable low-side and high-side pressures, motor rating, refrigerant and oil charge are printed on the machine and they define your acceptance limits.
Roughly ninety percent of a compressor diagnosis on a plant machine can be done through the service valve gauge ports without breaking the sealed circuit. That is what the valves are for.
A machine with sound electrics that cannot make discharge pressure has a mechanical or valve-plate problem, and on a semi-hermetic that is a consumable repair, not a scrap event.
Compare measured performance against published capacity data at a defined rating point. A hand-tight guess about "not cooling well enough" becomes a decision once you have kilowatts and amperes on paper.
Fix the cause before you replace the machine. A compressor destroyed by liquid slugging or by a dead crankcase heater will destroy its replacement in the same way.
Open the terminal box only after the supply is locked and tagged out and you have proved it dead at the terminals. A compressor terminal box is fed from the other side of a contactor, and a contactor that has welded or a control circuit fed from a different phase will both surprise you.
Then read the plate before you measure anything. On a Blait semi-hermetic the nameplate carries the model designation, the refrigerant the machine was built for, the motor output in kilowatts, the voltage and phase supply, the allowable low-side and high-side pressures, the oil charge, the production licence mark and a serial number with build date. The wiring diagram sticker on the terminal box cover tells you the connection scheme, which matters when you later measure windings.
The terminal box carries the wiring diagram, the production licence mark and the nameplate with the refrigerant, the motor rating and the allowable low-side and high-side pressures that define your test limits.
Those pressure limits are not decoration. The subcritical CO2 series is built to a design envelope of 3.0 MPa low side and 5.3 MPa high side, which is why its gauges, valves and relief settings differ from an R404A machine. A technician who tests a machine against the wrong envelope will call a healthy unit faulty. Note the serial number as well: it is how we trace a machine back to its factory run-in record.
The appliance diagnostic hierarchy is: relay, capacitor, then compressor. Small domestic compressors are single-phase split-phase motors with a PTC relay and an overload disc bolted to the shell, and those relays fail constantly, so the hierarchy is correct where it came from.
Our machines are three-phase induction motors, typically 380-420 V, 3 phase, 50 Hz, with a double-star (YY) winding connection on the T-Series frames. There is no start relay, no start capacitor, no run capacitor and no potential relay. Larger screw units may use a part-winding start, which is a contactor and timer arrangement, not a capacitor. Some plants run compressors on a variable frequency drive.
So the "cheap external culprit" list changes. On a plant machine, before you suspect the compressor, the same role is played by:
The contactor: burnt or pitted contacts, or a coil that chatters on a low control voltage, will produce a machine that hums, trips or never starts.
The overload relay: set below the motor's full load current it trips on a healthy machine; the setting must be checked against the nameplate, not against whoever last touched it.
The control logic: low-pressure cut-out, defrost termination, anti-short-cycle delay and pump-down settings all stop a good compressor.
Motor protection devices: winding thermistors or PTC elements and their relay module will hold a machine off with the windings intact.
The crankcase heater: if it has been off while the machine sat, refrigerant has migrated into the oil and the first start is a wet start.
That is the honest translation of the "rule of exclusion" into commercial equipment. It still means test the cheap things first; it just means a different list of cheap things.
Supply quality first. Measure line voltage at the compressor terminals with the machine attempting to run. Voltage imbalance above one percent deserves investigation, and above five percent the motor should not be run at all, because the resulting negative-sequence current heats the rotor without any overload relay seeing it. Also check for a loose lug: a single high-resistance connection will produce a machine that trips, smells and tests "fine" on the bench.
Winding balance. With the supply isolated and the cables off the terminals, measure resistance phase to phase across the three terminal studs. On a healthy three-phase winding the three readings agree within a couple of percent. A wide spread means a winding problem. Do not expect the appliance arithmetic here: on a split-phase motor the start winding deliberately reads higher than the run winding, which is why the Common-Start-Run method works there and is meaningless here.
Insulation resistance. Use a megohmmeter at 500 V DC between each phase and a clean bare spot on the frame. A dry, sound winding reads very high, often tens or hundreds of megohms. A commonly used field floor is in the region of 1 to 2 MΩ, but the motor manufacturer's stated limit governs where one exists, and the trend matters more than a single number: a machine that read 200 MΩ last year and 4 MΩ now has a moisture problem even if it still passes. Two cautions. Never measure insulation with the control transformer, thermistor relay or drive connected, or you will destroy electronics. And a winding that has failed to ground on a semi-hermetic is usually a terminal, seal or moisture problem rather than burnt copper, so chase the cause before condemning the motor.
Protection devices. Check the resistance and continuity of the crankcase heater with the machine cold; on the T-Series M4 frames it is a 120 W unit and it is the cheapest component that saves a compressor. If the machine carries winding thermistors, measure them at the relay end and confirm the trip setting, because a nuisance trip looks exactly like a failed compressor to whoever is holding the toolbox.
Current under load. With the machine running, clamp each phase in turn. Readings should agree within a few percent of each other and sit at or below the nameplate full load current at the measured voltage. A single high leg points to supply imbalance or a winding problem; all three legs high points to the machine working too hard, which is a system fault, not an electrical one.
Rotation and bind. On a machine that has stood for months, try to turn the fan or the coupling by hand with the supply isolated. A semi-hermetic that is hard to turn is usually seized from moisture or from oil that drained away, and a machine that turns freely is not automatically healthy, because reed valves can be intact and still worn.
Oil level and condition. The sight glass on the crankcase has a marked level. It should sit there steadily. If the oil fizzes or foams after start-up, liquid refrigerant is coming back in the suction line, and that dilutes the oil, washes out the bearing film and destroys the machine from the inside. This is the single most common way a healthy compressor is killed in the field.
Crankcase oil level sight glass with the marked level. A steady film at the mark is normal; oil that fizzes or foams after start-up means liquid refrigerant is coming back and diluting the oil.
Noise and vibration. A reciprocating machine has a characteristic rhythm; a new knock, a rattle that changes with load, or a vibration that grows usually means a mounting bolt, a loose counterweight or a broken valve. Compare it against the same machine cold-starting and at steady load. Listening through a screwdriver on the casing is an old trade trick and it still works.
Temperature pattern. Feel the discharge line, the oil sump area and the motor shell. A compressor that pumps hard on discharge while the sump runs cold is behaving normally. A hot sump with a cool discharge line is not pumping. Do not use your hand on the discharge of a CO2 or ammonia machine; use the gauge ports.
Locked rotor. If the machine hums and does not turn, current on all three legs will be at the locked rotor value and the overload will trip. Stop trying to start it. Repeated attempts burn the winding that was still good, and this is the most common way a repairable machine becomes a scrap machine inside ten minutes.
This is the section that decides the diagnosis, and it is the one that needs no charge recovery. Screw your gauges or electronic probe into the service valve gauge ports. Our J2BS shut-off valves for screw compressors are cast iron rated 4.5 MPa with a working range of -30C to +150C on R22, R404A and R507A, and the angle and straight stop valves on the reciprocating machines do the same job. The isolation function is what lets you test, and later service, a compressor without pumping down the whole plant.
Suction and discharge service valves and a rotalock joint on a semi-hermetic machine. The gauge ports on these valves are where pressure evidence is taken without recovering the charge.
Record, in one sitting, the following: suction pressure and the corresponding saturation temperature, suction line temperature at the valve so you can compute superheat, discharge pressure and saturation temperature, discharge line temperature, condensing approach against ambient, liquid temperature against condensing saturation for subcooling, and current on each phase. That single page of numbers tells you more than any insulation test.
Then compare it against published performance data rather than against a feeling. Here is the T-Series M4 snapshot at one rating point, R404A and R507A at minus 10C evaporating and 40C condensing:
Published T-Series M4 capacity at minus 10C evaporating and 40C condensing on R404A and R507A. Measured power input and cooling effect should track these figures, and a machine that cannot reach them has lost compression.
A machine whose measured current implies the tabulated power input but whose cooling effect is clearly short has lost compression, and on a semi-hermetic reciprocating the usual reason is the valve plate. Conversely, if the electrical numbers are low and the machine is simply not being asked to do much, the fault is upstream in the controls.
The other comparison is against the operating envelope. Our published range for the M4 frames shows Z class models running from minus 5C down to minus 40C evaporating on R22 and to minus 45C on R404A and R507A, while G class models cover the warm end. Two consequences for diagnosis. If a machine is being asked to run below its published corner, you will see high discharge temperature and oil breakdown, and the compressor is the victim. And in the low-temperature corner these frames are used with an electronic liquid-injection controller, so a machine that is overheating should have that circuit checked before the compressor is blamed.
| What you observe | What it usually means | Check next | Related parts we supply |
|---|---|---|---|
| No hum, no trip, motor cold | Supply, contactor or control not calling | Voltage at terminals, control circuit, cut-out settings | Solenoid valve coils |
| Hums, draws locked rotor current, trips | Mechanical bind or seized from standing | Try rotation by hand, then teardown; stop repeated starts | Motor and rotating assembly spares |
| Runs, cools poorly, low discharge, high suction, warm casing | Worn or broken valve plate, lost compression | Capacity against the published table, then valve plate inspection | Valve plate replacement part |
| Oil foaming in the sight glass | Liquid return, flooded start | Superheat, receiver charge, level control, heater | Ball valves and regulating valves |
| Frost or oil weeping at a joint | Gasket or rotalock failure, over-torquing | Visual, leak check, torque | Paper gasket seal, ductile iron globe valve |
| Insulation resistance low or falling | Moisture in the machine or a damaged terminal seal | Heater operation, megger each phase, trend against last year | Terminal box and gasket sets |
| High head pressure, high amps | Condenser fouling, airflow, non-condensables | Approach temperature, purge, clean condenser | Isolation valves for cleaning and service |
Read that table from right to left and you get the commercial point: most of the failures on the left are fixed with a part on the right, and not with a new compressor. That is why semi-hermetic machines are still built this way, and it is also why we make the valves, the gaskets and the valve plates ourselves.
A semi-hermetic reciprocating compressor is designed to be opened. A valve plate set, gaskets and a thorough clean with an oil change is a routine overhaul, and on a machine with an intact crankcase and a sound motor it returns the unit to near its published capacity. That is the economic case for repair, and it is the reason our spare parts range exists alongside the machines.
Replacement becomes the right answer in a narrower set of cases than the appliance guides suggest: a cracked or eroded crankcase, a motor that has run hot long enough to carbonise the windings and acidify the oil, a frame damaged by a locked-rotor episode, or a machine on a refrigerant and control scheme you are retiring anyway. Between those two extremes, the decision is usually made by downtime rather than by part price: on a loaded cold store, an overhaul kit that arrives in days beats a new machine that arrives in weeks, and on a marine installation the class-approved replacement route is longer still.
Do the arithmetic on energy with the same tables you used for diagnosis. Two M4 frames at the minus 10C and 40C point differ in power input by a few hundred watts at similar cooling output, and a machine that has lost compression costs more than that. Efficiency claims without a rating point are not a decision, they are marketing.
Field diagnosis is only meaningful against a baseline, so it is worth stating what the baseline is. Every unit is run-tested before packing, leak-tested to its nameplate pressures, charged to the stated oil quantity of 2.6 litres on the M4 frames with its 120 W crankcase heater fitted and proven, and marked with a serial number that ties it to that record. The terminal box carries the production licence mark, and the marine-approved frames carry the China Classification Society type approval covering the eight M4 models on R507, which is what allows a surveyor to accept the machine on a fishing vessel. Details are in our certification article.
The practical consequence for you is that the serial number is a diagnostic tool. Send it to us with your readings and we can pull the test record, confirm the winding data and the original charge, and tell you whether what you are measuring is a deviation or a specification you were not aware of.
No. That method belongs to single-phase split-phase appliance motors, where the start winding deliberately reads higher than the run winding. A three-phase semi-hermetic has three identical windings, so you measure phase-to-phase balance, which should agree within a couple of percent, and insulation resistance to the frame at 500 V DC. Equal readings on a three-phase machine are a pass, not a symptom.
Measure at 500 V DC between each phase and bare metal on the frame. A dry, sound winding normally reads far above the pass threshold, often tens or hundreds of megohms, and a commonly used field floor sits around 1 to 2 MΩ, with the motor manufacturer's stated limit taking precedence where one is published. The trend matters more than the absolute number, and a sharp fall from the machine's own previous reading is the warning sign even if the value still passes. Isolate the drive, thermistor relay and control transformer first, or the test will damage them.
Most often lost compression from a worn or broken valve plate, confirmed by a discharge pressure that will not reach the value expected for the condensing temperature and a suction that stays high. Verify against published capacity data at a known rating point rather than by feel, then inspect the valve plate. On a semi-hermetic this is a consumable overhaul with a replacement valve plate and gaskets, not a new compressor. Rule out the system causes first: low charge, a blocked filter drier, an iced evaporator or a condenser too hot to reject against.
No, and you should not have to. Pressure and temperature evidence comes from the gauge ports on the suction and discharge service valves, and electrical evidence comes from the terminal box. The charge only needs to be recovered when you intend to open the refrigerant circuit. That distinction is the whole point of fitting proper isolation valves, and it is why we build them to 4.5 MPa with a temperature range that covers low-temperature duty.
Assess it, do not assume. Slugging breaks valve plates, bends rods and can score a crankcase, but it can also leave only a valve plate and a set of gaskets damaged. Open it, inspect, measure. More important, find why liquid came back, because the same fault will destroy the next machine: superheat, the crankcase heater, the liquid line level, the solenoid coil and the defrost termination are the usual suspects. If the oil went dark and acid after repeated hot trips, the motor is the part you should distrust.
The honest answer is that life is set by three things you can control: discharge temperature, whether liquid returns to the machine, and supply quality. A machine held inside its published operating envelope, with a working crankcase heater, correct superheat and balanced voltage, will run for decades and be overhauled more than once. A machine run below its envelope, flooded and started repeatedly against a locked rotor, will fail in months. Any guide that quotes a fixed lifespan without a rating point is guessing.
For our range, yes, and it matters most when a vessel is off-hire. The T-Series M4 frames hold China Classification Society type approval for fishing vessel duty on R507, the same factory supplies the service valves, gaskets and valve plates, and the spare parts are shared across the reciprocating range. See our after-sales page for how support is handled.
A diagnosis is a data problem, and we can usually tell you the answer from one page of it. Send:
Model and serial number from the nameplate, so we can pull the factory run-in record
Refrigerant and the oil type marked on the plate
Suction and discharge pressures, with the saturation temperatures and the corresponding evaporating and condensing conditions
Superheat at the compressor and subcooling at the receiver or condenser
Discharge temperature and, if fitted, the thermistor or winding sensor reading
Current on each phase, plus the measured supply voltage and any imbalance
Insulation resistance to the frame, each phase, with the megger voltage used
Oil level in the sight glass and whether it foamed on start
What changed first: temperature, noise, trips, defrost behaviour, or a valve or control that was adjusted
Reach us through the contact page or on +86-18072289720. If you want the underlying performance data before a failure happens, the full capacity and power tables for every frame are on the T-Series and YBF series pages, and the working principle of the machines covered here is in The Basics of Reciprocating Compressor for Refrigeration.
