A chiller that “does not cool as it used to” is a symptom with several possible causes. The heat exchanger may be fouled, the refrigerant charge may be low, air may have entered the circuit, the compressor may be worn, the water flow may have fallen or a sensor may be lying. Replacing the chiller before measuring these five items is an expensive way to guess.
The measurements below can be taken with the plant running, a calibrated thermometer, a pressure gauge and a clamp meter. They separate a maintenance fault from a capacity fault.
1. Chilled-water flow and temperature difference
Capacity is calculated from the water side:
Q = 1.163 × V × ΔT
where Q is kW, V is flow in m³/h and ΔT is the chilled-water temperature difference in K. A 350 kW chiller at 60 m³/h and 5 K gives 1.163 × 60 × 5 = 349 kW. If the same flow shows 4 K, the calculated capacity is 279 kW, a 20% loss.
But a smaller ΔT can also mean a smaller load, not a smaller capacity. Check the control valve position, the pump speed and the building load. The capacity test needs a known load or a full-load condition; a part-load reading is not a capacity verdict.
2. Condenser approach
The condenser approach is the condensing saturation temperature minus the leaving condenser-water temperature. A clean condenser typically runs 2–5 K; a fouled one rises. If the approach is 2 K above its commissioning value at the same flow and inlet temperature, the condenser is fouling or the water flow has fallen.
Measure the saturation temperature from the pressure, not from a dial gauge with a wide band. A 0.1 MPa error in pressure is several K in saturation temperature, which is larger than the trend you are trying to detect.
3. Evaporator approach
The evaporator approach is the leaving chilled-water temperature minus the evaporating saturation temperature. A clean evaporator typically runs 1–3 K. A rising approach at the same flow points to fouling, a low refrigerant charge or a restricted expansion valve.
If the evaporator approach rises while the condenser approach is stable, the problem is likely on the evaporator or charge side, not the condenser. If both rise, the water flow or the overall heat-transfer surface is the common cause.
4. Suction and discharge pressures, superheat and subcooling
| Measurement | Normal trend | What a deviation suggests |
|---|---|---|
| Suction pressure | Matches the design evaporating temperature | Low: charge, restriction, low flow; high: high load or valve overfeed |
| Discharge pressure | Matches the design condensing temperature | High: fouling, air, overcharge, high water temperature |
| Superheat | Stable at the valve setting | High: undercharge or restricted valve; low: overfeed or sensor fault |
| Subcooling | Stable at the design value | High: overcharge or liquid-backed condenser; low: undercharge |
| Discharge temperature | Below the compressor limit | High: high compression ratio, low charge, oil problem |
Air in the circuit is a common and underestimated cause. It raises the discharge pressure and the condenser approach without a corresponding water-side fault. A system that was opened for repair and not evacuated properly can lose several percent of capacity.
5. Compressor current against the baseline
Record the current at the same voltage, load and condensing condition as the commissioning baseline. A current that is higher than the baseline at the same capacity suggests a higher compression ratio or a mechanical problem. A current that is lower than the baseline at a lower capacity can suggest valve leakage or a worn compressor. Neither reading alone is a diagnosis; it must be paired with the pressures.
Also check the voltage and phase balance. A 3% voltage imbalance can raise the winding temperature and reduce the available torque. A phase loss will trip the overload, not gently reduce capacity.
6. The decision
If the approaches are high and the water flow is correct: clean the heat exchangers first. A condenser cleaning can recover 10–20% of the lost capacity in a fouled machine.
If the superheat is high and subcooling is low: leak-test, repair and charge by weight. Do not add gas without finding the leak.
If the approaches are normal, the charge is correct and the current is high at the design load: the compressor or the expansion valve may be worn. A compressor overhaul or replacement may be economic, but compare it with the residual life of the exchangers.
If the capacity loss is above 25% after cleaning and charging: replacement becomes a serious option. Use the lifecycle cost, not the age alone.
Common questions
Why does the chiller run continuously but not reach the setpoint?
Check the actual capacity against the load. A fouled condenser, a low charge or a low water flow can leave the machine at 70–80% of its design capacity while the load is unchanged.
Can a dirty condenser cause a 20% capacity loss?
Yes. Fouling raises the condensing temperature, reduces the compressor mass flow and increases power. The exact loss depends on the fouling severity and the compressor curve.
How do I know the water flow is correct?
Measure the flow and the ΔT, and compare with the commissioning data. A ΔT well above design at the same load suggests low flow.
Is a low suction pressure always a charge problem?
No. It can be a restricted valve, a fouled evaporator, low water flow or a low load. Check superheat and the evaporator approach before charging.
Should I replace a 12-year-old chiller with 15% capacity loss?
Not automatically. If cleaning and repair recover the capacity and the exchangers are sound, repair can be the lower-cost option. Replace when the repair cost or the efficiency loss no longer justifies the remaining life.
What is the single most useful baseline?
Capacity, condenser approach, evaporator approach and current at commissioning. Without it, every later reading is a guess.
What five measurements?
Water flow, entering and leaving temperatures, refrigerant superheat and subcooling, compressor current and condenser approach. The pattern shows whether the loss is water, refrigerant or mechanical.
When is replacement justified?
When the repair cost approaches a large share of a new unit and the efficiency is below the economic threshold. A worn compressor alone is not a reason to replace the chiller.
Further technical reading
- Chiller Evaporator and Condenser Heat Transfer
- Chiller Condenser Descaling
- Chiller Refrigerant Leak and Charging
- Chiller Part-Load Staging
LONGCOM can review the five measurements against the commissioning baseline and separate a fouling fault from a compressor or charge fault. For a capacity diagnosis, contact 400-878-0280 or 13002001280.
Related reading
Evaporator water-side cleaning · Chiller plant sensors · Water chiller sizing