A technician who only checks superheat is diagnosing half the refrigeration circuit. Subcooling is the reading that tells you what is happening at the condenser and how much liquid refrigerant charge is actually in the system — information superheat cannot give you on its own. This article sets out the exact measurement procedure, the arithmetic, a worked example, and why this particular number belongs to the high side of the system rather than the low side.
What Subcooling Actually Measures
Subcooling is the saturation temperature corresponding to the measured liquid line pressure, minus the measured liquid line temperature:
Subcooling (K) = Saturation temperature at measured liquid pressure (°C) − Measured liquid line temperature (°C)
Refrigerant leaving the condenser as saturated liquid sits exactly at the saturation temperature for the high-side pressure — subcooling reads 0 K at that point. As the liquid continues to lose heat to the condenser coil past full condensation, its actual temperature drops below the saturation point, and that drop, in kelvin, is subcooling. A healthy positive subcooling number confirms two things at once: that the condenser has enough surface area and airflow to fully condense the refrigerant, and that there is sufficient charge in the system to keep a liquid seal ahead of the metering device.
What You Need
- A manifold gauge connected to the high-side (liquid) service port, plus a P-T chart or app matched to the exact refrigerant in the system.
- A clamp-on temperature probe (thermocouple or thermistor) — not an infrared gun, for the same surface-and-reflection reasons that apply on the suction side.
- Insulation for the probe once clamped.
- Access to the manufacturer’s design subcooling figure for that specific condensing unit, from the nameplate, install manual, or commissioning sheet.
Step 1: Let the System Reach Steady State
As with superheat, take no reading until the system has run at least 10–15 minutes at a steady load. On a hot Cape Town afternoon (condensing unit ambient 30°C or higher) a system cycling on after being off will show climbing head pressure and a moving subcooling value for the first several minutes as the condenser catches up. Do not measure during a defrost cycle, during a hot gas bypass event, or in the first 5 minutes after compressor start.
Step 2: Measure Liquid Line Pressure and Find Saturation Temperature
Connect to the high-side service port — ideally as close to the condenser outlet or the liquid line service valve as your access allows, and definitely before the filter drier if you also intend to check for a pressure drop across it (covered in the diagnostic articles linked below). Record the pressure in kPa or bar gauge, and pull the saturation temperature for that pressure from the correct refrigerant’s P-T chart. For R410A at 2,400 kPa gauge, saturation temperature runs around 42°C; for R32 at the same pressure it is close to 40°C. A 2 K gap between refrigerants at the same pressure is enough to turn a borderline reading into a wrong diagnosis if you use the wrong chart.
Step 3: Clamp and Insulate the Probe on the Liquid Line
Clean the pipe to bare copper at the exact point you read pressure, clamp the probe with full contact and even tension, then insulate both the probe and a few centimetres of pipe on each side. Liquid line temperatures in direct Cape Town summer sun can run several degrees above the refrigerant temperature if the probe is left exposed — expect a 1–3 K high-side error from an uninsulated clamp in full sun on a black-jacketed line. Allow 60 seconds for the reading to settle before recording it.
Step 4: Subtract
Subcooling = saturation temperature from Step 2 − measured liquid line temperature from Step 3. Record both raw values alongside the result.
Worked Example
Liquid line pressure at the service valve: 2,500 kPa gauge. On the R410A chart, 2,500 kPa corresponds to a saturation temperature of 44°C. The clamped, insulated probe at the same point reads 36°C after settling.
Subcooling = 44°C − 36°C = 8 K.
That falls at the top of the generally accepted 4–8 K band for most split and packaged systems — but “generally accepted” is a starting point, not the answer. The number that actually matters is the manufacturer’s design subcooling figure for that specific unit, which can run anywhere from 3 K to 12 K depending on condenser sizing and the metering device fitted. Always check the nameplate or commissioning documentation before treating an 8 K reading as either “fine” or “a problem.”
Why This Reading Is a High-Side, Not Low-Side, Diagnostic
Superheat tells you about the evaporator and the metering device — whether the low side is being fed correctly. Subcooling tells you about the condenser and the charge — whether the high side is rejecting heat properly and whether there’s enough refrigerant mass in the system to keep the liquid line full ahead of the TXV or orifice. The two readings are diagnosing different halves of the same circuit, which is exactly why neither one, alone, gives you a full picture.
| Subcooling reading | Typical implication | What to check next |
|---|---|---|
| Below design figure by more than 2–3 K | Undercharge, or condenser not rejecting enough heat (dirty coil, fan fault, high ambient) | Confirm condenser coil is clean, fan running at rated speed, then consider charge |
| At or within 1–2 K of design figure | Condenser performing correctly, charge likely correct for that load | No action — log the reading |
| Above design figure by more than 2–3 K | Overcharge, or a restriction downstream backing liquid up in the condenser | Check liquid line temperature drop across drier and metering device before adding or removing gas |
Common Measurement Errors
| Error | Effect on reading | Typical size |
|---|---|---|
| Reading pressure at the service valve but temperature further down the line (or vice versa) | Invalid, non-comparable values | Any magnitude — discard |
| Uninsulated probe in direct sun | Subcooling reads low (temperature looks higher than actual) | 1–3 K |
| Measuring before 10–15 minute stabilisation | Value still drifting, not representative | ±2–5 K |
| Using design subcooling from a different unit or a generic 4–8 K rule without checking the nameplate | Correct reading misread as a fault, or a real fault dismissed as normal | Depends entirely on the specific unit — can exceed 5 K |
| Ignoring liquid line pressure drop on long vertical risers | Saturation temperature calculated from the wrong pressure | 1–2 K per several metres of rise, refrigerant dependent |
Subcooling on Zeotropic Blends
On a zeotropic blend with temperature glide — again, R407C is the standard example — subcooling must be calculated from the bubble point temperature, not the dew point. The bubble point is where the first vapour bubble would form as the liquid warms, and it is the correct saturation reference for liquid-side calculations. Using the dew point instead (the mistake that inflates superheat, described in the companion article) will make calculated subcooling read lower than it actually is, by roughly the glide value of the blend — commonly 4–7 K. Confirm your gauge or app is set to the bubble point reference for subcooling and the dew point reference for superheat; conflating the two is the single most common source of a wrong charge diagnosis on blended refrigerants.
Final Thoughts
Subcooling closes the loop that superheat opens: superheat confirms the evaporator and metering device are behaving, subcooling confirms the condenser is rejecting heat and the charge is sufficient to back it up. Measure both at the same visit, at steady state, against the correct P-T reference for the refrigerant in the system, and against the manufacturer’s design figure rather than a generic rule of thumb. For the complete framework tying both readings together, see the complete guide to superheat and subcooling diagnostics. If your subcooling reading comes back high while superheat reads low, that pattern points to an overcharged system rather than a condenser fault — see reading an overcharged system for how to confirm it before pulling gas off. And for how these two numbers are read together as a diagnostic pair rather than in isolation, see superheat vs subcooling: which one tells you what.
+ There are no comments
Add yours