11.3 Evacuation, Dehydration & Refrigerant Charging Procedures
Key Takeaways
- The industry standard evacuation target is 500 microns, at which water boils near minus 12 degrees Fahrenheit so residual moisture flashes to vapor and is pumped out.
- A decay test distinguishes the two failure modes: a vacuum that rises and then stabilizes indicates remaining moisture, while one that rises continuously indicates a leak or an open valve.
- Systems with a fixed metering device are charged by superheat; systems with a thermostatic or electronic expansion valve are charged by subcooling.
- Dry nitrogen must flow through the tubing during brazing to prevent cupric oxide scale that later plugs the metering device and filter drier.
- Pressure testing is performed with dry nitrogen and a regulator; oxygen, acetylene, and compressed air are never used, and refrigerant may not be vented as a leak-search gas.
11.3 Evacuation, Dehydration & Refrigerant Charging Procedures
[!IMPORTANT] The three enemies inside a sealed system are moisture, non-condensables, and particulate. Moisture combines with refrigerant and oil to form acids and freezes at the metering device. Non-condensable air occupies condenser surface and raises head pressure. Particulate - usually cupric oxide from brazing without nitrogen - plugs screens and driers. Every procedure in this section exists to keep one of the three out.
Step 1: Braze With Flowing Nitrogen
When copper is heated above roughly 800 degrees Fahrenheit in the presence of air, the inside surface forms a black, flaky cupric oxide scale. It breaks loose in service and plugs the filter drier, the screen, and the metering device.
- Set a nitrogen regulator to roughly 2 to 3 psig and let a slow, steady flow pass through the tubing during heating and while the joint cools.
- Purge with nitrogen before applying heat so the line is not full of air when the torch is lit.
- Never braze on a pressurized system, and never braze with refrigerant in the line - heated refrigerant can decompose into hydrogen fluoride and phosgene.
- Remove or heat-shield Schrader cores, service valve seats, and TXV power heads.
Step 2: Pressure Test With Dry Nitrogen
- Use dry nitrogen through a regulator - never a full cylinder pressure, and never oxygen or acetylene, which react violently with oil.
- Test to the pressure specified by the manufacturer or the applicable code for the system, hold, and watch for decay corrected for ambient temperature change. A 10-degree ambient swing moves the gauge on its own.
- Do not use refrigerant as a trace gas and vent it. Knowingly venting is a violation of the Clean Air Act venting prohibition regardless of the refrigerant class.
Step 3: Deep Evacuation
Why 500 Microns
Atmospheric pressure is about 760,000 microns. As absolute pressure falls, the boiling point of water falls with it:
| Absolute pressure | Water boils at |
|---|---|
| 760,000 microns (atmospheric) | 212 F |
| 25,000 microns | about 80 F |
| 4,600 microns | 32 F |
| 1,000 microns | about 1 F |
| 500 microns | about -12 F |
At 500 microns any liquid water left in the system is far above its boiling point at that pressure, so it flashes to vapor and the pump removes it. Pulling only to 1,000 or 2,000 microns leaves liquid water behind in a system that will never get colder than about 35 degrees at the evaporator.
Setup That Actually Reaches 500 Microns
- Two-stage vacuum pump with clean oil. Change the oil after every job; contaminated oil raises the pump's blank-off pressure.
- Large-diameter, short hoses. A quarter-inch charging hose is a straw. Use 3/8-inch or 1/2-inch evacuation hoses, as short as practical.
- Remove the Schrader cores with core removal tools. A Schrader core alone can double evacuation time.
- Evacuate from both the high and low sides so the vacuum is not drawn through a closed metering device.
- Place the micron gauge at the system, as far from the pump as possible. A gauge at the pump reads the pump, not the system.
- Replace the liquid line filter drier before evacuating; a drier installed after evacuation defeats the purpose.
Triple Evacuation
On a wet system - a burnout, a long open line set, or a flooded unit - break the vacuum twice with dry nitrogen. Pull to about 1,500 microns, break to a few psig with nitrogen, repeat, then pull the final vacuum to below 500 microns. The nitrogen sweeps water vapor that a single pull-down leaves clinging to the oil.
Step 4: Read the Decay Test
Isolate the system from the pump with a valve at the manifold - not by shutting the pump off - and watch the micron gauge.
| Behavior | Interpretation | Action |
|---|---|---|
| Holds below 500 microns | System is tight and dry | Charge |
| Rises, then levels off at a stable value | Remaining moisture boiling off | Continue evacuating; consider triple evacuation |
| Rises continuously without leveling | Leak or an open valve or core | Stop; pressure test with nitrogen and find it |
Step 5: Charge Correctly
Weigh In
The most accurate method whenever the factory charge and the line set length are known. Start from the nameplate charge, then add or subtract for line set length per the manufacturer's per-foot adjustment. Weigh-in is required on any system evacuated to a vacuum, and it is the only defensible method on a fixed-orifice system at low outdoor ambient.
Charge a Fixed Metering Device by Superheat
A piston or capillary tube does not regulate superheat, so the charge sets it.
The target comes from the manufacturer's charging chart, entered with outdoor dry-bulb temperature and indoor return-air wet-bulb temperature. A high indoor wet bulb and a low outdoor dry bulb call for a high target superheat; a low indoor wet bulb and a high outdoor dry bulb call for a low one.
Worked example: R-410A, suction pressure 118 psig, saturation temperature 40 F, suction line temperature 55 F. Total superheat is 15 F. If the chart calls for 10 F at the measured conditions, the system is undercharged and refrigerant is added until superheat falls to target.
Charge a TXV or EEV System by Subcooling
An expansion valve maintains its own evaporator superheat, so superheat tells you almost nothing about charge. Subcooling does.
Most manufacturers target roughly 8 F to 12 F of liquid subcooling; always use the value on the data plate or in the installation instructions.
Worked example: R-410A, discharge pressure 335 psig, saturation temperature 104 F, liquid line temperature 89 F. Subcooling is 15 F against a 10 F target - the system is overcharged, and refrigerant must be recovered, never vented, until subcooling reaches target.
[!WARNING] Do not diagnose charge on a system with an airflow problem. Low airflow across the evaporator lowers suction pressure and mimics an undercharge. Verify the filter, coil, blower wheel, and total external static pressure before touching the gauges, or you will chase an airflow fault by adding refrigerant and end up with a flooded compressor.
A technician pulls a system to 480 microns, closes the valve isolating the vacuum pump, and watches the micron gauge climb to 1,900 microns over ten minutes, where it stops and holds steady. What does this indicate?
An R-410A split system with a thermostatic expansion valve shows a discharge pressure of 335 psig (104 F saturation) and a liquid line temperature of 89 F. The data plate calls for 10 F of subcooling. What is the condition and the correct action?
Why must a low flow of dry nitrogen pass through copper tubing while brazing a refrigerant line set?