Charge Verification by Weight and Superheat

Purpose

Weight is how this shop puts charge into a system. Superheat and subcooling are how it proves the weight was right. Those are two different jobs and the damage comes from doing the second one first: a tech who reads a number, adds gas, reads again and repeats has no idea what is in the system and has usually corrected for something that was never a charge problem. A dirty condenser and an overcharge move subcooling the same direction, and only one of them is fixed by recovering refrigerant. This procedure makes the weight the record and the verification the check, so that a system leaving with a passing number also has a known quantity in it.

Scope

Covers charge determination and verification on residential and light commercial split systems and packaged units after any opening, and on a service call where charge is in question. Covers weighing in, the metering-device fork between superheat and subcooling verification, and the airflow and ambient conditions that have to be true before either number means anything.

Does not cover evacuation, which the evacuation SOP owns and which must be complete before charge is released. Does not cover leak repair. Does not cover topping off a system with a known active leak, which this shop does not do without a documented repair plan and the customer's agreement in writing.

Roles and responsibilities

Procedure

1. Identify the metering device before any gauge goes on, because it decides which number verifies the charge. A thermostatic or electronic expansion valve controls evaporator superheat itself, so on those systems superheat tells you about the valve and the evaporator load, not about how much refrigerant is in the system, and subcooling is the verification. A fixed orifice, piston or capillary system has no such control, so superheat against a target chart indexed by indoor wet bulb and outdoor dry bulb is the verification. Acceptance: the device type written on the ticket from the nameplate or a physical look, not assumed from the model year. Stop rule: unknown device type means you do not read a verification number yet.

2. Verify indoor airflow before you trust any refrigerant reading. Check filter condition, coil cleanliness, blower speed tap or setting, and measure total external static against the nameplate rating. Acceptance: static at or below the nameplate maximum and a clean coil and filter. Low indoor airflow puts less heat into the evaporator and drives superheat down, which on a fixed orifice system reads exactly like an overcharge, so a tech who skips this step will recover refrigerant from a correctly charged system. Hazard in this step: the blower wheel is turning during a static reading, so probes go into drilled test ports and hands stay outside the housing. Stop rule: static over the nameplate figure means the airflow is the finding and the charge verification waits.

3. When the system has been opened, start from zero rather than from a guess. Recover the remaining charge, weigh it, and write down what came out. Acceptance: a recovered weight on the ticket, and a system at atmospheric before evacuation begins. What wrong looks like: adding to a partial charge of unknown size, which makes the final quantity unknowable and destroys the value of the record. Hazard in this step: a recovery cylinder is never filled beyond 80 percent of its water capacity by weight, and it goes on the scale so you know when you are near that limit rather than judging by feel.

4. Compute the required charge from the nameplate, with the line set measured, not estimated. Read the factory charge, the line length that charge covers, and the adder per foot of liquid line beyond it, then measure the actual liquid line run. Acceptance: an arithmetic line on the ticket showing factory charge, feet beyond the covered length, the adder, and the total. What wrong looks like: using the factory charge alone on a long line set, which leaves the system light by a real margin on anything much past the covered length. Stop rule: if the nameplate is unreadable or the adder is not stated, get the figure from that manufacturer's installation manual before charging, do not interpolate from another model.

5. Weigh the charge in on a calibrated scale, and record the actual delivered amount. Tare the cylinder, deliver the computed quantity, and write down what the scale says was delivered rather than what you intended to deliver. Acceptance: delivered weight within the tolerance your shop sets against the computed figure, commonly a fraction of an ounce on a residential system. Hazard in this step: liquid refrigerant into the suction service port of a running compressor will slug it, so liquid goes in on the high side with the system off, or through a metering device made for the purpose, or it goes in as vapor. Gloves and eye protection stay on, because liquid on skin is a cold burn.

6. Let the system stabilize and check the ambient gate before you read anything. Run the system in the mode you are verifying long enough for pressures and line temperatures to settle, commonly on the order of 15 minutes with a steady load. Acceptance: readings that have stopped moving over a couple of minutes, and an outdoor ambient inside the range the manufacturer's charging chart is valid for, which for cooling mode is commonly not below about 55 to 65 F. What wrong looks like: reading a system two minutes after startup, or verifying cooling charge on a cool spring morning where the chart does not apply. Stop rule: outside the chart's ambient range, the weighed figure stands and the verification is deferred, noted on the ticket as deferred rather than recorded as passed.

7. Verify against the target the manufacturer states, using the method step 1 selected. For a TXV or EEV system, take liquid line pressure and liquid line temperature at the same point, convert pressure to saturation temperature off the correct refrigerant chart, and subtract measured liquid temperature to get subcooling. For a fixed orifice system, take suction pressure and suction line temperature, convert, and subtract saturation from measured to get superheat, then compare to the target for the measured indoor wet bulb and outdoor dry bulb. Acceptance: the value inside the manufacturer's stated band. Hazard in this step: hoses are on a high-side port at operating pressure, so eye protection stays on and the manifold is not left unattended with a customer or a child nearby.

8. Out of band means diagnose, not adjust. High subcooling has three common causes and only one of them is overcharge: a fouled or blocked outdoor coil and a restriction downstream of the condenser both back liquid up in the condenser and raise it the same way. Low subcooling can be undercharge, a leak, or a load higher than the system can meet. Acceptance: a named cause supported by a second reading before any refrigerant moves. What wrong looks like: recovering gas until subcooling lands in band on a system with a cottonwood-packed condenser, which leaves the system undercharged and the coil still blocked. Stop rule: no refrigerant is removed or added on a verification number alone; airflow and coil condition are ruled out first, and the reason goes on the ticket.

The record this produces

One charge block per system: metering device type; indoor static pressure against nameplate; recovered weight if the system was opened; the nameplate factory charge, the covered line length, the measured line length, the per-foot adder and the computed total, shown as arithmetic rather than as a single number; the delivered weight from the scale; outdoor ambient and indoor wet bulb at verification; and the verification value with the manufacturer's target beside it.

The computed arithmetic is the part shops leave out and the part that pays later. A leak search two years on starts by asking what should be in the system, and a recorded total with its derivation answers that in one line. The delivered weight matters for the same reason: the difference between what was computed and what the scale actually moved is the only honest figure, and a manufacturer reviewing a compressor claim will ask for it.

Worked pass: condenser and line set replacement, R-410A, TXV indoor coil

Outdoor unit and line set replaced, existing TXV evaporator retained. Step 1 confirmed a TXV, so subcooling is the verification and superheat is not a charge indicator on this system. Step 2 measured total external static below the air handler's nameplate maximum with a clean filter and coil, so airflow was not going to distort the reading.

Step 4 arithmetic, off the outdoor unit's data plate: factory charge 9 lb 4 oz, which is 148 oz, covering 15 feet of line set, with an adder of 0.6 oz per foot of liquid line beyond that. The liquid line measured 47 feet, so 47 minus 15 is 32 feet beyond, and 32 times 0.6 is 19.2 oz. Total required is 148 plus 19.2, which is 167.2 oz, or 10 lb 7.2 oz. Step 5 delivered 167.2 oz on a scale calibrated that month, recorded as delivered rather than as intended.

Step 6: ran 18 minutes in cooling with outdoor ambient 88 F, readings steady, well inside the chart's valid range. Data plate target subcooling 10 F with a plus or minus 2 F band, so 8 to 12 F.

Step 7 FAILED. Liquid line pressure read 340 psig, which off the R-410A chart is a saturation temperature of about 105 F. Liquid line temperature at the same point read 91 F. Subcooling is 105 minus 91, which is 14 F, above the 12 F top of the band.

Under step 8 the tempting move is to recover a few ounces until it lands at 10. That is the exact failure this SOP exists to prevent, because the system had a known, weighed, arithmetically derived 167.2 oz in it, and high subcooling has causes other than overcharge. The condenser was checked and the outdoor coil was packed with cottonwood seed on the entering face, invisible from the top. The coil was cleaned, the system re-run and re-stabilized for 15 minutes at the same 88 F ambient.

Second reading: liquid line pressure 317 psig, saturation about 100 F, liquid line temperature 89 F, so subcooling is 100 minus 89, which is 11 F, inside the 8 to 12 F band. Head pressure fell and subcooling fell with it, which is the direction a restored condenser moves both, and no refrigerant was removed at any point.

What the failure teaches: the verification was genuinely out of band and the weighed charge was genuinely correct, both at the same time. A tech who treats the verification as the method removes roughly the difference between 14 F and 10 F worth of charge, then hands over a system that is undercharged, still has a blocked coil, and now carries a record that disagrees with what is actually inside it.

References

• Manufacturer data plate and installation instructions for the factory charge, covered line length, per-foot adder, and the subcooling or superheat target, which govern over any general rule of thumb
• Refrigerant pressure temperature chart for the specific refrigerant, read on the liquid side for subcooling and the vapor side for superheat
• 40 CFR Part 82, Subpart F, for recovery requirements and the prohibition on venting
• See related: Evacuation and Vacuum Decay Verification; Callback Within 30 Days Investigation