Refrigerant — sample cylinder
Purity and contamination testing to AHRI Standard 700. An evacuated, tare-weighed cylinder, filled from the liquid valve of the source, leak-checked before it travels.
- Cylinder
- 400 psi
- Fill target
- ~350 g
// Sampling
The result is decided before the sample reaches us
A laboratory measures the sample, not the machine. Both of our procedures are published here in full — the cylinder draw and the oil bottle — along with the specific errors that make sound material fail and unsound material pass.
01Why technique dominates
The draw is the largest error term
Nothing on the bench can recover information the sample never carried. The contaminant limits in AHRI 700 are set at parts-per-million and fractional-percent scale, and at that scale what happens at the machine bounds everything that happens after it.
What a limit means in the hand
3.5 mg
That is the total mass of water permitted in a 350 g sample of R-22, whose limit is 10 ppm by weight. A small fraction of a single drop. That allowance has to survive your cylinder, your hose and your fittings before it ever reaches ours, which is why an unevacuated cylinder is not a small procedural shortcut.
Limits are per refrigerant — see the table →One sample cannot serve both phases
Liquid phase
6 characteristics
- Water
- All other volatile impurities
- High boiling residue
- Particulates and solids
- Acidity
- Chloride
Vapour phase
1 characteristic
- Air and other non-condensables% by volume @ 25.0 °C
A liquid draw cannot report non-condensables and a vapour draw cannot report the other 6. If your job needs both, it needs two vessels — and the cylinder procedure says where in the sequence the second one is taken.
02The two procedures
Different vessels, different failure modes
They share almost nothing. One is a pressure transfer where the enemy is air; the other is a decanting job where the enemy is time.
Oil — glass sample bottle
Condition and wear analysis for chiller and refrigeration compressor oil — mineral, alkylbenzene, POE and PVE. Standing oil discarded first, the bottle allowed to stop degassing, then sealed and sent without sitting on a shelf.
- Vessel
- Borosilicate
- Before sealing
- Degas
Worth being clear about: the AHRI Refrigerant Testing Laboratory programme certifies refrigerant testing. It has no oil scope, and the oil procedure does not borrow it. Refrigerant results are measured against a published standard; oil results are read as condition against the machine’s own history.
03The kit
What arrives, and what you already have
A kit is a vessel, its identification label, the written procedure and the form that travels back with the sample. Everything else the procedure calls for is standard service equipment you are carrying anyway.
| Item | Refrigerant kit | Oil kit |
|---|---|---|
| Sample vessel | Sample cylinder rated 400 psi | Borosilicate graduated glass sample bottle |
| Labelling | Included. Confirm current packaging, marking and declaration requirements with your carrier before you ship — the shipper is responsible for them and they change. | |
| Written procedure | Also published here | Also published here |
| Sample form | Filled in at the machine, and comes back in the box with the sample | |
| What the procedure assumes you have | Vacuum pump and a micron gauge, a scale that resolves single grams, a short charging hose, leak-detection solution, a way to read source temperature | A live sample point you can drain, somewhere clean to set the bottle down, and a marker |
| Identity | Register the sample before it travels, and label the vessel before it is full | |
The sample cylinder is laboratory property and comes back with the sample inside it. What is in a kit, in full.
04Rejected, or worse, reported
8 ways a sample lies
A rejected sample costs you a return trip. A sample that is accepted, tested and quietly wrong costs more than that, because it looks exactly like a result. Each row names the mechanism, what it does to the reported number, and the one action that prevents it.
| Failure mode | What it does to the number | Avoided by |
|---|---|---|
| Air drawn in during the draw | Reads as non-condensables, the one characteristic measured on vapour. Material that left the plant room in specification arrives out of it, and there is nothing in the result that distinguishes air you introduced from air the system had. | Purge the hose at the source before opening the sample cylinder, keep the hose short, and make the connection with the source valve — not the cylinder valve — as the last thing you open. |
| Sample cylinder not evacuated | The same failure as above, plus moisture. Whatever was in the cylinder is in your sample, and it is measured against a limit expressed in parts per million. | Pull the empty cylinder to 50 microns or below and confirm it on a micron gauge, not on the pump's running time. |
| Drawn from the vapour valve when liquid was required | Every characteristic except non-condensables is defined on the liquid phase. A vapour draw off a blend reports the composition of the vapour space, which is not the composition of the charge, and it leaves oil, high-boiling residue and acid behind in the vessel. | Draw from the liquid valve. If non-condensables are also required, that is a second sample, not the same one. |
| Cylinder overfilled | A liquid-full cylinder has no vapour space left to absorb thermal expansion. It is a hazard in a warm vehicle, and a cylinder that cannot be opened safely is a sample that has to be taken again. | Fill to 75–80% of cylinder volume by weight — about 350 g — with the cylinder on a scale. |
| Oil taken from a dead leg | A drain valve, a gauge line or a capped port holds oil that stopped circulating at some unknown point in the past. The result is a true reading of that fitting and tells you very little about the machine. | Choose a point where oil is moving, discard the standing oil before the bottle goes under, and use the same point every time so a trend means something. |
| Bottle left open, or sealed too soon | Left open, the oil takes on room moisture and the moisture result becomes a measurement of your plant room. Capped while still degassing, dissolved refrigerant pressurises the bottle and pushes oil past the cap. | Let the sample stop releasing gas, then seal it immediately. The gap between those two things is the whole technique. |
| Source temperature not recorded | Non-condensables are defined at a stated temperature. Without the source temperature at the moment of sampling, the result has no reference condition attached to it. | Write the source temperature, the sample point and the time on the record when you take the sample, not when you get back to the truck. |
| Vessel arrives unidentified | A cylinder with no label is a sample with no identity, and no laboratory can honestly attach a result to it. | Register the sample before it travels, and label the vessel before it is full — oil on the outside of a bottle will not take a marker. |
If something went differently at the machine than the procedure says, tell us when you register the sample rather than leaving it out. A known deviation is information we can report around. An unknown one becomes part of the result.
05What travels with the sample
The record is part of the sample
6 fields, written at the machine. Every one of them changes how a result is read, and none of them can be reconstructed later.
Refrigerant or oil type
Limits are per refrigerant, not one generic column. For oil it sets what the numbers are compared against.
Source temperature at the time of sampling
The non-condensables measurement is defined at a stated temperature, so the source condition is part of the record.
Sample point
Distinguishes a circulating sample from a standing one, and lets the next sample be taken from the same place.
Machine or vessel identity
A result without a subject is not a result. This is also what makes the second sample a trend rather than an anecdote.
Date and time taken
Moisture and acidity move while a sample waits. The interval between the draw and the bench belongs in the record.
Anything that deviated from the procedure
A disclosed deviation can be reported around. An undisclosed one is indistinguishable from a property of the material.
Before you send anything
Amalina Technologies shares ownership with Chiller Services Refrigerants, a refrigerant reclaimer in the adjacent unit, and a purity result can be an input to what reclaimed material is worth. We state it rather than leave it to be discovered, and we set out how the laboratory is kept separate from it.
Read the independence policy →See what comes back
The specification your sample is judged against is published in full, and so is the report form we issue — reproduced on an illustrative sample and annotated row by row.