Aircon commissioning readings: the record nobody asks for
Almost nobody asks an installer what the system read on the day it was commissioned, so almost nobody has those numbers. Their value shows up much later, when a fault has to be argued about and there is nothing to argue from.
By Team Snowflake | Updated 5 Aug 2026
What commissioning means, and what it is not
Commissioning is the point where a new system is proven to work correctly, not the point where it is switched on. Switching on shows that power reaches the unit and the fan turns. Commissioning shows that the sealed circuit was prepared properly, filled with the right weight of refrigerant, and running the way the maker intended. One is an action. The other produces figures.
A brand new aircon blows cold on handover day almost regardless of how carefully it was put in. A shallow vacuum, a charge estimated instead of weighed, a service valve left part open: none of those stop cold air arriving at the grille that first afternoon. What they do is shorten the life of the machine and raise what it costs to run, and both of those show up slowly.
That gap is why commissioning has to be written down rather than felt. A walkthrough at sign-off is an observation exercise, and a good one. It confirms each unit responds, drains and cools, which is what the installation handover checklist is for. It cannot confirm anything about the state of the sealed circuit, because a sealed circuit shows no symptoms on the day it is filled.
Neither side is hiding anything here. The figures simply never come up. Owners do not know to ask, installers are not routinely asked, and a record nobody requests is a record nobody keeps. The result is a large stock of Singapore flats running systems whose original condition was never captured by anyone.
The readings that belong on the record
Five figures carry almost all of the later value. Each one pins down a different part of the system at the one moment when everything was new, the installer was still in the flat, and the instruments were already out. Recorded together they describe a specific machine in a specific home, which is what makes them worth keeping.
The vacuum reached before charging is the only one that can never be recovered afterwards. After the gas goes in, the state the pipework was left in becomes unknowable, and no later test brings it back. A written figure, with a note on whether it stayed put after isolation, is the entire evidence trail for that step. What the reading proves and what a weak one costs is set out in the guide to the vacuum and pressure test.
The weighed charge is the figure most often missing and most often needed. A split system ships with a factory charge sized for a nominal pipe length, and any run beyond that needs refrigerant added by weight. A proper record shows the factory figure, the addition made for the actual run, and the total that went in. Without it, nobody can later say whether the system is low, correct, or was never filled right to begin with.
Running pressures and the air temperature drop describe how the system behaved once it settled. Pressures from both sides go on the sheet, with the outdoor conditions noted beside them. That pair is the reference a future set of pressures gets read against. The air temperature drop is the only line an owner can repeat alone. It needs a thermometer and no access to the circuit. Taking one properly is covered in the guide to measuring an aircon temperature split.
Compressor running current closes the electrical side. A compressor draws a characteristic current for a given load, and that figure drifts as windings age, bearings wear, or the system is pushed harder than it was meant to work. Captured at commissioning, it becomes the line every later reading is measured against. Captured for the first time during a breakdown, it is a number with nothing behind it.
| Reading recorded at commissioning | What it pins down | What a later departure from it suggests |
|---|---|---|
| Reading recorded at commissioningVacuum reached, and whether it stayed put | What it pins downThat the circuit was dry and sealed before charging | What a later departure from it suggestsNothing later, which is exactly why it has to be captured on the day |
| Reading recorded at commissioningWeight of refrigerant charged in, including the pipe allowance | What it pins downExactly how much gas the system started with | What a later departure from it suggestsGas has left the circuit, or was never fully put in |
| Reading recorded at commissioningRunning pressures on both sides, with outdoor conditions | What it pins downHow the circuit behaved when everything was new | What a later departure from it suggestsA charge change, a restriction, or a compressor losing capacity |
| Reading recorded at commissioningAir temperature drop across each indoor unit | What it pins downDelivered cooling per unit, repeatable by the owner | What a later departure from it suggestsCapacity is down somewhere, though not yet which part |
| Reading recorded at commissioningCompressor running current under settled load | What it pins downThe electrical signature of a healthy compressor | What a later departure from it suggestsMechanical wear, a failing capacitor, or a system working too hard |
Two more figures worth having if the technician took them
Superheat and subcooling belong on the same sheet whenever they were measured. They sit a level deeper than pressures, and between them they separate a genuine shortage of gas from a blockage that more gas will not clear. A separate guide covers what each one rules in and out.
Both figures move with operating conditions, so a bare number helps nobody. Whatever gets written down has to carry the conditions beside it: the mode, the fan speed, the set point, what the room was reading, and what the weather outside was doing. A figure stripped of its context has nothing to be measured against afterwards, which makes it decoration rather than evidence.
A fault is a departure from a baseline
Every diagnosis is a comparison. A technician looks at what a system is doing and decides whether that counts as normal. Where a commissioning record exists, normal is a documented fact about this machine in this flat. Where it does not, normal is an estimate drawn from other machines in other flats.
That estimate is looser than most owners assume. Two units of the same model and capacity will not read the same once fitted. Pipe run length, how the outdoor unit sits, how much afternoon sun the ledge takes, and what the room is doing all move the figures. A number that is perfectly healthy on one install sits at the edge of acceptable on another.
So a technician meeting an unfamiliar system has to rebuild the baseline before diagnosing anything. That rebuild is judgement, and judgement is where honest technicians disagree with each other and where a dishonest one has room to move. A commissioning record removes the rebuilding step. It swaps an opinion about what normal should be for a measurement of what normal was.
This is the difference between a complaint and a case. A room that does not feel as cold as it used to cannot be tested, and it earns an opinion in reply. A temperature drop several degrees narrower than the figure taken at commissioning, on the same fan setting with the room in a similar state, can be tested. The second version gets acted on.
Why the warranty window turns on the record
The warranty period is where an absent record costs real money. A fault under cover raises one question ahead of every other: was the part defective, or was the system fitted and filled wrongly? Makers stand behind the first. Installers stand behind the second. Nobody stands behind an argument that cannot be settled either way.
Without figures from the day, that question has no factual answer. A compressor that dies young might have failed on its own, or it might have spent its whole life running against a wet circuit or a charge that was never right. Both stories fit the same dead compressor equally well. The side that argues most confidently wins, and the owner is rarely that side.
With the figures, the question turns into arithmetic. Three entries settle it between them. The vacuum was reached and it held. The weighed charge matches the nameplate plus the allowance for the pipe run. The settled pressures sit inside the window the maker states. Together they establish that the system left commissioning in the state it was meant to be in. A fault after that points at the equipment. A record showing a charge that was estimated rather than weighed points somewhere else, and it points there in writing.
Ask for the record in a form that survives. A photo of the gauge, a filled commissioning sheet, or a written summary in the installer's own message all work equally well. Verbal confirmation on handover day does not, because it converts into nothing at all at the moment it is needed. Store it where the warranty card and the model numbers already live.
Ask at quotation, not at handover
Raise the readings while quotes are still being compared. An installer who knows the figures are expected turns up with the instruments and takes them as part of the job. An installer told at sign-off that a commissioning sheet is wanted has nothing to write on it, because the moment for capturing most of those figures has already gone.
Asking early also changes the quotes in a useful way. Proper commissioning takes instruments, attention and a technician willing to stand behind what the gauge said. A quote priced with that in it reads differently from one priced without, and the difference is the information. How a firm answers the question tells you far more than the finished job does, since a rushed install and a careful one look alike once the trunking is on.
Put the request in writing at the point of acceptance, and keep it to what the installer should already be reading:
- The final vacuum figure at the system, and how the gauge behaved once the pump was shut off
- Nameplate charge, the extra weighed in for the pipe run, and the total in the circuit
- Suction and discharge pressures once running had settled, with the outdoor conditions noted
- Intake and outlet air temperatures at every indoor unit, with fan speed and set point beside them
- Compressor running current, taken alongside the pressures rather than separately
- Model and serial numbers of each indoor and outdoor unit, so the figures tie to the actual machines
What a straight answer looks like
The reply worth hearing describes instruments and a sheet, not reassurance. Firms that commission properly treat the request as ordinary, because they are being asked to write down work they already do. Firms that do not are being asked to start doing something new, and the answer usually shows which of the two you are speaking to.
Push back on the answer that the system will be tested to make sure it is cooling well. Cooling well on the day is not a test of anything. It is the one outcome that happens whether or not the circuit was prepared, whether or not the charge was weighed, and whether or not the compressor is drawing what it should. The whole reason for taking figures is that the room cannot tell you any of it.
If nothing was recorded on your system
Most systems in Singapore flats were handed over with nothing written down at all. There is no way to go back for it. The vacuum figure is gone for good. The weighed charge is gone unless the installer kept a job sheet, which is worth one polite request and no more. Chasing figures that were never taken produces a story rather than a record, and a story is not something a later technician can measure against.
The workable move is a baseline captured now, while the system is behaving. A baseline taken from a healthy machine does nearly the same job as one taken from a new machine, since the purpose is identical: to be the thing a future reading gets compared against. Starting late is a weaker position than starting at commissioning. It is a far stronger one than never starting.
Part of it needs no technician. The air temperature drop across each indoor unit, taken at a fixed fan speed with the room conditions written beside it, needs one thermometer and no access to the sealed circuit. Repeat it now and then, and the record builds itself out of readings the owner took. Everything on the refrigerant and electrical side needs gauges and a clamp meter on a running system, which makes the next service visit the natural moment to ask.
Ask for those figures to be written down at a visit where nothing is wrong. A service report that captures a healthy system is among the most useful documents a homeowner can hold. It is also almost never requested. A visit where a fault is already present is the wrong occasion to work out what normal looked like. By then the system has stopped showing it.
What a late baseline cannot settle
A baseline taken long after installation says nothing about how the system was fitted. It cannot show whether the circuit was properly emptied, whether the charge was weighed, or whether anything was wrong from the first day. Those questions close for good once the original figures go untaken. No amount of later measurement reopens them.
What a late baseline does is protect everything from here forward. It turns the next fault into a comparison instead of a guess. It gives a second opinion something concrete to work from. It puts a number on any slow decline that would otherwise be argued about in adjectives. That is a smaller claim than a full commissioning record. It is still worth the request.
Ready to get started?
Tell us what’s going on. Symptoms, setup, photos, anything we should know. We’ll assess and come back with the right next step.