What PM2.5 Is, and What Your Aircon Filter Does About It

PM2.5 names a size range, not a material. The label covers any airborne particle measuring 2.5 micrometres across or smaller, whatever it is made of. Smoke from burning vegetation, soot from an engine, and fine road dust all fall into the same category once they are small enough to qualify.

That threshold exists because of how particles at this scale behave. Coarse grit settles under its own weight soon after it is stirred up, while particles this fine stay suspended and drift a long way from where they were released. Smoke reaching Singapore from a regional fire has already travelled before anyone here sees it.

Size also decides how far into the body a particle can get. Larger dust is mostly caught in the nose and upper airway, while particles in the PM2.5 range pass that point and reach deep into the lungs, which is why health authorities track this fraction separately. That depth of reach is what the number in the name marks.

This guide stops at what the particle is and what gets measured. Anything about your own health during a spell belongs with a doctor, not an aircon company.

PM10 is the same idea one step coarser, covering particles up to 10 micrometres. Every PM2.5 particle is also a PM10 particle, because the finer range sits inside the wider one. Both are reported here, and PM2.5 is the one that draws attention during a haze spell.

NEA publishes two figures during a spell, and they answer different questions. The 24-hour PSI is the headline index, computed from several pollutants at once, with PM2.5 among them. The 1-hour PM2.5 reading is narrower and faster, giving the concentration of fine particles measured over the most recent hour.

The averaging window separates them. A 24-hour index absorbs a sharp afternoon spike into the whole day behind it, so it can read lower than the air outside your window looks. The 1-hour figure tracks the air as it moves. When the two disagree, the shorter window describes the present and the longer one the day.

NEA attaches a descriptor to each PSI band, and those descriptors are what news coverage and weather apps quote.

• •0 to 50: Good
• •51 to 100: Moderate
• •101 to 200: Unhealthy
• •201 to 300: Very Unhealthy
• •Above 300: Hazardous

No. The washable filter in a wall-mounted split unit is a coil guard, not an air-quality device. It is a coarse plastic mesh built to stop hair, lint and larger dust reaching the evaporator behind it, so clear fins keep exchanging heat. That is the entire design brief.

The openings in that mesh are sized for that job, which is why fine particles go straight through. A mesh tight enough to hold a 2.5 micrometre particle would choke the airflow the unit depends on, so no maker fits one as the standard washable panel. Whatever comes out grey after a bad week is the coarse fraction.

A grey filter is easy to misread as proof the unit is cleaning the air, when it is proof of the opposite. The mesh held what it was designed to hold, and everything finer carried past it into the room and onto the coil. Anyone selling a haze package on the strength of a dirty filter photo is showing coil protection, not air purification.

Treat the filter as maintenance, then, and nothing more. A haze spell loads it faster than an ordinary stretch, and a loaded mesh weakens airflow before it weakens anything else. Washing it more often through a spell keeps the unit cooling the way it should. It does not change the particle count in the room, and it was never going to.

Three things do genuine work here, and only one of them is the aircon. Sealing limits how much new outdoor air arrives. A purifier with a filter rated for this particle size removes what is already inside. Running the aircon keeps the room comfortable enough that nobody opens a window, which is the contribution people underrate.

A split system is a closed loop on room air, and a separate guide works through that mechanism properly. The part worth carrying across is narrow: a shut, cooled room takes on fine particles more slowly than an open one, but nothing in the system removes the particles already sitting in it.

Size the purifier to the room, not the flat. A unit rated for a small bedroom will not turn over the air in an open-plan living area, and one that cannot cycle the room volume often enough does less than the box implies. Match it to floor area first, then put it where people actually sit.

Sealing is the cheapest of the three and gets skipped most often. Gaps under doors, an unsealed pipe penetration behind the indoor unit, and a bathroom exhaust left running all draw outdoor air in. A purifier working against a leaky room is refilling a bucket with a hole in it.

Fitting a denser filter panel into a split unit costs airflow. Airflow is what the cooling rides on. The indoor blower was sized around the resistance of the coarse mesh the unit shipped with. A tight media in that slot leaves the same fan moving less air across the coil.

Less air across the coil means the coil runs colder than it was meant to. Moisture condensing on the fins stops draining and starts freezing. Ice then narrows the gaps between fins, cutting airflow further and growing faster. By the time the room notices, the unit is blowing weakly and barely cooling.

The melt is the second cost. A coil that ices and then thaws sheds more water at once than the drain pan and pipe were sized to take, and the overflow comes out of the front edge of the indoor unit onto the wall below. A filter upgrade that ends in a wet wall was not an upgrade.

One distinction matters more than the rest. A filter the maker designed and rated for a specific model has had the airflow penalty engineered around it, with the blower matched on the bench. A dense aftermarket panel trimmed to fit the same slot has had none of that. If a shop offers a purifier-grade element for a wall unit, ask what the blower was rated for and whether the maker lists that element against that model.

The aircon and the purifier are two machines because these are two jobs, and folding one into the other tends to degrade the machine that was already working. Run the unit for cooling, keep its mesh clean, and let a purifier deal with the particles.

Two complaints get mixed together during a spell, and they need opposite answers. One is that the room does not feel clean. The other is that the room does not feel cold. The first is an air problem the aircon was never going to solve. The second is a unit problem, and a haze spell makes it more likely to surface.

Weak airflow that arrived during the spell usually traces to a loaded mesh or a coil starting to cake. Wash the mesh first, because it is the fastest thing to rule out and the most common cause by a distance. If airflow comes back, the mesh was the answer. If it does not, the buildup has already moved past it.

A smell that shows up on startup and fades as the unit runs is a different signal. Damp buildup on the fins and in the drain path gives that off, and a spell adds material to both. Neither is somewhere a filter wash reaches, so a clean mesh and a returning smell are not a contradiction.

Cooling that lags with a clean mesh and normal airflow points away from the filter entirely. Refrigerant charge, the outdoor coil and the compressor all sit behind that symptom, and none of them are haze-specific. A bad spell is often just the load that finally exposes a unit that was already marginal.

Push back on any diagnosis that jumps from a haze spell straight to a gas top-up. Weak cooling from a caked coil and from low refrigerant feel identical from the sofa, and the coil is cheaper to check and far more likely after a heavy spell. Ask for the filter and both coils to be looked at first, and treat an earlier refrigerant answer as premature.