How to check and measure
CO₂ levels in your room

Buying a CO₂ meter is easy. Getting a number from it that means anything is slightly harder, and there are four or five ways to get a confident, precise, completely useless reading. This is the practical version.

Step 1: buy the right kind of sensor

This is the decision that determines whether everything afterwards is worth doing, and it is where most money gets wasted.

NDIR — the real thing

NDIR stands for Non-Dispersive Infrared. Carbon dioxide absorbs infrared light at a specific wavelength, so the sensor shines an IR beam through a small chamber of air and measures how much arrives at the far end. Less light through means more CO₂. It is a direct physical measurement of the gas itself.

A decent NDIR sensor is accurate to somewhere around ±(50 ppm + 5% of reading) after a short warm-up, and it stays that way for years. Every professional air-quality instrument, every building management system, and every device used in published research uses NDIR. The VentPlus Main Unit uses an industrial NDIR sensor reading 400–5,000 ppm.

“eCO₂” — the thing to avoid

Many inexpensive air-quality gadgets, and almost every sub-₹3,000 device, display something labelled eCO₂ or “equivalent CO₂”. That prefix is doing a great deal of work. These devices contain a metal-oxide VOC sensor, which detects volatile organic compounds, and then infers a CO₂ figure on the assumption that VOCs and CO₂ rise together because both come from people.

Sometimes they do. But spray a deodorant, open a bottle of sanitiser, cook, clean with anything, or apply nail polish, and an eCO₂ sensor will report a CO₂ spike into the thousands with no carbon dioxide involved whatsoever. Equally, it will sit at a comfortable 600 in a genuinely stuffy room whose VOCs happen to be low. It is not a CO₂ sensor with a tolerance; it is a different measurement wearing a CO₂ label.

The test before you buy: find the specification and look for the words “NDIR” and a stated accuracy in ppm. If the listing says “eCO₂”, or gives no sensor type at all, or quotes accuracy as a percentage with no ppm term, it is a VOC sensor. In India this single check eliminates most of the cheap market.

Step 2: put it in the right place

Sensor placement is where good hardware gets wasted. The goal is to measure the air people are actually breathing, which is not the same as the air at any convenient mounting point.

• Height: roughly 1.1 to 1.5 m — seated or standing breathing height. In a poultry shed or anywhere the occupants are at floor level, measure at their level, not yours.
• Away from the fresh-air inlet. Mounting next to an AC vent, an open window or a supply grille gives a permanently reassuring reading that describes the incoming air rather than the room.
• Away from the exhaust. The opposite error, and it makes ventilation look worse than it is.
• Not directly in front of a person. Exhaled breath is around 40,000 ppm. A sensor 30 cm from someone's mouth reads their breath, not the room. Keep at least a metre of clearance from where people habitually sit.
• Not in a corner or behind furniture. Dead air pockets lag the room by many minutes.
• In the occupied zone. Corridors, storerooms and lobbies tell you about corridors, storerooms and lobbies.

Step 3: read the trend, not the number

A single reading is nearly worthless. Any NDIR sensor needs about ten minutes of warm-up to be within specification, and more importantly, a room's CO₂ is a moving quantity. What tells you about the building is the shape of the curve.

Three specific readings are worth deliberately capturing:

The unoccupied baseline. Leave the meter running in the empty room for an hour or two — early morning, or a weekend. Where it settles is the most diagnostic number you will get. A room that returns to 450–550 ppm has real air exchange. A room that sits at 800 ppm with nobody in it is barely connected to the outdoors at all, and will start every occupied day already halfway to trouble.

The occupied peak. The worst moment of a normal day: the full meeting, the packed class, the evening in the bedroom with the door shut. This is what the room does under load.

The recovery slope. Open the window or start the fan and watch how fast it falls. Quick recovery means adequate ventilation capacity that simply is not being used — the easiest and cheapest problem to have, because the equipment already exists. Slow recovery means the capacity itself is inadequate.

Step 4: interpret the result honestly

Set the numbers against the standard bands (the full chart is in safe CO₂ levels in a room):

The four mistakes

1. Buying an eCO₂ device. Covered above, and it is the big one. A wrong number is worse than no number, because you will act on it.
2. Measuring once. A spot reading at 11 a.m. tells you about 11 a.m. Rooms are worst at the end of a long occupied stretch, which is exactly when nobody is walking around with a meter.
3. Measuring in the wrong place. Beside a window, under the AC, or a foot from someone's face. All three produce confident nonsense.
4. Stopping at measurement. This is the expensive one. A meter creates awareness for about a week, and then it becomes furniture. The room is not improved by being observed.

Why a controller beats a meter

Almost every room we see in India already has the equipment to fix its own air — an exhaust fan, a fresh-air unit, a window. What it does not have is anything that decides when. So the fan runs all day (wasteful, and it fights the air conditioning) or never (which is the usual outcome), and the meter on the wall quietly records the failure.

A CO₂ controller closes that loop. It measures continuously, and when the level crosses a threshold you have set, it switches the fan on. When the air is clear again, it switches it off. Nobody has to notice, remember, or care. That is the difference between knowing your room is at 2,200 ppm and your room not being at 2,200 ppm.

A quick self-check

If you have not bought anything yet and want a sense of the scale of the problem, run your room through the free CO₂ build-up calculator. Enter the floor area, ceiling height, number of people and a rough guess at ventilation, and it will show you the curve and the steady-state level. It is a model rather than a measurement — but if the model says your bedroom settles at 2,100 ppm overnight, it is telling you something real about the room, and it costs nothing to find out.