Learn / Measuring

Can I trust my CO2 monitor? Four checks that take one evening

Check the sensor type, its stated accuracy, the outdoor reading and the way you use it.

Jump to this week’s step

A monitor on a shaded porch table in open outdoor air, away from people and exhaust.
Use a shaded spot in open air, away from people, cars and exhaust. Follow the maker’s instructions.

At a glance

In this guide 8 sections

If your monitor has a true CO2 sensor and reads close to outdoor air when you check it outside, it's accurate enough for a bedroom. Good home monitors are rated within about 55 to 90 ppm at 800 under lab conditions, and the bedroom goal has a margin for that. In a real bedroom, the bigger risks are a monitor that has learned the wrong "fresh air" level (Check 4) and, at high elevation, thin air.

Check 1: Is it measuring CO2 at all?

Not every device with a CO2 number measures CO2. NDIR (non-dispersive infrared) sensors detect it directly, from the way CO2 absorbs infrared light. The UK Health and Safety Executive recommends NDIR monitors as the most suitable type for checking ventilation. Some newer sensors use a photoacoustic form of NDIR, sometimes labeled PAS, and measure CO2 directly too.

Other devices estimate a CO2-like number from other gases, usually labeled eCO2 or CO2eq, often next to a TVOC figure. The datasheet for one common chip of this kind, Sensirion's SGP30, says its CO2eq signal is calculated from ethanol and hydrogen measurements. It can move for reasons that have nothing to do with fresh air, so it won't give you a fair bedroom answer.

Look for NDIR or photoacoustic on the box, in the manual or on the product page. The Aranet4 HOME and SwitchBot Meter Pro CO2 both use NDIR sensors.

Check 2: How close is close enough?

Makers state accuracy as a fixed amount plus a percentage of the reading. The Aranet4 lists ±(30 ppm + 3%), and the SwitchBot Meter Pro CO2 ±(50 ppm + 5%). At the readings that matter for a bedroom, that works out to:

ReadingAranet4 HOMESwitchBot Meter Pro CO2
800 ppm (the goal)±54 ppm±90 ppm
1,000 ppm±60 ppm±100 ppm
1,400 ppm±72 ppm±120 ppm

These are lab ratings. Aranet's applies at 15–35 °C and 0–80% humidity, and some monitors reach theirs only after about a week of settling; Airthings says so of its Wave Plus.

That's one reason the goal sits at 800. The 2025 review behind it found sleep slightly but measurably more disturbed from about 1,000 ppm and no measurable difference at 850 or below, and its authors set the goal at 800 partly to allow for sensor accuracy. Even at the edge of its stated accuracy, a good monitor reading 800 is well short of 1,000, as long as it passes the outdoor check. Why 800, and not 1,000.

Check 3: The outdoor check

Outdoor air is the one reference everyone has. If you can, do this in the afternoon, when outdoor CO2 is lowest and steadiest; AirGradient recommends noon to 4 pm. Put the monitor in a shady spot away from people, cars, doors and exhaust vents, step at least 3 ft (1 m) away, and give it 15 to 30 minutes, until three readings in a row agree. Monitors update slowly, so plug a SwitchBot in or press its top button (on batteries it updates only every 30 minutes), and set an Aranet4 to a 1- or 2-minute interval.

  • About 400 to 450 ppm: fine. NOAA's latest monthly averages are about 426 ppm, and many monitors treat fresh air as 400 (the Aranet4, about 420), so they read a little lower.
  • Below about 380: it's reading low, most often because automatic calibration has learned a stuffy room as fresh air (Check 4).
  • Above about 500 in clean air: it's reading high. Try a spot farther from roads and the house first; AirGradient notes that a busy road or an exhaust vent can push a reading hundreds of ppm above background. SwitchBot recommends recalibrating above 500.

These cut-offs are HouseExhale's rules of thumb. If the reading is off, recalibrate with the maker's steps, and only ever outdoors: calibrating tells the monitor "this is fresh air."

Check 4: Does it recalibrate itself?

Most CO2 monitors recalibrate themselves. Every day, week or month, depending on the model, they treat the lowest level they've seen as fresh outdoor air, about 400 ppm, and shift every reading to match. That works in a room that gets fresh air now and then. In a bedroom that never airs out, it doesn't: if the room's lowest level is really 600, the monitor treats 600 as 400 and slowly starts reading about 200 ppm low, so a stuffy room can look fine. Senseair, which makes sensors used in several home monitors, says the feature doesn't suit spaces where CO2 never falls back to outdoor levels.

  • Aranet4: ships with automatic calibration off, so it won't drift low in a stuffy bedroom, but it won't correct itself either. Aranet suggests calibrating about once a year.
  • AirGradient and INKBIRD IAM-T2: let you turn it off, in AirGradient's dashboard or the INKBIRD app.
  • Airthings and Awair: recalibrate themselves with no documented off switch, and ask for fresh air at least once a week.
  • SwitchBot and others: the makers don't say.

If your bedroom never gets near outdoor levels in a normal week, turn automatic calibration off where you can, or give the room or the monitor a few hours of fresh air once a week. Then repeat the outdoor check every month or so, and before any before-and-after comparison.

At high elevation

CO2 sensors read low in thin air unless they correct for air pressure, and an outdoor calibration doesn't fix that. Aranet says the Aranet4 corrects for it down to 750 hPa, roughly 8,000 ft. Sensirion's sensors assume sea-level pressure unless the monitor's maker supplies the real figure, and none of the other makers on our list says whether theirs does. By our arithmetic from standard air pressure (about 82% of sea level's in Denver, 74–77% at 7,000–8,000 ft), at Denver's elevation a monitor that doesn't correct for air pressure shows a bedroom that is really at about 900 as about 800; at 7,000–8,000 ft, one really at about 950. If yours isn't an Aranet4, aim for about 700 on its screen and treat 700–800 as borderline. In the HouseExhale app, add your elevation in Home details (tap the gear at the top of any screen) and the room report shows the converted number.

Why two monitors disagree

Put two good monitors side by side and they'll rarely match exactly. SwitchBot's own help page says "different brands use different sensors and technologies, so direct comparisons may not provide meaningful results." At 800 ppm, an Aranet4 and a SwitchBot can both be within spec and still be about 140 ppm apart, and one may have learned a different fresh-air level. To compare them fairly, put them side by side outdoors in the afternoon for the same 15 to 30 minutes. Comparing one on the dresser with another by the pillow tells you about placement, not accuracy.

Telling a real test from a "tested and ranked" list

Many monitor roundups say they tested monitors without saying how. A real test names the reference instrument it compared them against and the conditions, and shows its numbers.

A CO₂ monitor on a dresser across the bedroom, away from the pillow and window.
Measure the room: keep the monitor away from faces, direct drafts and sun.

What matters more than accuracy

Placement and consistency. A monitor beside your pillow catches your breath directly; HSE notes that breath close to the sensor can make readings misleadingly high. Keep it at about head height, at least 2 ft (60 cm) from anyone's face, about 3 ft (1 m) from windows, doors and vents, and out of direct sun. Where to put your monitor.

Then use the same monitor in the same spot every night. If yours reads a little high, it reads a little high every night, so a before-and-after comparison still shows whether a change helped, as long as it hasn't recalibrated itself in between. That's how HouseExhale compares nights: same room, same monitor, one change at a time, each night scored by its average while you slept.

This week: Check your monitor

  1. Look for NDIR or photoacoustic (PAS) on the box, manual or product page. If it only says eCO2 or CO2eq, or shows CO2 next to TVOC, the number is an estimate.
  2. Get it updating often for the check: plug a SwitchBot into USB or a power bank, or set an Aranet4 to a 1- or 2-minute interval.
  3. In the afternoon if you can, take it outside to a shady spot away from people, cars, doors and exhaust vents. Step at least 3 ft (1 m) away and wait 15 to 30 minutes, until three readings in a row agree.
  4. About 400 to 450 ppm is fine. Below about 380, it's reading low; above about 500 in clean air, it's reading high. Recalibrate only outdoors, with the maker's steps.
  5. If your bedroom never airs out in a normal week, turn automatic calibration off if your monitor allows it, and repeat this check every month or so.

Sources