As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
Most CO2 monitors that appear to have failed are working correctly and are being read wrongly. The three common causes are a stale baseline (the sensor has not seen fresh outdoor air in days or weeks), a placement problem (the unit is in a dead zone, in direct sun, or next to a breathing person), or a display that has frozen on the last good reading. Before replacing anything, take the unit outdoors for twenty minutes and watch whether the reading falls toward the outdoor background of roughly 420 ppm. If it does, the sensor is alive and the fault is in the room or in the calibration history. If it does not move at all, the sensor or its firmware has failed. The products referenced here, such as the wall mount air quality monitor with CO2 detector, illustrate the different sensor classes, but no monitor is immune to the baseline problem described below.
The reading is stuck at zero, or at a number that never changes
A CO2 reading that sits at exactly zero is almost never a real measurement. Outdoor air contains roughly 420 ppm of carbon dioxide, and indoor air is nearly always higher. A true zero means the sensor is reporting no absorption in its optical path, which happens for one of four reasons.
- The sensor has failed or its lamp has aged out. NDIR sensors use an infrared source and a detector; when the source dims, the instrument can lose span and read low or zero.
- The unit is in a startup or warm-up state and is showing a placeholder. Many consumer monitors show zero or dashes for the first thirty to sixty seconds.
- The firmware has crashed and the display is frozen on the last value written. Power cycling is the test.
- The unit is not actually a CO2 monitor. Some inexpensive air quality devices labelled as CO2 meters use a VOC sensor and estimate CO2 from it. Those estimates can sit at zero in clean air. The listing for the 16-in-1 indoor air quality monitor states CO2 and PM2.5 measurement but does not publish the sensor type, so a buyer cannot confirm from the bullets alone whether the CO2 channel is NDIR or estimated.
A reading frozen at a plausible number, such as 800 ppm, is more common than a zero. The display looks healthy and the owner assumes the room is fine. The test is to breathe directly on the sensor for thirty seconds. A working NDIR unit will climb visibly within a minute. If nothing moves, the sensor or the display driver has failed.
Why the baseline matters more than the accuracy figure
NDIR sensors do not measure absolute CO2 concentration directly. They measure the absorption of infrared light at a wavelength CO2 absorbs, and they compare that against a reference. Consumer instruments establish that reference through a process called automatic baseline correction, or ABC. The instrument assumes that at some point in the recent past it saw fresh outdoor air at roughly 420 ppm, and it slowly shifts its zero point to match.
This is why a monitor that has lived in a closed bedroom for a month can read low. It has never seen the fresh-air reference, so its baseline has drifted upward or downward depending on the algorithm. The displayed number is internally consistent but not anchored to reality. The fix is to put the unit outside, or in a well-ventilated space with the windows open, for twenty to thirty minutes on a regular schedule. Some units do this automatically overnight; the listing for the portable 8-in-1 monitor states a 1600mAh battery and up to 72 hours of use, which makes it practical to carry the unit outdoors for a baseline reset without losing the session.
The accuracy figure in a listing, where one is published, describes the sensor’s performance against a reference gas at the moment of factory calibration. It does not describe what the instrument will read after six months on a wall. The listing for the 5-in-1 monitor with the Swiss-made sensor publishes a range of 400 to 5000 ppm and an accuracy of plus or minus 50 ppm. Those are useful numbers, but they are not a guarantee that the unit is currently anchored correctly. Baseline drift is a separate error source, and it is usually larger than the quoted accuracy.
Placement faults that look like instrument faults
CO2 is heavier than air and tends to stratify in still rooms, but the effect is small at the concentrations found indoors. Placement matters more for other reasons. A monitor mounted at head height near a window will read lower than one at floor level in the same room, because the window is a source of fresh air. A monitor placed directly above a radiator or in direct sun will read high on temperature and can drift on CO2 as the optical bench heats.
The most common placement error is putting the unit within a metre of where someone sits. Exhaled breath contains roughly 40,000 ppm of CO2, and a monitor close to a person’s face will read several hundred ppm above the room average. The number is not wrong; it is local. A wall-mounted unit such as the GOWENIC wall mount monitor avoids this because it sits away from the desk, but only if it is not mounted directly behind a seated person.
A second common error is placing the monitor in a room with a gas hob, a wood stove, or an unvented heater. Those appliances produce CO2 as a combustion product, and the monitor will read high. That reading is correct and is a ventilation warning, not a fault. It is also not a carbon monoxide alarm. A CO2 monitor does not detect CO, and a room with a combustion appliance needs a separate CO alarm regardless of what the CO2 number says.
When the alarm never sounds, and what that tells you
Many monitors publish an alarm threshold, often around 1000 or 1500 ppm, and owners report that the alarm never triggers even in stuffy rooms. There are three explanations. The first is that the room genuinely stays below the threshold, which is common in small rooms with any ventilation at all. The second is that the alarm is disabled in a settings menu, which is common on units sold for bedroom use where the owner does not want a night-time beep. The third is that the alarm depends on the same baseline that has drifted, so the unit believes the room is at 700 ppm when it is at 1200 ppm.
The listing for the wall mount unit states audible and visual alarms when levels exceed safe thresholds, but does not publish the threshold values. That is a real limitation for a buyer who wants to set the alarm to a specific figure. The listing for the 16-in-1 monitor describes a seven-level colour AQI indicator rather than a CO2 threshold alarm, so a buyer should not assume it will beep at a chosen CO2 concentration.
If you need an alarm you can trust, test it deliberately. Hold the unit in a small closed room with the door shut and breathe normally for ten minutes. A working alarm should trigger if the threshold is below the resulting concentration. If it does not, check the menu before concluding the unit is faulty.
Where a CO2 monitor stops being the right instrument
A CO2 monitor is a ventilation indicator. It tells you how much of the air in a room has been breathed by the people in it, and therefore how well the room is being flushed. It is not a general air quality instrument. It does not measure carbon monoxide, radon, mould spores, or fine particulate. A room can have excellent CO2 readings and dangerous PM2.5 from a candle or a nearby road.
If your concern is smoke, traffic pollution, or wildfire, a CO2 monitor will not help and a particulate monitor is the right tool. The guide to choosing a PM2.5 monitor covers that decision. If your concern is a gas appliance, a CO alarm is the right tool and a CO2 monitor is not a substitute. If your concern is humidity and condensation, a temperature and humidity logger is the right tool; the Qingping temperature and humidity monitor review covers one option.
If your concern is whether the CO2 number itself is trustworthy, the answer usually lies in calibration history rather than in the sensor specification. The guide to calibrating a CO2 monitor explains the fresh-air and reference-gas methods, and the explainer on CO2 monitor accuracy sets out what the published figures do and do not cover.
The order to check things in
Work through the cheap explanations first. Power cycle the unit and watch whether the display changes at all. Breathe on the sensor and see whether the number climbs. Take it outdoors for twenty minutes and see whether it falls toward the outdoor background. Check the menu for a disabled alarm. Only after all four tests have failed should you conclude that the sensor has died. A monitor that passes the outdoor test is working, and the problem is in how it has been placed, baselined, or read.
Frequently Asked Questions
My CO2 monitor reads zero even in a crowded room. Is it broken?
Probably, but not certainly. A true zero is physically implausible indoors, so the sensor is either failed, in a warm-up state, or not actually an NDIR CO2 sensor. Breathe on it and watch for a climb. If nothing moves within a minute, treat the CO2 channel as failed and use the rest of the unit only for whatever else it measures.
Why does my monitor read much lower than a friend's in the same room?
Two monitors in the same room can differ by several hundred ppm if one has a stale baseline and the other does not. The difference is usually baseline drift, not sensor accuracy. Take both outdoors for twenty minutes and compare again. If they converge near the outdoor background, both are working and the indoor disagreement was a calibration artefact.
Do I need to calibrate a consumer CO2 monitor?
Most consumer units calibrate themselves through automatic baseline correction, which requires periodic exposure to fresh air. If the unit lives in a closed room, that exposure never happens and the baseline drifts. The practical fix is a regular outdoor or open-window session, not a calibration gas kit. Reference gas calibration is for instruments where the published accuracy has to be maintained against a traceable standard.
The listing says the sensor is high precision. Why does the number still drift?
Precision and long-term stability are different properties. A sensor can be precise at the moment of calibration and still drift over months as the infrared source ages and the baseline algorithm accumulates error. The listing for the 5-in-1 unit publishes plus or minus 50 ppm, which describes the sensor against a reference gas, not against the room six months later. Drift is a separate error source and is usually larger.
Will a CO2 monitor warn me about a gas leak or carbon monoxide?
No. A CO2 monitor measures carbon dioxide only. It does not detect carbon monoxide, natural gas, or propane. A room with a combustion appliance needs a separate CO alarm, and a CO2 reading that rises when the appliance runs is a ventilation warning, not a leak detection.
The alarm never sounds even when the room is stuffy. What is wrong?
Check three things in order. Confirm the alarm is enabled in the menu. Confirm the threshold is set where you expect it; many listings do not publish the threshold, so you may have to test it. Then confirm the baseline is sound by taking the unit outdoors. If the alarm still does not fire in a deliberately stuffy closed room, the alarm circuit or the firmware has failed even if the display still works.