Why breathalyzers show different readings even minutes apart

A breath reading is a picture of one moment, and the subject of that picture keeps changing. In the period after drinking, alcohol is still passing from the stomach and small intestine into the blood, so the level climbs. Later, the liver removes it at a fairly steady pace and the level falls. Two tests taken twenty minutes apart are therefore measuring two genuinely different states.

The direction of change surprises people. A second reading taken half an hour after the first is often higher, not lower, if the drinking ended recently. A meal slows absorption and stretches the rising phase out, while drinking quickly on an empty stomach makes the peak arrive sooner and sit higher. Neither pattern means the device is wrong.

The breath to blood relationship also varies between people and within the same person over the evening. Body temperature, breathing rate and the stage of absorption all nudge the ratio slightly, which is one more reason two people who drank the same amount rarely see the same number. For the underlying science, our guide to blood alcohol concentration explains what the figure represents.

This is an estimate. It cannot tell you whether you are safe or legal to drive. If you have been drinking, do not drive.

A handheld tester wants air from deep in the lungs, the part of the breath that has been in contact with blood. The first air you push out comes from your mouth, throat and windpipe, where little exchange happens. If you stop early, the sample is weighted toward that shallow air and the result reads low.

Holding your breath before blowing has the opposite effect. Air that sits in the lungs longer picks up more alcohol, so a held breath can read higher than a normal one. Breathing fast and hard just before a test tends to push the result down a little. Each of these habits changes the sample, which is why a consistent routine matters more than any single trick.

• Take a normal breath, not a huge gulp and not a held breath.
• Blow steadily for the full time the device asks for, without a burst at the start and a fade at the end.
• Keep a firm seal on the mouthpiece so outside air does not dilute the sample.
• Repeat the same technique every time you test, so readings can be compared fairly.

Alcohol that has not yet cleared from the mouth adds to the breath sample and inflates the result. A drink finished a few minutes ago, a rinse with an alcohol based mouthwash, or a burp that brings vapor up from the stomach can all push a reading well above what the blood actually holds. That inflated value then fades quickly, so a second test a little later appears to drop sharply.

This is the most common reason people see a high first reading followed by a much lower one. Waiting before you blow removes most of the problem, and our post on how long to wait before using a breathalyzer covers that routine. If you use a mouth rinse often, the article on whether mouthwash can affect a breathalyzer is worth a look.

Two devices can disagree simply because they are built differently. A semiconductor sensor reads a change in the resistance of a heated metal oxide, and that surface responds to more than just ethanol. A fuel cell sensor converts alcohol into a small electric current and is much more selective. When you compare a budget tester with a fuel cell unit, some of the gap is the sensor itself. Our fuel cell vs semiconductor breathalyzer page lays out those differences side by side.

Sensors also need a moment to recover. After a sample, a fuel cell has to consume the alcohol it captured, and a semiconductor surface has to return to its baseline. Blowing again straight away can give a reading that is too high or too low because the sensor is still working through the previous sample. Many manuals suggest waiting a few minutes between tests, and longer after a high reading.

Every breath sensor slowly drifts. Its response changes with use, age and storage, so a device that matched a reference when it left the factory may read differently a year later. Two testers bought at different times, or used at different rates, will often disagree because they have drifted by different amounts and in different directions.

Makers handle this with recalibration, sensor replacement or, on some low cost models, no correction at all. If your device is overdue, any gap between it and another tester says more about calibration than about your blood. Our guide to calibrating a breathalyzer explains what makers typically ask for, and the broader question of how accurate are breathalyzers is covered in its own guide.

Breath testers are designed to work within a stated temperature range, usually around room conditions. A device carried in a cold car or left in a hot glove box may read differently until it settles. Very humid air, a nearly empty battery and a dirty or wet mouthpiece can also affect the sample or the electronics.

Strong vapors in the room can matter for semiconductor sensors in particular. Paint, solvents, hand sanitizer and cleaning products release compounds that a less selective sensor may pick up. Testing in a well ventilated space, away from those sources, removes one more variable.

You cannot remove every source of variation, but you can stop adding new ones. A steady routine makes two readings comparable, which is usually what people actually want to know.

• Wait at least 15 minutes after drinking, eating, smoking or using mouthwash before the first blow.
• Let the device reach its working temperature and use it indoors at normal room temperature.
• Use the same breath technique and the same length of blow every time.
• Leave a few minutes between tests so the sensor can recover.
• Keep the device calibrated or replace the sensor on the schedule the maker gives.
• Compare readings from the same device, not from two different ones.

If two readings still differ after all that, treat the higher one as the more cautious guide and remember that neither number decides whether you can drive. When a device is due for replacement, a fuel cell unit holds calibration better and reacts less to outside substances, so it tends to produce more repeatable numbers.