How do breathalysers work?
You can't measure blood without a needle, so how does blowing into a small box tell police your blood alcohol level? The trick is your lungs. Here's the science behind breath testing — and why nothing but time gets you under the limit.
The short version
A breathalyser doesn't measure your breath for the sake of it. It measures your breath to work out how much alcohol is in your blood. When you drink, alcohol is absorbed into your bloodstream, and as blood flows through your lungs, a tiny, predictable fraction of that alcohol evaporates into the air in your lungs. Breathe out, and the device reads the alcohol in that air and converts it to an estimated blood alcohol concentration (BAC).
Step one: alcohol moves from blood into breath
Deep in your lungs are millions of tiny air sacs called alveoli, where oxygen and carbon dioxide swap between air and blood. Alcohol is small and water-soluble, so it makes the same journey: alcohol in the blood passing through the alveoli evaporates into the air there until the two reach a stable balance, or equilibrium.
This is why a breathalyser wants the last portion of a long, steady exhale — the "deep-lung" or alveolar air. The first air out of your mouth has been sitting in your windpipe and mouth and doesn't reflect your blood as reliably. Modern instruments are built to sample only that deep-lung air, which is part of what makes them accurate.
Step two: the 2100:1 conversion
Once the device knows how much alcohol is in your deep-lung air, it converts that to an estimated blood figure using a fixed ratio. The internationally used breath-to-blood ratio is about 2100:1 — meaning 2,100 millilitres of deep-lung air is assumed to contain the same amount of alcohol as 1 millilitre of blood.
That ratio is an average. Real people vary a little around it depending on body temperature, breathing pattern and other factors, which is one reason a roadside screening reading is treated as an indication rather than the final word. Evidential instruments are calibrated conservatively and, in Australia, you can request a blood test as a check.
What's inside the box: fuel cells vs sensors
Not all breathalysers use the same technology, and the difference matters for accuracy.
Fuel cell devices (police and quality personal units)
The evidential instruments police use — and the better personal breathalysers — rely on an electrochemical fuel cell. Your breath passes over two platinum electrodes separated by a porous, acidic layer. The platinum oxidises any alcohol present, producing acetic acid, protons and electrons. Those electrons create a small electric current, and the size of that current is proportional to how much alcohol was in the sample.
Fuel cells are prized because they react specifically to alcohol, stay accurate even at low concentrations, and respond in a straight line across a wide range of readings. They also won't be fooled into a false positive by the acetone on the breath of someone with diabetes or on a low-calorie diet.
Semiconductor sensors (cheap consumer gadgets)
Many low-cost keyring and novelty breathalysers use a semiconductor (metal-oxide) sensor instead. These are cheaper and fine for a rough "have I had too much?" nudge, but they're less specific — they can react to other substances on your breath and drift out of calibration faster. They're not the standard for anything with legal or safety consequences.
| Feature | Fuel cell | Semiconductor |
|---|---|---|
| Used by | Police, quality personal units | Cheap consumer gadgets |
| Specific to alcohol | High | Lower — reacts to other vapours |
| Accuracy at low BAC | Good | Variable |
| Calibration stability | Holds well | Drifts sooner |
How random breath testing works in Australia
Random breath testing (RBT) has been part of Australian roads for decades and is credited with cutting drink-driving deaths sharply. Under RBT, police don't need a reason to test you — as Transport for NSW puts it, every police car is a mobile RBT, and refusing a test is itself an offence.
There are two stages:
- The roadside screening test. You blow into a hand-held device that quickly indicates whether you're likely over the limit. It's fast, but it's a screen, not the final reading.
- The evidential breath analysis. If the roadside test is positive, police take you to a station or a "booze bus" for a second test on a larger, calibrated instrument. This one prints a precise result and is what's used in court.
Australia's limits are low: under 0.05 for most full-licence drivers, and zero for learner and provisional drivers. Our guide to Australia's drink-driving limits breaks down which limit applies to whom.
Mouth alcohol, and why there's a wait
Breathalysers are designed to read alcohol coming up from your lungs — not alcohol lingering in your mouth. Alcohol-based mouthwash, a breath spray, a very recent sip, a burp or acid reflux can all leave traces of "mouth alcohol" that briefly inflate a roadside reading.
This trace clears in roughly 15 minutes, which is exactly why evidential testing includes an observation period before the reading that counts. It's not a loophole — it's built into the process so the number reflects your blood, not your last mouthful.
Personal breathalysers and apps: useful, not a green light
A good personal fuel-cell breathalyser can give a helpful ballpark, and a BAC tracking app like Drink Buddy can estimate where you're at from what you've logged. Both are great for awareness — for realising "I've had more than I thought" and deciding to hand over the keys.
But neither should ever be used to decide you're okay to drive. Personal devices can drift out of calibration, your BAC can still be rising for up to two hours after your last drink, and impairment begins well below 0.05. If you've been drinking and you're near a limit, the safe reading to act on is simply: don't drive.
Frequently asked questions
How does a breathalyser measure blood alcohol from breath?
Alcohol in your blood evaporates into the deep-lung air until it balances with your blood. The device measures the alcohol in that air and multiplies by a fixed ratio of about 2100:1 to estimate your blood alcohol concentration.
Can mouthwash or a burp set off a breathalyser?
Briefly, yes. Alcohol-based mouthwash, sprays, a recent drink, a burp or reflux can inflate a roadside reading. This mouth alcohol clears in about 15 minutes, which is why evidential testing includes an observation period first.
Can you trick or beat a breathalyser?
No. Mints, coffee, holding your breath or eating won't lower a properly conducted reading. Your liver clears alcohol at a fixed rate — roughly one standard drink per hour — so only time lowers your BAC.
Are personal breathalysers accurate?
Quality fuel-cell units can be reasonably accurate; cheap semiconductor gadgets are rougher and drift out of calibration. Either way, use them for awareness, never to decide whether you can drive.