Modern police DUI checkpoints rely on a two-stage testing approach. Officers first use a passive or preliminary breath tester (PBT) to screen drivers quickly. Those who fail the initial screening are then subjected to an evidential breath test using a certified fuel cell analyzer.
The evidential device produces a printed or digitally logged result that is legally admissible in court. These devices undergo regular calibration using certified gas standards to ensure accuracy is maintained.
Data logging capabilities also allow police departments to maintain comprehensive records of all tests conducted, supporting prosecution and department compliance audits.
Why two stages rather than one
Screening and evidence are different jobs, and one device rarely does both well.
A screening device has to be fast, cheap enough to deploy in numbers, and quick to recover between drivers. It answers one question — is there alcohol here — and it is allowed to be occasionally over-cautious, because a positive screen leads to a better test rather than to a charge.
An evidential device has to produce a number that survives a courtroom. That means a fuel cell sensor, a documented calibration history, and a result recorded in a form nobody can alter afterwards. It is slower and more expensive, and it is used on far fewer people.
Running both is not duplication. It is what lets a checkpoint move at the pace of traffic while still producing evidence that holds.
What happens between the two tests
The gap matters more than it looks. Alcohol still in the mouth — from a recent drink, mouthwash, or some medicines — can raise a screening reading well above what is actually in the blood.
Standard practice is a short observation period, typically around fifteen minutes, during which the driver takes nothing by mouth. The confirmatory test then measures deep lung air, which reflects blood alcohol rather than the last mouthful.
Skipping that wait is the most common way a technically correct positive becomes a contestable one.
The record is the case
An evidential reading with nothing behind it is weak. The same reading with a printed slip, a timestamp, the device serial, and a calibration certificate covering that date is very hard to argue with.
A checkpoint that logs consistently ends up with something a department can also use internally: how many drivers were screened, how many escalated, how many confirmed, and whether a device is starting to behave differently from the others in the fleet.
Practical setup
Equip screening officers with rugged handheld units that recover quickly and are cheap enough to have several. Keep at least one evidential fuel cell device with a printer at the checkpoint, with its calibration certificate on hand.
Carry spare mouthpieces in sealed packs, keep devices out of direct sun and out of a hot vehicle, and check batteries before the shift rather than during it. A checkpoint stalls on small things far more often than on big ones.
Above all, apply the same sequence to every driver. A procedure applied consistently is what makes each individual result defensible.