are designed on the expectation that the person being
tested wants the most accurate result possible.
They are familiar with the device and know how to blow
(time & pressure) in order to get the most consistent
readings. The accuracy readings given by the
manufacturers (see below) are usually only
acheived after considerable practice.
assumption is that the person being tested may well try
to provide a false result, either by giving an
insufficient sample or failing to give honest answers.
The certification process assumes an unwilling and
unco-operative participant yet still ensures accurate
and repeatable results over a wide range of levels from
very low to high.
mostly rely on a timer, occasionally with a minimum blow
pressure. They measure alcohol from the entire
breath sample - the majority of which has not been
anywhere near the blood vessels in the lung (alviolae
air). Only this "deep lung" air can
give accurate & consistent results. (see "how
a breathalyzer works")
have sophisticated sensors for volume, pressure,
temperature and humidity to ensure an exact sample is
measured. They sample a full lung of air - around
1.5ltrs - but only measure the last 20ml, which
is the deep-lung air that has been in direct contact
with the blood in the lungs.
They assume the user will look after the device and not
test themselves after smoking, drinking or eating.
The sensors are not able to cope with this type of abuse
and will quickly degrade or fail if misused in this way.
Lack of regular use will also "dry out" the
sensor leading to drift or failure.
The testee may well deliberately smoke or drink in an
effort to provide a false reading; they have no regard
at all for any damage it might do. The Fuel Cell
sensors are therefore designed to be robust enough to cope with
such abuse and not fail or provide false readings.
user knows if they've been using (say) mouthwash and
would not use the breathalyser accordingly. To
protect the user, they will provide a positive result
over a wider range of substances - just in case it
might be alcohol.
person doing the testing has no idea whether the person
tested has, or has not, been using mouthwash, gum etc. They have to be VERY specific to pass certification
and will NOT display a result for anything other than
Usually use a semi conductor or low-cost fuel cell that is
rarely more accurate than to within +/- 20% when
new. Semi conductors in particular will degrade over time, particularly if mis-used.
Generally cheap and mass-produced, they are rarely
Certification requires large, high
quality Fuel Cell Sensors that will tolerate
considerable misuse while maintaining a high degree of
accuracy over long periods of time. They are
expensive to produce, and impecable quality control is
an integral part of the certification process.
Often expressed in BAC%, such as "£ 0.02% " 0.08%"
This is in fact £ 25% of the actual reading and it
gets signicantly worse away from the calibration point
(usually 0.08% BAC) - as much as £ 100% at levels of
around 0.02% BAC. Assuming the user to be a
driver, they will therefore over-read to "err on the
side of caution" - hence false positives.
Standard Draeger accuracy is a maximum of
£ 1.75% of the reading value. Typically accurate
to within less than 0.001% BAC they will only give
exact, accurate results and will not provide false
positive readings. They will consistently
reproduce their readings during repeated testing no
matter what level of alchol is tested.
Typically the reading is displayed for a short period
and then clears, unable to be recalled.
Occasionally they have a basic memory function but the
lack of an internal time-stamp means the test cannot be
wide range or recall options are available including
hard copy printouts to mobile printers, dowload of data
to a PC, alphanumeric input of the name of the
person tested and even GPS time-place-date stamps to
Used 4 or 5 times a month and exactly as per the user
guide, they will typically last 2-3 years. Sensor drift
is the biggest issue but so long as the breathalyser is
being used by the same person, they will recognise the
pattern and know to have it checked or calibrated.
Designed to withstand mistreatment both in handling and
sampling from the rough enviroment of a police traffic
unit, they will last 5-6 years as a police device and in
a commercial use they should last in excess of 10 years
if calibrated on schedule.