This is only a preview of the November 2013 issue of Silicon Chip. You can view 23 of the 104 pages in the full issue, including the advertisments. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Items relevant to "Build A GPS Tracker & Record Your Journeys":
Items relevant to "Dual-Channel Audio Delay For PA Systems":
Items relevant to "SiDRADIO: Integrated SDR With DVB-T Dongle, Pt.2":
Purchase a printed copy of this issue for $10.00. |
Highly accurate, low cost alcohol analysis . . .
ALs
AL
sCOLYSER
You’ve probably seen a few breath alcohol monitors
in the past but this one is quite different – this
simple system measures the amount of alcohol in
the drink before it ever reaches your mouth (or
stomach (or kidneys, or whatever!).
It determines the percentage of alcohol
in any drink with two very important
parameters: high accuracy and low cost.
Compare it with commercial units
costing tens of thousands of dollars!
by Allan Linton-Smith
The components
of Al’s Alcolyser: a
Digitech Fuel Cell
Alcohol Tester with supplied
mouthpieces (Jaycar QM7302), a
heated Travel Mug (Jaycar GH1301),
an accurate thermometer (or a thermocouple
and DMM), a caulking gun nozzle and some lengths
of plastic tube. The 12V power supply at left is for the Travel
Mug, while the Digitech electronic scale at rear (Jaycar QM7264)
is for accurately weighing the alcoholic drink sample and water.
72 Silicon Chip
ABOUT THAT NAME:
Well, it is an alcohol analyser
designed by Al . . .
so what else would you call it?
siliconchip.com.au
N
ot only is it easy to build, it
works with all drinks (many
analysers only work with specific drinks) and it also works over a
very wide range – from no alcohol at
all (0%) right through to nearly pure
alcohol (95%).
And it can even be used to calibrate
one of those cheapie breath alcohol
analysers as well, so you can be pretty
much assured that when that is telling
you you’re still under the limit, you
ARE under the limit.
One word of warning: if you do happen to earn the attention of the boys in
blue, don’t try to use this as evidence
in a court, because it won’t impress the
magistrate one little bit. As they say
in all good disclaimers, it’s for educational and/or entertainment purposes
only and no responsibility for its use
or consequences of its use will be accepted by the management . . .
Having said that, if you’re accurate
with your samples and temperature,
the accuracy of this simple system will
be right up there with the big boys!
Introduction
This unit uses fuel cell technology
(see panel) to evaluate the alcohol
content of drinks. Not only is it accurate but it is $0000’s cheaper than
most alternatives.
You might think there’s a zero or two
too many there but that’s the amount of
money you’d need to spend on a commercial equivalent (and then spend a
fortune to keep it calibrated).
Not only will it assist makers and
consumers of wines, spirits, liqueurs,
mixers, RTDs (ready-to-drink beverages) and cocktails, it may be used by
those seeking to identify alcohol in
beverages when it is not wanted!
For example, we tested a “lemon,
lime & bitters” (a commonly-ordered
“non alcoholic” drink), bought over a
bar, and this indicated that it contained
over 1% alcohol (from the Angostura
bitters, which contain 44.7% alcohol!).
Our system is quick, easy and
requires minimal training. Part of it
uses a commonly available fuel-cell
meter and a temperature-controlled
travel mug (which heats the sample
to the required temperature) while the
rest is easy to build, requiring only a
drill and a few pieces from your local
hardware store!
How do you measure alcohol?
If you’ve ever dabbled with maksiliconchip.com.au
BREATHE
INTO HERE
(ALCOHOL FREE BREATH!)
MOUTHPIECE
(ALL DIMENSIONS
IN MILLIMETRES)
THERMOMETER
(OR THERMOCOUPLE
AND DMM)
EXIT TUBE
ASSEMBLY
GOES INTO
HOLE AFTER
THERMOMETER
REMOVED
10mm PVC OR
NEOPRENE
FLEXIBLE TUBING
5mm DIAM
x 150 LONG
HEATSHRINK
TUBING
165
105
BREATH TUBES
(SUPPLIED WITH
BREATH ANALYSER)
155
LIQUID
UNDER
TEST
THIS END INSERTS INTO
JAYCAR BREATH ANALYSER
5.0
THERMOSTATICALLY
CONTROLLED TRAVEL MUG
ing home brew, we can hear you saying “that’s easy, just use an alcohol
hydrometer.” But it’s not quite that
simple.
Those instruments are perfectly
good for beer and even wine and
pure spirits. But what if you have to
measure drinks which are a mixture
of alcohol & sugars or juices – such
as “mixers” (eg, a bourbon and cola),
RTDs, ports, liqueurs and cocktails?
Or what if you want to check the
teenagers’ (theoretically non-alcoholic!) fruit punch that might have
some – ahem – surreptitiously added
“extras”? A hydrometer simply won’t
work at all!!
The problem is that sugars add to the
density of a drink. A hydrometer used
for beer and wine works on densities
lower than water but sugars in drinks
are greater than water.
Once you have any significant
amount of sugar in an alcoholic
beverage, your hydrometer becomes
basically useless.
The measurement of alcohol content
is often a problem for makers of fortified wines such as port, sherry, RTDs
and liqueurs or any alcohol which
contains sugars.
Microdistillation
One of the methods used for determining alcohol content relies on micro
This graphical representation of the
photo at left shows how the pieces fit
together to analyse alcohol content.
distillation, where a predetermined
test sample is put into a still and the
condensed alcohol is weighed. This
has to be done by trained personnel
and is expensive!
One of our contacts in the liquor
industry recently paid $14,000 for an
automated instrument of this type!
Other techniques such as infrared
spectroscopy are also very expensive,
typically $25,000 and they have a limited range – you need at least two of
these instruments to cover 0-20% and
20-60% alcohol, whereas ours covers
0-95% alcohol.
Those instruments are extremely
accurate, down to 0.02% and if calibrated correctly will stand up in court.
They are highly recommended if you
have $50,000 to spare!
But if you want something 300 times
cheaper you may want to consider our
setup which is typically accurate to
±0.5% – maybe not accurate enough
for a court of law but more than accurate enough for most users.
How it happened
Necessity being the mother of invention, we wanted to help a friend who
makes wines and other beverages so he
could quickly check the alcohol in his
liqueur, which contained a lot of sugar.
He wanted to know how many
“standard drinks” was in a bottle. (We
November 2013 73
Here’s a close-up
of the heart of
the system, the
Digitech (Jaycar)
QM7302 Breath
Analyser. The
opaque circular
piece at the top
is actually the
breath tube into
which you blow
to get a reading
of blood alcohol
content (%BAC).
We’re using it
in a different
manner than the
manufacturer
intended!
make no comment about his readiness
to allow us to sample his wares. . .)
Accurate breath analysers, incorporating fuel cells are readily and
fairly cheaply available. We thought
we might try the one offered by Jaycar
(QM7302 <at> $119.00) to see if we could
determine the alcohol content of various beverages.
We reasoned that if you bubbled
some (sober) breath through a warm
test sample (body temperature) and
measured the outcome, you would be
able to calculate the alcohol content.
Providing you knew the concentration of that sample you would be able
to convert the reading to the percentage of alcohol by volume (ABV).
It worked well – in fact, much better
than we expected… so here it is!
How it works
This project uses the fuel cell alcohol tester combined with the Jaycar
GH-1301 thermostatically controlled
travel mug (to accurately set the temperature), connected by a short tube.
The operation is very simple: a 1-4g
sample of an unknown alcoholic beverage is made up to 200ml with water
and is placed in the travel mug, which
is set to reach a temperature of 39-40°C
Once the sample solution reaches
40°C you blow into the mouthpiece,
bubbling air through the liquid and
then read the blood alcohol content
(%BAC) from the fuel cell breathalyser.
By referring to a conversion chart
you will be able to convert a %BAC
74 Silicon Chip
reading to %ABV (alcohol by volume)
for your test sample.
Naturally the dilution of the test
sample has to be in the range of the
breath tester which is specified at
0.00-0.4%.
So you will have to make a weaker
solution for strong alcoholic drinks
(1g/200ml) and stronger solutions for
weak drinks (4g/200ml) and then just
read off the chart for that particular
concentration.
We took quite a few measurements
using the Jaycar unit with various
alcoholic solutions to establish the
conversion chart and were happy that
it was accurate and more importantly,
the results were repeatable.
Some losses will occur mainly in
the exit tube where there may be some
condensation. Most of it will be water
but a tiny amount of alcohol is lost
(distillers call this the “angels’ share”).
We could heat the tube to prevent
condensation but we feel this is not
necessary and will complicate the
project.
Those of you who wish to use the
device in the tropics need not worry
too much but may find problems if you
live in Antarctica or high on Everest’s
slopes!
Keeping the exit tube short helps
considerably and we have set up the
chart using a standardised length of
tube because a longer tube will incur
higher losses.
It is easier to take measurements
with a longer exit tube but the results
will be unreliable.
Parts List –
Alscolyser
1 fuel cell alcohol breath tester
(eg, Jaycar QM7302)
1 heated travel mug
(eg, Jaycar GH-1301)
1 105mm length of 10mm PVC or
Neoprene tubing
1 unused caulking gun nozzle
1 150mm x 5mm diameter heatshrink
tubing
1 laboratory (mercury) thermometer
(or thermocouple and DVM)
1 accurate digital scale
(eg, Jaycar QM-7264)
1 12V 3A power supply (with lead and
RCA plug to connect to travel mug)
or use a 12V car battery with the
lighter plug lead which is included
with the travel mug.
Other applications
You can also use the setup to check
and/or calibrate other breath testers.
Many of these are not accurate or they
have changed with age, abuse (or simply don’t work at all!).
If you are using a personal breath
tester, we urge you not to rely on it
before you drive, unless you have had
it checked. Better still just don’t drink
and drive at all!
Even wall-mounted units in pubs
and clubs can be way off unless they
are frequently calibrated according to
the manufacturers’ instructions. (Once
again, that’s why they have those disclaimers!).
We checked a $10 unit and at 0.05%
it gave a reading of 0.04%, and at
0.06%, a 0.0% result – so don’t trust
them!
Calibration for these devices is usually done by putting a standardised
alcohol solution, usually 0.38% ABV
(alcohol by volume), in a calibrating
vessel at 38-40°C.
Then the operator blows into it bubbling air into this standardised mixture
which feeds to the breathalyser via a
short tube.
You can check the 0.05% level by
weighing 2.5g scotch plus 197.5g
water.
We have to hope that the operator is
not too drunk otherwise the machine
will be set incorrectly and will always
read too low! In fact, this applies just
as much to our system – any alcohol
in the operator’s breath will obviously
skew the results.
So beware – there is no guarantee
that any breath analyser, large or small
is correct. Unless, of course, it is one
in one of those vans with flashing red
and blue lights – these are regularly
calibrated and are declared (by law) to
be scientific instruments which means
you can’t argue with their results.
To calibrate another breathalyser
you can compare a reading from the
Jaycar unit to the breathalyser under
test.
You should try different concentrations of alcohol because one reading
is really not enough to verify that it
is working.
Some breath analysers do not have
provision to fit a tube and you only
need blow over the top of it for a few
seconds. To check these you will only
need to aim the exit tube to the receptor and keep it very close.
siliconchip.com.au
This method is also used by various
Police departments to check their hand
held breathalysers and they use rather
expensive devices to do this!
As an aside, this device would have
been very handy a couple of decades
ago when a certain teenager (who must
remain nameless, Craig) developed a
taste for his father’s Jim Beam. Said
father never did work out why said
JB tasted so “weak” until many years
later, son admitted to father that he
used to refill the bottle behind the
bar with water to mask his nefarious
activities.
So if you think your drinks are
being “watered down” at your pub
or club, here’s the way to accurately
check them!
Construction
Cut a piece of tubing to the specified length of 105mm and fit a plastic
mouthpiece (supplied with the Fuel
Cell Alcohol Tester) to each end. Two
mouthpieces are provided with each
breath tester and you can purchase
extra ones separately if needed.
Drill two 7mm holes in the lid of
the mug as per the diagram and seal
the slider with some insulation tape.
One of these holes does double duty
– it’s used for both a thermometer (or
thermocouple) and (when at the right
temperature) for the outlet tube.
Many thermometers and thermocouples will also fit through the outlet
tube anyway, so if yours does the outlet
tube can be left in situ.
Then fit 165mm of 5mm diameter
heatshrink tubing over the end of the
caulking gun nozzle and push it into
one of the holes.
Push your thermometer (or thermocouple) into the other hole.
You are now ready to calibrate and
test!
Operating and calibrating
We have made things pretty easy by
doing all the calculations and conversions for you but we do recommend
a check to ensure the breath tester is
working correctly.
Firstly warm it up by blowing
directly into it to check if the tester
reads zero.
Obviously, you need to be absolutely
stone cold sober for this, otherwise you
will have to wait until you zero it. And
if you remember that your body only
“loses” 0.015% of blood alcohol per
hour, that could be a rather long wait!
siliconchip.com.au
Breath Analysis: the players
There are three major types of breath alcohol testing devices, each based on different
principles. Regardless of the type, each device has a mouthpiece, a tube through which
the person blows air and a sample chamber where the air goes to be analysed.
Older style “Breathalyzer”# – Uses a chemical reaction involving alcohol that produces a colour change
Intoxilyzer – Detects alcohol by infrared (IR) spectroscopy
Alcohol Sensor – Detects a chemical reaction of alcohol in a fuel cell
Breathalyzer
The Breathalyzer device contains:
• A system to sample breath
• Two glass vials containing the chemical reaction mixture
• A system of photocells connected to a meter to measure the color change associated with the chemical reaction
The breath sample is bubbled in one vial through a mixture of sulphuric acid, potassium dichromate, silver nitrate and water. Sulphuric acid removes the alcohol from the
air into a liquid solution and the alcohol reacts with potassium dichromate to produce
chromium sulphate, potassium sulfate, acetic acid and water.The silver nitrate is a catalyst
which makes accelerates the reaction without participating in it. During this reaction, the
reddish-orange dichromate ion changes color to the green chromium ion when it reacts
with the alcohol; the degree of the color change is directly related to the level of alcohol
in the expelled air. To determine the amount of alcohol in that air, the reacted mixture
is compared to a vial of unreacted mixture in the photocell system, which produces a
voltage proportional to the colour change which is calibrated for alcohol concentration.
Intoxilyzer
This device uses infrared (IR) spectroscopy and identifies molecules based on the
way they absorb infrared light.
The various bonds within molecule absorb IR at different wavelengths. In ethanol
the bonds (C-O, O-H, C-H, C-C) absorb the IR light at different wavelengths and these
wavelengths help to identify the substance as ethanol and the amount of IR absorption
tells you how much ethanol is present.
A lamp generates a broadband (multiple-wavelength) IR beam, which passes through
the sample chamber and is focused by a lens onto a spinning filter wheel which contains
narrow band filters specific for the wavelengths of the bonds in ethanol. The light passing through each filter is detected by the photocell, where it is converted to an electrical
pulse and then is relayed to a microprocessor which calculates the BAC based on the
absorption of this filtered infrared light.
In general, the larger table-top units found in Police stations and in some “booze bus”
installations are IR spectroscopy types.
FUEL cell type
Modern fuel-cell technology (the same as which may power our cars and even our
houses some day) has been applied to breath-alcohol detectors. Devices like the one in
our project use this technology.
The fuel cell has two platinum electrodes with a porous electrolyte material sandwiched
between them. As the exhaled air from the breath flows past one side of the fuel cell,
the platinum catalyses the oxidation of any alcohol in the air to produce acetic acid and
in the process moves electrons across the electrolyte.
A voltage is generated between the two electrodes proportional to the amount of
alcohol and this is measured and converted to BAC by a microprocessor.
The protons move through the lower portion of the fuel cell and combine with oxygen
and the electrons on the other side to form water.
# “Breathalyzer” is the registered brand name and trademark of a particular manufacturer (specifically Dragerwerk AG, a German conglomerate). However, like many words in
our language, it has become synonymous with a wide range of breath analysis devices.
We are using the name in this context.
November 2013 75
0.25
CONVERSION CHART FOR JAYCAR QM-7302
FUEL CELL BREATH TESTER
0.2
EXAMPLE:
4g/200ml sweet wine – 0.11% BAC
measures 7.4%
0.15
BREATH
TESTER
READING
%BAC
at 40°C
0.1
0.05
% ALCOHOL BY VOLUME
0
0
0
0
2.5
5
10
5.0
10
20
Then connect the breath tester to the
exit pipe making sure that all connections are secure.
If you have a balance or electronic
scale which is accurate to 0.1g, weigh
out exactly 1.0 gram of scotch whiskey
(blended scotch is always 40% ABV;
you may also use vodka or brandy as
long as the strength is 40%) and make
this up to 200g with water. Ideally, that
would be pure or distilled water but
in most circumstances, ordinary tap
water will be satisfactory.
Note that “normal” kitchen scales
will not be good enough unless they
are accurate to one gram. To maintain
sample accuracy with one gram scales
you should multiply everything by ten
(eg, 10g/2litres).
The Jaycar QM7264 digital scales
we used have a resolution of 0.01g.
Place the alcohol/water mixture
in the travel mug and hook it up to a
power source (12V/3A) or to the 12V
cigarette lighter socket in your car and
set the temperature to 40°C (104°F).
When the sample has reached 40°C
on your separate thermometer you can
pull the plug out because the mug may
exceed the dialled in temperature.
Conversely, if it doesn’t reach 40°C
you may need to set it to a higher
temperature.
You may wish to leave the mug in
its default mode which shows °F, in
which case you can do all your measurements at 100°F
Every time you remove power to the
mug, all settings will be lost.
(Every mug will be different so you
may need to set it a bit higher if it
76 Silicon Chip
7.5
15
30
10
20
40
12.5
25
50
15
30
60
won’t go to 40°C.)
The mug we used is fairly well
insulated so it will maintain the temperature for a few minutes, enough to
do two or three repeats. For accuracy,
you must check the temperature of the
sample between tests with a separate
thermometer such as a digital thermometer or glass thermometer because
the mug is only designed to heat coffee
– it is not super accurate.
Bear in mind that if it is too hot or too
cold your readings will be incorrect.
Follow the instructions provided
with the QM7302 alcohol tester and
warm it up as recommended.
When everything is ready and the
tester beeps for you to “blow” hold
down the button and blow gently and
Do’s and don’ts
• DO read all the instructions which
accompany the tester and the mug.
• DON’T switch on the mug without
liquid
• DO remember temperature is important, so keep the sample as close to
40°C as is possible
• DO give the breath tester time to
recover after four samplings
• DO dry the mug and tubes before
adding the sample
• DO accurately weigh the sample and
water – weigh 1g (or 2g or 4g) then
add water to 200g – ie, 1g sample
+199g water
• DO measure the sample immediately –
if it sits too long alcohol will evaporate.
17.5
35
70
20
40
80
22.5
45
90
SAMPLE
4.0g
2.0g
1.0G
steadily into the mug for six seconds
or until you hear a second beep. Don’t
blow too hard or you may risk blowing
big droplets into the tester.
If all is well, you should get a reading close to 0.20%.
Jaycar’s specifications indicate that
the accuracy of this tester is ±0.008%.
Our measurements of alcohol were
typically accurate to ±0.5%, which is
almost good enough for establishing a
wine alcohol label!
The conversion chart shows how to
calculate the %ABV of a beverage from
a %BAC reading on the Jaycar unit for
1, 2 or 4 gram samples in 200g water.
Note that we have specified grams
of water here instead of the usual “ml”
because you are weighing the sample
on the scale, not measuring it by volume. Of course, 200ml of pure water
will weigh 200g.
Examples
Example 1: a 4.0 gram test sample of
sweet white wine, labelled 8.0%,
gave an average reading of 0.11% so
tracing down we will get a measurement of 7.8% ABV. (yellow line)
Example 2: a 1.0 gram sample of an
unknown spirit gave a reading of
0.215% and again tracing down we
get 70% ABV – quite a strong spirit!!
If you have no idea what the approximate alcohol content is then do
a rough reading using 1g and then
increase it to 2g or 4g to get higher
readings (and hence more accuracy)
on the breath tester.
You may need to do a few practice
runs to be sure, to be sure!
SC
siliconchip.com.au
|