Here's a link to
an old article by David Vizard on using Colortune.
Hmm, I don't like the font. So I'll just copy and paste it here:
Mixing it on your Own Dyno
by David Vizard
The average home tuner's most frequent problem, if the letter
we get are anything to go by, is getting the fuel/air ratios
(mixture) right after having tinkered wit hthe motor. After
printing this article, we do not expect to get, ever again, a
Technical Query asking what needles or jets to use on a
modified motor, becuase this really should be the end of your
mixture problems!
When tuning the engine, it is more than likely that the
fuel/air demand of the engine is changed. Becuase of this it
is neccessary to "calibrate" a carburetter so that it passes
fuel into the engine in the correct quantities for the amount
of air being consumed. Over its operating range, a
carburetter may have to pass a volume from say two to two
hundred cubic feet of air a minute! At tickover, the
consumption of air is very low, but at full throttle the
opposite applies.
At every point between the two extremes the air flow
requirement is different, yet at all these points the
carburetter is expected to mix the correct amount of fuel with
the incoming air. There is little point in trying to set up
the mixture at tickover becuase very few, if any, of us drive
on tickover and there is absolutely no guarantee that the
mixture is right throughout the range.
How do we know or how can we tell what the mixture is like
anyway? Up until now, the most common method has been to
"read the plugs". If you have had years of practice and are
an expert plug reader (not the print on it) you can just about
pull off the job of mixture setting. The trouble, however,
with a plug count is that one is never too sure what the plug
is saying.
For instance, a dark, very slightly sooty plug can look like a
mixture that is just a shade rich, but it could be that the
mixture is just right and the plug type is too hard (runs too
cold) for the motor. Unless you are an expert then, reading
the plug is, to say the least, just shade dodgy.
An alternative to reading a plug is to get your engine set up
on a dynomometer; either the rolling road type or the engine
type. When an engine is on a dyno, it can be run at various
rpm and against various loads, thus simulating the conditions
met on the road. While all this is going on, an electronic
gas sampler can be analysing the exhaust and indicating the
fuel/air ratio.
Going from reading plugs at a dollar a time to using three
grand worth of dyno might seem like going from one extreme to
the other or to use a comparison, going from an abacus to an
electronic computer. What is needed to use the analogy again
is a slide fule ie: a method between the abacus and computer,
or to come back to reality, a method of setting the mixture
which is more akin to the dyno and mixture analyser than the
plug reading method.
A dynomometer measures horse power, commonly called brake
horse power becuase a dyno is nothing more than a brake,
calibrated to read out the work absorbed and the rate of
absorbtion. Here we have a clue; all cars have brakes (or
should have) so what we have, in effect, is an uncalibrated
dyno fitted to our car. To simulate road conditions whilst
standing still, all we need do is to jack up the driving
sheels of the car, put it in gear and use the throttle as if
we were driving along the road.
To simulate road levels we need only apply varying pressure on
the brake pedal. Such action will, of course, get the brakes
hot. For our purposes we need to be able to hold full power
for about fifteen seconds maximum. If your brakes cannot cope
with this from the heat point of view, then you are sadly
lacking in that department so see to it. It's probably better
brakes you need and not more power!
Okay, so we have our dyno. The fact that it does not read out
in horsepower is, for our purposes, irrelevant. What we need
now is a mixture analyser. Up until a few years back this
would have cost a tidy sum. These days we have a device known
as a "Colortune" and within the price range that can be
afforded by the enthusiast, this is the _only_ device we know
of that will do the job in hand.
For those who man not know, a Colortune is a device which
replaces the sparkplug in the cylinder. The top of the
Colortune is made of a Borosilicate glass and this allows you
to see what is going on in the mixture combustion chamber.
Different mixture strengths burn at different colours.
By looking into the combustion chamber through the Colortune
we can get a good indication of the mixture strength
prevailing. Inspection of the flame color shows that four
fairly distinct stages occur. when the flame colour is
blue/white, the mixture ration is between 16 and 14:1; a blue
colour indicates a mixture strength between 14 and 12,5:1; a
blue/orange colour indicates 12,5 to 11:1 and orange indicates
11:1 or less.
The Colortune, then, is indicating at the colour transition
points the mixture strength of the ingoing charge. Maximum
power occurs when the fuel/air ratio is between about 12,5 and
13,5:1. The exact point varies from engine to engine, but
most cast iron tuned production engines seem to be best around
12,8:1. The best economy is achieved on weaker mixtures than
that giving maximum power, and fuel/air ratios between 14 and
16:1 seem to be the easiest on the pocket.
To set up the mixture in the manner about to be described you
will need an accomplice. The first and essential step is to
part the car in some place which is poorly lit, so that you
can see the combustion colours. Jack up the driving wheels of
the car until they are just clear of the ground and in the
interest of safety, securly blockthe car so that it cannot
move under any circumstances.
At this point, warm up the engine, the remove a spark plug and
replace it with a Colortune. Set up the mixture so that you
have a fuel/air ratio of about 12 to 12,5:1 (orange/blue) at
normal tickover revs. If you have multiple carbs you will
have to do this for each cylinder or set of cylinders having a
carb.
Next, get your accomplice to put the car into gear, usually
third gear is best, and increase the throttle opening but at
the same time put on the brake. (Continue opening the
throttle and increasing breaking pressure until your
accomplice ends up with the throttle wide open and the revs
pulled down by braking to 2000 rpm). You can now look ath the
Colortune and at this point it will reveal what the fuel/air
ratio is under the prevailing conditions.
It can then be noted, preferebly by colour rather than
reference to its fuel/air ratio. After this, let the brakes
cool for a few minutes, then repeat the procedure at 3000 rpm
then at 4000 rpm and finally at 5000 rpm, stopping to let the
brakes cool between each run.
As far as brake overheating is concerned, it should not take
more than fifteen seconds to ascertain the mixture ratio at
each rpm interval, so they will be well within their capacity.
Once you have an indication of the state of affairs of the
fuel/air ratio up the rev range to 5000 (5000 rpm is the limit
on the Colortune) the necessary corrections can be made to get
it right. Not only can the full throttle conditions be
catered for in this manner, but so can part throttle and
transient conditions. For instance, a hesitant pickup when
going from parth throttle to full throttle could indicate that
the mixture is too weak during the transition from one state
to the other. During the transition period and for a short
while after the Colortune should show a rich mixture
condition. If it doesn't, then you can bet your life that on
a fixed jet type carb, the accelerator jets or pump stroke are
inadequate.
On carbs like the SU and Stromberg CD variety, a lean mixture
during the acceleration phase would indicate that the damping
is insufficient. A thicker oil is usually required to
compensate this.
A couple of cars were used as guinea pigs to test the method.
Both cars were modified and therefore required different carb
settings. In each case the carburation has been originally
set up by the owners who had only an average working
knowledge of what was required. After use of the Colortune by
the mothod just described, both power and economy were better
between 5-8bhp and 10-15 miles per gallon.
The reason that consumption was so much better after use of
the Colortune stemmed from the fact that the mixture was
originally set rich for maximum power and unfortunately
becuase of the guesswork method of setting, it was too rich.
The performance increase was better than the power increase
alone suggesting that the mixture was right (within limits)
throughout the rev range. As a side effect this led to a
smoother running engine with a snappier throttle response.
To sum up, the Colortune proves to be a very useful device.
It can, at a price of 4-87,5 UKP easily justify its place in
the tool kit of any self-respecting enthusiast, and its
intelligent use can only bring about an increase in
performance.
Editor's note:
Whilst David Vizard has used this tuning method successfully
with both a Mini and Austin 1100 its use on many other types
of car could possibly be dangerous. Firstly therefore we
recommend that the suspension characteristics and drive
shaft-prop shaft geometry of the car in question be checked
carefully ebfore even considering use of this method.
On rear wheel drive IRS cars which may have large wheel
angularity, attempt to jack the car at points on the
suspension that will allow the wheels to assume a position in
angle similar to that which exists in normal use.
On a non-IRS rear wheel drive car jack under the springs on
either side at the axle location point.
Make sure jacks are very secure and will not move under
testing vibration, also make sure any car to be tested has
alternate props underneath in case of jack failure -- ie:
spare wheels and tyres which should be first tested using the
full weight of the car. One advantage of the DV method is of
course that during load tesing the wheels on the ground have
the brakes applied.
On a Mini the best method Vizard found was to put a piece of
wood on top of trolly jack lift point and jack up from the
Mini sump. Once having jacked car up, jam wheels and tyres
underneath car for safety. Car does rock but cannot thus fall
over. Drive shafts will assume peculiar angles during testing
and this can be minimised by supporting bottom suspension arms
on axle stands --these can however move and you may have to
take the risk of the odd shaft angles for the few moments of
testing.
This method does not apear to be practical with swing axle IRS
cars ie: Herald, Spitfire, Vitesse Mk 1, GT6 Mk1.