This article was first published in 2008.

Unlike petrol engine cars that tend to rev to the
redline with little if any dropoff in power, diesels usually die early. That
doesn’t apply to all diesels  it obviously depends on where the manufacturer
wants to put the redline and rev cut – but a diesel’s forte is definitely in
making torque. To put that another way, diesels develop good power low in the
rev range.
So when you’re driving a diesel for maximum
performance, you have a much greater choice as to when to change gear. Do you
change early and let the torque do the work? Or do you change later and let the
topend power give you best acceleration?
In many diesels the answer’s not at all obvious,
especially when you consider that in turbo diesels, shortshifting better loads
up the turbo, so potentially giving more boost than is achieved in a lower gear.
Deciding when to shift gear is even harder when
you’re driving a modified diesel. Have a look at this dyno graph for our Peugeot
405 SRDT. The upper red line on the graph shows the standard power output. The
power flatlines from about 3500 rpm, holding at 56kW until 4500 rpm, where it
abruptly starts to fall. The sharp decline in power can be easily be felt from
behind the wheel, giving a very natural 4500 rpm shiftpoint. (Redline on the
tacho is 5000 rpm.)
The upper green line shows the power output gained
with exhaust, intake, rev limiter, boost and fuelling mods. As you can see, the
power again peaks at about 3500 rpm, holds until 4200 rpm and then starts to
drop fairly quickly. So does that mean that in the modified car the gear shifts
should actually occur earlier – at 4200 rpm rather than the standard car’s 4500
rpm?
But hold on! At 4500 rpm the modified car is
making more power than the standard car, so wouldn’t it be faster with a later
gearchange – say at 4750 rpm?
Finding the Answer
You could get all mathematical at this stage,
consulting gear ratios, final drive ratios and wheel diameters to calculate
tractive effort at the wheels. But then you need the torque curve – and that
might vary with boost pressure and so load...
But with a few simple measurements, it’s deadeasy
to find the answer on the road. In fact, to find the answers – you can
easily measure the best revs for each gear change.
The way you do it is to measure the actual onroad
acceleration in each gear. Rather than indirectly working out acceleration from
seeing how long it takes the car to accelerate to a certain speed, with this
technique you measure the actual acceleration that is occurring at full throttle
at every speed and in every gear. Well, you don’t do it for every speed – you
simply do it in increments of (say) 10 km/h.
But isn’t an accelerometer an expensive
instrument? Nope! Just use a mechanical accelerometer – easy to use and cheap to
obtain. Compared with an electronic accelerometer performance computer, it
requires much more input from the user  but there are always going to be some
tradeoffs!
A mechanical accelerometer can use either a
vertical pendulum that is deflected across a scale by acceleration, or a tube
shaped in a semicircle in which a small ball bearing is moved. A US company
called Analytical Performance many years ago produced one of the best of the
latter type of accelerometers, which was called the GCurve. Their accelerometer
consisted of an engraved alloy plate into which was let a long curved glass
tube. The tube was filled with a damping fluid and a small ball bearing was
sealed inside. A very good handbook was also provided with the instrument.
Unfortunately Analytical Performance is no longer
in business but a substitute accelerometer can easily be assembled. Boat and
yatching supply companies sell clinometers that are designed to measure the
angle of boat heel. One such clinometer is the 'LevOGage', which in
construction is very similar to the GCurve. (However, because it is designed to
measure heel angles, the scale is calibrated in degrees rather than g units.)
Like the GCurve, the glass tube is filled with a damping fluid to prevent the
ball overshooting.
Here’s another boating clinometer. This one uses
two scales, the top one being extremely sensitive.
So how does the accelerometer work? When the car
accelerates, the ball climbs up one arm of the curved tube, showing how hard the
car is accelerating. To convert the degrees reading of the clinometer to g
readings, simply use a scientific calculator to find the tangent ("tan") of the
number of degrees indicated. This means that if the car is accelerating hard
enough to move the ball to the 20 degree marking, the acceleration is about
0.36g (tan 20 = 0.3639). However, note that converting the degree readings into
g's isn't necessary for gear shift testing, where you just want an indication of
the relative strength of the acceleration.
Testing
You’ll need a flat empty road and an assistant.
You’ll also need a way of attaching the accelerometer so that it’s parallel with
the long axis of the car and is horizontal. Suction capping it to the side glass
works well. Level it (ie set the accelerometer so it reads zero when you’re
stationary) and then face the car the other way. If the accelerometer is now no
longer showing zero, the road isn’t flat.
Start off by doing the testing in third gear, or
if the car is very powerful, in a higher gear that lets things happen relatively
slowly. Drive along at the slowest speed the car will travel at in that gear and
then plant your foot. At each 10 km/h increment, yell out “Now!” and have the
assistant read the accelerometer and write down the figure. Continue right to
the redline. Then do the same in each of the other gears.
If in lower gears things happen too quickly, do
the odd increments first (eg 30, 50, 70, etc km/h) and then go back and do the
even increments (20, 40, 60, etc km/h) on another run. In really quick cars, do
it one speed at a time. The assistant should see a clear pattern in the figures
for each gear – a mentally drawn graph shouldn’t show acceleration rising then
falling then rising, for example.
If mistakes get made, or the figures don’t seem to
make sense, do more runs.
Graphing
It’s best to graph the data – that way, the
results are quite simple to understand. Here are the graphed results for the
Peugeot 405 SRDT. To make this graph, the acceleration (in g’s) has been plotted
against road speed in km/h. The readings for each gear have then been joined by
straight lines. (Clearly, a smoothed curve would give a better representation of
the real acceleration but for these purposes, a straight line connecting the
points is fine.) Each gear’s acceleration was graphed from the slowest road
speed possible to the highest road speed possible, the top end of 4^{th}
gear (and the missing 5^{th} gear) excepted.
By dropping a line down to the speed axis from
where the acceleration lines for each gear cross with the next higher gear, the
optimal change points can be found. So, for maximum acceleration, the
first/second change should occur at 38 km/h, the second/third change at 65 km/h
and the third/fourth change at 94 km/h.
And what revs do these correspond to? In first
gear 38 km/h corresponds to 4750 rpm, in second gear 65 km/h is 4250 rpm, and 94
km/h in third gear is also 4250 rpm.
Let’s look again at the dyno graph. The best gear
change points have been marked on the graph, the green square showing the
optimal 12 shift and the blue square showing the 23 and 34 shift points. As
can be seen, these optimal gear change points are NOT what you would assume from
looking at the dyno curve!
On the Road
Even though you can clearly feel power falling
away at the top end, once you get used to holding 1^{st} gear to near
the redline, using the optimal shift points feels good. It takes only a few
acceleration runs to realise that what works on paper also works on the road.
However, on this car, the gains made by using the optimal shift points is not
shown in a faster 0100 km/h time – the improvements in acceleration are offset
by the time taken for the gear change from 3^{rd} to 4^{th} at
94 km/h. As the graph of onroad acceleration shows, holding 3^{rd} gear
to 100 km/h saves the shift time and drops acceleration only a little for that
last 6 km/h.
But then this testing is about improving real
world acceleration – not just acceleration to an arbitrary figure like 100
km/h.
Conclusion
Especially in a turbo diesel car with a peaky
power curve and lots of bottomend torque, finding the optimal gear shift points
can make a real difference to onroad performance. You can’t easily calculate it
from a dyno curve of engine power but you can do it by using a simple and cheap
instrument, onroad testing and the services of an assistant.