To avoid possible confusion by using the words ‘high’ and ‘low’ which mean different things to different people when talking gear ratios, I am going to use the word ‘tall’ to mean a gear that gives more speed for a particular engine speed and ‘short’ for a gear that gives a lower road speed for a given engine speed. I am considering a 4 speed box, and ‘top’ for a transverse box may well be 4th, particularly on the slower circuits..

The first thing to know is that power= rotational speed * torque. It is important to know the power curve of your engine, from which you can work out the torque curve. More of this later.

Years ago, before data logging was available, I worked out the car speed, acceleration, and torque at the wheels for every ten yards around Snetterton. You spend the vast majority of a lap time accelerating in ‘top’ gear. This means the single most important gearing decision is to install a final drive that in ‘top’ gear gives maximum revs at the end of the longest straight. When I say maximum revs I should qualify this to mean the maximum engine speed where there is usable power. Using gearing that is too short will give away lap time with restricted speed at the end of the straight. Using a gear that is too tall will mean that all that time accelerating in ‘top’ you will have reduced torque at the wheels, roughly by the percentage difference in the overall gearing. If we are talking which diff to install in an A series axle, the decision is reasonably easy because the difference between them averages about 8%. With a Hewland gearbox the difference is more like 5% and then wind direction and strength can make the difference. I have done back-to-back testing on this at Snett with two different cars with different engines. One car had a peaky race engine and poor body shape, and the other had a torquey engine and a slippery body. With the former it was quicker to gear to feather the throttle for the final 50 yards at the end of the straight. For the latter it was quicker to not quite reach max revs. Luckily for A Series axle users the majority of circuits we use demand the same tall diff. Of the others, Brands Club and Mallory need a shorter diff, and Thruxton possibly extra tall. Oulton is marginal between the tall and short.

For comparison I work out the road speed at maximum revs for the particular engine for each diff.

Take the diameter of the tyre in inches, multiply by 3.142 to get the circumference, divide by 36 to get it into yards and then by 1760 to make it miles, multiply by the engine revs and multiply by 60 bring that into hours, then divide by the overall ratio (just if the diff with a 1-1 top). You then have the speed. I stand to be corrected here, but I think something in the order of 110mph should work for those majority circuits.

The next gear to consider is third. The main consideration here is the rev drop when going up into top. Third is too short if the revs drop much below the maximum torque. With a transverse box this is a real concern and the inlet and exhaust manifold lengths have to be chosen to give decent torque at this point. With a Hewland or equivalent the choice of third might be influenced by optimising for a particular corner, perhaps the one before the long straight.

Choice of second gear is more flexible because a car accelerates much faster in second. This is for two reasons. Firstly the shorter gear gives much more torque at the wheels. Secondly below 50mph the aero drag is far less. This means if there is a choice of second gear it can be chosen to suit a particular slow corner.

First gear is used for driving around the paddock and for a couple of seconds at the start. Even the tightest hairpin, perhaps the short circuit shortcut at Cadwell, going into first takes up time to change up again, and you have much more torque available than traction will allow. For a similar car to a 750 I reckoned the optimum ratio for starting is one that gives 28-30mph at max revs. With the transverse box Billy found starting in second did not cost too much time, so this may be a sensible option in the wet.

If you have a complete power curve for your engine, or you can interpolate a minimum of about 4 known points, you can work out the torque and speed at the wheels for each gear. The written formula for speed is the starting point. The units of torque don’t matter for comparison. This spreadsheet shows the ideal point to change gear, where the torques match. In the low gears it is certainly best to change up at the red line, but with the wide spread of torque of the Fiat taking it to max revs in third may not necessarily be quickest way. It is also useful to see if 5th is a useful option for the transverse box. From what I hear, for this engine there is a huge variation as to what people rev to, and presumably where peak power is, so I suggest you make decisions based on your own data. On the other hand, the engines I have seen on the rollers peaked at about 6500rpm. This speed/torque chart will also help with those marginal decisions about ‘top’ gearing.