Race Cam Selection - Rev Range

In this article we're looking at why the rev range of a cam is so important for racing. The information in here is applicable to all forms of racing but the example data is all for MG Midget / AH Sprite A series engines as used in Lackford Engineering Midget and Sprite Challenge, CSCC Swinging Sixties and similar series.

Of the many parts of an engine , its the camshaft choice that really determines the characteristics of power delivery. A series engines are single cam, non-VVC engines so are limited in that the tradeoff is always between power at the high-end or torque in the low / mid-range.  So when you're building an engine to win races, where is the best place (best compromise) to have the power?

It's all about where the torque is.

Having said that cams are a balance of low-end torque Vs high end power, we are now going to throw out the concept of power completely! Power is derived from torque and engine speed so although it is useful as a comparison tool in many applications it can actually hide what's important in the cam selection for racing.

Getting a car around a circuit as quickly as possible is a combination of grip forces and acceleration forces. The engine can only affect forward acceleration so the easy trap to fall into at this point is to think that more peak power (or torque) at the engine will translate into faster laps but its not true and here's why:

1) The engine has to apply force through the gearbox and wheels.

2) The engine will spend only a fraction of time at peak output.

So (without the complicated maths and with all other things on the car being equal) the laptime becomes dependant on the area under the usable torque vs speed curve at the wheels. 

Sometimes its better to think of it like this - As you drive around the circuit you'll go through all speeds from 0mph to VMax (more of VMax later) the (forward) acceleration you experience at each speed is due to the force applied at the wheels pushing you along. It can't be applied in any other way - if the tyres have zero friction you just won't move. The more force your tyres can apply then the faster you will accelerate and the faster you will therefore go around the circuit. So to maximise the acceleration you need to add together the force at every speed you use. Now, not all speeds are equal. Depending on the circuit, the Vmin (at racing speed  so not including the start) is much more than zero, in fact its typically ~50mph.

Thought experiment...

Consider, a 'torquey' engine in a car with only 1 gear. The acceleration at lower speeds will be excellent as the lower speed torque is strong but the compromise is that the higher speed torque is low (thats why it doesn't 'make power'). In the opposite engine, the acceleration at lower speeds will be poor but it will continue to accelerate more at higher speeds when the low-revving engine does not. Which will be faster around a circuit? Well, it depends on the circuit and what the ratio of that 1 gear is! But if we look at a range of circuits and gears what we can ascertain is that the engine that can achieve most force at the wheels for the most amount of time around the circuit will be the faster. 

Moving into the real world

At this point you would be forgiven for thinking that this doesn't amount to much but some useless theory. But the good news is that the theory becomes constrained when we apply it to the real world and we can create a simplified model that will allow us to compare options directly.

Constraints:

Vmin - circuit dependent but available from data analysis if you've been there before.

Vmax - as above

Gearbox Ratios - This is usually fixed for club racers so we have a definite set of values

Final Drive Ratio - Options are 3.7, 3.9, 4.22, 4.5  :1

Rolling circumference of tyre (or Revs/ mile) - Available from the tyre manufacturers website

Extra (sccr box specific constraint - you don't use 1st gear once you are moving!)

So to make a decision on what engine might be best we can now create the mathematical model that takes the engine torque curve, multiplies it through the gearbox and diff ratios and the diameter of the tyre and calculates the applied wheel torque at each speed in each gear.

Real examples!

Luckily we have the data logged from years of racing spridgets on UK circuits so we can use the following:

Vmax  - 120mph (Brands - 114mph, Oulton - 116mph, Donington - 118mph, Castle combe - 119mph, Cadwell - 119mph, Silverstone GP - 121mph)

Gearbox ratios are: 2.573, 1.722, 1.255, 1

Tyre revs per mile: 967 for Avon ZZR 185/55-R13

And we have dyno data from multiple engine builds using different cams available to us.

So lets look at a real example and see what we get.

Example 1 - Conventional race cam Vs tight-lash cam

Engine 1:

Professionally built 1380 engine by a top engine builder using their best race cam. 

Peak Power: 147bhp@8400rpm

Peak Torque: 105lbft@6000rpm

Engine 2:

This was actually one of our builds. Still a nicely built engine but had some cost constraints on it reusing parts. VP8 cam. Peak Torque: 114lbft@6100 rpm, Peak power: 142bhp@7700rpm. 

Comparing the two we see that the conventional long-duration (>300deg) cam shaft gives a very flat torque curve and maintains performance to beyond 8500rpm leading to 147bhp peak power output to be proud of. In comparison the MMVP8 engine has increased torque in the mid-range but can't sustain performance far beyond 8000rpm due to the ~285deg inlet duration which gives it that lower 142bhp power figure.

Lets have a look and see how it fairs in a car.

Interestingly, the mid-range torque of engine 2 (blue) gives a significant performance increase over large chunks of the speed range. The conventional cam (orange) is really great to drive and very rewarding as the engine gives more as you keep the revs climbing. It also sounds fantastic. But over the course of a lap it looses out on torque. However, it also exceeds the required Vmax. Lets try it with a 4.5 diff:

Now the 2 cars are in a fight! the blue car has the edge after each gearchange but the orange car keeps on pulling. Overall the blue car has the beating of the orange in 2 ways: firstly, more torque for the majority of the time. And secondly, it doesn't require a multiweb billet crank (£3000+) because it only just revs over 8000rpm.

Example 2 - Tight-lash camshafts optimised for different rev ranges

Engine 1:

This is a commercially available race engine - very nice, has all the trick bits on it and has a very torquey race cam with characteristics similar to the MMVP8 camshaft but optimised further down the rev range. Peak torque: 125lbft @5500rpm and Peak power : 144bhp@6800rpm.

Engine 2: is the same engine from the 1st example

So we can see that actually both engines make almost the same peak power. Both engines make good torque but engine 1 is VERY torquey and makes peak power and torque further down the rev range so we can see that the area under the engine's torque curve looks really good. If you are doing a dyno shootout this engine is going to win every time. What about when we get it on track - does it translate to a race winning car?

Lets put the torque curves into the model with a 4.22 diff and see. 

When we have the torque applied through the wheels we can see that the 2 engines are very similar in performance through the gears but the higher revving engine (blue) has little bit of an advantage when it is using those higher revs which enable it to hold a gear longer and cancel out some of that excellent mid-range performance of engine 1 (orange). The torquey engine has an advantage under 50mph though so should be strong off the start and out of hairpins although traction can be an issue in these areas so it might not be able to use the extra torque in the real world. So on balance, these engines are going to be pretty evenly matched although not necessarily at the same points on the track.

But this isn't the full story. 

With a 4.22 diff Engine 1 has a Vmax of 111mph @ 7500rpm compared to Engine 2's 121mph @ 8200rpm. So at the end of the straight at all major UK circuits, Engine 1 has become gear limited. We can fix that though by changing to a 3.9 diff for E1:

Now we have 2 engines that can race against each other for the whole track. The shift points are very equal but E2 in blue now has an advantage over E1 at almost all speeds, despite making 10% less peak torque at the engine. 

So what have we really learned in all this?

Well, we can see that by looking at the applied wheel torque rather than just the engine output we can get a much clearer view on how the car is likely to perform on the track. We can also see that looking solely at peak numbers doesn't give us the whole story. When we are looking for optimum performance we must be prepared to look past the headlines and really analyse how the car will perform. And I guess thats a metaphor for all engineering really. If you want to get the best solution you have to be prepared to look in depth at the information instead of just looking taking things at face value.

The Mamba Motorsport MMVP8 cam is available here. We have built engines far exceeding the performance of the one detailed above but the characteristics of the camshaft remain the same with a good torque spread offering strong torque without requiring either a long diff or an expensive billet crank.