Torque

Torque is simply a measure of the twisting force that is applied in an attempt to rotate an object. It is a force that is applied to a lever arm, and is measured in Newton Metres (Nm). Newtons are a unit of force, and at the surface of the earth a 1kg object will exert a force on the ground of 9.8N, due to gravitation. Torque is, in effect, the product of the force and the length of the lever arm. Understood in this way, it is clear that there are two ways of increasing torque; either increase the force or increase the length of the lever arm. Think about a door that you wish to open. You can either apply a small force at the outer edge (where the lever arm is long), or far more force close to the hinge (where the lever arm is short). In this way, you can have the same torque, but it is developed by applying different forces at different lever lengths. "But what has all of this got to do with cars?" I hear you ask. Well, think about some of the engines you know. You know that increasing the lever length can increase the torque. Now think of those long stroke engines, and how, for their capacity, they seem to generate a lot of torque (such as the Ford 4-litre engine). The capacity, in effect, gives you the force, and the lever multiplied by that force gives the torque. So, to increase torque, increase the stroke, or increase the capacity, or both. Of course, there are negative aspects to increasing the stroke, which we will come to later. There is one facet that is very important to realise: you can be exerting a lot of torque but not be doing anything. Think of all the time that you have skinned knuckles attempting to loosen a rusted bolt. You have definitely applied a lot of torque, but you have not done any work. We now come to the aspect that causes a lot of angst among many enthusiasts - power. Most people think that they know what power means; unfortunately, many do not. POWER Power is defined as the rate of doing work, and has units of Kilowatts (kW - named after James Watt) or horsepower in the old Imperial units. To see what power actually is, let's consider the experiment that James Watt did a couple of hundred years ago. He wanted to know the rate at which draft horses could raise coal from a coal mine. So he measured the mass of coal brought up, the distance that the coal was raised, and divided this by the length of time that it took to do this. He found that the horses would lift 33 000 pounds 1 foot in one minute (or 1 pound 33 000 feet in one minute). He called this unit the horsepower. In metric, the Watt is defined as the power to do one Joule of work per second. One horsepower is equivalent to about 746 Watts. Now, another aspect to realise is that power and torque are intimately related. Remember how power is the rate of doing work? So, with engines, power is the torque multiplied by the radial velocity. Without getting into the physics in detail (we will forget about radians etc), the power of an engine is given by the following relationship:
Torque (Nm) x RPM

Power (kW)= 9549

Or, in the old Imperial units:

Torque (lb/ft) x RPM Power (hp)= 5252

So, as you can see, power and torque are very intimately related... but it is important to realise that they are different. Remember that you can be applying a lot of torque for no result. Well, looking at the fact that power is the rate of doing work, it is obvious that if you are doing no work, you generate no power! So, no matter how hard you push that spanner, if the bolt doesn't turn, you generate no power. There are two ways of increasing the amount of torque generated by an engine - either increase the capacity (or, more correctly, capacity times volumetric efficiency), or increase the length of the lever arm (or stroke). For increased power, you can increase either (or both) the torque, or the revs at which that torque is generated. Here we revisit the issue of stroke. Increasing stroke will increase torque, so theoretically it would be good to have very long stroke engines. The problem is, if the stroke is too long, the volumetric efficiency decreases, particularly with increasing revs (which is why long stroke engines don't like a big rev, apart from the rotating friction and harmonics). Now this decrease in revvability more than compensates for the torque increase, which is why very high power output engines tend to have very short strokes (once again, engine strength issues ignored). Okay, now we have all that physics stuff out of the way - we understand the meanings of power and torque in their correct contexts. So next week we can get into the interesting stuff - looking at engines, cars, and how power and torque relate to performance in terms of acceleration and top speed. Last week we looked how power and torque are defined and related. This time we apply some of that knowledge to seeing how hard cars accelerate when endowed with engines with either lots of peak torque, or lots of peak power. Which is required to give the best acceleration? Let's say that we have three engines:

Engine A, which is representative of the old Ford 4.1 carby engine. Engine B, which has the same torque curve, only it is spread out, so that the torque numbers are at one and a half times the revs of A.

Engine C, which has the same power curve as A, but spread so that the power figures are one and a half times the revs of engine A, but it only has two-thirds of the torque.

Revs 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000 7500

Torque (A) 190 255 280 300 280 255 225 180 120

Power (A) 20 40 59 79 88 93 94 85 63

Torque (B) 170 190 245 265 280 295 295 280 260 235 225 195 160 120

Power (B) 18 30 51 69 88 108 124 132 136 135 141 132 117 94

Torque (C) 113 127 163 177 187 197 197 187 173 157 150 130 107 80

Power (C) 12 20 34 46 59 72 82 88 91 90 94 88 78 63

Past Redline

To shows how this data looks graphically, here are the power curves of the three engines:

and the torque curves of the same three engines: Okay, so now we have some numbers to work with to determine which of these engines will give the most performance in terms of acceleration. If peak torque were the most important characteristic, you would anticipate that engines A and B would perform similarly (they have near identical peak torque), and that engine C would trail. If peak power were the important factor, you would expect that engine B would be quickest, with engines A and C level pegging. For argument sake, let's assume that each of the cars has a gearbox with a 2:1 first ratio, and a direct second, and that in first the car does 10 km/h per 1000rpm, and 20 km/h per 1000rpm in second. Also assume for argument's sake that the diff ratio is 1:1. Use the basic equation handed down from old Isaac Newton, F =ma, where F is the force (related in this case to the torque at the driven wheels), m is the mass and a is the acceleration. The force at the driven wheels is related to the torque by considering factors such as wheel radius, but these are constant so we will leave them out, and basically say that the torque at the tyres is directly proportional to the engine torque. So, in turn, the acceleration is directly proportional to the torque at those driven wheels. Let's have a look at what happens when we compare the cars.
Speed Torque at the Wheels (A) 10 20 30 40 50 60 70 380 560 560 450 240 280 255 Torque at the Wheels (B) 340 490 560 590 520 450 320 Torque at the Wheels (C) 254 314 394 374 346 300 214 Torque at the Wheels (D) 380 560 560 450 240 280 255

80 90 100

225 180 120

295 280 260

197 187 173

225 180 120

Now suddenly things look very different! Notice that car B pretty much always has the acceleration advantage, but that car C trades blows, so to speak, with car A. Car C has a lot torque at the wheels to 40 km/h, more from 50 km/h to 60 km/h, less between 70 and 80, and more from 90-100 km/h. On the whole, car C would accelerate roughly equally to car A, except over the first part of the rev range, where car A would have a decisive advantage.

But here is where gearing plays its part. If we put in a diff of ratio 1.5 in car C (now call it car D), you will have the same torque figures at the wheels throughout the range as car A (which is what you would expect, given that the torque of engine C was only 2/3 of that of engine A). So, it is clear that you can compensate for a lack of torque if you have power. Simply change the final drive ratio. The question is, can we change gearing to compensate in the case of an engine that has a relative lack of power? Let's have a look by comparing engines A and B. The first problem that you are stuck with is that if you attempt to lower the final drive ratio, the car will no longer reach its top speed (it runs out of revs). The second point to realise is that you could quite easily change the gearing in engine B (use a 1.5 diff ratio again) to get significantly more torque at the wheels than engine A. So, in short, you can compensate for a lack of torque in a powerful engine by shortening gearing. But you cannot compensate for a lack of power by gearing.

(An interesting example - by appropriate gearing, you can get a human being to generate a lot of torque at the driven wheels at extremely low (like about 1 km/h!) speeds. The ability to hold this torque at higher revs is power, and the fact that a human being does not generate much power is the reason that a human-powered vehicle are not able to accelerate quickly at any but the very lowest speeds.) Earlier I mentioned that, by knowing engine power at a certain speed and mass alone, you can determine the acceleration. The equation is: P = Fv Where P is the engine power, F is the force and v is the velocity. Now, remembering that F = ma, P = mav
P a= mv

So all you have to do at a certain speed is look at the tacho, look at the power graph and determine how much power you develop at those revs, and you can determine the acceleration at that speed (assuming no drivetrain losses). You will note that you cannot do the same for engine torque without knowing the gearing to determine the actual torque at the wheels, rather than the torque at the engine, which is modified by the gearing. As you'll now realise, gearing is a torque multiplier, not a power multiplier.

It is therefore clear that in order to maximise your acceleration, it is very important to have high power, rather than high torque. You can get high torque at the wheels by using appropriate gearing, but you cannot get high power from gearing. In short:

for best acceleration, maximise power for best driveability, a high torque, non-peaky engine is better

An engine with high torque throughout the rev range will have good low-rev and part throttle response; this is the reason that cars like 4.1-litre Cortinas feel that they are very quick, even though they are not in absolute terms. That 4.1 has a very flat torque curve, with a maximum torque similar to the Commodore Ecotec V6 - but I know which engine I would prefer in outright acceleration terms (the V6 has about 50% more power)!

Conclusion

So power is the critical determining factor for maximum acceleration, and torque is necessary for driveability. This can quite clearly be seen in looking at the types of engines used for certain applications. Trucks and industrial vehicles use engines with large amounts of torque with a fairly flat torque curve, but relatively little power. Racing and sporting vehicles use high revving engines, with high power and relatively little torque in comparison to their power outputs. That is why you have F1 having engines revving to over 18000 rpm. And there's no use doing that if to maximise acceleration all you needed to do was maximise the amount of torque you had!

Top Speed
The top speed of a vehicle has almost nothing to do with the mass of the vehicle (apart from bearing friction and tyre losses which, in terms of power consumed at top speed, are minimal). The two major factors are power and aerodynamic drag. The equation to get a pretty good approximation of top speed of a vehicle is that relating power to drag:-

where P is the power in Watts, r is air density (1.3), C Dis the coefficient of drag, A is the frontal area in square metres, and v is the velocity in m/s. To see how well this works, lets take, for example, a modern Falcon XR6 and an old 4.1 Cortina. The XR6 produces about 110kW at the wheels, has a CD of 0.31, and a frontal area of about 2.4 square metres. In the case of the Cortina, the numbers are 60kW, 0.48 and 2.2. To get top speed:

For the XR6, this equates to 66.6m/s or 239km/h. The XR6 has been tested at over 230km/h, so this seems pretty reasonable. For the Cortina, the speed is 48.4m/s, or 174km/h, which is pretty lineball with the tests for the car. An interesting exercise would be to test the numbers for your own favourite car, remembering to use power at the driven wheels, not at the flywheel. And again, note here that it is peak power, not peak torque, which is the important engine performance variable.