Economical engines
Start talking the fuel economy of different petrol engine types and designs and things get complex, fast. Amongst other factors, fuel economy is affected by internal friction, pumping losses, combustion inefficiencies and the air/fuel ratio that is used.
The biggie with internal friction is, literally, how big the engine is. A larger engine has longer internal bits rubbing other bits, and so a 5.7 litre V8 is always going to have poorer fuel consumption that a 2-litre four cylinder. That statement applies when both engines are producing the same low power required for cruising, but may not be the case when the power demand is high – climbing a hill while pulling a trailer, for example. In the latter situation, the smaller engine will have to rev very hard to develop adequate power, and the higher the speed of the engine, the greater the power loss through friction. Because of the high loads to which it is being subjected, the small engine might also move out of closed loop (ie ~14.7:1 air/fuel ratio) to a much richer mixture. So as the power demand increases, the practical on-road fuel consumption may not so clearly favour the smaller engine over the larger engine.
Another way of seeing this is to look at the fuel economy gained from a small engine car that always has to have the ring driven out of it to keep up with traffic. In this situation, the fuel economy is often poorer than the larger engine car that is always just loafing along.
Pumping losses refer to the drag caused on the movement of the pistons on their intake and exhaust strokes. Any restriction on the intake – including, critically, the partly closed throttle – will lower the pressure of the air that fills the cylinder on the intake stroke. Rather like drawing down a syringe that has the needle opening blocked, power is needed to overcome this partial vacuum. On the exhaust stroke, anything that restricts flow out of the cylinder – from a poorly flowing muffler to bad port design – will again require power that’s subtracted from what is available at the flywheel.
Some BMW engines dispense with the throttle – and instead change intake flows by varying valve lift and timing – however the intake pumping losses remain. (Diesel engines, of course have no throttle and so much smaller pumping losses.) A better approach to reducing pumping losses is to adopt the Atkinson or Miller cycles, where the closing of the intake valves is much delayed at lower engine rpm. This poorer intake flow requires the driver to more widely open the throttle for a given power output, so reducing pumping losses. However, the engine also develops less power because its volumetric efficiency is much lower than an engine with conventional valve timing. Atkinson/Miller cycles are therefore used only when there is forced induction at low revs (eg a supercharger as in the Eunos 800M) or power is available from an electric motor (Toyoya Prius and most other current hybrids).
The lower the temperature of the exhaust, the more heat energy that has been extracted by the engine from the fuel. Turbochargers use the heat and flow in the exhaust that’s usually wasted – therefore, a turbo can improve the fuel efficiency of an engine. However, a turbo imposes greater exhaust backpressure (pumping loss on exhaust stroke) and a turbo is usually fitted to an engine which has had its compression ratio lowered over standard (reducing efficiency). Also, a turbo that is matched to provide light load boost (the vast majority of driving occurs at light loads, so that’s where it’s critical for fuel economy) provides even higher exhaust backpressure at high loads, although that also depends on the wastegate size.
But the big advantage of a turbo is that the turbo engine has frictional losses that correspond to its size but a maximum power output that corresponds to an engine 1.5 times or more larger. Modern turbo engines with well-matched turbos can also provide boost low in the rev range, resulting in a greater availability of power while keeping revs down. Over a non-turbo engine of the same size developing the same power, this reduces the actual frictional power losses.
Of course, the actual on-road fuel economy is also hugely dependent on the car itself – rolling friction, aerodynamic drag, mass – and the driving style that is used.
With so many interrelating variables and so much which is hard to really pin down, what’s the outcome?
Firstly, a modern engine with good breathing will reduce full-load losses. Secondly, the engine should be small enough in capacity (eg 2 litres) and in cylinder number (eg four) that it has low frictional losses and requires larger throttle openings – but isn’t so small that it’s always struggling. If it is turbocharged, a greater peak power output will be able to be gained with only a negligible reduction (if any) in cruise fuel consumption. The turbo should be small enough that peak torque occurs low in the rev range – eg less than one-third of redline rpm. This will probably result in a modest peak power output but with a very high average power across the rev range.
The air/fuel mixtures should stay in closed loop either the whole time, or alternatively, until very high engine outputs. This requirement will probably necessitate the placing of the cat converter a metre or more from the engine. Cat warm-up strategies may then be required on cold start.
The vehicle itself should be aerodynamically slippery and have a very spacious interior when compared with its exterior dimensions. Its mass should be low in comparison to the size of the car.
You can do your own search but fitting all these criteria is, for example, the Volvo S40 2-litre turbo (8 litres/100km in the Australian government test) and the 2-litre turbo Saab Linear 9-3 (8.1 litres/100km). Even the Saab Aero 9-3 has a rated economy of 8.2 litres/100 km – and that’s with a power output of 155kW and a huge 300Nm of torque.