[Viper2000]

Originally Posted by Crumpp

It was....

Both Daimler-Benz and BMW were forbidden from even being in the aviation market.

AFAIK they had to quit for 3 years. Thereafter they didn't get back into aerospace because they didn't have a market rather than anything else.

However, Daimler has quite a big stake in EADS, whilst BMW started a joint venture with RR to make turbofans in Germany from 1990, though now this is 100% owned by RR.

Originally Posted by Crumpp

And injecting fuel directly into the combustion chamber is even better, Viper. How hard is that to understand?

Why do you think it's better?

A supercharger is a pretty effective way to homogenise a mixture. The intake manifold is going to end up at roughly charge temperature, which for a Merlin at high power is going to be about 90ºC. You are very unlikely to see condensation of the fuel onto the manifold at that temperature. FAR will therefore be pretty constant from one end of the manifold to the other.

Charge distribution may well vary, which would modify CHT somewhat, but the same argument applies to air distribution.

FAR will become variable when supercharger delivery temperature is low, and this will affect acceleration behaviour, especially from low boost & revs. But aero-engines spend most of their time at fixed, relatively high, power settings, and so this sort of transient behaviour is far less of a problem for an aero-engine than for a car engine.

Originally Posted by Crumpp

And that it is much more efficient to realize the power gains by directly injecting fuel into the combustion chamber than it is by dumping it into an intake manifold......

If you are supercharging then you'll win by injecting into the supercharger and thereby reducing supercharger work.

The supercharger is basically adiabatic if you're not injecting fuel or water into it. However, isentropic efficiency of superchargers tends to be much lower than the isentropic efficiency of the compression stroke of a piston engine.

In any case, you're always going to gain more by reducing temperature as early in the compression process as possible, because compressors (whether steady-flow or non-flow) produce temperature ratios in exchange for pressure ratios, whilst the absolute work required for the compression process is proportional to deltaH, i.e. Cp*deltaT.

If you reduce the starting temperature then you reduce the deltaT all the way down the chain, and the benefit multiplies. Therefore, if your fuel is liquid, you really want to inject it at or before the start of the compression process in order to maximise the thermodynamic benefit associated with its latent heat of evaporation.

Clearly for a naturally aspirated engine you might as well go for direct injection, especially if the number of cylinders is small.

The cylinders & pistons are very far from being adiabatic, but are very efficient at performing compression work. The limiting factor is the rate at which they can pass non-dimensional flow through their intake & exhaust valves at any given rpm. Hence supercharging; pre-compressing the air allows you to get more absolute mass flow rate into the fixed non-dimensional mass flow capacity of the piston engine. That's the objective of the exercise.

You use a steady flow machine upstream of the unsteady flow machine because unsteady flow machines are inherently bigger than steady flow machines, and therefore you can shrink the physical size of the engine in relation to its effective flow capacity.

Originally Posted by Crumpp

No, it is attractive and if we had the technology to do it on a cost effective basis, we would have done it. It is the ultimate fuel metering method for a piston engine in terms of power and efficiency. A single point injection simply cannot maintain a stoichiometric mixture in all the cylinders. That is why the EGT and CHT will always be different in each cylinder unless you have direct fuel injection.

That's only true for naturally aspirated engines.

EGT and CHT will be different anyway because that's life; holding FAR constant is great but it's not magic; airflow into the cylinder depends upon induction manifold design and engine speed. Induction manifold design is quite a complex business, and compromises are inevitable.

DI is very useful if you want to vary non-dimensional power setting over a wide range, but this isn't so important for an aero-engine, and so the higher design-point efficiency offered by injecting into the eye of the supercharger is a pretty compelling argument, before you even consider the cost, mass and complexity advantages.

Modern GA engines are going DI because they're going CI (in order to burn Jet-A and save money), and also because they don't have a lot of cylinders, which means that the cost of injectors is inherently less important.