From the discussed upper limit in cylinder- and piston wall temperature improved efficiency calls for improved cooling.
In speed engines a power increase will normally go together with a more than proportionally increasing fuel consumption to give more internal cooling.
It is not by chance that in the last years more and more cylinder heads with improved heat transfer from the cylinder and combustion chamber to the outside were seen in teamracing.
The Russians, Larsson-Rylin, Nelson and last year Bugl too, changed from the traditional steel contra-piston, directly fitted in the cylinder (fig. 15a), to an all aluminium alloy head.

In the old system the cylinder is hardly cooled because at rising temperature there is no direct contact between cylinder and crankcase due to the more expanding aluminium crankcase.
There's no way to keep a metallic contact above 150°C (with a steel cylinder). It would need an interference fit at room temperature of 0.025 mm or more, which is quite impossible for different reasons.
The all aluminium head with a small moving contra-piston, gives a good thermal contact between the cylinder and outside because the head insert expands more than the cylinder.

In fig. 15b the FMV version of this head is drawn. The compression is adjusted by a push-pull mechanism, avoiding the necessity of a very accurate tolerance on contra-piston fit.
An interference fit of about 0.005 mm is used with a moving contra-piston of 18% SI-aluminium alloy to counter seizure and to get a more constant fit over a wide temperature range.

Since the conductivity of heat is better in aluminium than in steel, this may turn out to be a strong argument to change to aluminium cylinders. A better heat conductivity also means smaller internal temperatures differences, also an advantage with respect to thermal deformation. A logical next step to a better cooled cylinder will possibly be a cylinder with integral cooling fins like HGK and the small Cox engines.
Thermal deformation of the cylinder wall due to temperature differences caused by cooling air entering from one side is the problem to be solved.
We are sure, that when all the mechanical problems in the motor will be solved, only a better control of the piston and cylinder wall temperature will enable further improvements in efficiency.

5. The cooling of the motor in the model.

The goal of all cooling systems is to extract enough heat from the motor and limit thermal deformation. Temperature differences have therefore to be minimized. In the FMV model of '78 the cooling air supply was divided in 5 separate parts.
Fig. 16 shows the way the air was supplied (a bit different form Dave Clarkson's interpretation in Aeromodeler Nov. '78. It really was your own idea Dave, and probably not a bad one too!).

The 4.5 mm i.d. "pipe" directing to the carburetor is extremely important in our view. Air from a constant temperature is the only way to get constant needle settings. The only air of constant temperature is air directly from outside and this is also the coolest air we can get.
Vapour lock problems, being more common than most people like to think, are solved by this approach. In the FMV model the crankcase bottom (ballraces etc.) was not directly cooled for the earlier mentioned reasons. For an all aluminium crankcase cooling is absolutely necessary.
The crankcase between the cylinder fins and the bottom gets a moderate amount of air. Some of this air is directed to the exhaust and leaves the model there through a hole near the exhaust.
The air to the cylinder fins is pressed through by making the cooling duct very closely fitted to the motor.
The air is directed to the head by a tube and pressed through the fins in the same way as for the cylinder. All this cooling air openings are more or less independently variable by changing the tube diameters or the setting of the air baffle thus changing the distribution of the air between the cylinder fins and the exhaust/ middle crankcase part.

This way of a regulated cooling for each part of the motor will only work, if the main air resistances in the system are the ones we can vary. So for the rest of the cooling ducting an air resistance as low as possible is necessary.
A relatively large air outlet is therefore used.

The problem of temperature difference between the front and the rear of the cylinder is not solved in our system. Until now it doesn't seem to be important (for a side exhaust engine like the FMV at least!) because the thermal deformation of the crankcase (bending and getting an egg-like section) doesn't "touch" the steel cylinder at running temperatures. With an aluminium cylinder and/ or integral cooling fins, this problem may become more important.

-oOo-