Induction system

The only real mistake in the design of the FMV was (and still is) the size of the crankpin: 5 mm dia times 3 mm length.
The reason to choose a short pin was to minimize the effect of non perpendicularity caused by crankshaft bending (during ignition up to ~0,4° rotation) and crankcase-bending due to temperature difference between front and rear of the cylinder (approximately 20° C, giving about 0.03° forward bending).
With a short pin a small crankcase volume is obtained which is generally considered to be an advantage in a two stroke engine.
Unfortunately because of the relatively small projected aerea (15 mm2) compared with other motors (Rossi Æ4,5 x 4,5; Nelson Æ4 x 4,7; Super Tigre Æ5 x 4,5; Bugl Æ6 x 4,5) the generated heat per mm^2 is relatively high. (The specific generated heat is proportional to r.p.m., pin diameter, pressure, and coefficient of friction).

However, considering the small difference in aerea, it was difficult to justify why the FMV turned crankpins blue and ruined big-end bushes in the con-rod, while the Nelson kept their crankpins and con-rods looking brand new even after hours of running.
It must be said that the first ruined con-rod we saw in a Nelson (King-Rudd's) brought us the answer. King-Rudd had changed the induction system to a Natalenko type, having the same disadvantage as the Russian bell-valve type we used at the time, to direct no fresh gas flow to the big-end.
An easy solution seemed to make a disc-valve on ballraces, admitting the fresh gases right in the bottom of the engine. But surprise-surprise, no improvement was found.
So at last we tried the Nelson (K&B) rotor, and it solved all our big-end problems. The explanation is quite simple after all.
In a disc-valve all the heavy parts (droplets of fuel) in the fuel-air mixture were centrifuged to the outside without hitting the big-end. In the case of the K&B Nelson-rotor the particles will hit the crankpin on their way, also guided by the groove in the rotor front, cooling and lubricating the pin.
The big-end of a motor really being a "hot spot", could be used to help vaporizing the fuel that arrives there, thus giving more efficient combustion (hope, hope!!).

At Woodvale, we had the opportunity to talk to Don Jehlik (World Champ. T.R. '66 and '68), more or less the supposed "inventor" of this induction system used in his ETA and HP and it turned out that he had always known that this really helped to make those motors superior to everything flying around at the time.
A few things became clear now:

• Bugl had to go to a 6 x 4,5 mm crankpin to solve his problem;

• Rossi RV's give more problem than Rossi FI's, especially at the big-end side;

• Our Rossi front-exhaust stopped to be a 50-lapper after changing to a Natalenko drum.

The gas-flow of the Nelson-K & B rotor is identical to that of a front induction system, so why not use an FI-engine in TR?
We think there are four main reasons not to do so:

• The crankshaft on a FI-engine is a small monster, stiffness and strengthwise.

• It makes the use of a bush between the inner rings of front and rear bearing nearly impossible.;

• The carburetor will probably be exposed to turbulent air all the time and also to dust while on the ground;

• A too long distance between tank and spray-bar will cause acceleration troubles and fuel pressure differences between flying and ground running, depending on the position of the fuselage.

Getting back to the size of the big-end, there are indications that a length-diameter ratio around 1, like Rossi and Super Tigre have, is a good compromise between sensitivity for the shaft's oscillation and that for end effects of an excessively short pin (pressing oil to the sides and shortening the length of the lubricated part considerably). Maybe this effect made the FMV big-end extra critical for the type of induction systems.

A few words about the material of the pin. For still difficult to explain reasons the hardness of the pin has turned out to be of extreme importance.
The needle bearing rollers used in Nelsons, the FMV's and in some other motors are very hard (over 1000 Vickers), which is more than any other crankpin in one-piece crankshafts.
After the Worldchamps and a very interesting talk to Don Jehlik we plated with hardchromium the pin of one FMV (layer thickness 0.001 mm) and found about 0.5 secs/10 laps improvement.
A hard chromium plated surface, being extremely hard and also porous seems to have significant advantage.
Be sure that the hardness of the chromium plating is at least 1000 Vickers. For example, the chromium plating on S.T. G15 and X15 big-end pins is far too soft to be used with a hard type bronze or cast iron bush in the con-rod, and fell off during a few experiments we did.

Since we really believe that a lot of power of our motors is spoiled in the big-end, our future investigations will be focused on materials selection and design of this part, including improved cooling and lubrication.

4. The top end of the motor.

Our ideas on con-rod, piston cylinder assembly and cylinder head, as well as on the thermal aspects related to cooling will be descried here.

4.1 The con-rod.

Although being a simple part, the con-rod plays an important role in the engine. Since its weight is affecting the balance in the motor, it has to be made as light as possible.
Not having done experiments in thinning down con-rods to the point they will brake, we still use our original rod. It's main body has a cross section of 7 x 2,5 mm^2, the small end has a Æ4 x 4,5 journal bearing and the big-end one of Æ5 x 3 mm. Its weight is 1,9 grammes.

Until now bronze bushes were used with a wall thikness of 0,25 mm.
There is absolutely no reason to use thicker bushes and it leaves more room for the dural (AISI 2024 type).
Our bushes were mounted with a 0.01 mm interference fit and loctite hot retaining compound.
Minimizing vibrations is an important part of our design philosophy so we gave a lot of attention to the free-play of the con-rod in all possible directions.
A play of ~ 0.025 mm on both crankpin and piston fit was found to be right and a few experiments on tighter and looser fits gave no dramatic effects. This is probably due to the fact that the difference in thermal expansion between the steel pin and the con-rod is about ~ 0.01 mm at 150°C. The fit of the crankpin into the big-end shouldn't be too tight anyway, to allow the crankshaft bending during ignition.

In the description of the crankshaft positioning it was clearly stated that we want the con-rod as well as big-end exactly on the piston axis.
Limiting the con-rod lateral play on a big-end and small end location was the only solution we could find for the problem.
At both ends a lateral play of 0.06 mm has been used.
See also fig. 10.
Looking at con-rod small ends and piston pins told us, that in normal type piston the axial movement of the con-rod is considerable, introducing perpendicularity errors and increasing heat generation of both big- and small-ends.