LINEAR TO ROTARY CONVERTER

What is wrong with the good old crankshaft? Nothing at all, with the shaft. The problem with the 'system' lies with the fact that the connecting rods flip from side to side with each cycle, effectively shortening their length in the process. If we assume that the crank is revolving smoothly, then the 'big-end' bearings will be describing a sinusoidal directional motion. The piston at the other end of the con-rod, therefore cannot be describing a sinusoidal motion, since the "flipping" has modified it.

So what? – Well, several things actually. In the first instance, the two con-rod bearings are being subjected to alternating lateral forces. Forces which become more damaging as engine speed rises. If the frequency is allowed to coincide with that of any component's 'resonant' frequencies, and if this persists for long enough, catastrophic damage will result. Spin up a well run-in engine and monitor it with microphone type sensors and you will find many engine speeds which will give rise to "resonance amplification" (as is exploited in a trumpet, or even a laser). To make matters even worse, the relationship of the piston velocity curve and the crank motion, by being only marginally different, creates a second harmonic hazard. The French piano accordion or the twelve string guitar are common examples of this 'near match' tuning creating additional, and fairly spectacular, harmonic interactivity. Put an engine under load such as on a steep hill with a strong head-wind, drop the gears and push up the revs in order to maintain vehicle speed, and you have created the classic "big-end" failure scenario. With luck, it will only be the bearings and the crank surface which are seriously damaged. If you are not so lucky, a con-rod smashing through the block is by no means that rare a result. One only has to consult any engine hand-book and note the number of different size bearings, which allow for multiple re-grinds, to realise that the problem is no secret.

Is the conventional crankshaft system really so bad? No, if we don't want higher engine speeds, and we can take steps to avoid sustained critical resonance. After all, it has been around for years - Perhaps even as long as the blunderbuss, longer than the scrubbing-board, and most definitely longer than the wind-up gramophone!

Increased engine speeds mean increased power - without an increase in weight. The average crankshaft engine is pushing its luck at 12k RPM - the VLB engine is comfortable with 20k. The VLB 'transfer oval' is essentially a "Scotch Yoke" and is by no means a new idea; furthermore, it too has been in widespread use in engineering for more than a century - at least. Pad-saws and sewing-machines are pretty common examples of its use, but it was also common in much heavier machinery such as steam powered pumping systems as well.

The mechanism is efficient, it's smooth and it's easy to engineer. It reduces the number of significant moving parts, it doesn't require high-pressure oil lubrication, it's less expensive, and it requires less skill, and takes significantly less time, to replace. It ensures that the pistons share the precise 'sinusoidal' motion as the crank, so it doesn't generate spurious harmonics, and the piston movement is more efficient, a 'plus' which rises with speed, to become very significant at 20k.

The oval track and the roller bearing are precision engineering parts, but since these can be produced by existing precision production facilities, and the engine is designed to allow their replacement with such ease that one could teach the requirements to any mechanic in about ten minutes, it has to be a significant improvement over the regular 'white-metal' half bearings which seldom get put in properly. Then there is the fact that the design needs only one bearing for each pair of pistons.

The mechanism runs much more smoothly than most people expect from looking at the sketches. The bearing is actually rolling along a perfectly aligned 'circular' path. In fact, during the 'empirical engineering' phase of the VLB engine development, we found that we could substantially reduce the flywheel size and weight from that required by the conventional crank system.

OK, let's look at the negatives. "It looks funny", seems to be the most common reaction, but we will skip that one. It does need lateral stabilising 'oil-slides'. This is because a piston must have a little 'play', that is to say that it must not scrape the piston walls, and because the 'rings' which are employed to effect this gap without loss of compression have to allow for heat expansion, a tiny amount of 'slap' can occur at mid-stroke, just when the roller bearing changes direction. This can be adequately dampened out by the slides (which are constantly fed with oil) except when the dreaded resonance occurs. However, because the assembly has a very simple and fundamental resonant frequency, and because the 'whole' design of the engine recognises the need to avoid this problem, the system has been equipped with 'resonance dampers' – the universally accepted 'best method' of dealing with unavoidable resonance problems.

A quick look at the WCY.1 sketch will reveal all. What you might have thought was just a combustion chamber with an air inlet and exhaust gas outlets, a fuel injector and a couple of spark plugs (with an excess of air) - is also a 'tuneable' damper. There is a theoretical 'resonance' hazard, but by 'tweaking' the fuel input, and the exhaust port timing and duration, resonance amplification is avoided.

The conventional crankshaft is a relatively simple device to build, and with its soft bearings and high-pressure oil system, it can be made to work well enough for most applications. On the other hand, the transfer oval has to be precision-engineered, with significant attention given to stress relief. Even then, without the damping, and without the oil slides, the resonant hazard can be considerably worse with the transfer oval system, than with a conventional crank. Which is quite possibly the reason why the scotch yoke never caught on (for IC engines) in the first place. On the other hand, our production engineering abilities have improved a bit over the past 100 or so years, have they not?

John Allen