LOL all, sorry but tuning for resonance has everything to do with it. First we are dealing with side draft carburetors and the complexities of a prop on one side. The issue of making the manifolds all equal length would prevent a lot of problems with installation and I agree that would be the nice way to do it all however we need to consider it has to fit into an engine cowl and also fit within the design concept. Naturally a equal length intake and the same for exhaust runners would be ideal.
Rotax had to make some compromises. it is understood that engines are not flying normally at low power settings so that range for tuning was not the concern. The most important areas are the 65% to 100% power ranges. We can see this in a fully instrumented engine. For use financially challenged people a good set of EGT will also give us a good window into what is happening. In testing a number of the carb engine 914 types you can see this very well on an engine with 4 EGT sensors. For the most part we need to understand that there is some imbalance for temperature due to airflow considerations, the prop has a big influence on this part in cooling the ½ side barrels and heads as a tractor and a smaller effect on the cooling of the 2/4 barrels as a pusher. This is due to the fact that air is compressed on the thrust side of the prop as opposed to the suction side, the compressed or thrust side depending on your cowl or lack of cowl can have a big influence. That said the obvious difference in the manifold length of the runners of the ¾ cylinders and the ½ cylinders will always show 2 cylinders run richer that the other. This will always show an EGT split due to the slight fuel imbalance due to the tuning pulse.
Given we have a reference now we pick up the throttle and watch the EGT, as it advances the flow will change due to the wavelength of the intake valve pulses back into the manifold. (resonance) At about the 65% power area is reached we get to the designed tuning length and front that point the EGT now drift toward equal readings. At the design operational speeds EGT is almost always completely balanced. Give it a try, it never fails to amaze. When you land and taxi back to the hanger a plug check will show the same old rich cylinders due to the run at lower speed as you taxi in and the fuel imbalance again shows up.
In short, it is designed that way. The injected engines have greatly reduced any of those kinds of imbalance to the point that nobody complains about EGT differences anymore. So Jeffry is right, we have to look at what the turned lengths were designed for and it was never for idle speeds.
I worked for many years with 2 stroke race engines and can tell you it is all about tuned waves and pulses. Most 4 strokes in aircraft have huge differences with intake systems and for the most part exhausts are almost unexplored territory for a lot of engine companies.
Cheers
Just a PS on the original 9 series design. This engine was laid out on paper, not a computer, in the mid 1980s. I had the pleasure to see some R&D runs before serial production and can tell you for a fact that all these concerns were major discussions. There were many trials and many failures before the production series were made. And yes they have great beer in Austria.
