So, now that my bike is completed and running well I can turn my mind to more theoretical and unimportant things. I've often wondered why the weber manifold is split instead of open inside to make a common plenum for both barrels of the carb to feed. Seems to me that it would eliminate the need for synchronization if they were both playing for the same team. Has anyone ever experimented with removal of the webbing inside the manifold to create a common plenum chamber? What part of the carburetion black magic equation am I missing?

I've heard that runner length can impact the speed of the column of air and thus its momentum into the cylinder, impacting on its filling of the cylinder. Or at least people I know who design for OEM have told me. It's an aerodynamic trade off thing.

but that said I'm with you, i would have thought there is merit in creating a crossover to account for carb imbalance. It would mean that you can't run a different tune on different cylinders though. on the later injected bikes they do run a slightly richer mixture on the rear cylinder for additional cooling. I've played with the tuning on my 907 (myECU) and with my laser thermometer I don't notice a major difference running the rear cylinder 2% richer.

Someone out there must have tried it. what was the outcome?

Just spoke with my brother who used to be a nascar cyinder head guy and now makes custom race exhausts and also happens to be a bit of a mechanical genius. He reminded me that even in a well built engine every cylinder has it's own unique air/fuel requirements. For example, he said, that in a nascar v8 the number 7 cylinder is almost always leaner than the others. considering the very different conditions that the 2 ducati cylinders must operate in because of their different orientation, it's easy to see why they would have different tuning requirements.

Yeah but .... the whole idea of individual manifold runners is that there is no interference from adjacent cylinders. The sort of manifold you speak of is called an interference manifold and usually any 2 ( or more ) barrel carb usually have a primary and secondary barrels to provide adequate velocity at low and high revs respectively , which is its advantage. It would probly be less of an issue on something with an even firing order .... like the old British 360 deg twins . I also think you would have a shit of a job getting the weber to behave without some serious #ucking about

I accept everything you said and certainly wouldn't suggest an interference manifold would provide "performance" benefits... but it occurred to me that my injected 907 for which i can customise the tune for each cylinder runs nominally the same tune. Of course the injection and spark is timed for each cylinder and with the Ducati not having an even firing order then looking at a pulse view you would think that the manifold would potentially go from an overly lean to a rich mixture with each engine cycle (if, as in injection, the amount of fuel is fixed by the computer) but in a carb the vacuum would pull more fuel out of the carb... (and assuming the carb can fuel instantly in response to vaccum)

So why would you do it? Potentially you could use smaller primaries because both carbs are providing capacity for the engine... so if you had air speed issues (such as poor throttle response at low rpm/low accelerator) then the response of two "smaller" carbs (primaries) might give you more consistent response than a single larger carb (in the carb per cylinder model). At WOT you would still be limited by the secondaries can provide (and assuming they are the limiting factor).

As a counter argument though... i've played with intake runner length on old valiants and with everything else being the same (engine, carb) I got better drivability (throttle response) with a long runner intake manifold than i did with the short runner intake manifold.

yep , check out this Mopar cross ram manifold for dedication to the long intake runner

You Just Need to buy another mainfold and cut a part of the diaphram between folds to see what happens.

I doubt 'opening up' the manifold would aid performance in any way. Imagine that this were so, then air speed thru each throttle body would be reduced, leading to fuel mixture problems due to the reduced vacuum.
Incidently, running the rear cylinder richer (cooler) is most likely due to it being heated by the radiator whereas the horizontal cylinder is not.

Yes, I'd agree you would probably end up looking for smaller primaries to better match the operating situation. I was just wondering if that an interference manifold and suitable carbies would address the flat spot issue.