I guess the relative failure of Ducati's 'frameless' MotoGP bike, combined with the imminent release of a flagship road bike with a conceptually similar chassis design, has caused people to consider the implications of the motorcycle engine as a frame component.
A good post on the Superbikeplanet.com blog quotes one of Rossi's mechanics saying that if you take the engine out of most MotoGP bikes, you can still push it around the paddock, while if you take the engine out of a contemporary Ducati MotoGP bike, you're left with a fork, swing arm, and seat lying separate on the ground.
IE, the Ducati motor is a fully stressed frame member.
That's not to say that in all other contemporary racing designs (and all modern street bikes) the motor isn't part of the frame; it is. It's just that it's semi-stressed. IE, it contributes to the stiffness of the overall frame but does not completely replace any frame member.
This is an idea with a long history in motorcycling, stretching from the Vincent Black Shadow to the Britten among other innovative designs (as I've noted in earlier posts.) But it occurs to me that given the relatively conventional suspension design of the Ducati MotoGP bikes - they have the same telescopic front fork as other entries, and a similar rear swingarm - it may be that the fully stressed engine design has inherent problems that a semi stressed design would not have.
I note the conventional suspension components because one big 'issue' with modern sport bike design is that at maximum lean the conventional fork and swing arm arrangement essentially stops working as a way of keeping the tire in contact with the track over bumps. Obviously, as lean angles get well over 50 degrees, the suspension travel is sideways, not up-and-down.
The idea of ultra-stiff-and-light carbon chassis components and the idea of fully-stressed engine-transmission packages are both well proven in race car design. But because race cars ride essentially level on the track, ultimate stiffness is the goal. All the bump compliance in the suspension system comes from the suspension components. Motorcycles, because they lean into corners, require a completely different approach.
Motorcycle chassis designers have, thus far, sought bump compliance by tuning the stiffness of the total package, so that the total package flexes to maintain traction.
Even in a more conventional twin-spar frame design, making minor tweaks to flexibility is a challenge. But the beauty of a semi-stressed design is that while you have two expensive, complex-to-engineer components -- the frame and the motor -- contributing stiffness to the backbone of the chassis, you can tweak chassis stiffness without changing either of them. Rather, you can adjust stiffness by changing the way those two components are attached to each other.
I've talked to engineers developing modern sport bikes, who've told me they made just such tweaks by doing things as subtle as changing the bolts used to attach the motor to the frame.
Some people have speculated that the Ducati MotoGP project is having problems because of the material (carbon fiber) used to make the subframe/airbox that attaches the steering head to the motor. That's an area of the chassis that would typically be made of aluminum alloy in other bikes. But Nicky Hayden mentioned that he didn't think it was the material, per se, that was the issue. He said that he'd tested swingarms made of both aluminum and carbon fiber and found that both types could be made stiffer or more flexible.
The shape and structure of a typical swingarm conspire to make it a little easier to have different stiffness in the plane of suspension travel than at right angles to the suspension travel. IE, it's a little easier to have some lateral flex at maximum lean angle in the swing arm than it is in the front fork.
The shape of front fork components -- the tubes are pretty much radially symmetrical -- makes that differential stiffness much harder to achieve at the front. The front fork has to be stiff enough fore-and-aft to prevent the wheel rubbing on the radiator under braking (and to allow fork travel under heavy braking, too.) That means it's also stiff laterally, and max-lean bump compliance at the front end of the bike comes from tuned flexibility in the frame. Note again, as Rossi's mechanic pointed out on the Soup blog, the motor is the frame. The motor has to be as stiff as possible, to handle the enormous forces generated by the pistons changing direction hundreds of times per second. There's no taking a bit of stiffness out of that structure!
And indeed, it's been at the front, not the back, that the Ducati's lacked feeling and eaten tires. It's perhaps slightly less of a problem for Hayden than Rossi, and was less of a problem for Stoner last year, because both of those wild colonial boys are relatively more comfortable riding the rear tire than Rossi is.
Burgess, before the season, was certain he (and Rossi) would be able to sort the Ducati's handling problems out in the first few races, but handling and tire wear issues have bedeviled them all season. They seem to make a bit of progress at some races (like the recent one at Motegi) but I think those are the races where new paving reduces the problem of bump compliance at maximum lean. And since the series uses control tires, there's no chance to look for a little compliance in the one other spot they could seek it out -- in the tire carcass.
The first partially aluminum solution that Ducati delivered to Rossi was clearly (based on the handful of pics I saw) just an aluminum component added to the basic carbon fiber front subframe/airbox. I presume they reduced stiffness in the steering head area and then put some back in by bolting on that alloy component. That would allow them to seek some torsional flex up there by changing the alloy bit, or even the way it was attached to the air box, without re-engineering they air box itself.
I guess the next step is a full twin-spar solution we'll see at some future race. My guess is that if they choose to go that route, they'll find that the easiest way to adjust the stiffness of the overall package is not by making new frames (or new engine castings, which are a huge problem under current 'six-motor' rules anyway) but by tuning the connection between the two.