I've been working on the aircooled jugs some more, and hit upon a novel solution to the lubrication issue I was running into. With the fork-in-blade design, lubricating the piston was turning into a problem. Difficult to align the lubrication holes in the crankshaft, and the casting for the shaft was more difficult due to the lubrication channels.
Then a bit of serendipity struck, and I'd found my old Suzuki GT380 manual (long lost the bike). The GT380 was an interesting engine in that instead of feeding lubricant through the crankshaft, it used direct oil injectors inside the engine cylinders themselves. I could accomplish this because the pistons were just a bit longer than the stroke of the engine. So, I looked at the pistons I'd designed, and if I extended them by 2mm, they'd also fit the same bill. Simplifies the casting of the crankshaft, while retaining strength.
yes, it did, just like the GT250, GT550 and GT750, but those bikes were all 2 strokes. the oil injection went into the crancase to lubricate the pressed-together roller crankshafts as well as the piston skirts.
in a 4-stroke, you still have to run oil thru the crankshaft to lubricate the main and rod journals, as long as you're using plain bearings. If you want to look at 4-strokes with roller cranks, study the japanese 4-cylinder 4-strokes from the 1970s.
roller bearing cranks are fine up to a point, but they aren't as torsionally strong as even cast single piece plain bearing cranks, for a given journal diameter.