Those are neat projects. Taking on a project like a self made piston supercharged bike is commendable.
I thought I would throw out some ideas about the piston supercharged bike although it is apparent you have put a lot of though into this project.
If you have not already done so, you may consider making timing charts of the engine process. These are helpful for almost any process especially ones that involve gas flow. They do not need to be overly technical or even include hard data (like pressures) that are difficult to determine.
The way I would do it is just make a line from left to right on paper that is chronological. Mark on the line each pistons TDC and BDC with sufficient distance between these events to add more stuff. The two pistons event's are placed in the correct order on the same line but labeled power piston (PP) and compressor piston (CP). Or you could give each cylinder its own line one above the other. Then on the line fill in all the valve open and closing events including your guesstimates for the reed valves. For a 90 degree V twin with both rods on the same throw, the two cylinders are 90 degrees in crank rotation apart in their cycles. The attached image is an example for a simple single cylinder 4 stroke. http://www.tuks.nl/WFCProject/Les_Banki_Project/img/4_stroke_engine_timing_cycle.png
Below this line I would try to make another time line that guesstimated each cylinders pressure as it goes thru the four strokes. This chart may give you an indication of the actual pressure of the power piston chamber when the intake valve closes but before the piston comes up and further compasses the gas. Obviously you want it above atmospheric. You of coarse have the excellent advantage of the power piston is operating as a 4 stroke and the compressor piston is acting as a 2 stroke so the whole process shows significant pressure benefits.
One potential trap I see in a process like this engine is generating higher pressures in the compressor cylinder, that do not get transferred to the power piston. This could use energy to compress, that is wasted. Also the volume of any transfer runners must be considered. The overtemp problem got me thinking about this.
Due to your familiarity with your creation this exercise may be unneeded because you already have your mind wrapped around the process. For me visual representations almost always help.
I appreciate the interest.
I think I see what you are saying... but the limitations of the current equipment make that largely a guessing game:
The 2 valve Ducati cams have a relatively large amount of overlap... packaging is currently the problem on the exhaust length... I think it is right, but tuned length should mean something. Very different now as compared to 2014 when all the pictures are from.
The compressor "plenum" has a variable diameter down through the intake valve.
The CP isn't really compressing in the piston, it is compressing in the plenum. This doesn't seem to matter after the first few seconds of running.
Even with a terrible CP head and a broken "exhaust" reed valve, it made a consistent 25+ psi and ran 84 mph on pump gas. I'm hoping "better" design of the air path in the CP head will alleviate some of the temperature-generating effects... plus a switch to methanol for fuel to help cool the charge and add that alky smell to the ambiance.
Design changes:
CP air path was a Y shape. "Lots" of unswept volume.
CP air path was terribly unsmooth. In fact so unsmooth as to not qualify as smooth compared to the surface of the moon. Hard to access with any kind of tool to make smoother.
Throttle plate after CP broke the "exhaust" reed valve.
New air path "insists" that air enter the swept volume. "Significantly" (unmeasured) less unswept volume.
New air path is "smooth" and accessible to make smoother.
Move the throttle plate forward of the intake for the CP.
Theoretically I am doing this all wrong anyway... I should have just blanked off a cylinder and run a turbo.