I like the idea about two rotary valves stacked.

But instead of rotating they could just oscillate back and forward controlled by a couple of fast and precise step/servo motors like the x and y mirror on an industrial engraving laser. The motors used to move the beam is very fast and accurate.

Then you would have fully programmable intel duration controlled by a combined ECU also controlling, ignition, exhaust valve, power jet and step motor for variable intake lenght.

Will start on a simple sketch of the layout.

i would of thought 2 X valve covers that could be moved separtely,

but how to overcome the force needed to move the covers quickly against the resistance of oilseals or o-rings (or what ever to seal against air leaks)

After some more thinking one of the rotary discs could be removed.

One disc controlled by servo motor would be enough. It would still provide a good flow until fully opened and upon fully opened the flow is perfect.

Of course some calculations on torque requirements is needed. But I'm quite sure the motor speed is available based on the current design of galvano mirrors on lasers. But how much torque is need to overcome the friction caused by the suction and pressure on the disc is a good question.

@Frits -> Perhaps skip the variable intake length in the first prototype to simplify the design

A simple idea how to measure the torque need to move the rotary disc. Don't know if it's good or bad.

Take a tandem twin, remove ignition on one cylinder and make a dyno run. Then remove the disc on the cylinder not firing and make a new run. Then the difference in the runs would be an indicator of the torque needed.

Is it needed to run the disc for overreving under braking?

The motor isn't pulling anyway so why not just leave the inlet constant open until the engine speed drops down to 0.8*peak power rpm.

For RSA this would mean that the disc is only needed until 10.400 rpm. Above that constant open.

The reason for I considered the rotary valve solution for you 24/7 system was because of the perfect flow.

You could have any hole shape and with a perfect smooth inlet.

I reckon that my idea is more complicated than a simple reed based solution.

Just tried to come with a different solution

The "swing" motor for the reedcase would also have to react very fast in order not to disturb the flow with a half open reedcase during change from reed to 24/7. If it takes to long the flow rate would drop along with the power during the change over.

And when I retarded inlet closure 2 or 3 degrees my engine lost a lot of power at 10.500 revs!
This might work on an aircraft engine with constant rpm.
Or with a flow diode!
A small diameter and long trackt would destroy power.

I gained above 12500 rpm