I can easily see the point of using software to tune the final design vs. trying different mechanical combinations.  that doesn't address air flow routes thru the intake manifold tho, just fixes the individual air/fuel mixtures arriving at each rotor.

 

next wonderment - why are exhaust manifolds typically designed to get larger dia. as more tubes attach downstream (as with the waste plumbing in buildings), when, in reality, the flow is thru only one tube at a time?    kevin (turrentine renesis coming soon!)

In the intake design, At least for normally aspirated engines, there is never more than 14.7 pounds of pressure available (at sea level) to push air into the engine. And that would only be the answer if there were a genuine vacuum in the engine. There never is such a vacuum. Even so there may be places in the RPM band where a slick intake with proper tuned length may see a bit above the ideal tuned length assist in the flow.

And this may produce (typically two) RPM where slightly more than 100% of the engines displacement is ingested. The rats nest of tubing and valves in the modern rotary is evidence of the quest to capture this "tuned effect" over a broader range of RPM.

 

 

Forget that. We have been using 2350 for so long that now most props work fine at that RPM and except for us and a few more car engine nuts that 2350 is not all that bad. Learn all about props and tip speeds and such.

 

In any case the cruise RPM we be in that area and we want to be closer to the largest diameter we can swing for efficiency and disc area.  

 

Wet flow???

Air with fuel droplets in it does not perform like plain air. It tends to stay on the outside of turns. The droplets falling out of suspension and sticking to surfaces, act as turbulence generators and make more flow cling to that surface. Bad mojo. So, perhaps injecting into a straight run closer to the port would be better than before a turn in the tubing.

 

Maintaining a uniform velocity through the inlet tract is better than changing speeds over its length. A change

up or down costs energy and slows total flow. So, turns in the ideal tube would be shaped like a "D", with the flat of the "D" on the inside of the turn. Look at the exhaust ports on a Chevy head. "D" shaped on the bottom. Right there is a few hundred million votes for "D: shaped. Make patterns in graph paper and fit them through the bends. Alter the bends to maintain or increase slightly the cross section as flow moves toward the port. Who knows, you might find that a pair of burs stood up with a three sided chisel just ahead of the floor of that "D" shaped turn might stick flow to it like glue............................

 

Now that you have completed your flow bench from the free plans, you have noticed that all of this goes out the window, if the entrance into the runner tube is not bell shaped. Or God forbid a sharp edged tube stuck through the wall of a box. Read Vena-Contracta.

 

On the exhaust side, there is plenty of pressure available when the port opens. In fact there is so much that the opening port results in a supersonic shock wave leaving the area. Again the finish of the port has a bigger effect than you would expect. A small imperfection at very high velocity makes a big difference. It acts just like an amplifier circuit. Imagine a boat wake in the flow behind the imperfection. Like a poorly fitted manifold or gasket. This supersonic flow is maintained so long as the tubing diameter is maintained less a small amount for surface drag. So blending the two pulses from the two housings is done in a collector and the collector exit is bigger than either primary tube. This slows the shock wave a bit and makes muffling slightly more manageable. In the race car the collector diameter is about 3" at the very end, and then a megaphone expands up to 4". The expanding pulse drops to subsonic, and a less exotic muffler will work. The mufflers however, are very large and heavy and tend to look like a water heater of a large suit case.

 

For airplanes, muffling is just going to be done at the supersonic level for space and weight reasons, and has been seen up to now, a very big problem.

 

Primary lengths in multiples of about 11" to 12". A straight run out of the port as long as possible before the first turn. Both primaries the exact same length, and entering the collector in exactly the same angle.

 

Primary ID the same as, or just slightly larger than the port liner. Use NA housings for less noise.

Spring powered slip joints to account for engine movement and length changes from heating. Make the primaries thick wall stainless and wrap them with heat tape, held on with safety wire. Make everything in stainless. A little time at peak EGT is bad for apex seals and will destroy even thick walled carbon steel tubing.

 

 

Lynn E. Hanover