Since you seem to be something of an expert on the subject, would there be a good reason not to build a Savonious Rotor with a horizontal axis? I think making the support structure could be simplified if oriented that way.
I believe you already figured this one out.
Also, what did you mean by exhausting out the center? Most designs I've looked at are open to the opposite side in the center (they overlap). Is that sufficient?
Maybe you can point to some information on an efficient design for one of these?
The thing to understand is that, as the direction of flow is changed by the initial impeller shape, the speed is increased by compression into the center of the alignment. When it is exhausted "out the other side," you are actually vampiring energy from your torque to feed the combination of temperature drop and expansion (V = nRT/P -- to rewrite the
Ideal Gas Law). This statement applies to a
canisterred unit that
also protects you from the C
v(air density)(area of impingement)(air velocity²)/2 (where C
v is the
Velocity Coefficient of Drag -- approximately = 0.5 for a hemispherical face) drag on the "return" side of your rotor. In a
canister variation of a Savonious rotor, the pressure drop between the inlet and exhaust is the driving energy. If you return your pressure to ambient, you are losing 50% of your energy. If you draw it from the (low-pressure/high velocity) center, you gain back a significant portion of those losses. If you use a NACA-7 aerocurve for your
capture geometry you gain (nearly) 15% over a semicircle shape.
Basically, you want to look into aero mass-flow equations and their related geometry in an aeronautical engineering handbook. The problem is that they are written to
protect the high priests of aerodynamics and
not to instruct those seeking to learn today. You need to go back to handbooks & texts written in the 1920's & 1930's to find truly
explanatory texts. The version of
Wing Theory Sections that credit
Bruno as the author (not the "revised" ones crediting Abbott & von Doenhoff) is quite good. They were readily available in used technical book stores (at least) into the mid-1970's.
One of my mentors was Maj Gen, Dr. Joseph Black. One of his statements was that, to really understand a subject, you need to go back before there was a
priesthood established to keep the
hoi polloi from intruding. This is one of the best pieces of advice it has ever been my pleasure to receive.