I was curious as to what the side 'rails' are

that are located on each side of the inlet ramp to the duct.

 

Those are boundary layer deflectors.  The original NACA papers on these scoops talk about improving NACA ducts by having those deflectors.  The boundary layer is the real problem with submerged inlets, especially toward the aft end of the fuselage where the boundary layer is fully developed.  The boundary layer is, as you probably know, slower moving air near the surface where it has been slowed down by friction with the surface.

 

 I think one of the major problems w/ the Naca

ducts used in this application is that they should still be placed in an

area of high pressure. I my humble opinion, right behind the thickest part

of the fuse isn't really a high pressure location. However having said

that..  I'm not sure where else he could have put them unless he had wanted

to do alot of ducting. I think that's why he put those little VG's where

they are. It would appear that he was having problems w/ airflow velocity

[boundary layer thickness] and by adding the vg's he re-energized or speed

up/thinned down the boundary layer and improved the airflow into his ducts.

 

Experiments is recent years have found that strategic placement of VGs in front of a NACA scoops make significant improvement.  By setting up two vortices (spirals in the air flow) going in different directions (notice the two VGs are placed at opposite angles), you can separate the boundary layer and spiral some of the faster moving air into the duct.

 

Placing NACA scoops toward the front of the fuselage generally pits them where the boundary layer is thinner and the pressure is higher.  It depends a lot on the fuselage shape, but is generally good surface pressure recovery toward the aft end of the fuselage (or aft edge of the wing), so a NACA scoop can work reasonably well toward the back of the fuselage where the boundary layer flow will be forced into the scoop by a pressure differential between the scoop inlet and wherever the air is making its exit, and/or if you can do something about the boundary layer, like using the VGs.

 

Of course you will never get a submerged scoop to give you the efficiency of a ram air scoop, but you might get up to about 80%.

 

Al