I took out the first splitter/duct that was shown in my 5/27 posting photo 1678.  Cooling was than back to slightly marginal not critical!

I than built a rig with 3 turning vanes to direct air into the radiator near the front of the wedge. (See attached photos)  I achieved about a 16 deg. F drop in oil and water!  This told me that most of the inlet air was going through the radiator at the small end of the wedge.  Bobby Hughes said he had seen data indicating a wedge duct had the highest pressure near the end of the wedge and lower pressure near the opening.  I now am considering modifying the lower cowl - inlet duct bottom -to include a curved ramp to direct air upward over the first 30 - 40 % of the radiator and than transition into a wedge the rest of the way back to the trailing edge of the radiator.

I have not found any information I feel comfortable to use to design the curved ramp and may try to build a wind tunnel in the cowl on the plane. I should be able to insert several ramps and wedge shapes in the test area and measure air pressure on the back side of the radiator using a sensitive manometer.  I have two electric leaf blowers that may provide enough air to run it.  I'm planning on a 3 in wide x 4 in high inlet size.   Any comments or suggestions?

Radiator to Duct Sealing:
My radiators are sealed to the duct with silicone baffle rubber strips used on cowl-to-baffles on spam cans.  I believe my seals are very good.   But - after several E mails concerning sealing the duct I looked more closely at the total air flow sealing package.  On the Griffin radiator the fins between the tubes do not extend to the water tanks.  There is gap of 1/4 to 1/2 inch!!  This is on both ends of the radiator. The area of the holes is  0.375 in X 12.5 in X 2 ends = 9.375 sq in of holes just in the one radiator!  The Mazda oil cooler also has gaps but they are only about 1/8 inch wide. I also found about 2 more sq. in of other potential leaks I can fill using Leons light testl   Thanks for the info. to make me look harder at this area!   If I get the wind tunnel working I will first test with the holes and no ramp/wedge - fill the holes and retest to determine how much the holes decrease the outlet air pressure.

Air Speed in Duct at Radiator:
The inlet duct is 51 sq. in. (per James Cowls) and expands to about 17.25 X 8.5 inchs or 146 sq . in. at the start or forward end of  the radiator wedge.  This gives a velocity of about 42 mph for the cooling air at the front entrance of the wedge at the radiator at a climb speed of 120 mph.   The duct is a wedge from that point on the the rear of the radiators.