[wildman10]

I couldn't find anything in those references that affected my write-up.

I dug up some old work on aircraft radiators here.

The optimum cooling air speed for minimum drag depends on the radiator design, and these probably haven't changed much since then; the optimum speeds were those that gave pressure drops across the radiator of 250-750 Pa. I don't know what that translates to as a speed for my intercooler and radiator, so I can do 2 things:
1. Estimate order-of-magnitude optimum speeds, where 1/2RhoV^2 = 250/750 Pa gives V = 12/35 m/s = 26/78 mph when the pressure behind the radiator is atmospheric.
2. Measure the pressure drop across the intercooler and radiator, say by strapping them to a roof rack and measuring the drop at various speeds using my Magnehelic gauges.

With the heroic assumption that pressure behind the radiator and intercooler is atmospheric, for a maximum continuous speed of 100 mph and a maximum velocity at the cooler face of 70 mph, I'm looking at a minimum reduction ratio for the front duct of 100/70 = 1.4; similarly, for a cruising speed of 70 mph and the middle point of the 26-78 mph range, 52 mph, I'm looking at a cruising reduction ratio of 70/52 = 1.4. The duct length and shape should be chosen to give this ratio.

Of course, drag may not be as important on our cars as on aircraft and the pressure behind the coolers needs to be atmospheric for this 1.4 ratio to hold, and this needs careful assessment and design of the exhaust route. It's too late in the day to think this through, so that will do for now.