Update on cowl flaps.  I closed off the exit area about 20% by changing the cowl flap exhaust ramp shape (added some curved sheet metal).  Temperatures at 65% cruise 50F LOP rose modestly to above 300F during flight conditions that were 9F above standard at 8500F.  Speed up about four knots.  Looks like I can reduce exit area even further!  This would be consistent with the observations of big radials with closed cowl flaps operating in economy cruise.

Full power climb with best power mixture (worst heating conditions), cowl flaps open at Vy  produced maximum temperatures of about 340F starting with 65F ambient at sea level take off.  I would expect about 380-390F maximum with a 100F sea level departure.  Further cooling can be obtained by accelerating from Vy (135 knots) to 160-170 IAS.  Cooling with cowl flaps open is thus more than adequate even with abusive conditions.

Larger inlets are GOOD (assuming you use cowl flaps) because they permit pressure recovery in front of the inlet as the flow slows and then spreads around the inlet.  Small inlets require flow deceleration inside the inlet which is much harder to do without flow separations.  The rule of thumb that seems to work and is fairly widely used is to figure your cruise air flow requirements (actual cubic feet per minute) and size the inlets so that the velocity at the throat of the inlet is about 50% of the free stream velocity.   This gets about 75%  of the available pressure recovery in front of the inlet, and with some good internal design should get up to 90+% total pressure recovery.  This gives plenty of surplus pressure to accelerate the hot flow out the cowl flap nozzles and recover  most of the lost momentum thus reducing cooling drag.

Fred Moreno