In my plane the largest radiator I could fit under the cowl and not under
 the engine was an MR2 radiator.  Its core size is 26 x 12.  What radiators
 are the high HP p port guys using?  Primarily I would be looking for
 size,material and number of cores.  If you used a remote fill container with
 the radiator cap, did you buy it or make it?

 Aluminum welding isn't easy to learn and even more difficult to have the
 equipment to do it in some places, so an off the shelf version would be
 nice.  However, if you cant get one find a way to make it.  That requires
 dimensions.  Next time you are in the hangar, take a pic of the cooling
 system and send it over.  Notes help, and a measuring tape in the picture is
 worth its weight in gold.  KA The ideal core is out of a Dodge pickup.
 Probably overkill.  With our asymmetric motor mount and motor plate the rad
 core will fit vertically on the plug side of the engine.  The turbo will
 also fit with the asymmetric motor mount.

KA

 I don't recommends using the stock Cont. mount. The motor mount is the last
 thing to be designed.  First one fits all the major components.  Then one
 builds the mount around them.  A good 3D program like Rhino is essential.
 The stock cowl shown will work fine, No room under the engine with a nose
 gear for the rad core.  Several airplanes are flying with this arrangement.
 Nothing new here.

 I still need the dims on the firewall hard points to finish this..  A stock
 Corvette pressure can works well.  http://tinyurl.com/d6p3s9p Paul Lamar I
 will confirm the dims of the mounting points today after the sun comes up.
 The stock Cont mount has a few more bars than some of the other mounts, but

 the only ones that are in the way are the ones for the retractable nose gear
 brackets.  I am holding out hope that I will be able to make the retracts
 work at some point.  Now that I have the radiator mock up in the cowl, the
 tubes aren't in the way much at all.

 I cant get a Dodge truck radiator in the cowling- 14 inches high is about
 the max and be able to close the hood.  The MR2 rad gives me a few inches
 either end, and room above and below tor get my hands around it.

 KA

 All rads are aluminium now a days. Back about 5 years ago everybody had
 cooling problems before we got a handle on it.  It takes about 600 cubic
 inches of rad to cool a 200 HP rotary.  For a 250 HP turbo it takes about
 750.  That is for a 3 inch thick rads.  Knock off 25% for a 1 inch thick
 rad.

 BTW Here is a simple way of looking at cooling drag as it affects rad
 thickness.  Lets say we have a 1 square foot cooling scoop.  If the scoop
 was completely blocked the drag would be full dynamic pressure.  Dynamic
 pressure at 200 MPH is 100 pounds per square foot.  200 MPH is about 300
 feet per second.

 Therefore the HP consumed is 300 x 100 or 30,000 pound foot per second.  550
 pound foot per second is one HP so our blocked scoop is consuming 54 HP.  If
 the scoop was totally open and the air went straight through with no
 restriction the drag and HP consumed would be near zero.  Only a little
 bit from skin friction. Thinner rads have less restriction but more frontal
 area for the same volume.  The job of the diffusers is to increase the area
 of the duct.  If you increase the area of a duct the air velocity is reduced
 but the volume of the air passing through is the same.  Since drag is air
 velocity squared.....By now you can see where I am going with this.

 This is why Kays & London wrote: "the use of highly compact [efficient]
 surface geometries is that the resulting core shapes are characterized by large flow frontal
 area and short flow lengths for the gas-flow
 path".

-------------------------------------------------------------------
Close the NACA scoop. NACA scoops don't work when feeding restricted ducts. The air merely bypasses the opening due to positive pressure in the duct.

Ideally you want the nose scoop half way down the rad. Moving it down
from where it is. Nose scoops are best as they see full dynamic pressure.

Here is a pictures of the MR2 rad mock-up sitting in the cowl.

The rad can be 4 or 5 inches higher. Every little bit helps.

The motor mount is workable but the aluminium plate between the pan and the block will flex causing oil leaks. Making it thicker will help.
The rad comes close but the mm tubes do not significantly block the flow.

I don't think the stock RX7 oil cooler is going to work with a turbo.

Your on the right track with the diffuser for the rad.

Paul Lamar