Ed,
Bernie's comments on weird have prompted me to start my investigation on a round radiator, PWR locally has an engineer who is interested in my ideas and we may move to develop a prototype.
It seems the easiest way to manufacture this type of radiator is along the lines of the stacker radiator, used mostly for water to oil and are very efficient for this purpose. I've seen where people are using them for radiators in racing. To optimize the suitability for my application the air flow channels would be modified for reasons of maximum surface area, turbulence etc.
Naturally a round radiator need to be thinner rather than thicker ( unlike the round intercoolers). In my design the air flows from the centre outwardly and taken back toward a butterfly to control the exit air flow.
The inlet would need to be shaped to maximize pressure as would the internals. I can't easily explain everything but I would appreciate some help with sizing fore hp etc. I do have notes on the subject and will have to find them again but I think a rule of thumb is either 3 sq" per hp or 3 cu" per hp and radiator incidence in excess of 55 deg and if the angle is taken from the vertical or horizontal.
Any help is appreciated!
George (down under)