Turbulent flow in the radiator actually increases the rate of heat transfer! Heat exchanger designers actually try to achieve turbulance in the coolant flow as it passes thru the tubes. Turbulance is a function of the tube design and shape.

The only reason to slow down the speed of the water pump is to reduce cavitation....period! It's a band-aid.

NASCAR guys run some very special water pumps with custom impeller designs that allow very high flow without cavitating. From the discussion on the open wheel site by some very savy racecar engineering types:

Quote:
Q = M x C x Delta T

Where Q = rate of heat transfer
M = mass flow rate
C = Constant (specific heat of water) 4.186 joule/gram °C
Delta T = difference between temp out/temp in.

So, increasing flow rate increases rate of heat transfer.

Quote:
That slower flow gives better heat transfer IS an old wives tale. Faster flow always gives better heat transfer, and increased flow also keeps coolant temperature more uniform throughout the system, helping keep the head cool and improving heat transfer at the cool ends of the radiators and the exit of the engine (the front of the head). The old wives tale about slower flow originates from poorly designed flow paths and hot spots inside the engine that exacerbated localized boiling (cavitation), and slowing the flow (while also increasing coolant pressure in the engine) with a restrictor reduced overheating due to reducing that localized boiling. Thus, it was thought (by hands-on, non-engineering types) that slower flow was good. It was actually the increased pressure in the head caused by the exit restriction that was good, NOT the reduced flow (except for the minor effect of reduced cavitation in the head as a result of the slower flow).

Quote:
From the equation above, you CAN'T make the flow 'too fast', you always get better heat transfer with higher flow.

Eventually you would need a pump so big that the engine wouldn't be able to turn it, or the heat generated by the pump would be more than the engine produces, which would skew results somewhat, but I think we can leave those extremes out of this discussion.

Quote:
That is, of course, correct. The coolant flow can NEVER be too fast for heat transfer. The "fly in the ointment" is the engine internal cavitation/local-boiling problem, and as you said, the HP loss from the pump.

This matches my real world observation of heat exchanger designs for the hydraulic industry. More flow will always result in better heat transfer and more cooling.