Or have Al do it on his dyno.  He can load his engine at operating loads, disconnect his EDWP and let the EWP (with flow meter in the circuit) do all the cooling.  With just-less-than-adequate cooling, the engine will gradually heat up to max temp.  Just monitor the flow at each temperature as this happens.  If the viscosity drops off as much as is being postulated, there will be lots of evidence as the flow increases a lot with temperature.  Having to make its way around the stationary water pump and other flow restrictions will generate unnaturally low flow rates and not be a particularly valid data point for climb performance, but the issue is how much the flow CHANGES, not how much it IS. 
Might be interesting to see how fast the water pump (with no pulleys, belts, etc.) rotates trying to get in harmony with the flow from EWP, but really just a curiosity.

At worst, it's better data than testing EWP in series with an OPERATING EDWP ... Jim S.

Al Gietzen wrote:

After all this talk of viscosity changes with temp, I plan to put the flowmeter (not arriving until next week) in the cooling loop on the engine.  I'll run the Davies Craig EWP and record the flow rate, then run the engine enough to bring the water up to normal temp, shut the engine off, and check the flow rate again.  I just can't believe this is going to be significantly greater than when cold, but I'll certainly be happy if it is, especially if it will shut Al up :-) 

 

On the contrary; I’ll be cheering.  I think it’s great that you are willing to make the effort to some data.  We always benefit from good data.

 

Al