Hi Al,

 

Could you say a few words about how you determined what the optimum flow

requirements for aircraft use shoud be.  If it's in the archives, just

a pointer to the message(s) would be enough

 

Bob W.

 

Trying to put me on the spot, ehJ.

 

I don’t know how to respond; it would take much more time than I have right now.  Much of the information has been covered, but to pull it together from the archives would be very tough as it has been scattered through here and ACRE over the last 4 years.

 

In general, the design is driven primarily by factors outside the engine.  The design of the cooling jacket and the engine internals is what it is; Mazda has done that, and we can’t change it.  All we know is we don’t want the coolant to boil in there, and we know that there are potential hot spots, particularly around the plugs that can be the limiting factor.  We also know that heat load is much higher on the plug side of the engine than on the other, which can lead to some thermal expansion issues if the temperature drop across the engine gets too high; how high is to high, I don’t know.

 

Designing for an aircraft has some different priorities than for a car, mostly related to drag, weight and configuration.  I first looked at radiator design to get an idea of configuration for minimum cooling drag and weight. These factors are coupled with configuration issues.  I also had to pick a design point, which I chose as 200 hp climb from sea level on a 90F day.  This is for my Velocity with a 20B.

 

A few years back Fred Moreno posted results of a whole series charts from detailed software analysis on rad performance.  I studied those and whatever other reference seemed worthwhile.  From that I determined approximate best core thickness (2 to 2 ½”), and temp drop across the radiators (20 to 30F).  Obviously there are judgments and tradeoffs involved- your results may vary.

 

I also studied the airflow side, looking at fin spacing, air side pressure drop etc., to determine roughly what core configuration and airflow velocity through the core made most sense.  Putting this stuff together with the coolant flow needed for 20-30F drop allowed design of the custom radiators.  Then there was design of scoops to get the needed pressure recovery and air velocity.

 

The builders of the rads also provided data on the flow and heat transfer characteristics.  When I had the rads and scoops I did flow tests to measure and confirm pressure drops.  Similar design and analysis and testing for the oil cooler. 

 

The dyno tests should me that the flow was lower than needed for 20F drop, but adequate to get 30F; without a thermostat.  With a thermostat (in open condition) the flow was considerably lower.

 

Overkill? Maybe.  I did it because I could; and I wanted to maximize the probability of going flying and having adequate (or better) cooling.  Did I get it right? You design and you test.  Some of the things I can’t test until I fly; so it remains to be seen.  I just know from my years of engineering analysis in industry that it can be a very powerful tool and save a lot of trial and error. 

 

Al