Emissivity of Aluminium (Engines)

[Dafydd Llewellyn]

Yes you would be better off with everything matt black, but not by that much. And you could do more harm than good if the paint acted even slightly as an insulation layer.In a Jabiru engine at 170 degrees C, if you were to paint it black, there would be a gain in radiated energy from about 200 watts (at emissivity = 0.2) to about 1000 watts (at emissivity = 1) . Now at 12kg/hour avgas usage, there is 147,000 watts of energy input. Some of this energy goes out as work and some out the exhaust pipe as hot gas, but a lot goes into heating the engine, which is mainly cooled by transfer of heat from the metal surface to the flowing air at the junction between the air and the cooling fins. The cooling gain from the black paint would only be 1 or 2 percent of this airflow cooling effect, so if you want to improve your cooling, the airflow past the fins is the place to look at.

I wrote an article about this for the magazine last year asking the editor how it could be improved for publication, but didn't get a reply. It was obviously too technical and therefore boring. You get better technical stuff here anyway.

 

regards, Bruce

Thank goodness for some technical nous at last. Bruce is right; conduction from the metal surface to the passing air is about 25 times as effective as radiation - and the shape of a cowled engine is hardly suitable for efficient radiation. The rough rule of thumb, is 26 square inches of fin surface area per horsepower - but that assumes unobstructed flow through the fins, whereas there are things like cylinder head bolts (on, for example, VW derivatives) that muck up the airflow. Also, the air may be cool when it first hits the fins, but it heats up on its way through, so on the downstream side of the cylinder, not only is the airflow difficult to direct to stay in the fin slots, but the air is a lot hotter - so you get uneven cooling around the cylinder periphery. Good baffling is essential. The conduction of heat from metal to air is very complex, it depends on the boundary layer, the conductivity of the metal surface (oxide on the surface doesn't help), and it's driven by the temperature difference between the metal and the air - and overall, it's a difficult design issue. Water cooling isn't without its problems; the radiator (wrongly named - it hardly loses any heat by radiation) is also a metal surface trying to get rid of its heat into the air - but at about half the temperature difference of an air-cooled cylinder head. In a liquid-cooled engine, there are three fluid/metal interfaces - the first between the water and the cylinder head metal, the second between the water and the inside of the radiator tubes, and the third, between the outside of the tubes and the air. Essentially, a liquid-cooling system exports its problems from the engine to the radiator; however testing back in the 1930s showed there was almost no difference between the exhaust valve head temperature of a liquid-cooled engine and that of an air-cooled engine. It's the temperature of the exhaust valve head and the tips of the spark plug electrodes that usually trigger pre-ignition and hence detonation, so this is a fundamental barrier to better engine efficiency.

 

 

[Guest Maj Millard]

Water-cooling is 30 times more efficient then air cooling.........................

 

 

[facthunter]

I've never heard it expressed in that way, but certainly when heat treating metal it could easily show that sort of result. Liquid cooling has a few downsides including Water is vaporised at below the optimum temperature, It is a more complex system which makes it less reliable (as a system). The castings have to be of a higher quality as any porosity is a problem and you have to regulate the heat more closely to avoid freezing and boiling. At higher altitudes the boiling point drops considerably. Water cooling is just using the liquid as a means of transferring the heat from the engine to a radiator which is then cooled by air. ( as Dafydd said) A marine installation without a heat exchanger would be water cooled.

 

Above a certain specific HP/Litre of engine size it would be difficult to cool by air flow around the engine directly. Probably about 60 HP/ Litre would see it out. without getting exotic. Nev