-OCT. 19, 1921 PARSONS AND HARPER: ENGINE RADIATORS 415 



of direct application, without any intervening step, for the nose ra- 

 diator, a mounting which is today probably the most common in 

 American practice. The designer who is interested in such a mount- 

 ing must first secure a measurement or estimate of the air flow to be 

 expected in the core of his radiator, taking into account the size and 

 shape fuselage, engine, propeller, and all the objects which will de- 

 termine this air flow, and then with such data in hand he can apply 

 all the results of this paper. 



The paper is carried one step beyond expressing radiator perform- 

 ance in terms of air flow through the core, and states the results of 

 the measurement in terms of flying speed for a mounting in an un- 

 obstructed position. Comparative performance is stated in terms of 

 figure of merit which is the ratio of the power dissipated as heat to 

 the power required to sustain the weight of the radiator and overcome 

 its head resistance in pushing it through the air. The relative fig- 

 ures of merit of two cores in an unobstructed position do not, of course, 

 furnish an index of value in an obstructed position. The compara- 

 tive behaviors of various classes of core are discussed in some detail 

 for the unobstructed position and it is shown that for this mounting 

 and high speed of flight, the flat plate core construction has a figure of 

 merit sufficiently above other types to deserve attention. Its chief 

 disadvantage is inherent mechanical weakness, a feature of extreme 

 importance in selecting a design for use, and the data presented 

 should afford the designer definite information regarding the sacri- 

 fice in figure of merit which must be balanced against increased stur- 

 diness, when choosing between types of core. 



For flying speeds of 60 to 200 miles an hour' it seems certain that 

 an unobstructed radiator plus a streamline nose to the fuselage offers a 

 figure of merit so much above that of a nose radiator installation as 

 to justify considerable effort toward overcoming the structural difii- 

 culties attendant upon such a mounting. 



The measurements made on a very wide range of variety of cores 

 indicate that vaneS, holes in the air-tube walls, and all devices 

 intended to ncrease turbulence in the air tubes are detrimental 

 to radiators in an unobstructed position, lowering the figure of merit 

 for a given flying speed. The effect is to decrease the air flow more 

 than is compensated by the increased heat-dissipating power at a 

 stated rate of air flow. On the other hand, in obstructed positions, 

 where the air flow is determined more by the radiator surroundings 

 than by its own core construction, the cores with turbulence devices 



