284 



TEMPERATURE AND HUMIDITY 



objects by the physical transfer of the Hquid 

 or gas with which the objects are in contact. 

 This type of heat transfer depends upon the 

 actual streaming movement of warm mole- 

 cules from the warm object to the cooler 

 one. The transport of heat by such 

 streams (convection) is to be distinguished 

 from heat conduction through a gas or 

 hquid, which is the type of heat transfer in 

 which there is no streaming and in which the 

 molecules of the medium remain essentially 

 in the original locations, passing the heat 

 energy from one to another by molecular 

 vibration and collisions. The heating of a 

 building by hot air and the cooling of a 

 surface with an electric fan are examples 

 of forced convection, whereas the streams of 

 air rising about a warm surface, the human 

 body for example, are called natural con- 

 vection. Natural and forced convection can 

 contribute to heat transfer at the same time, 

 although it is likely that natural convection 

 will become decreased as the forced convec- 

 tion is increased. 



In this discussion we shall confine our 

 description to forced convection, since under 

 the conditions which normally affect human 

 heat loss there is usually some degree of 

 forced air movement. This treatment is 

 also recommended by the fact that in experi- 

 mental studies on humans it is most feasible 

 to determine by difference methods an 

 empirical convection constant as a function 

 of measured air movement. This constant 

 yields the total convection loss when multi- 

 phed by the temperature difference between 

 unit area of a subject and the ambient air. 



Forced Convection 



An analytical consideration of the basic 

 concepts concerning forced convection will 

 naturally lead to a statement of the factors 

 upon which this heat-loss channel quantita- 

 tively depends. It is obvious that the 

 greater the velocity of the air stream, the 

 greater the difference in temperature 

 between the gas and warm surface, etc., 

 the greater will be the convection. Stating 



this mathematically, we can say that the 

 convective heat, He, is a function of these 

 several variables. That is, 



(9) Ho = f{D, V, n, p, AT, K, C„ t), 



where He = heat loss by convection, / = the 

 functional relationship which is not known 

 a priori, but can be determined from experi- 

 ment, D = the characteristic dimension of 

 the object (for example, the diameter of a 

 sphere or a cylinder), V = velocity of the 

 gas, n = viscosity — a factor concerned in the 

 mobility of the gas molecule, p = density, 

 K = thermal conductivity, Cp = specific 

 heat, T = temperature difference between 

 the warm surface and the air, AT" = Ts — 

 Ta, and t = time. 



How these values can be combined to 

 make a sensible formulation of convection is 

 largely a matter of experiment, but it is also 

 a matter of clever guessing. The final 

 arrangement which best fits both the experi- 

 mental facts and theoretical considerations 



is: 



(10) 



go-r^v 



,+^[--jr)yT,, 



(DVp" 



where a and b are constants depending upon 

 the particular units used. It is convenient 

 to reduce all surfaces to equivalent cylinders 

 or spheres, smce most of the experimental 

 work has been done by engineers who are 

 interested in convection losses from pipes. 

 Neglecting everything but convection, the 

 adult human body loses heat like a cylinder 

 7 cm. (c. 3 inches) in diameter (a = .407, 

 h = .00123, if velocity is expressed in miles 

 per hour, diameter in inches, and convective 

 heat loss in kg. cal./mVhr./°F) or like a 

 sphere 15 cm. in diameter (33), and the 

 study of convection losses from the formulae 

 developed by the engineers can be most 

 easily made by considering the body as a 

 3-inch cyhnder or a 6-inch sphere. 



- is called Nusselt's Number {NU) by 



H 



