162 Prof. Barnard on the Elastic Force of Heated Air. 
been to heat the air with any power of furnace, since radiant heat 
produces no sensible impression on it, since there is little time for 
the process of convection, and since but a limited portion of it 
can come in contact with the surface of the heating chambers, it 
would be advantageous to fill the fire boxes above the fuel with 
meshes of wire similar to those of Ericsson’s regenerators. These 
to the steam engine. : 
These arrangements not only afford a good security against 
waste of heat through chimneys, and by radiation, but they put 
the fire most completely under the command of the engineer, 
and they secure what has not yet been sécured, the certainty that 
the air will be heated to the temperature desired. 
There remains no serious objection to be considered except that 
which arises from the resistance of the supply cylinders. This 
difficulty is entirely of a mechanical nature, and it is to be sut- 
mounted by mechanical expedients. There is abundant force, 
but this force happens to be greatly in excess at one time, and in 
deficiency at another. I have already pointed out, in a former 
number of this Journal, in what manner the difficulty may be ob- 
viated in engines designed, like Ericsson’s, to carry not more than 
two atmospheres of pressure. That method will answer, witha 
supply cylinder of. three-fourths the capacity of the working cyl- 
inder, to an extent of more than four atmospheres, preserving 4 
predominance of positive power at every point of the revolution. 
By attaching three piston rods to the same crank, the cylinders 
being placed at angles of 60° from each other, the method will 
answer for still higher pressures. But that other and still more 
data it appears that the oxygen contained in 13 cubic feet, or one pound of air will, 
t 
by the combustion of a due amount of carbon, raise the temperature of one pound 
of air 4152 A refere summaries presented further on in the text will 
show that t gements theoretically the most-eligible, do not require a larger 
os le som 200°, 
and while those which are likely to be most convenient in practice will make neces- 
sary only about 250°. One pound of air will therefore furnish oxygen enough to 
heat about fourteen , or more than 180 cubic feet of air through the largest re- 
ired range. It is consequently quite safe tosay that one pound of fresh air intro- 
uced at each stroke would heat at least ten pounds to the necessary extent, besides 
heating itself and making good all waste. If three cylinders, therefore; work in con- 
nection with one crank cylinder capable o ting ten pound ir of the or- 
dinary temperature and density, they will hold under the pressure of five atm 
pheres, fifty pounds each, same tem e. If two of these cylinders be 
supplied from a reservoir into which air has been compressed to this ee, 
one, working against the atmosphere alone, w oxygen enough to keep uP 
combustion, If compression in the reservoir is less, the supply of oxyge® 
will be proportionably in excess. pply of oxys 
