8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 76 



are still run under the realm of guess-work. Even some of the larger plants 

 that are well equipped with various instruments to determine the efficiency 

 of their plants and their unit cost of steam are operated with such an inferior 

 class of help and with no intelligent supervision that these advantages are 

 entirely lost. 



Page 7 gives a graphic representation of the combustion reactions 

 appHed to industrial furnace practice. The coal fired on the grate 

 plus air in the form of draft yields carbon dioxide in the form of 

 waste gas plus or minus free air or free carbon monoxide gas, depend- 

 ing upon how the ideal reaction ( i ) has been approached. It will be 

 noticed that the hot gases from combustion are forced to take a round- 

 about course on their way to the stack by the introduction of a series 

 of baffles. This is to give the maximum opportunity for the boiler 

 tubes to absorb the heat. In spite of this, the gases are still hot as 

 they enter the chimney which means that they carry away a very 

 appreciable amount of heat. Too much air fed in as draft results in 

 an unnecessary volume of gases and a correspondingly preventable 

 loss of heat. Insufficient air affords insufficient oxygen, resulting 

 in incomplete combustion forming carbon monoxide instead of 

 dioxide and the carrying off of two-thirds of the heat units as lost 

 potentiality. 



Two general lines of procedure suggest themselves for subjecting 

 the industrial furnace to chemical control, namely, measuring the 

 raw materials, coal and air, and measuring the results. The first 

 named is impractical, if for no other reason than that of the ever 

 varying composition of the coal, as will be discussed more in detail 

 later. The second line of procedure, that of measuring the results, 

 may seem, offhand, open to the objection of locking the door, as it 

 were, after the horse is stolen. Given an indication of satisfactory 

 versus unsatisfactory results, however, we are in a position to gage 

 our procedure accordingly. This brings to mind the relationship 

 already brought out, to the effect that the best results are, in general, 

 attendant on the highest percentage of carbon dioxide in the flue 

 gases, and points at once to the importance of flue gas analysis as a 

 medium of control. Along with this, the importance of steam flow 

 measurement is self-evident inasmuch as steam is what is most 

 desired. These two together afford a check against one another in 

 the determination of results and the fixing upon the procedure best 

 calculated to bring them about. All that now remains is to interpret 

 these results in terms of the conditions existing in the furnace and 

 provide a means for maintaining those conditions uniformly. Thus 

 with the co-ordination of three sets of records we are in a position to 



