abstracts: technology 77 



The results are given of an extended series of investigations of the 

 various factors that may affect the accuracy of heating value determina- 

 tions with flow calorimeters, such as: Completeness of combustion; 

 accuracy of the temperature measurements; the magnitudes of the vari- 

 ous heat losses from the calorimeter, particularly as affected by the vol- 

 ume of the entering air and of the products of combustion and by the 

 atmospheric humidity; and the accuracy of measurements of the quanti- 

 ties of water and of gas. The effects of certain other factors, which 

 depend on the nature of the gas tested and are particularly noticeable 

 in the testing of illuminating gas, are briefly considered. 



The results are given of an extended series of experiments with natural 

 gas and with nearly pure hydrogen which show that, when the necessary 

 corrections for losses of heat from the surface, for the effect of atmos- 

 pheric humidity, etc., were applied to the observed heating values 

 obtained with flow calorimeters of suitable design and construction, the 

 total heating values thus found were in agreement to about 0.3 per cent 

 with the total heating values obtained with calorimeters of theBerthelot 

 bomb type, provided due allowance was made for the difference between 

 the heat of combustion at constant pressure and the heat of combustion 

 at constant volume. The results obtained with illuminating gas indi- 

 cated that the heating values obtained with the Berthelot bomb type of 

 calorimeter were probably in error, being too low (about 1 per cent), 

 but further investigation will be required to determine the cause of the 

 discrepancy found. 



The summarized results are given of a critical investigation of eight 

 flow calorimeters, representing types widely used in this country or 

 abroad, and of one calorimeter of the comparison type. 



The investigation has shown that several of the calorimeters that were 

 investigated could be used, when proper precautions were observed and 

 suitable corrections were applied to the observed heating values obtained 

 with them, to determine the total heating values of most kinds of gases 

 to an accuracy of about 0.3 per cent. Some of the calorimeters are, 

 however, subject to constant errors that cannot be eliminated or be read- 

 ily determined except by comparison with a standard instrument, such 

 errors for one calorimeter amounting to nearly 2 per cent. 



Examination of a very large amount of data obtained during the in- 

 vestigation indicates that with those calorimeters free from significant 

 constant errors, total heating values should be determinable, under 

 conditions obtaining in the practical testing of gas, to an accuracy of* 

 the order of 1 per cent. C. W. W. and E, F. M. 



