Errors in Thermal Measurements

Abstract

IN some recent investigations on adsorption¹ it seems to have been established that erroneous conclusions have been reached on account of errors unsuspected by the experimenters and arising from the time lag of thermometric apparatus. Workers with experience in thermal measurements are well aware of the possibility of such errors, and in particular the extreme care which must be exercised in the use of platinum resistance thermometers, more particularly when these are wound on insulating supports and enclosed in tubes. The application of such thermometers requires not only extreme care in avoiding thermometric forces in the circuit, but also in making certain that the very appreciable thermal lag is eliminated. Generally speaking, such thermometers are quite unsuitable for use when fairly rapid changes of temperature are involved, and in any case are best avoided, except by experienced workers. For practically all purposes, a properly chosen mercury thermometer is much more trustworthy.