Abstract :
IN most engineering measurements the readings obtained at a given time depend upon a number of conditions, such as temperature and barometric pressure, existing at the time of the test. In some eases the effects are very small, compared to the accuracy desired in the result, and can be neglected. In other cases prevailing conditions have a decided effect on the numerical values of the test results. For instance, in a determination of the boiling point of water, it is very necessary to consider the barometric pressure. In measuring the flashover voltage of an insulator, the temperature, pressure, and humidity each has a measurable effect on the observed value. Therefore, it has become necessary to select standard conditions to which certain measurements may be referred.
