Abstract
The last fifteen years have seen the introduction of several new types of high-pressure sodium lamp designed, over a range of power ratings, to give either very high efficacies or better colour characteristics. In order to design any of these lamps it is necessary to specify eleven variables. This paper shows how for one type of lamp (the standard high-pressure sodium lamp) a multiple regression procedure has been used to derive equations from experimentally observed relationships between efficacy, power, voltage, bore, arc tube length, arc tube wall thickness and sodium mole fraction. The equations have been brought together in a computer program so that new lamp designs can be obtained by keying in values for five variables; namely power, voltage, bore, wall thickness and sodium mole fraction. The program then calculates the current, arc length for maximum efficacy and the magnitude of the efficacy for each chosen combination of the variables. The optimum lamp design and ballast combination can then be selected from the calculated data.
Get full access to this article
View all access options for this article.