The Production of a Vacuum in an Air Tank by Means of a Steam Jet

Abstract

The paper consists mainly of an account of experiments conducted by the author on a steam-jet operated air ejector with the object of investigating certain quantities which, although they affect the vacuum produced, are not included in the theory of the ejector as it exists at present. The chief of these quantities was found to be the distance from the outlet of the steam nozzle to the throat entrance of the diffuser, and in order to examine the effect of varying this length a special form of air ejector with a sliding diffuser was designed and provision was made in the design for the use of a search tube which could be passed right through the diffuser and up to the nozzle outlet. In addition to the work conducted on this apparatus the author was obliged to carry out a separate set of experiments, with the diffuser removed, and in which an air pump created the vacuum, in order to examine the change in the stationary waves in the steam jet when it was discharging at different vacua. This was done by viewing the jet through a plate-glass window fitted into a part of the ejector. Observations were made and photographic records were obtained at different outlet pressures ranging from atmospheric pressure to 27·5 inches of vacuum. These photographs are probably of interest in connexion with steam nozzles in general, as they show the form of the jet when it was over-expanded and under-expanded as well as when it expanded to the designed pressure. The ejector was also operated normally with a glass mouthpiece forming the entrance and parallel throat of the diffuser, and this arrangement was used to view the waves in the throat at the higher vacua.

Another quantity which affects the vacuum is the length of the diffuser throat and two throat lengths were used: (a) inch, and (b) a length equal to twice the throat diameter, or 1½ inches. Two different forms of diffuser entrance were tried, and the vacuum and steam-air ratio at different initial steam pressures for these two different designs have been recorded in the paper.