stimson: mercury vapor pump 481 



condensable gases very much less than a ten-thousandth of a 

 millimeter of mercury as was shown on a MacLeod gage, which 

 was the only low pressure measuring device available. 



On account of questions that have been raised from time to 

 time, which threw doubt on the assumption that the vapor flows 

 through the intermediate pump with a velocity comparable 

 with the velocity of sound, verificative experiments were made. 

 A temporary rubber connection with pinch cock on the lower end 

 of the tube K below the intermediate pump allowed the lique- 

 fied mercury to be caught and measured, thus giving the term 

 m/t in the dimensional equation. The pressure difference 

 supported by the pump was observed in the manometer arms K 

 and the cross-section of the tube is known. Computations of 

 velocity from the dimensional equation give a value within a 

 few per cent of the best available value for the velocity of sound 

 under these conditions. The vapor density in the boiler is an 

 independent datum and furnishes another means of computation, 

 whose result indicates more conclusively that the assumptions 

 of flow are verified as nearly as could be expected from the pres- 

 ent incomplete knowledge of the properties of mercury vapor. 



Some of the advantages of a pump of this design are apparent, 

 namely, that it has a single boiler run by a single gas burner, a 

 single line of cooling water, and is effectively a single unit with a 

 direct air tight connection between the high vacuum stages. It is 

 also noted that several paths conveniently open for free exhaus- 

 tion with a fore pump are closed as soon as the mercury vapor 

 pump is started. The advantage of not requiring a primary 

 pressure less than 4 cm. is also apparent. 



This type of pump could be employed in nearly all places where 

 high vacuum is necessary. If mercury is objectionable either on 

 account of its vapor pressure or its being an undesirable element 

 in the region to be evacuated, the apparatus can be protected 

 by a liquid air trap or possibly a gold foil trap. The speed of 

 the pump when working on high vacuum can be increased by 

 increasing the dimensions of the high vacuum unit. However, 

 high speed, as has often been pointed out, is sometimes illusive, 

 for a connecting tube 1 cm. in diameter and 0.5 meter long is 



