264 ANNUAL REPORT SMITHSONIAN INSTITUTION, 193 3 



sphere. Thus the attainable voltage depends upon the size of the 

 spherical terminal. 



The current, on the other hand, is simply equal to the rate at 

 which electricity is carried to and from the sphere by means of the 

 belts, and this in turn depends upon the size, speed, and number 

 of belts and the quantity of electricity which can be placed on a 

 unit area of the belt. This latter quantity is also limited by the 

 breakdown voltage of the surrounding air, to an amount of about 

 5X10"^ coulombs per square centimeter of belt area. Under these 

 conditions it is readily shown that a belt running at 6,000 feet per 

 minute could theoretically carry a maximum current of 150 micro- 

 amperes per inch width of belt. Actually, the best adjustments have 

 given about half of this theoretical maximum, probably because the 

 breakdown strength of the air is reduced by the mechanism whereby 

 charge is sprayed onto the belts. 



Theory and practice also show that these belts may be placed as 

 close together as is geometrically possible, in fact, practically in 

 contact, without interfering with their capacity to carry charge. 

 By packing many belts together it is therefore possible to produce 

 very sizeable currents. For example, a small laboratory model for 

 demonstration purposes, constructed this year in the laboratories 

 at the Massachusetts Institute of Technology, develops IV2 million 

 volts between a pair of 2-foot spherical terminals and delivers a cur- 

 rent of 600 microamperes carried on two 8-inch belts in each sphere. 



Even in this small model the currents are approximately a million 

 times greater than those which have been obtained in the high-speed 

 ion source designed by Lawrence. 



The first model of such a generator that was actually constructed 

 was built in Princeton in the fall of 1929, being built out of a tin 

 can, a silk ribbon, and a small motor, at no expense. This model 

 developed 80,000 volts, being limited by the corona discharge from 

 the edges of the can. 



The next model was designed and built for operation in a vacuum 

 tank for reasons to be outlined later. 



The third model was built to give a quick and easy demonstration 

 of the possibilities of the machine, using 2-foot spherical terminals 

 supported on pyrex rods, and supplied by current carried on silk 

 belts 2^ inches wide, driven by small motors. This apparatus was 

 demonstrated successively in Princeton, New York, Washington, 

 Boston, and elsewhere. Although built at a total cost of less than 

 $100, it developed more than twice as high a voltage as any direct 

 current generator of which we have knowledge. 



Encouraged by the success of this model, plans were immediately 

 made for the construction of as large a generator as seemed practical 



