A precision investigation of some aspects of parallel-plate free-air chamber geometry.

An account is given of a series of precision experiments to investigate cut-off effects in a parallel-plate type of free-air chamber, and the relationship of these effects to the chamber geometry is analysed. The discussion is extended to enable estimates to be made of the errors associated with other parallel-plate chambers. In conclusion, it is shown that deficiencies in the geometry of the standard chamber in use at present at the National Physical Laboratory in England have led to errors approaching 3 per cent in the realisation of the rontgen at ordinary deep-therapy qualities. The corresponding error in the National Bureau of Standards chamber in the United States is estimated to lie between 1 and 2 per cent.