UIGII Q RESONANT CAVITIES 419 



pressed to a point where their presence does not interfere with the normal 

 operation of the cavity. In this latter case, the amount of suppression is 

 naturally dependent upon the use to which the cavity is to be put, and is 

 conceivably different for a high Q cavity used as a frequency meter, for 

 example, and one used as a selective filter. 



Experience has shown that certain families of modes are much more diffi- 

 cult to suppress than others, and are to be avoided, if at all possible. The 

 feasibility of doing this can be determined by sliding the operating area 

 (a suitable opening in a sheet of paper) around on a large mode chart until 

 the most favorable operating region, consistent with other requirements, 

 has been found. 



Once a suitable operating area has been chosen, the cavity diameter is 

 fixed and length and frequency scales added to the mode chart make it read 

 directly in quantities readily measured. 



Cavity Couplings 



To be useful the cavity must be coupled to external circuits. The prob- 

 lem here is to get the correct coupling to the main mode and as little coupling 

 as possible to all others. Since the electric field is zero everywhere at the 

 boundary surface of the cavity for the TE 01 » mode, coupling to it must 

 be magnetic. This may be obtained either by a loop at the end of a coaxial 

 line or by an orifice connecting the cavity with a wave guide. 



The location for maximum coupling to the main mode is on the side of 

 the cavity, an odd number of quarter-guide wavelengths from the end, or 

 on the end about halfway (48%) out from the center to the edge. Correct 

 orientation of loop or wave guide is achieved when the magnetic fields are 

 parallel. This requires the axis of the loop to be parallel to the axis of the 

 cylinder for side wall feed and to be perpendicular to the cylinder axis for 

 end feed. Wave guide orientation is shown in Table IV. 



The theory of coupling loops and orifices is not at present precise enough 

 to yield more than approximate dimensions. Exact sizes of loops and holes 

 have therefore been obtained experimentally for all designs. 



On the basis of rather severely limiting assumptions," coupling formulas 

 for a round hole connecting a rectangular wave guide and a TE Oln cavity 

 are given in Table IV. The assumptions are that the orifice is in a wall of 

 negligible thickness, its diameter is small compared to the wavelength, it is 

 not near any surface discontinuity, and that the wave guide propagates 

 only its principal (gravest) mode and is perfectly terminated. In echo box 

 applications, this theory leads to a computed diameter that is somewhat 

 smaller than experiment shows to be correct. 



The coupling to other modes can be analyzed, at least qualitatively, from 

 the field expressions of Table I. This has been of value in making final 



