INCREMENTAL SHEATH THICKNESS MEASUREMENTS 361 



about 10 feet long) and are maintained mechanically in the testing posi- 

 tion h}^ the probe assembly (see Figs. 3 and 5(a)). 



In the design of the probes and their assembly, various difhculties 

 had to !)(' overcome. The probes operate on cables subjected to some 

 luunoidable swings and vibrations while moving with speeds up to 80 

 feet per minute. The capacitance from either of these probes to the metal 

 cable core, in equivalent conditions, should match each other within 

 approximately one-thousandth of a mmF. This capacitance should not be 

 appreciably affected by limited displacements of the probes with respect 

 to the cable plane of symmetry, such as may occur in actual operating 

 conditions. 



The first experiments with probes of a conventional design, having 

 flat, or nearly flat, contact surfaces, were quite discouraging. The probe- 

 to-core capacitances fluctuated to an intolerable degree as a result of 

 even minute cable displacements. 



Eventually, probes were developed which met all the requirements. 

 Each of these probes is in the form of a cut-off segment of a toroid. The 

 major axis of the cut-off elliptical plane is oriented in the direction 

 essentially parallel to the cable axis, while the convex center part of the 

 probe slides on the cable sheathing. This form of probe has the advantage, 

 common with the spherical form, that the capacitance from the probe 

 to the cable core varies but little as a result of displacements and changes 

 of position caused by the cable motion. But the toroidal form has the 

 following advantages over the spherical: first, for the same residual 

 capacitance to the cylindrical cable core, the transverse dimensions of 

 the former are smaller; and, second, the capacitance of the toroidal form 

 with respect to a cj^lindrical cable core can be conveniently adjusted by 

 the simple expedient of twisting the probe element in a plane parallel to 

 the cable axis. (Adjustments with a precision exceeding one-thousandth 

 of a iifiF were actually performed). 



The probe electrodes, surrounded (except for the contacting face) 

 by the B-shielding, are mounted on mechanically balanced light alu- 

 minum arms [Figs. 5(a) and 5(b)]. There might be one, tw^o, or four 

 probes to an assembly, which can be turned over 3G0° around the cable 

 axis. For eccentricity measurements two probes can be simultaneously 

 used, having a spacing of 180° (foi- measurement of eccentricity across a 

 diameter) or of 90° (for measurement of elipsoidal eccentricity). Also 

 for eccentricity or direct thickness investigations and process settings one 

 probe only may be used, with th(! other bridge measuring arm connected 

 to an auxiliary standard. 



