220 CURTIS and grover: alternating current resistances 



where R is the ohmic resistance of the coil, L its inductance, 

 C, the capacity between the two wires, and p is 2tt times 

 the frequency. From this we see that R' =R and 6 = only 

 under the condition that both L and C are zero. While this 

 cannot be accomplished in practice, yet if L is made as small 

 as possible and if L — CR~ = 0, then the phase angle will be zero 

 and the change in resistance with frequency negligible. With 

 coils having low resistance (less than 10 ohms) it is difficult to 

 make the capacity large enough to make L — CR 2 = 0; but with 

 coils of high resistance (1000 ohms and over) the capacity is usu- 

 ally much larger than it should be. Between these extremes 

 satisfactory coils can be constructed without great difficulty. 



In case alternating current is used, the absorption of a poor 

 dielectric between the wires may produce an apparent leakage 

 between turns where the direct current insulation resistance is 

 sufficiently high. The magnitude of the effect of this upon the 

 resistance of the coil will depend upon the resistance of the coil 

 and upon its capacity as well as upon the absorption. Hence 

 while it is entirely negligible for coils of low resistance, it becomes 

 of importance in coils of high resistance. By so designing the 

 coil that the capacity is small, it may usually be made negligible. 



By keeping in mind the facts which are stated above, it has 

 been found possible to design coils of denominations from 0.1 

 to 10,000 ohms in which the change in resistance is less than a 

 part in a hundred-thousand for frequencies up to 3000 cycles per 

 second, and in which the phase angle is not more than 3.5' at 

 3000 cycles. For xo-ohm, 1 ohm and 10 ohm coils strip man- 

 ganin, wound in a bifilar manner is used. For 100-ohm coils a 

 single-layer bifilar winding is satisfactory. For 1000-ohm coils 

 fine 200-ohm single-layer bifilar coils are wound on an insulating 

 spool (porcelain is satisfactory) and connected in series. For 

 10,000-ohm coils a special winding was devised. A tube of bis- 

 cuit porcelain is cut along a diameter for two-thirds of its length. 

 The winding is then made as follows: One turn is made with a 

 single wire, then the wire passed thru the slit, and a turn made in 

 the opposite direction. The wire is then passed thru the slit 

 and the third turn is made in the same direction as the first, 

 and the process repeated until sufficient wire is upon the spool. 



