172 THE BELL SYSTEM TECHNICAL JOURNAL, JANUARY 1951 



As none of the other energy losses in the core or winding vary with line 

 current strength, it became convenient to consider the attenuation loss 

 caused by hysteresis as an ''excess" loss, when referred to the attenuation 

 that would result if the hysteresis should be vanishingly small or zero. The 

 theoretical studies above mentioned not only showed the "excess" attenua- 

 tion in long loaded, repeatered, circuits to vary as a function of the magni- 

 tude of the input current and frequency, as above noted, but also showed it 

 to vary as a function of the length of repeater section, the position of the 

 repeaters in the line, the weight of loading, and the over-all circuit length. 

 Since it is not possible to offset the effects of loading coil hysteresis by means 

 of distortion corrective or equalizing networks at repeater stations or circuit 

 terminals, the piling up of the "excess" losses along the line in very long 

 loaded, repeatered, circuits could reach values that would be large relative 

 to the desired over-all working equivalent, if the individual loading coils 

 should have large hysteresis losses, as did the soft-iron dust-core loading 

 coils. 



Comparative theoretical studies of the excess loss due to hysteresis effects 

 in 65-permeability core loading coils showed these coils to be greatly superior 

 to the soft-iron dust-core coils in this feature. On the other hand, the wire- 

 core coils were relatively unsatisfactory from the inductance-stability stand- 

 point for use on long repeatered circuits. 



6.3 Compressed, Unannealed, Powdered-Iron Core Loading Coils 



It was very fortunate that the development work on the compressed, 

 unannealed, powdered-iron core-material, previously mentioned, approached 

 commercial fruition at about the time the unsuitability of the soft-iron 

 dust-core loading coils for very long repeatered circuits became apparent. 



As noted in Table III, the effective volume permeability of this improved 

 core-material was closely that of the 65-permeability wire-cores. The new 

 standard loading coils using this improved material had cores generally 

 similar in dimensions to those of the older coils used on 19 and 16 ga. toll 

 cables, and their over-all dimensions were sufficiently similar to avoid the 

 need for developing new loading coil cases. 



In general terms, the new coils combined the best qualities of the soft- 

 iron dust-core loading coils with the best performance characteristics of the 

 65-permeability wire-core loading coils. Actually, they were much better 

 than the soft iron-dust core coils with respect to stability of residual induc- 

 tance, and susceptibility to magnetization by superposed steady currents. 

 In these respects they were also substantially superior to the 65-permea- 

 bility wire-core loading coils. However, they were not quite so good as the 

 low permeability wire-core coils with respect to hysteresis losses and telegraph 

 flutter transmission impairments. On the other hand, they were substan- 



