I x has been shown that power losses are observed when a type II superconductor is subjected to a low frequency magnetic field, even when the peak value of the field is lower than He~. I-3 These losses have been linked to a local hysteresis of magnetization in the surface region. 4
The following general behaviour has been observed. Power losses vary linearly with frequency up to frequencies ranging between 10 and 50 kc/s. Losses vary as high powers of the strength of the a.c. field (H 2"5 to H 4"5) and increase when an additional static field is applied. We also found it useful to investigate the following points: influence of surface roughness; influence of trapped flux; influence of the normal state electrical conductivity.
These experiments were performed on niobium samples which had been given such treatments as electropolishing and vacuum outgassing.
Experimental Technique
Losses were measured in the calorimeter shown schematically in Figure 1. Its principle is close to that described by Buchhold and Molenda. 1 The measurement principle is as follows: a carbon resistance thermometer and a heating resistor are glued in the hole shown on the Figure . The superconducting specimen provides magnetic shielding of the thermometer. The sample is located inside a nylon chamber that can be immersed in liquid helium. The specimen mounting and high vacuum within the chamber provide correct thermal insulation.