A BRAUN TUBE HYSTERESIGRAPH 287 



The hysteresis curves of Plate I (1) are those of a ribbon of Armco 

 iron. The curves were made with 25 cycle magnetization and at 

 four different magnetizing fields, the highest field being sufficient 

 to saturate the sample. The effects of frequency and of permeability 

 on the eddy current loss in materials are shown in the next two figures 

 of this plate. Two similar ribbons were used. The one was permalloy 

 magnetized to saturation, Plate I (2); the other, Plate I (3), was 

 Armco iron at a low magnetizing field. The curves were made at 

 three different frequencies, those marked with the letters a, b and c 

 corresponding to magnetization at 25, 200 and 1,000 cycles, respec- 

 tively. A comparison of the magnetic properties of pure iron and 

 permalloy is made in the two figures, (4) and (5). In each figure the 

 letter a indicates the curve for iron, b that for permalloy. The former 

 figure was made at a field strength high enough to saturate the iron, the 

 latter at a field strength which saturated the permalloy but not the iron. 

 The last figure of this plate, (6), shows the influence of tension on the 

 magnetic properties of a ribbon of permalloy containing 65 per cent 

 nickel. The curve a was obtained without tension, b with a moderate 

 tension on the ribbon. The curves of Plate II reproduce the magnetiza- 

 tion cycles of a sample of Armco iron at various temperatures, from room 

 temperature to the recalescence point. At the temperature of about 

 790° C, ferromagnetic properties were gone and the only flux that is 

 indicated existed in the uncompensated air space of the search coil. 



I. Description of Method 



A change in magnetic flux in a specimen is usually measured either 

 as a change in the field in a relatively short air gap, or as the time 

 integral of the potential set up in a search-coil surrounding the speci- 

 men. The second method is illustrated by the use of the ballistic 

 galvanometer as an integrating instrument, employed in most magnetic 

 measurements. In the present case, however, the integration is 

 accomplished by a purely electrical circuit. The integrating element 

 consists of a resistance and condenser in series with the search coil 

 surrounding the sample of material.^ 



^ This type of integrating circuit is one of at least ten simple combinations of 

 resistance, capacity and inductance which can be used for obtaining the cyclic 

 integral of current or potential. Some of these have been described in connection 

 with hysteresis measurements by E. L. Bowles, //. A. I. E. E., 42, p. 849, 1923; 

 O. E. Charlton and J. E. Jackson, //. A. I.E. E., 44, p. 1220, 1925; W. Kaufmann, 

 Zeits.f. Phys., 5, p. 316, 1921. W. Kaufmann and E. Pokar, Phys. Zeits., 26, p. 597, 

 1925. K. Kruger and H. Plendl, Zeits.f. Hochfr., 27, pp. 155-161, 1926; W. Neu- 

 mann, Zeits.f. Phys., 51, p. 355, 1928. The circuit chosen here was used by Bowles 

 and by Charlton and Jackson in connection with a mechanical oscillograph, and by 

 Kruger and Plendl in connection with a Braun tube. 



