630 Dr. W. M. Thornton on the 
These relations can be illustrated from figs. 3 & 4 as 
follows :— 
| Cast Steel Ring. Transformer. | 
Area A, (by planimeter).| 6 sq. in. 10°75 sq. in. 
oa), ) Aare Laat ocniaanewagels iy Oss cage e 
(AAS ecimiestanas Je aah | 15°42 ,, 
| (A Aa pdus tenes ore gs ee GOs | 
Ratio Brn/ By: dese. vx: 31 / 2°5 
Sectional area C ......... 75°1 sq. em. 420 sq. em. 
‘urns on winding ...... 777 ) 97 
| Resistance in circuit ...| 1-9 ohms. 5 ohms. 
PR Pe ene 10 amp.=1 inch. | 4 amp.=1 inch. 
AE ae NENT ET oh 5 | °125 sec.=1 imch. ‘05 sec.=1 inch. 
We ints hoa see eee | 23X10" persq.in. | 1°0X10" per sq. in. 
Bm (from eqn. 9)......... 17,900 per sq. cm. 18,900 per sq. cm. 
Be es SO ea ea as 7,960 = 
FA mag, 63k toe 295 ;; 67 
Comparing the values of B,, and B, obtained in this way 
with those of § 1 for the ring, it is seen that there is a fairl 
close agreement for the maximum induction, but about 
15 per cent. difference in the residual. Of the two methods, 
the latter is the more likely to be correct in determining the 
residual flux. 
These relations suggest a very simple method of measuring 
the induction through the windings of large electromagnets, 
such as the exciting circuits of large dynamo-electric 
machines, which may be in use but has not come to the 
writer’s notice. It is to record the rise of current when the 
magnetization is suddenly switched on, and to measure the 
area corresponding to abo in fig.3. The curve will in general 
shape resemble curve A. in fig. 5; and since most large exciting 
currents take from 5 to 60 seconds to reach their full value, 
the speed at which the record must be made is not excessive. 
Proceeding in this way, the magnetic leakage between the 
windings and the armature can be measured by ordinary test- 
room instruments. 
A still further simplification suggests itself in the case of 
machines where the time constant is above five seconds. 
Instead of recording the current rise, the times, taken by a 
stop-watch, for the current to reach given values in turn 
are plotted. The first rush cannot be measured, but after 
about three-quarters of a second readings can be obtained 
which coincide with the recorded curve. To measure the 
magnetism of a large generator or motor it is then only 
necessary to have a deadbeat ammeter and a stop-watch. 
