352 
DRS. J. AND E. HOPKINSON ON DYNAMO-ELECTRIC MACHINERY. 
across their terminals. The power absorbed must be due entirely to local currents in 
the core of the armature, and the energy for the reversal of magnetisation of the core 
twice in every revolution of the armature. 
No. 13 gives the results of the experiments on the friction of the bearings and in 
bending the belt already referred to. 
It will be observed that tire figures in column XIII. are calculated by deducting the 
power absorbed in the armatures and magnets, and extraneous resistances from the 
total power given to the combination as measured by the dynamometer. They must 
therefore include all the energy dissipated in the core of the armature, whether in 
local currents or in the reversal of its magnetisation ; also the energy dissipated in 
local currents in the pole-pieces, if such exist; also the energy spent in reversing the 
direction of the current in each convolution of the armature as they are successively 
short circuited by the brushes. Further, it will include the waste in all the con¬ 
nexions of the machine from the commutator to its terminals and the friction of the 
brushes against the commutator. A separate experiment was made to determine the 
amount of this last constituent, but it was found to be too small to be capable of 
direct measurement by the dynamometer. Moreover, from the manner in which the 
figures in this column are deduced, any error in the dynamometric measurement will 
appear wholly in them. Since, undoubtedly, the first two components enumerated 
are the most important, and the conditions determining their amount are practically 
the same throughout the series, the close agreement of the figures in the column are a 
fair criterion of the accuracy of the observations. Probably 100 watts is the limit of 
error in any of the measurements. Such an error would affect the determination of 
the efficiency when the machines were working up to their full power to less than 
5 per cent. 
It has been assumed that the sum of these losses is equally divided between the 
two machines. This will not accurately represent the facts, as the intensities of the 
fields and the currents passing through the armatures differ to some extent in the two 
machines. The inequality, however, cannot amount to a great quantity, and if it 
diminishes the efficiency of the generator it will increase the efficiency of the motor by 
a like amount, and contrariwise. In No. 11 of the series the effect of the sparking at 
the brushes of the generator is very marked, the power wasted amounting to at least 
250 watts. 
If it be assumed that the dissipation of energy is the same whether the magnetisa¬ 
tion of the core is reversed by diminishing and increasing the intensity of magnetisa¬ 
tion without altering its direction, or whether it is reversed by turning round its 
direction without reducing its amount to zero, a direct approximation may be made to 
the value of this component. (J. Hopkinson, Phil. Trans., vol. 176 (1885), p. 455.) 
The core has about 16,400 cubic centims. of soft iron plates, hence loss in magne¬ 
tising and demagnetising when the speed is 800 revolutions per minute = 16,400 X 
3 g- 0 p - X 13,356 ergs per second = 292 watts. 
