Journal of Mechatronics and Artificial Intelligence
Vol. 3 No. 1 (2026): JMAI: June 2026

Efficiency Evaluation of a DC Generator Using Experimental Loss Measurements

Elysa Nensy Irawan (Shibaura Institute of Technology)
Chisaki Osumi (Shibaura Institute of Technology)



Article Info

Publish Date
30 Jun 2026

Abstract

This study evaluates the efficiency of a DC generator using experimental loss measurements based on a two-identical-machine method. Two identical DC machines were mechanically coupled, with one machine operated as the driving motor and the other as the generator under test. The total loss was determined from the difference between the motor input power and the generator output power, while the copper loss was calculated using the armature resistance obtained from a locked-rotor test. Since the generator did not have a field winding, only armature copper loss was considered. The friction loss was estimated from the no-load test, and the iron loss was determined using a residual loss method. The experimental results showed that the generator output power increased from 0.0064 W at 1.5 V to 1.8032 W at 5.0 V. The generator efficiency also increased with operating voltage, reaching a maximum value of 63.96% at 5.0 V. Copper loss increased significantly with load current and became a dominant loss component at higher operating conditions. The residual iron loss showed positive values at low voltage levels but became negative at higher voltages, indicating limitations in separating friction and iron losses using the current method. Overall, the proposed experimental approach was effective for evaluating the efficiency trend and identifying the dominant loss behavior of the DC generator, although further refinement is required to separate mechanical and magnetic losses more accurately.

Copyrights © 2026






Journal Info

Abbrev

jmai

Publisher

Subject

Description

The Journal of Mechatronics and Artificial Intelligence (JMAI) (E-ISSN 3048-4227 P-ISSN 3062-729X) serves as a platform for disseminating scholarly research related to the fields of mechatronics and artificial intelligence, as well as related sub-disciplines. We extend an invitation to researchers, ...