Consider this scenario:

  1. A 3-phase permanent-magnet generator is being driven by a strong force and at varying speeds e.g. 1000-6000 RPM
  2. Its output is rectified via a 3-phase bridge rectifier
  3. The resulting DC is fed to a DC-DC converter which powers something (e.g. charges a battery) - consider it a constant power load.

You'd have some capacitance in front of the DC-DC converter (e.g. it may be needed for stability). In this case the classical conduction angle issue of bridge rectifiers emerge, especially if the capacitance is large, and this setup specifically exacerbates this:

  • when you increase the generator speed, the amplitude of its output voltage increases (so the amplitude of peaks and valleys of the rectified waveform increases)
  • the period between peaks is smaller
  • higher voltage = less current drawn by the DC-DC, thus the capacitor can hold out for longer

I've made a simulation of what happens electrically: simulation

  • the three dim traces are the phases of the generator
  • the red trace is the rectifier output (what it would look like if it were unloaded)
  • the cyan trace is the capacitor voltage
  • the yellow trace is the generator current

I'm simulating winding resistance and cap ESR, but not winding inductance. Values taken from a specific application I was considering.


I'm worried that this pulsating current causes uneven load to the generator, and may cause issues with the generator itself - e.g. vibrations, uneven load to the bearings, or similar. On the other hand, rotor inertia may be capable of smoothing things out. Is my concern well founded, or I'm worried about the wrong thing?


1 Answer 1


Concern about the harmonic content in the generator load current is well founded, but the severity of the end effects will be difficult to quantify. Your simulation needs to include all of the inductance in the circuit. You should also investigate the effect of adding inductance. In addition to torque pulsations, you should consider losses in the generator due to harmonic content.

  • \$\begingroup\$ I was considering simulating the inductance as well, but I'm not sure what's the proper model to use. Can I approximate the generator as a perfect AC voltage source, in series with a resistor, and in series with an inductor (each phase)? Or something more complicated is needed? \$\endgroup\$
    – anrieff
    Commented Mar 12, 2020 at 14:26
  • \$\begingroup\$ Yes, that would be a good model. \$\endgroup\$
    – user80875
    Commented Mar 12, 2020 at 14:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.