I have a DC motor, which is used as an electromagnetic brake. However I want to leave certain frequencies uninhibited, so I thought this should be possible by adding a shunt filter on the DC motor mains. The Idea is that if I high-pass the current through the shunt, that for a sinusoidal velocity input below 3 Hz no current flows through the motor coils resulting in no back-emf torque.

The circuit below consists of a sinusoidal voltage source as a result of the emf of the DC motor, the armature inductance (L1) and resistance(R1) and a still unresolved shunt circuit across the actuator leads.

What I want to accomplish with the shunt is to obtain a certain frequency response of the current in the shunt circuit, but I'm struggling with the design. Most of what I've found describes voltage filtering behaviour and I'm wondering if it's possible to filter the current in a shunt circuit.

The schematic currently shows a low pass filter arrangement, which in my understanding will result in a high current for high frequencies and a low current for low frequencies as the capacitor will have a low and high impedance respectively. The inductor acts as a low pass filter on the current, but it's filtering higher frequencies I don't really care about.


simulate this circuit – Schematic created using CircuitLab

What I would like to achieve is a frequency response somewhat like the picture below. The important part is the relatively close proximity of pass and stopband frequencies.

I would like to know if it is possible to obtain such a filter on the total current and whether this could be done without any need for active control (aside from opamps and "dumb" active components)

Frequency response magnitude plot

  • \$\begingroup\$ Welkom bij SE.EE! What is the quantity and unit on the vertical axis of the drawn frequency response? I would guess the current through the filter in dB? And why do you want -80 dB (?) from 0 Hz to 1 Hz? Since it is an actuator, could you tell how it is driven? \$\endgroup\$ – Huisman Apr 10 at 11:09
  • \$\begingroup\$ I think I'll redraw that picture for more clarity. I don't need -80 dB specifically, it was just to show the higher order decent (ascent?) between 1 and 3 Hz. I don't really have another way of deciding the exact slope other than "as steep as possible". The chosen slope will probably depend on the physical size of the electrical components required. Currently the actuator is not driven electrically, but mechanically so it acts as a generator. The idea is that a shunt circuit controls the back emf torque to resist mechanical torque inputs dependent torque disturbance frequency. \$\endgroup\$ – Hugo van der Kort Apr 10 at 11:26
  • \$\begingroup\$ I think you should rephrase your question, first stating what you really want. Something like: I have a motor that acts as generator. I want it to resist [describe to what extend] rotation dependent of the frequency of the applied torque. It should [block/allow] torques with frequency of 0 Hz to 3 Hz, it should [block/allow] torques with frequency of 3 Hz to 10 Hz.. etc \$\endgroup\$ – Huisman Apr 10 at 12:43
  • \$\begingroup\$ If you apply a short circuit instead of the filter, you'll see R1 probably prevents what you seem to like: resist rotation of the generator. R1 allows slow rotation of the generator with a short circuit, so also no matter what filter you apply. Because of R1 spoiling it, I think you rather want an active motor control based on (very accurate) position/velocity measurement of the motor/generator. \$\endgroup\$ – Huisman Apr 10 at 12:45
  • \$\begingroup\$ Maybe I made an error, but I understand that electrically you model a DC motor as a voltage source (emf) and the coil resistance and inductance in series. Thats what R1 and L1 are for. \$\endgroup\$ – Hugo van der Kort Apr 11 at 8:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.