I am implementing a SVPWM (Space Vector Pulse Width Modulation) scheme for a 3-phase inverter (3 H-bridges, nothing fancy) to control a BLDC Motor.

Since one can choose which zero vectors (000 or 111) to use, a huge variety of implementation alternatives are present. The main division seems to be between "2 leg active 1 leg completely ON or OFF" and "3 legs active" versions.

In the first variant, its clear that there is 33% less switching and this will lead to less switching losses. The second approach has more switching losses (3*2 compared to 2*2 per PWM period) but in many papers I have read its presented as "it has less neural point voltage swing".

I have no problem implementing any of the alternatives but I am having a hard time understanding the pros and cons of "less switching losses" vs "less neutral point voltage swing".

I am driving Gimbal motors which are designed to operate at lower speeds and I plan to have a full-fledged F.O.C (Field Oriented Control) at some point. My aim is precise control at lower speeds. The PWM frequency is at 20 Khz (center aligned) and i am using IFX007T from Infineon for the H bridges.

Any help is much appreciated on deciding which variant to use for my goals.

  • \$\begingroup\$ For slow-speed servo applications I would guess that less neutral point voltage swing would be preferable. However, it seems like it should be relatively easy to implement both in the code, and switch between them while performance testing to see if the neutral shift is an issue. If not, you could improve efficiency by switching to the other scheme. \$\endgroup\$ – John D Jan 28 at 15:32
  • \$\begingroup\$ Neutral point voltage swing is inevitable - and necessary to achieve the higher modulation indexes that SVPWM enables. Also, the neutral (along with at least two of the phase terminals) is still going to swing with every PWM transition, so the additional swings you get with each sector change with the various DPWM strategies aren't that important. \$\endgroup\$ – Phil G Jan 28 at 17:23
  • 1
    \$\begingroup\$ A FOC control is more suitable for PMSM rather than BLDC whose don't produce sinewave back EMF nor they are made to be feed with sinewave current. \$\endgroup\$ – Marko Buršič Jan 28 at 17:52

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.