1
\$\begingroup\$

I've implemeneted a 3rd harmonic injection SVPWM (Space Vector PWM) scheme and it performs quite well. One peculiarity i have noticed was, when i set the modulation index to 0, the motor stops as expected but after leaving it like that for a couple minutes the H-bridges (Infinieon IFX007T) on my custom 3 phase inverter board got quite hot - okay to touch but still you wouldnt want to keep your finger on it for too long..

The modulation index at 0 corresponds to all 3 PWM duty cycles being at half the max value. In essence that should be "zero power transfer" but the H-bridges getting hot clearly indicates a power transfer and considering you can measure the current at this state ( when the motor is running ) got me wondering.

A quick "hack" would be just manually setting the duty cycles to 0 or shutting down the inverter entirely but i would like to know the inner workings behind it and if thats normal behavior.

TLDR; Setting modulation index to 0 returns duty cycles that are half the max duty cycle. Would it be ok to leave the motor like this for hours or would it still drain some power ?

\$\endgroup\$
1
\$\begingroup\$

The only reason to continue switching with the motor at standstill is to use the motor as an active brake to prevent external forces from moving the load mechanism. An SVPWM scheme is not likely to be capable of doing that. If the motor is to remain at standstill for hours at a time, it would be good to consider the use of a "sleep" mode which cuts power to a minimum. Even if active braking is needed, a mechanical brake might be preferable for holding the load for hours at a time. Selecting a stop mode is very much dependent on the power level involved. Reducing the idle power is a lot more significant if the motor power is in the kilowatt range rather than the milliwatt range.

\$\endgroup\$
  • \$\begingroup\$ I accepted the answer, but realized something peculiar in the meanwhile; when the motor is at full rpm or at least spinning for a long time, the h-bridge IC's are basically very cool temperature-wise. but when i let the motor stop, i.e modulation index = 0, they become quite hot, even hard to touch. I am pretty sure something is off in my implementation, any comments about this peculiarity ? \$\endgroup\$ – Firat.Berk.Cakar May 21 at 15:29
  • \$\begingroup\$ Yes. That does sound like a problem with implementation. It seems like there must be current in the motor at stop. If there is DC current or an AC current trying to rapidly transition between directions of rotation, the motor would be carrying current without rotating. Braking torque may be produced. \$\endgroup\$ – Charles Cowie May 21 at 16:08
1
\$\begingroup\$

When MOSFETs switch they dissipate a small amount of power and, because the motor is still loading the output, this power (dynamic switching power) will still be present irrespective of motor speed. There will also be some power dissipated in the motor even though the duty cycle (50%) prevents forward or reverse motion.

\$\endgroup\$
  • \$\begingroup\$ Can i interpret your answer as this is an acceptable and expected thing and i shouldnt bother with cutting the power entirely when i am at a stand still ? \$\endgroup\$ – Firat.Berk.Cakar May 20 at 7:36
  • \$\begingroup\$ I can't tell you that on several levels. First is that I don't know what "getting hot" really means. The second relates to the circuit itself and not knowing that circuit I can't guess. Cutting the power (or not) depends on other factors like not wishing for things to get above some arbitrary temperature or not wishing to consume watt-hours (joules) in standby and these are questions you need to address yourself. \$\endgroup\$ – Andy aka May 20 at 7:42
  • \$\begingroup\$ I should have provided more info. "Getting hot" means okay to touch but you wouldnt want to keep your finger on it for too long. The IC is infineon IFX007T and max temp is listed as 150 ° celcius so it should be okay. The circuit is a custom 3 phase inverter, nothing fancy. What actually wanted to know was if what i experience is common in motor control. \$\endgroup\$ – Firat.Berk.Cakar May 20 at 7:47

Your Answer

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

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