Limits of switching frequency in motor control

[Oliveira]

Hi everyone

I have heard many times about the effect of high switching frequencies in motor control, due to the interaction with the machine internal inductance.

Does anyone know how to calculate the maximum switching frequency "allowed" for a certain motor drive? Or is it just an "intuitive" decision?

Thank you in advance!

Oliveira

 

[Grossel]

Hi.

The limit is often decided by the switching transistors rather than the motor itself.

When you increase switching freq, the rise and fall times for the transistors will be greater compared to t_on + t_off. As a result, transistors will consume more power (heat) and thus the efficiency decreases.

 

[ronv]

One other thing to keep in mind is the "break away" friction. If there is a lot of friction in the system and low speed is desired a lower frequency will work better because it will allow the current (thus torque) to be higher on each pulse. A lot of times at high frequency the motor will not move for quite a bit of the total pwm range and then break loose and run at a higher speed than you would like.

 

[crutschow]

Why would you want to use a higher switching frequency then the motor requires for control? I see no advantage to that.

 

[misterT]

One other thing to keep in mind is the "break away" friction. If there is a lot of friction in the system and low speed is desired a lower frequency will work better because it will allow the current (thus torque) to be higher on each pulse. A lot of times at high frequency the motor will not move for quite a bit of the total pwm range and then break loose and run at a higher speed than you would like.

Low switching frequency is a bad solution for that kind of problem. Better solution to overcome static friction is to briefly drive the motor with 100% PWM. There is no limit for the switching frequency.. the higher the better.

Why would you want to use a higher switching frequency then the motor requires for control? I see no advantage to that.

To reduce current ripple in the motor. Of course there is no point to use higher freq. than "required", but that is true for everything. Then, how do you define the "required" frequency?

The PWM frequency should be selected so that it does not create too much current ripple in the motor windings. Current ripple results in heating of the motor. If you know the terminal inductance L and the terminal resistance R of the motor, you can calculate the electrical time constant of the motor (t = L/R). With the time constant, t , calculated you should make the PWM frequency (much) higher than 1/t [Hz]. Good servomotors can have time constants as low as 50 to 150 microseconds, requiring PWM frequencies up to 40 kHz to 60 kHz.

If you want to use low switching frequencies, or your motor heats up too much, you can add inductance (current choke) in series with the motor. This increases the electrical time constant of the system and therefore lower freq. can be used.

A good estimation for current ripple with 2-level PWM is:
dI = V / (2*f*L)

Where:
dI is current ripple amplitude
V is drive voltage
f is switching frequency
L is motor inductance

The limit is often decided by the switching transistors rather than the motor itself.
When you increase switching freq, the rise and fall times for the transistors will be greater compared to t_on + t_off. As a result, transistors will consume more power (heat) and thus the efficiency decreases.

This is true. When you think of the upper limit of switching frequency, the limiting factor is the control electronics and switching losses in power mosfets. The motor itself does not limit the frequency.

 

[smanches]

The main reason to control the frequency is to reduce or eliminate audible noise caused by the switching. I've heard many people say to stay above the audible frequency (~20khz) but noticed that you can use lower frequencies, and just find a "quiet spot" for that particular motor/drive combo.

 

[misterT]

The main reason to control the frequency is to reduce or eliminate audible noise caused by the switching. I've heard many people say to stay above the audible frequency (~20khz) but noticed that you can use lower frequencies, and just find a "quiet spot" for that particular motor/drive combo.

This is the "usual" reason, but for most small motors 20 kHz is too low. The main reason to control the frequency is to reduce current ripple and make control loops more stable.

 

[ronv]

What is your application? Is it open loop or a closed loop servo? Hardware or micro processor control?