Motor max RPM Control

[Nuts&Volts]

Ok so I have been reading up on motors laterly and I have pretty understanding of their general theory, construction and operation. My focus here is an Axial flux PM motor. I read through "Doin the Math" but I am still struggling on what determines max rpm.

So lets take a 20 pole (10 pole pairs) 2 rotor axial flux motor. Then a Battery supply voltage of 300V. Lets say I want to fun the motor up to 4000 RPM. So I need a controller frequency of
4000RPM= (freq/10)*60
freq=666.6 Hz this is the frequency I need from the controller to run at 4000RPM.
I do realize that certain Voltage needs to be applied to push more current at these speeds.

First question is this frequency (666.6 Hz)the PWM frequency or some other frequency?
Also how does voltage play a role in max RPM? Will 300V be enough to run at 4000RPM?

Thanks
Kyle

 

[Jeremy Harris]

The frequency you've worked out (correctly) is the commutation frequency, which is the frequency of the three phase rotating magnetic field in the motor.

The PWM frequency will be much higher, typically 10 to 30kHz, with 15KHz being fairly typical for motors with a modest amount of inductance.

If you want to run at 4000 rpm and 300V, then the motor needs to be wound so that it has a Kv of 4000 / 300 = 13.33.

Jeremy

 

[Nuts&Volts]

Ok I think I am gettin it some.

So the commutation frequency has to do with how fast the phases are switched on and off (ie the transistors on and off). This controls how quickly the rotor is pulled around, thus the RPM.

Does the PWM frequency depend on the commutation frequency in an AC drive? It seems to me that is does because the faster you run the motor the higher your commutation frequency. This means thou that there is a higher back EMF (from some reading). So the PWM frequecny has to increase so that it is "on" more and thus a higher voltage is seen on the phases?

I hope I am stating that clear enough.
Also feel free to direct me to links where I can find this out for myself if thats easier

Edit: So the Kv must relate to the Back EMF then?
Edit 2: Figured out it does, Kv is an easy way to understand back EMF.

 

[liveforphysics]

Ok I think I am gettin it some.

So the commutation frequency has to do with how fast the phases are switched on and off (ie the transistors on and off). This controls how quickly the rotor is pulled around, thus the RPM.

No. The rate they are turned on and off is the PWM freq, or carrier freq. The commutation freq is the freq that they switch to the different blocks to PWM.

The PWM is to enable current control, nothing more. If the PWM isn't fast enough for the motor's dI/dT (current rise time), then you don't have current control. For some powerful RC motors, this means even 25-30khz is still too long of a period to keep current overshoot from hitting the kila-amp range (on a high pack voltage at least). AKA, the colossus motor, the 8800watt HXT motor etc.

Does the PWM frequency depend on the commutation frequency in an AC drive? It seems to me that is does because the faster you run the motor the higher your commutation frequency. This means thou that there is a higher back EMF (from some reading). So the PWM frequecny has to increase so that it is "on" more and thus a higher voltage is seen on the phases?

No. The PWM freq is the same always. If you're stalled starting from a stop, or if you're at the very last stage of current limiting, the PWM freq is always the same. What changes is the duty cycle % of the PWM waveform. This percentage is what bucks the phase voltage down by roughly exactly the inverse of the percentage of PWM.

Edit: So the Kv must relate to the Back EMF then?
Edit 2: Figured out it does, Kv is an easy way to understand back EMF.

Yes.

 

[Nuts&Volts]

Thanks for letting me know I'm wrong. After doing some more reading and thinking. I believe I have it.

So you have 3 blocks of FETs in your controller. Each connected to a phase on your motor. The PWM freq is a set value which turns the FET "on" a set number of times per second. The commutation freq tells when to switch to the next block. The duty (or amount of time the PWM stays "on" each cycle) is used to control the applied voltage on the motor and in turn controls the current through the phase. The commutation freq can be set by throttle position and is used to control the block switching->magnetic field speed->motor speed.

In general it seems that to increase the speed of my motor the commutation freq must be increased AND the duty cycle must be increased because the additional torque needed to get to a higher speed.

Also it then makes sense that if you increase you input voltage you can lower your duty cycle for the same motor speed?

Pardon my redudancy and thanks for the help
Kyle

 

[Jeremy Harris]

Thanks for letting me know I'm wrong. After doing some more reading and thinking. I believe I have it.

So you have 3 blocks of FETs in your controller. Each connected to a phase on your motor. The PWM freq is a set value which turns the FET "on" a set number of times per second. The commutation freq tells when to switch to the next block. The duty (or amount of time the PWM stays "on" each cycle) is used to control the applied voltage on the motor and in turn controls the current through the phase. The commutation freq can be set by throttle position and is used to control the block switching->magnetic field speed->motor speed.

In general it seems that to increase the speed of my motor the commutation freq must be increased AND the duty cycle must be increased because the additional torque needed to get to a higher speed.

Also it then makes sense that if you increase you input voltage you can lower your duty cycle for the same motor speed?

Pardon my redudancy and thanks for the help
Kyle

Almost right. The commutation frequency is only set by the trigger system that measures or detects rotor position, either zero crossing detection of the back EMF (for sensoreless control) or from Hall sensors (for sensored control).

The throttle controls the duty cycle of the PWM, the on to off ratio. This determines the effective voltage at the motor, which in turn determines rpm (because of the fairly fixed relationship between motor voltage and rpm).

Commutation is only driven by feedback from the motor itself - it's a "natural" process where the motor finds it's own speed based on load and applied voltage.

Jeremy

 

[Nuts&Volts]

That makes sense. Thanks Jeremy. I was assuming a lot of this, because I havent had too much hands on with this stuff.

Ok so now I have a question about input voltage versus motor voltage.
Lets say I have 300V battery side and that my motor has a Kv of 13.33. So how does my phase voltage relate to Kv or does it not. I have seen a controller spec that says 450V max battery side, but 320V max motor side. Would the Max RPM then be 450*13.33= 5998 RPM or 320*13.33=4265 RPM?

I know some of this has to do with RMS Voltages, but Im not sure how to apply that.

Thanks

 

[Biff]

Have a look at this datasheet page 17 bottom 3 drive outputs. (about 3/4 down the page)

http://www.onsemi.com/pub_link/Collateral/MC33033-D.PDF

you can see how instead of being on all the time, the last "on blocks" of each of those 3 signals are pulsed at some frequency rather than on continuously. That is the PWM frequency. The image shows about a 50% duty cycle because it illustrates the purpose. but as mentioned above, you varry the on / off ratio to set the amount of current through the coil, and hence torque.

To get advanced control, you can varry the PWM duty cycle during the on pulse to create more sinusoidal current waveforms, and alter the timing of those on pulses in relation to the magnetic field to compensate for the coil inductance.

-ryan