Output voltage of a VFD

[bpk]

I was always under the assumption that a VFD voltage output stayed constant, but today I was talking to coworker saying that the output voltage varies with the output frequency, is this true?

 

[winnie]

For all practical purposes, the output of a VFD is also variable voltage, and the voltage is controlled to match the operating state of the motor. To a rough approximation, the output voltage of the drive is proportional to frequency.

Note that there is some truth to both views.

The most common VFDs take the input AC, rectify it to DC, and then switch this DC to produce the output.

If you look at an output terminal relative to the DC rail neutral (which itself is approximately AC source neutral, and approximately ground for a grounded system), you will find a variable duty cycle square wave where the peak to peak value of the square wave is the internal DC voltage. Since this internal DC voltage is approximately constant, the amplitude of this square wave is approximately constant.

But the voltage actually 'seen' by the motor is not the square wave described above, but instead the voltage _between_ the inverter output terminals. The modulation of the square wave duty cycle is what varies the voltage seen by the motor, and this modulation both creates the sine wave voltages actually seen by the motor, and changing the modulation allows adjustment of both the frequency and voltage seen by the motor.

-Jon

 

[Besoeker]

I was always under the assumption that a VFD voltage output stayed constant, but today I was talking to coworker saying that the output voltage varies with the output frequency, is this true?

It is true*. The V/f ratio is constant.
The voltage has to vary with frequency to keep the magnetic flux constant.
Flux is the product of voltage time time. At lower frequencies, the time period is longer. If you didn't reduce the voltage, the flux would increase. This would cause magnetic saturation resulting in very high currents.

*If the motor runs above base speed (60Hz in your case) the frequency increases but the voltage stays the same as it was at 60Hz - the VFD can't give any more.
Many VFDs also have the option of a pump or fan fluxing curve where the V/f ratio changes with frequency to get better efficiency.

 

[bpk]

So lets say I had a VFD powered with 480 incoming voltage and the VFD was controlling a 480v motor. Would the motor ever have more than 480v applied to it across the drives frequency range?

 

[winnie]

In general, no. With some VFDs, you might get a few % over 480V on the output, but no more.

(With the caveat that the DC switching voltage described above is being ignored.)

-Jon

 

[Jraef]

In general, no. With some VFDs, you might get a few % over 480V on the output, but no more.

(With the caveat that the DC switching voltage described above is being ignored.)

-Jon

And therein lies the probably source of confusion on this issue, I have seen it a lot. An AC drive's output is a PWM signal, Pulse Width Modulation. It is not true AC, it is a series of DC pulses in strings, first positive, then negative. Each pulse is ALWAYS the same voltage potential, that of the DC bus. So in a 480V drive for example, the DC bus voltage is the same as the peak voltage of the incoming sine wave, or 480 / .707 = 678V. But an inductive load, such as a motor, cannot really respond to a DC pulse very fast because of hat is called the Inductive Time Constant (look it up), so by the time the pulse has an effect on the motor magnetic field, it has gone away and a new pulse has replaced it. That string of pulses then gets "averaged" out by the motor's reaction to it into a pseudo sine wave, called an RMS voltage. This is very close to looking like AC, in fact so close as to make the motor work fine. So then the rate at which the strings of pulses change direction (+-) determines the output frequency, the width of the pulses determines how high that RMS pseudo AC voltage gets in the motor. Combine them and you can change speed and maintain torque.

But STRICTLY speaking, the pulses that make up the PWM output are in fact all the same "voltage", albeit DC. How the motor reacts to is, essentially the effective RMS AC voltage, is what is changing. So what happens some times is that someone will get a cheap meter that cannot properly interpret the PWM signal, then leave it on "Peak" reading, and come to the conclusion that the output voltage never changes. but what's happening is that they are reading the DC pulses.