# What's the point of converting AC to DC and then back to AC?

1. Feb 24, 2012

### Femme_physics

This is from wikipedia's article about Variable-Frequency Drive

[snip]
Controller:
Variable frequency drive controllers are solid state electronic power conversion devices. The usual design first converts AC input power to DC intermediate power using a rectifier or converter bridge. The rectifier is usually a three-phase, full-wave diode bridge. The DC intermediate power is then converted to quasi-sinusoidal AC power using an inverter switching circuit. The inverter circuit is probably the most important section of the VFD, changing DC energy into three channels of AC energy that can be used by an AC motor.
[/snip]

I wonder what's the point of doing such a thing? Why not leave the AC current as it originally is? Is is purely a way to regulate the current?

2. Feb 24, 2012

### Travis_King

3. Feb 24, 2012

### M Quack

No, the point is to control the frequency. Once you have DC power, you can pretty much drive the inverter at any frequency you want.

4. Feb 26, 2012

### Femme_physics

Thank you, that pretty much covers it-- in short and in depth as well :)

5. Feb 26, 2012

### OldEngr63

This is a very common way to get motor speed control, using both an AC source and an AC motor. It has some obvious drawbacks in that the PWM waveform is not nice for the motor, but it gives you really good speed control if done well. It avoids the need for DC generation with commutation and associated problems.

6. Feb 28, 2012

### Femme_physics

Is DC power more easy to control than AC?

7. Feb 28, 2012

### M Quack

It is difficult to convert one frequency to another, but it is relatively easy to convert DC into any frequency you want (and to rectify AC power to DC).

Think of the inverter as a switch that reverses polarity. If on one side you have DC, then you can switch back and forth as you want. The faster you go, the higher the frequency on the output side. You can pick any frequency you want, there are no "hard" or "easy" frequencies.

They try to do the same thing with AC power on the input. Unless the frequencies match up, the output will be an ugly mess.

AC power is very easy to convert from one voltage to another, using a transformer. But the frequency on the input and output side of the transformer are the same.

8. Feb 28, 2012

### Travis_King

The biggest thing to understand here is that the voltage isn't really being switched back to AC before it goes to the motor. It is being inverted to mimic AC as M Quack described. A positive pulse is sent, then a "negative" pulse is sent, and this is done at whatever frequency you would have liked your AC current to be.

9. Feb 28, 2012

### OldEngr63

It is not quite as crude as Travis King's description suggests. It is not a single positive followed by a single negative followed by a single positive, etc. Rather, it is a step wise approximation to a sine wave, composed of typically six or more square steps up and down on the positive pulse and the same number down on the negative pulse.

There are problems associated with PWM drives in terms of "ringing" ie wave propagation, on the line connecting the PWM drive to the motor windings. There is a lot of very high frequency current propagting back and forth there, and when these wave re-enforce positively, the over-voltages that result can destroy the motor winding insulation.

10. Feb 28, 2012

### jim hardy

to add to oldengr's point -

high frequency can be induced into the motor shaft which eats up the bearings.

this used to be a problem only above ~500 hp, now it's in washing machines.
Fortunately it's recognized and there are ceramic bearings that electrically insulate.

11. Feb 28, 2012

### Averagesupernova

Never knew that. Interesting.