In my current electrical engineering course the concept of power factor has been a large topic, defined as the ratio of real power to apparent power. It is a measure of resistive power consumption in a circuit. Other kind of loss present in the circuit is the reactive or "parasitic" power.

In my material it is presented as a quantity that always should be maximized. But what I don't understand is why resistive power consumed is always a good measure of efficient use of power? There are lots of components where reactive power consumption is desirable, like transformers. If I connect a power source to a transformer, it would have mainly reactive losses and resistive power loss would actually be undesirable. But still, if we use power factor as a measure of circuit performance, it would be poor in this case as we would have mainly "parasitic" losses.

Not that I think of it, I can't really come up with many applications where purely resistive losses are good (in a resistive heater for one). So what do we really consider as consuming "real power" in this sense? What do we count as consuming "real" power and "parasitic"/reactive power when the circuit receiving power is more complicated than a simple example circuit consisting just a resistor, capacitor and an inductor?

## Best Answer

Looking at it from outside the box, a device with a power factor lower than unity causes power to be wasted in the distribution of the power between the generation and the device itself. That's assuming resistive losses in the wiring dominate, so the \$I^2R\$ losses are higher than the minimum for a given amount of real power that has to be delivered if there is a reactive component. If there is a lot of reactance, the related energy sloshes back and forth between generator and device each cycle, causing unnecessary losses along the way, and not actually transferring any average power to the devices.

What goes on inside the device is another question altogether, but it certainly needs to receive at least the real power that it consumes (assuming no energy storage) to satisfy conservation of energy.

So if you were designing, say, a large switchmode power supply, you would optimize the power conversion and regulation portion to minimize the losses in the supply (within all your other constraints), but would also likely add power factor correction so that the input current is minimized in normal operation, as much as practical, but certainly enough to meet any relevant standards.