A nonlinear load put through a AC source produces current harmonics and since only the fundamental frequency carries power you can get a power factor which depends on the amplitude of the current harmonics and the amplitude of the main frequency and describes how much power is delivered to the load from the maximum power possible(if the load was linear).

That power factor is inductive,capacitive or nothing of those?Does it make sense to think of it like that?

And if it is inductive or capacitive does it change with what current harmonics are present?

 

The most basic definition of power factor is the ratio of the true power to the product of the rms Voltage and the rms Current. This accounts for all possibilities: both non-linear load and the phase shift of the current with a reactive load.

 

If a non-linear load is restive than the power factor remains at 1.
It's only inductive or capacitive loads that reduces the power factor.

Waveform distortion is a different parameter from power factor.

 

A nonlinear load put through a AC source produces current harmonics and since only the fundamental frequency carries power you can get a power factor which depends on the amplitude of the current harmonics and the amplitude of the main frequency and describes how much power is delivered to the load from the maximum power possible(if the load was linear).

That power factor is inductive,capacitive or nothing of those?Does it make sense to think of it like that?

And if it is inductive or capacitive does it change with what current harmonics are present?

Forget harmonics. A reactive load causes a shift in phasing between voltage and current, upon which a power factor is calculated.

 

If a non-linear load is restive than the power factor remains at 1.
It's only inductive or capacitive loads that reduces the power factor.

Waveform distortion is a different parameter from power factor.

Not so.
https://en.wikipedia.org/wiki/Power_factor

 

Not so.

How not so?
You can have distortion in the AC waveform without changing the power factor.
Power factor is the relation between real power and apparent power.
Distortion of the waveform is a different parameter, which may or may not affect the power factor.

 

How not so?
You can have distortion in the AC waveform without changing the power factor.
Power factor is the relation between real power and apparent power.
Distortion of the waveform is a different parameter, which may or may not affect the power factor.

Let's take a typical distorted waveform, like a transformer - rectifier - smoothing capacitor system.
The output power is the DC rectified voltage x DC current.
The input power is the output power divided by η.
Increase the size of the capacitor. The transformer primary current becomes shorter but higher spikes, and the rms current increases. So the apparent power (rms voltage x rms current) increases, but the output power remains the same.
So the power factor has reduced.

 

If a non-linear load is restive than the power factor remains at 1.
It's only inductive or capacitive loads that reduces the power factor.

Waveform distortion is a different parameter from power factor.

Actually you cannot have a non linear load that is only resistive ( excepting perhaps a single diode rectifier into a resistive load ) - your comment is mis-leading, consider an R in series with an L or C ( or both ) the power factor is not one, as the Watts/VA is not one, a power factor of one means that all the VA ends up as useful work or heat.

Now considering distortion on the mains supply - only non-linear loads create this distortion ( by their very definition ) for e.g. a FWR cap input filter - this creates flat topped mains. A resistive load ( PF = 1 ) has current that has exactly the same shape as the volts - in this case flat topped too - the current is full of harmonics ( as are the volts ) but the PF = 1, this is because the volts and currents for each harmonic are exactly in phase with each other = real power ( no displacement angle for any harmonic ).

However if we have a load that draws wiggles in its current on top of the main sinusoid ( or fundamental ) then we need to analyse every frequency component to determine whether these wiggles draw real power or watts + VA, or just watt-less VA.
If the fundamental is phase shifted this introduces phase shift loss ( displacement ) of PF too, the two types are summed vectorially to give overall distortion and hence overall power factor. Displacement power factor is only part of the equation when there are harmonics in the current.
If you have no harmonics at all then you can talk about correcting lagging PF ( electric motors ) with capacitors to move closer to PF = 1 for the total system ( ironically the caps supply local high frequency harmonics which often lead to their overheating and demise if not fitted with limiting chokes ).

 

Actually you cannot have a non linear load that is only resistive - your comment is mis-leading, consider an R in series with an L or C ( or both ) the power factor is not one, as the Watts/VA is not one, a power factor of one means that all the VA ends up as useful work or heat.

...

 

Do we have to wade through these videos to see if they are factually correct ?

 

Do we have to wade through these videos to see if they are factually correct ?

If you want too. non-linear is the relationship of voltage waveform to current waveform. A plasma (arc, beam, etc...) load that's totally resistive (ion and/or electron currents) can be non-linear, and can have dynamic negative resistance in some power ranges.