PDA

View Full Version : PFC thoughts

ashiekh
03-13-2018, 02:53 PM
What current profile gives the best Power Correction Factor (the least losses)?

A sine wave in-sync with the voltage?

ridgid13579
03-14-2018, 10:52 AM
What current profile gives the best Power Correction Factor (the least losses)?

A sine wave in-sync with the voltage?

Yes, when the current is in phase with the voltage, you will have a power factor of 1, which is the most ideal.

A resistive load (say a good old wire heater) should have a power factor of 1.

ashiekh
03-14-2018, 11:13 AM
I tried a few current shapes, and so far none have beat the sine wave for losses. I even think I now have a proof of why the current needs to follow the voltage (for minimum losses), that works for all voltage shapes (not just sinusoidal); using ideas from functional analysis.

A strange exception to a resistive load being ideal:
Imagine a bulb whose filament resistance increases with temperature, it is pure resistive, but has higher resistance at high voltages, and so will have a non-sinusoidal current curve and so higher cable loses than an equal load with constant resistance.

Interestingly, the larger the primary capacitors, the worse the power factor issue (assuming no correction), due to reduced conduction angle; by the same reasoning the smaller the load the worse the power factor issue.

gdjacobs
07-11-2018, 03:07 PM
Well, P = I dot V. If the objective is to minimize I while P and V are constant, obviously the phase between any components should be zero. Also, any harmonics from a non sinusoidal current waveform will be orthogonal to a sinusoidal voltage waveform and will have no contribution to the overall power output and be waste as well.

P = sum ( I_i V_j )
= sum ( |I_i| |V_j| int sin w_i t sin w_j t dt )
= |I_i||V_i| int sin w_i t sin w_i t dt

ashiekh
07-11-2018, 03:14 PM
Nicely argued