I'm doing some modeling of LV networks, using meter data (AMI) giving me instantaneous values for real and reactive power, in 5-minute intervals. I am seeing a lot of negative reactive power, which I am interpreting as indicating leading power factor. This is not what I expected to see. This is happening on residential customers without solar. Has anyone done some investigation into typical power factor in an urban setting for residential customers, with and without solar, over a typical day?

  • 1
    \$\begingroup\$ What is "a lot"? Can you share some if this data? \$\endgroup\$ – Jeroen3 Feb 23 '18 at 6:48

You found that modern homes behave like capacitive loads. That does not surprise me at all. A "lagging" power factor is caused by big inductive loads (ie. electric motors). Look around your house now - how many big electric motors do you have?

Motors at my home:

  • air conditioning (has a VFD, so it is an SMPS with capacitors, which can behave like a capacitive load)
  • washing machine (mine has a BLDC, so it also has a VFD to drive it)
  • fridge (this one may actually have a "bare" motor so it can be an inductive load)
  • some small fans (same as the fridge)

Now look at all other loads, especially the ones with their own power supplies:

  • TV - big fat power supply with capacitors
  • computers
  • LED lighting
  • chargers
  • routers? switches?
  • everything that has a power supply "brick"

They all behave like capacitive loads and are plugged in all the time, while my fridge cycles and other motors have their own power supplies. Of course the power supplies must have PFC, but I guess that it makes them just "less bad" than "perfectly resistive". When you add my fridge power factor against everything else it turns out that my house behaves like a capacitive load.

  • \$\begingroup\$ Or.. the data I have has some problems. I'm not convinced its just the switch mode power supplies looking like capacitive loads. I'll do a bit more analysis of the data. Thanks \$\endgroup\$ – Bill Tarlinton Feb 26 '18 at 21:15

I believe that the only way for a load to have a leading power factor is to have an excess of power factor correction, or perhaps LED lighting that uses capacitors to limit the LED current. With rectifiers, electronic voltage buck and boost circuits and other electronic power conversion circuits the low power factor is due to harmonic currents drawn by those loads. Perhaps the negative power factor reading of the instrumentation is due to harmonic current rather than leading load current. Excessive filtering to eliminate harmonics could also cause a leading power factor.

Power Factor and Harmonics

Power factor can be defined in two ways: The total power factor is true power / true RMS VA. The displacement power factor is true power / fundamental VA. Phase angle has meaning only for the fundamental components of the waveform.

The relationship among real power, reactive volt-amperes and apparent power is often illustrated using a "power triangle." The effect of harmonic distortion can be illustrated by adding a third dimension for form a "power box" as shown below. The effect of distortion VA is similar to the effect of reactive VA, but they can not simply be added together. They add vectorial as shown. In considering the effect on the total system, the VA from all of the undistorted load currents can be added together. All of the distortion VA components must be added together separately.

enter image description here

  • \$\begingroup\$ Thanks Charles - can the distortion send the apparent power "south of real power axis" - ie create a leading power factor? Can the distortion actually act as a power factor correction mechanism, or does it only move along the z-axis as in your box above? \$\endgroup\$ – Bill Tarlinton Feb 26 '18 at 21:23
  • \$\begingroup\$ Distortion can not act to correct the displacement power factor. It only acts on the Z axis. \$\endgroup\$ – Charles Cowie Feb 27 '18 at 0:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.