Power Factor & Correction
Solve the power triangle from any two of kW, kVA, kVAr and power factor, then size a correction capacitor bank in kVAr and in microfarads — delta and star side by side at 50 or 60 Hz, each printed with the voltage it assumed — and see the current, loss and capacity it buys, plus the parallel-resonance check that decides whether a plain bank is safe to fit at all. Fully offline.
Open Power Factor & Correction →What is the power factor calculator?
A free power factor and capacitor sizing calculator that runs entirely in your browser. The first tab solves the power triangle: give it any two of real power in kW, apparent power in kVA, reactive power in kVAr and the power factor itself, and it returns the other two along with the phase angle, the lead or lag designation, and a note saying which pair it solved from. The second tab sizes a correction capacitor bank — the rating in kVAr from Qc = P (tan φ1 − tan φ2), and then the capacitance in microfarads for a single-phase capacitor, a three-phase delta bank and a three-phase star bank at 50 or 60 Hz. Every capacitance is printed with the voltage it assumed and the words per capacitor, and the delta and star answers are always shown together because one is exactly three times the other. The third tab shows what the correction buys: line current before and after, the I²R loss reduction upstream, the transformer capacity released, and the parallel-resonance check. Nothing is uploaded.
How to use Power Factor & Correction
- Solve the triangle if you need to — If you know the load in kVA and its power factor rather than in kW, enter those two on the Power triangle tab and read off the kW. Set the lead or lag toggle — it carries the sign of the reactive power, and four of the six pairings need it for a unique answer.
- Size the bank — On the Capacitor sizing tab enter the load in kW, the existing power factor and the target. The kVAr answer needs nothing else. Add the voltage and frequency and you also get the capacitance, the voltage across each capacitor and the currents.
- Check the voltage you typed — On three-phase the voltage field is line-to-line, and the label says so. Entering a line-to-neutral voltage instead inflates the capacitance by exactly three, which is the commonest error in calculators of this kind. The bank line current is identical for delta and star, so it is a useful cross-check that nothing has been swapped.
- Enter the bank you will actually buy — Capacitor banks come in steps, so pick a step size or type the figure into Bank actually fitted. The tool then reports the power factor genuinely achieved, says LEADING plainly if it overshoots, and runs the resonance check against that real bank rather than the ideal one.
Frequently asked questions
Why does the tool insist on the line-to-line voltage?
Because the three-phase capacitance formulas are written for it. A delta capacitor sees the full line-to-line voltage; a star capacitor sees that divided by the square root of three. Since reactive power goes as the square of voltage, a star bank needs exactly three times the capacitance of a delta bank for the same kVAr. Feed a line-to-neutral voltage into a line-to-line formula and the answer is out by that same factor of three, which is why every figure here is labelled with the voltage it assumed.
What is the resonance check, and why does it matter?
A capacitor bank and the supply inductance form a parallel tuned circuit at h = √(S_sc / Qc). If that order lands near the 3rd, 5th, 7th, 11th or 13th harmonic — the ones drives, rectifiers and electronics actually produce — an ordinary harmonic current can be amplified into a destructive voltage. Enlarging a bank lowers h, so adding capacitance can walk a healthy installation onto a resonance. The estimate assumes an infinite bus, so it is an upper bound and an indicator that a harmonic study is needed, never a clearance.
Is a target of unity a good idea?
Rarely. It is arithmetically legal and the tool will compute it, but it leaves no margin: real load varies continuously, and a fixed bank sized for unity goes leading as soon as the load drops. A leading power factor raises the busbar voltage at light load and many utilities penalise it as heavily as a lagging one. Between 0.95 and 0.98 lagging is the usual recommendation, with step-switched automatic compensation wherever the load varies.
Why will it not size capacitors at a motor's terminals?
Because doing so safely depends on tables of maximum kVAr per motor rating, from NEMA MG-1 and from manufacturers, that exist to stop a motor self-exciting as an induction generator when it is switched off with capacitors still connected. Those tables could not be verified against a primary source, so none is reproduced here. A flagged gap is safer than a confidently wrong number.
Does this correct the power factor my harmonic-rich supply shows?
Only part of it. This is displacement power factor — the cosine of the fundamental phase angle. Where there is significant distortion the true power factor is lower, by the distortion factor 1/√(1 + THD²), and shunt capacitors cannot fix that difference. They can make it worse, by resonating with the harmonics already present.
Tips
- The kVAr multiplier the tool prints is the same number as the published correction charts tabulate — it is computed from tan φ1 − tan φ2 rather than transcribed, so you can check it against the chart on your wall.
- Current reduction depends only on the ratio of the two power factors: going from 0.75 to 0.95 always cuts the line current by 21.05%, whatever the voltage or the size of the load.
- The loss saving applies only upstream of where the capacitors sit. The cable between the capacitor and the load carries exactly the same current as before.