APFC Panel Sizing

Required capacitor kVAr = kW × (tan φ₁ − tan φ₂), where φ₁ is the angle of the existing power factor and φ₂ the angle of the target power factor. For a 500 kW load being improved from 0.75 PF to 0.98 PF, that gives approximately 339 kVAr — calculated from actual kW demand, never from transformer kVA.

The formula is the easy part. The engineering decision is how that kVAr is divided into steps, because the smallest step must be small enough to follow the plant's minimum reactive demand without pushing the bus into leading power factor. Size the total bank from maximum demand, and size the first step from minimum demand.

The Typical Enquiry

A customer says:

“We have a 1000 kVA transformer. Please quote a 400 kVAr APFC panel.”

But is 400 kVAr actually required?

Maybe.

Maybe the plant really needs only 200 kVAr.

Or maybe it needs 450 kVAr.

Transformer kVA alone cannot answer that question.

The correct APFC size depends primarily on:

• Actual active power demand in kW
• Existing power factor
• Target power factor
• Load variation
• Minimum and maximum plant load
• Transformer reactive-power requirement
• Existing capacitor compensation
• VFD/UPS and other nonlinear loads
• Harmonic conditions
• Future expansion

The basic power-factor correction formula is straightforward:

Required capacitor kVAr = kW × (tan φ₁ − tan φ₂)

where:

• φ₁ corresponds to the existing power factor
• φ₂ corresponds to the target power factor

But calculating one kVAr number is only the beginning.

A properly engineered APFC panel must also decide:

How should that kVAr be divided into steps?

Should it be normal or detuned?

What capacitor voltage is required?

Will the bank overcompensate at low load?

What happens when the plant operates on DG?

Are VFD harmonics present?

This guide explains the complete practical process.

First Understand kW, kVA, kVAr and Power Factor

Before sizing an APFC panel, understand the power triangle.

kW — Active Power

This is the power doing useful work.

Examples:

• Turning a pump
• Driving a conveyor
• Producing compressed air
• Running machinery

kVAr — Reactive Power

Reactive power supports magnetic fields required by equipment such as:

• Induction motors
• Transformers
• Reactors

It does not represent useful mechanical work, but many electrical loads need it to operate.

kVA — Apparent Power

This represents the combined effect of active and reactive power.

For a simple sinusoidal system:

kVA² = kW² + kVAr²

Power Factor

Power factor indicates how effectively apparent power is being converted into useful active power.

For the displacement component:

PF = cos φ

Why Does Low Power Factor Matter?

Consider two plants both consuming:

500 kW

Plant A operates at:

PF = 0.98

Plant B operates at:

PF = 0.70

Both produce the same useful active power.

But Plant B requires substantially more current and apparent power from the upstream system.

Higher current can increase:

• Cable losses
• Transformer loading
• Voltage drop
• Switchgear loading
• Distribution losses

Depending on the electricity tariff and utility rules, low PF may also have commercial implications.

The applicable utility tariff should always be checked locally because PF incentives/penalties are not identical everywhere.

What Does an APFC Panel Actually Do?

APFC means:

Automatic Power Factor Correction

An APFC panel switches capacitor steps ON and OFF according to the reactive-power requirement of the electrical system.

The controller continuously monitors the electrical condition and connects or disconnects capacitor stages to maintain the selected target PF.

The word Automatic matters.

A plant load changes continuously.

A fixed 300 kVAr capacitor connected all day may be correct at peak load and badly oversized during light load.

An APFC panel adjusts compensation as the requirement changes.

The Basic APFC kVAr Formula

Use:

Qc = P × (tan φ₁ − tan φ₂)

where:

• Qc = required capacitor reactive power in kVAr
• P = active power in kW
• PF₁ = cos φ₁ = existing PF
• PF₂ = cos φ₂ = target PF

This is the basic engineering calculation used for power-factor correction.

Rather than working out tangents every time, the bracketed term can be read off a multiplier table. Multiply the plant's kW by the figure at the intersection of existing PF and target PF:

Existing PF	Target 0.95	Target 0.98	Target 1.00
0.65	0.840	0.966	1.168
0.70	0.692	0.817	1.019
0.75	0.553	0.679	0.881
0.80	0.421	0.547	0.749
0.82	0.369	0.495	0.697
0.85	0.291	0.417	0.618
0.90	0.156	0.281	0.483
0.95	—	0.126	0.327

So 500 kW at an existing 0.75 PF, corrected to 0.98, needs 500 × 0.679 ≈ 339 kVAr. The same 500 kW plant already sitting at 0.90 PF needs only 500 × 0.281 ≈ 141 kVAr for the same target — which is why the existing PF must be measured rather than assumed.

Two things this table makes obvious. First, the last stretch is expensive: going from 0.95 to 1.00 needs a multiplier of 0.327, more than double the 0.156 needed to go from 0.90 all the way to 0.95. Second, these are fundamental-frequency figures only — they say nothing about harmonics, minimum load or capacitor voltage, all of which still have to be checked before the bank is ordered.

Example 1 — 500 kW Load, PF 0.75 to 0.98

Suppose:

Actual load = 500 kW

Existing PF:

0.75

Target PF:

0.98

Using the formula:

Required compensation ≈ 339 kVAr

In practice, the engineer might then evaluate a standard bank around:

350 kVAr

subject to:

• Load profile
• Step design
• Harmonics
• Existing compensation
• Future growth

Notice something important:

The calculation used 500 kW actual demand.

It did not begin with:

Transformer = 1000 kVA → therefore 400 kVAr.

Example 2 — 100 kW Load, PF 0.80 to 0.95

Suppose:

Load = 100 kW

Existing PF:

0.80

Target:

0.95

Required capacitor compensation is approximately:

42 kVAr

A practical automatic bank could therefore be designed around the required operating range rather than automatically selecting 100 kVAr because a larger standard panel is available.

Example 3 — 750 kW Plant, PF 0.82 to 0.98

Suppose:

Plant load = 750 kW

Existing PF:

0.82

Target PF:

0.98

Approximate required compensation:

371 kVAr

A practical design might evaluate:

375 kVAr

or:

400 kVAr

depending on load variation and future requirement.

But automatically selecting 500 kVAr without studying minimum load could create unnecessary overcompensation.

Why Should You Not Size APFC Only From Transformer kVA?

This is one of the most common mistakes.

Suppose the transformer is:

1000 kVA

but the actual maximum plant demand is only:

450 kW

Sizing a large capacitor bank based only on transformer rating could result in a bank much larger than the operating load actually requires.

Transformer kVA tells you the transformer's capacity.

It does not tell you:

• Actual kW demand
• Existing PF
• Minimum load
• Reactive load profile

Therefore:

Transformer kVA is important input data, but it should not replace the actual PF correction calculation.

When Transformer kVAr Does Matter

A transformer itself consumes reactive power.

Even when downstream equipment is lightly loaded, the transformer requires magnetizing reactive power.

In some installations, engineers may therefore provide:

• A fixed capacitor at transformer secondary

or:

• Compensation through the APFC system

depending on the design philosophy.

But do not assume a universal transformer compensation percentage.

The appropriate value should preferably be based on:

• Transformer manufacturer's no-load current
• Transformer impedance/loss data
• Actual operating condition
• Project requirements

This is more accurate than simply stating:

“Always install X% of transformer kVA as fixed capacitor.”

Use Actual Maximum-Demand kW — Not Connected Load

Suppose a factory has:

Connected load = 1200 kW

but due to diversity, maximum simultaneous demand is:

700 kW

If you calculate APFC using all 1200 kW, the bank may be substantially oversized.

Use the load condition that represents the expected operating requirement.

Useful sources include:

• Utility meter data
• Energy-monitoring system
• Main incomer multifunction meter
• Historical demand data
• Load study
• Expected production profile

Connected load is not always equal to actual operating load.

Also Check Minimum Load

Maximum load tells you the total bank capacity required.

Minimum load helps determine:

The smallest useful capacitor step.

This is equally important.

Suppose:

Peak reactive compensation requirement:

300 kVAr

But at night the factory requirement falls to only:

20–30 kVAr

If the smallest step is:

50 kVAr

the controller may be unable to regulate PF properly at light load without overcompensation.

Therefore, APFC design should consider:

maximum kVAr + minimum kVAr + rate of load variation

not simply total bank size.

We have opened existing APFC panels in pharma plants where a 300 kVAr bank was built as 6 × 50 kVAr and the night-shift reactive demand was around 25 kVAr. The controller had only two possible answers — nothing, or 50 kVAr — so the plant spent every night alternating between lagging and leading PF, and the first-step contactor had worn out long before the capacitors. Our recommendation: ask for at least one week of logged kW and kVAr from the incomer meter, including a weekend or a shutdown shift, before the step schedule is frozen. Redesigning the first two steps at drawing stage costs nothing; changing them after the panel is built costs a shutdown.

What Is APFC Step Size?

An APFC bank is divided into multiple capacitor stages.

For example, a:

300 kVAr bank

could theoretically be arranged as:

Option A

6 × 50 kVAr

or:

Option B

25 + 25 + 50 + 50 + 50 + 100 kVAr

or many other combinations.

The correct arrangement depends on the load profile and controller capability.

Commercial APFC banks commonly use mixed step sizes rather than equal steps.

This is not a universal template—it simply illustrates that step arrangement is an engineering variable.

Why Smaller Initial Steps Can Be Useful

Suppose reactive demand changes gradually between:

20 kVAr and 250 kVAr

If the smallest step is 50 kVAr, regulation resolution is relatively coarse.

A smaller first step may allow the controller to match low reactive demand more closely.

Possible benefits include:

• Better PF regulation
• Less overcompensation
• Reduced switching hunting
• Better light-load operation

However, too many small steps also increase:

• Components
• Contactors
• Protection devices
• Panel space
• Cost

The correct design balances resolution and simplicity.

What Is the C/K Setting in an APFC Controller?

Traditional APFC controllers may use a C/K relationship to determine the sensitivity of capacitor switching.

Conceptually:

• C represents the capacitor-step current
• K relates to the CT ratio

Modern digital controllers may automatically detect step sizes or simplify commissioning. The practical lesson is:

CT selection and controller configuration are part of APFC engineering.

Do not install a perfectly sized capacitor bank with an incorrectly selected CT/controller setup.

Where Should the APFC CT Be Installed?

The controller needs to measure the system condition that includes both:

• Load current
• Effect of the capacitor bank

The CT location and polarity must therefore match the controller's intended connection scheme.

Incorrect CT installation or configuration produces a small set of very recognisable symptoms:

Symptom at site	Likely cause	What to check first
All steps switch ON and stay ON	CT installed downstream of the capacitor bank, so the controller never sees its own correction	CT position on the SLD — it must see load current plus capacitor effect
No steps switch at all, PF stays poor	CT ratio wrongly entered, or C/K set far too high	Controller CT ratio setting against the actual CT nameplate
Steps hunt in and out continuously	Smallest step too large for the present load, or switching delay too short	Minimum-load kVAr against smallest step size; then the switching delay
Controller reads leading PF with all steps OFF	CT polarity reversed (S1/S2 swapped)	CT secondary polarity and the phase the CT is on
PF corrects during the day, goes leading at night	Bank sized for maximum demand only; light-load resolution inadequate	Minimum-load reactive demand; a smaller first step is usually needed
Steps switch but measured PF barely improves	CT on the wrong phase relative to the voltage sensing	Phase correspondence between CT input and voltage reference

Always follow the selected APFC controller's wiring instructions.

In our FAT experience, reversed CT polarity and a CT positioned on the wrong side of the capacitor takeoff account for the large majority of "APFC controller not working" complaints — the bank itself is usually correct. We simulate a lagging load at FAT and confirm that the controller both adds and removes steps in the right direction, because a reversed CT will still switch steps and can look convincing on the display until the plant load actually changes.

Should Target Power Factor Be 1.00?

Not automatically.

It may appear logical to target:

PF = 1.000

But in a real automatic system:

• Load fluctuates
• Capacitor steps are discrete
• Measurement has tolerance
• Switching takes time

A target too close to unity can increase the risk of temporarily operating at leading power factor.

Many projects therefore use a slightly lower target according to utility/project requirements and controller philosophy.

The correct target should be selected based on:

• Utility tariff
• Client specification
• Operating profile
• Step resolution

Do not blindly program 1.000 because it looks perfect on the display.

What Is Overcompensation?

Overcompensation occurs when the capacitor bank supplies more capacitive reactive power than the load needs.

The system can then move into a:

Leading Power Factor

condition.

Possible concerns include:

• Unwanted voltage behaviour
• Generator/utility operating issues
• Utility penalty depending on tariff
• Unnecessary capacitor switching
• Poor control stability

This is another reason APFC bank sizing should consider minimum load.

A large bank can always switch steps OFF—but only if its smallest step is small enough for the actual low-load condition.

Fixed Capacitor vs Automatic Capacitor Bank

Not every capacitor needs automatic switching.

Fixed Compensation

May be useful when reactive demand is relatively constant.

Examples can include:

• Individual motor compensation
• Transformer magnetizing compensation
• Constant inductive load

provided the application is technically suitable.

Automatic Compensation

Better where plant reactive demand changes over time.

The APFC controller then adds/removes steps according to actual demand.

For a typical industrial main LT bus with varying production load, automatic compensation is usually far more flexible.

Individual Motor Capacitor vs Central APFC

Another engineering decision is where compensation should be located.

Individual Motor Compensation

Capacitor connected specifically for one motor.

Potential advantage:

Reactive current is compensated closer to the load.

But the design must ensure:

• Capacitor does not remain connected incorrectly
• Motor self-excitation issues are considered
• Switching is coordinated

Central APFC

One automatic bank at the main distribution point.

Advantages include:

• Central control
• Easier maintenance
• Flexible step switching
• Simpler monitoring

Many industrial installations use a combination depending on the electrical architecture.

What Happens to Line Current After PF Improvement?

Suppose active power remains constant.

Three-phase current is approximately related by:

P = √3 × V × I × PF

Therefore, improving PF reduces the current required to deliver the same active power.

For example:

A plant requiring:

500 kW at 415 V

will draw considerably more line current at PF 0.70 than at PF 0.98.

This is why PF correction can release capacity in:

• Transformer
• Cable
• Busbar
• Switchgear

provided those assets were limited by current loading.

But APFC does not reduce the actual kW energy required by the mechanical process.

A 100 kW motor does not suddenly become a 70 kW motor because capacitors were added.

Does APFC Reduce Electricity Consumption?

This needs a careful answer.

APFC primarily reduces reactive-power flow and therefore current.

Lower current can reduce:

I²R losses

in upstream cables and transformers.

So some technical-energy loss reduction can occur.

But a capacitor bank does not directly reduce the useful mechanical kW consumed by a correctly operating motor.

Therefore, avoid sales claims such as:

“Install APFC and your electricity consumption will fall by 30%.”

unless supported by a detailed study.

The financial benefit may instead come from:

• Reduced utility reactive charges/penalties
• Improved capacity utilization
• Lower distribution losses

depending on the site and tariff.

How Does Capacitor kVAr Change With Voltage?

This is very important when selecting actual capacitor units.

Capacitor reactive power varies approximately with the square of applied voltage.

Qm = (fm/fn) × (Um/Un)² × Qn

where:

• Qn = rated capacitor kVAr
• Un = rated capacitor voltage
• fn = rated frequency
• Um = actual voltage
• fm = actual frequency
• Qm = actual available kVAr

Therefore:

A capacitor rated:

25 kVAr at 440 V

does not necessarily deliver 25 kVAr when operated on a 415 V bus.

This is a frequent quotation mistake.

Example: 440 V Capacitor on 415 V System

Suppose:

Capacitor nameplate:

25 kVAr at 440 V

Actual system:

415 V, 50 Hz

At the lower voltage, actual capacitor output will be lower than its 25 kVAr nameplate rating.

Therefore, if the project requires:

300 kVAr effective correction at 415 V

do not simply count:

12 × 25 kVAr = 300 kVAr

without considering the capacitor's rated voltage and actual operating voltage.

The effective kVAr at the bus must be checked.

Applying the (Um/Un)² relationship to a 440 V unit on a 415 V bus gives a factor of approximately 0.89, so those twelve units deliver closer to 267 kVAr than 300 kVAr at the bus. This is the single most common reason a commissioned bank falls short of its nameplate. It is also why comparing two APFC quotations on kVAr alone is misleading: one bidder may be quoting nameplate kVAr at 440 V and the other effective kVAr at the actual bus voltage, and the second offer will look more expensive for the same delivered compensation. Our recommendation: state "effective kVAr at 415 V" in the RFQ and ask each bidder to show the capacitor rated voltage on the offer.

This Becomes Even More Important in Detuned APFC

In a detuned APFC panel:

Reactor + capacitor

operate as a matched system.

The reactor increases the fundamental-frequency voltage appearing across the capacitor.

Therefore the capacitor may need a higher voltage rating.

Also:

capacitor nameplate kVAr ≠ necessarily effective bank output at the main bus

The complete reactor-capacitor combination must be designed for the required effective compensation.

This was covered in detail in the related WTS guide:

Normal APFC vs Detuned APFC — 7% vs 14% Reactor

Should APFC Be Normal or Detuned?

Calculating:

339 kVAr

does not tell you whether the bank should be normal or detuned.

That requires a harmonic assessment.

A normal APFC can be suitable in an electrical network with sufficiently low harmonic pollution.

Where significant nonlinear loads exist, evaluate:

• VFDs
• UPS
• Rectifiers
• Chargers
• Welding systems
• Other power electronics

A detuned capacitor/reactor system may be required to reduce resonance risk.

Do not calculate the kVAr correctly and then ignore harmonics.

How Much VFD Load Requires a Detuned APFC?

There is no universal rule such as:

“If VFD load exceeds 20%, use 7% reactor.”

Different manufacturers publish application guidance, but the technically strongest approach is to evaluate:

• Nonlinear load percentage
• Actual harmonic spectrum
• THDv
• THDi
• Transformer impedance
• Existing capacitors
• Network configuration

A harmonic measurement or study provides more reliable information than one generic percentage threshold.

What About APFC on DG Supply?

This deserves careful attention.

Suppose a factory normally runs on utility power with:

300 kVAr APFC

During utility failure, the same plant transfers to:

500 kVA DG

Should the entire 300 kVAr APFC bank remain available?

Not automatically.

Generator systems can be more sensitive to:

• Leading PF
• Voltage regulation
• Reactive-power flow

The APFC philosophy should consider:

• DG rating
• DG loading
• Generator AVR
• Minimum load
• Reactive-power capability
• Operating PF requirement

Possible control strategies include:

• Limit APFC steps during DG mode
• Block selected steps
• Use separate PF setpoint
• Coordinate through PLC/DG controller

The exact strategy should be agreed with the generator OEM/project designer.

This is the item we push back on at clarification stage, because APFC panels are routinely specified with no mention of DG mode at all. Consider the case above: 300 kVAr of capacitors available on a 500 kVA generator carrying perhaps 200 kW of essential load overnight. If the full bank stays enabled, the machine can be pushed into leading PF, and generator AVRs generally regulate voltage far less comfortably in that region than a utility supply does. A simple DG-mode digital input to the controller — selecting a second, lower PF setpoint and blocking the larger steps — costs very little at manufacturing stage and is close to impossible to retrofit tidily once the panel is on site.

APFC and Solar Inverters

Modern industrial plants increasingly contain:

• Grid-connected solar
• BESS
• Active-front-end converters

These systems can change the reactive-power profile seen at the utility incomer.

For example:

Plant motor load may remain constant.

But solar generation reduces the net imported kW.

If the APFC controller and reactive-power strategy are not coordinated properly, the plant's operating PF at the grid incomer can behave differently from the pre-solar condition.

Therefore, large solar/BESS plants should review the overall:

• kW flow
• kVAr flow
• inverter PF capability
• APFC operation
• utility requirements

rather than assuming the old APFC settings remain optimal.

How Many APFC Steps Should Be Used?

There is no universal answer.

Step count depends on:

• Total bank size
• Minimum reactive requirement
• Load variation
• Controller capability
• Switching frequency
• Available space
• Cost

For example:

A 50 kVAr bank may need only a few steps.

A 1000 kVAr bank may require considerably more resolution.

More steps generally allow finer correction.

But they also mean:

• More contactors
• More protection devices
• More wiring
• More panel size
• More maintenance points

The best design balances control resolution and simplicity.

Step Ratio Examples

Possible arrangements can include:

1 : 1 : 1 : 1

equal steps

or:

1 : 1 : 2 : 2

or other combinations.

There is no universal best ratio.

If the smallest step is:

25 kVAr

the controller may create combinations such as:

25 50 75 100 125

150...

depending on the physical step arrangement and controller strategy.

The correct design should follow the measured load profile.

Example — 400 kVAr Bank With Variable Load

Suppose the factory requires:

Peak compensation:

390 kVAr

Minimum operating compensation:

35 kVAr

A simple:

4 × 100 kVAr

bank would be poor at light load.

A more flexible bank might provide smaller first steps.

For example, conceptually:

25 + 25 + 50 + 100 + 100 + 100

= 400 kVAr

This is only an illustration.

The final step arrangement should be based on:

• Load data
• Switching frequency
• Controller
• Component ratings
• Harmonic design

Rapidly Changing Loads May Need Faster Switching

Conventional APFC panels often use capacitor-duty contactors.

These are appropriate for many industrial applications.

But some loads change very rapidly.

Examples:

• Welding
• Cranes
• Rapid cyclic machinery
• Certain process equipment

If reactive demand changes faster than conventional contactor switching can practically follow, engineers may consider:

• Thyristor-switched capacitor banks
• Static PFC
• Hybrid systems

depending on the application.

This should not be confused with simply increasing the number of contactor steps.

Why Capacitor-Duty Contactors Matter

Capacitor energization can create high inrush current.

Therefore, capacitor switching should use devices suitable for capacitor duty rather than selecting an ordinary motor contactor purely from steady-state current.

This improves control of:

• Switching inrush
• Contact wear
• Reliability

The switching device must be coordinated with the capacitor manufacturer's requirements.

Capacitor Bank Current

For a three-phase capacitor bank, approximate current is related by:

I ≈ Q / (√3 × V)

where:

• Q = capacitor reactive power in VA reactive
• V = system line voltage

But do not size:

• Busbars
• Contactors
• Cables
• Breakers

at exactly the nominal capacitor current without considering applicable capacitor tolerances, harmonic current, equipment manufacturer guidance and standards.

PFC components often require additional current capability because real capacitor-bank operating current may exceed the ideal fundamental nameplate value.

APFC Incomer Sizing

The APFC incomer should be selected based on the complete capacitor-bank electrical duty.

Consider:

• Total effective kVAr
• System voltage
• Capacitor current
• Harmonic current
• Component tolerance
• Short-circuit level
• Switching duty
• Applicable manufacturer/standard factors

Do not select the APFC incomer using:

300 kVAr → calculate ideal current → choose the next breaker rating

without checking actual capacitor-bank duty.

Protection of Each Step

Each capacitor step requires suitable protection.

Depending on architecture, this can include:

• Fuse
• MCCB
• MCB/specialized protection
• Contactor
• Reactor thermal protection in detuned systems

Protection should account for:

• Capacitor inrush
• Continuous current
• Fault current
• Harmonics
• Manufacturer recommendations

A capacitor bank is not simply another resistive feeder.

Ventilation and Temperature Matter

Capacitor life is highly sensitive to thermal conditions.

A large APFC panel may contain:

• Capacitors
• Contactors
• Reactors
• Busbars

all generating heat.

A detuned APFC panel has even greater thermal considerations because reactors add significant losses.

Therefore, panel engineering should consider:

• Ambient temperature
• Internal heat
• Natural ventilation
• Forced ventilation
• Fan redundancy where required
• Thermostat
• Reactor thermal trips

Do not finalize enclosure size only from component physical dimensions.

How Should APFC Be Sized for Future Expansion?

Suppose present requirement is:

300 kVAr

but the factory expects substantial future expansion.

Should you install:

500 kVAr today?

Not necessarily.

Options may include:

Option 1

Install 300 kVAr now with physical space for future stages.

Option 2

Install larger busbar/incomer capability with only current required stages.

Option 3

Install part of future capacity where growth is imminent and minimum load allows it.

The correct decision depends on:

• Expansion timeline
• Minimum current load
• Capex strategy
• Panel space
• Future harmonic conditions

Oversizing an installed capacitor bank merely for possible future demand can reduce control quality today.

Practical APFC Calculation Workflow

Use this sequence.

Step 1 — Collect Actual Load Data

Measure:

• kW
• kVA
• kVAr
• PF
• Maximum demand
• Minimum demand

preferably across representative production conditions.

Step 2 — Establish Existing Power Factor

Example:

0.78

Step 3 — Establish Target Power Factor

Example:

0.98

according to utility and project requirement.

Step 4 — Calculate Required Fundamental kVAr

Use:

Qc = P × (tan φ₁ − tan φ₂)

Step 5 — Check Existing Compensation

Subtract any healthy existing capacitor compensation where appropriate.

Step 6 — Review Minimum Load

Determine smallest useful step.

Step 7 — Review Harmonics

Decide whether:

• Normal APFC
• Detuned APFC
• Additional harmonic mitigation

is required.

Step 8 — Select Capacitor/Reactor System

Check:

• Effective output
• Capacitor voltage
• Reactor %
• Harmonic duty

Step 9 — Design Step Combination

According to load variation.

Step 10 — Design Panel

Select:

• Incomer
• Busbar
• Protection
• Switching
• Cooling
• Controller
• Metering

Step 11 — FAT

Verify the complete automatic sequence.

A Realistic Calculation Example

Suppose an industrial plant records:

Maximum active demand = 600 kW

Existing PF:

0.78

Target PF:

0.98

The basic PF calculation gives the required reactive compensation.

But before ordering the bank, also collect:

• Minimum load = 180 kW
• VFD load = 250 kW
• Transformer = 1000 kVA
• Existing capacitor = 50 kVAr
• Plant operates on DG during utility failure
• Harmonic measurement available

Now the engineering team can decide:

1. Actual net additional kVAr
2. Normal vs detuned
3. Step sizes
4. DG blocking philosophy
5. Capacitor voltage
6. Panel ventilation

This is much stronger engineering than:

1000 kVA transformer → 400 kVAr APFC

Common APFC Sizing Mistakes

Mistake 1: Sizing From Transformer kVA Alone

Actual kW and PF are more important for the main compensation calculation.

Mistake 2: Using Connected Load Instead of Actual Demand

This can oversize the bank.

Mistake 3: Ignoring Minimum Load

This leads to overcompensation and poor step resolution.

Mistake 4: Targeting PF = 1.000 Blindly

A practical automatic system should avoid unnecessary leading PF operation.

Mistake 5: Forgetting Existing Capacitors

You may install more kVAr than required.

Mistake 6: Assuming Nameplate Capacitor kVAr Is the Same at Every Voltage

Capacitor kVAr changes approximately with voltage squared.

Mistake 7: Selecting Detuned Capacitor and Reactor Independently

They should be engineered as a matched combination.

Mistake 8: Ignoring Harmonics

Correct kVAr with the wrong harmonic design can still produce a bad APFC system.

Mistake 9: Using Equal Large Steps for a Highly Variable Load

Light-load correction may become poor.

Mistake 10: Ignoring DG Mode

The compensation strategy may need to change when running on generators.

Mistake 11: Assuming APFC Automatically Saves Large Amounts of kWh

Its primary purpose is reactive-power compensation, not reducing the process's fundamental active-power requirement.

Mistake 12: Ignoring Solar/BESS Effects

Net grid-import conditions can change even when plant loads remain unchanged.

APFC RFQ — What Information Should You Provide?

Instead of:

“Need 300 kVAr APFC panel.”

provide:

Electrical System

• Voltage
• Frequency
• Transformer kVA
• Transformer impedance
• Fault level
• DG operation
• Solar/BESS connection

Load Data

• Maximum kW
• Minimum kW
• Existing PF
• Target PF
• Existing kVAr
• Connected load
• Maximum demand

Harmonics

• Total VFD load
• UPS/rectifier load
• THDv
• THDi
• Harmonic spectrum if available

APFC Requirement

• Required effective kVAr
• Normal/detuned
• Step arrangement if specified
• Controller requirement
• Communication
• Approved makes

Installation

• Ambient temperature
• Indoor/outdoor
• IP rating
• Cable entry

Quality

• Applicable standards
• FAT
• Customer witness
• TPI
• Documentation

This produces much more comparable quotations.

Indian Standard Specifically for APFC Panels

This is particularly important for WTS and Indian buyers.

BIS currently lists:

IS 16636:2017 — Automatic Power Factor Correction (APFC) panels for voltage rating up to and including 1000 V

and notes that the standard was reviewed in 2022.

This is highly relevant to Indian APFC panel specifications and should be considered alongside applicable component and assembly standards.

IEC Standard for LV PFC Banks

IEC currently lists:

IEC 61921:2017 — Power capacitors — Low-voltage power factor correction banks

The standard applies to LV AC shunt capacitor banks intended for power-factor correction and notes that, where applicable, such banks comply with relevant IEC 61439-1 and IEC 61439-2 requirements.

Indian Standard for Power Capacitors

BIS currently lists:

IS 13340 (Part 1):2024, adopting IEC 60831-1:2014 for applicable self-healing AC shunt power capacitors.

Therefore, an Indian APFC specification can be much more precise than simply writing:

“APFC panel as per IEC.”

The relevant panel, capacitor and assembly requirements can be identified deliberately.

APFC FAT — What Should Be Checked?

A proper FAT should not end after manually switching all capacitor contactors ON.

Document Verification

Check:

• GA
• SLD
• BOM
• Datasheet
• Step schedule
• Capacitor/reactor data

Components

Verify:

• Capacitor make/model
• Rated voltage
• Rated kVAr
• Effective bank kVAr
• Capacitor-duty contactor
• Reactor where applicable
• Protection
• APFC controller
• CT

Wiring

Check:

• Power circuit
• Controller wiring
• CT polarity
• Phase sensing
• Step sequence
• Reactor thermal contacts

Functional Tests

Simulate:

• Lagging PF
• Step addition
• Further reactive demand
• Step removal
• Leading PF condition
• Alarm
• Manual/auto
• Fan/temperature logic

Electrical Tests

As applicable:

• Insulation resistance
• HV/dielectric
• Continuity
• Internal wiring verification

The controller should demonstrate that it can add and remove compensation logically, not merely energize every contactor.

WTS In-House APFC Testing

Wisdom Techno Solutions performs applicable in-house panel FAT using calibrated testing instruments and equipment.

Depending on project scope, testing can include:

• HV/dielectric testing
• Insulation resistance testing
• Protective continuity
• Busbar/current-path related checks
• Internal wiring verification
• APFC controller operation
• Step switching
• Interlocks
• Metering
• Alarm functions
• Reactor thermal-protection logic
• Cooling-system operation

We support customer or consultant-witnessed FAT.

Where required, we also facilitate Third Party Inspection according to the project Inspection and Test Plan.

How Wisdom Techno Solutions Approaches APFC Sizing

At Wisdom Techno Solutions, an APFC enquiry should ideally begin with the customer's actual electrical data rather than a preselected standard kVAr panel.

Wisdom Techno Solutions is a licensed partner for three design-verified switchgear platforms: Rittal Ri4Power (Rittal Solutions Partner since 2021 — the Ri4Power system carries Intertek ASTA design verification to IEC 61439-1 and -2 Edition 3, with copper and aluminium busbar systems up to 6300 A), C&S CX (Licensed Partner since 2023, up to Form 4B Type II) and Siemens SIEPAN Elite/8PU (IEC 61439 Solutions Partner since 2024). We build to each system's verified configuration and routine-verify every assembly in-house at our Vadodara, Gujarat facility, for industrial, EPC and infrastructure projects across India. That places the capacitor-bank busbar and switching-device rating inside a verified design envelope.

Useful inputs include:

• Actual maximum kW
• Minimum kW
• Existing power factor
• Required target power factor
• Transformer details
• Existing capacitor bank
• VFD/UPS loading
• Harmonic measurement
• DG operation
• Solar/BESS connection
• Future load

From this information, the required compensation can be calculated and the panel architecture can be selected.

Depending on the project, we engineer and manufacture:

• Conventional APFC panels
• Detuned APFC panels
• Multi-step automatic capacitor banks
• APFC integrated within PCC/PMCC systems
• PLC/monitoring-integrated solutions
• Customer-specific capacitor-bank arrangements

The objective should not be:

“Install the largest possible capacitor bank.”

The objective should be:

Supply the amount of reactive power the plant actually needs, in steps fine enough to follow the load, without creating leading PF, resonance or unnecessary equipment stress.

APFC Sizing Checklist

Before ordering the panel, confirm:

Load

✓ Maximum kW known ✓ Minimum kW known ✓ Existing PF measured ✓ Target PF defined ✓ Existing capacitors checked

Calculation

✓ Required kVAr calculated ✓ Transformer compensation reviewed separately ✓ Future expansion considered reasonably

Harmonics

✓ VFD loading known ✓ UPS/rectifier loading known ✓ Harmonic condition assessed ✓ Normal/detuned decision justified

Capacitor

✓ Rated voltage correct ✓ Effective kVAr checked at actual bus voltage ✓ Capacitor duty suitable

Steps

✓ Smallest step suitable for minimum load ✓ Total steps suitable for load variation ✓ Controller supports arrangement

DG / Renewable

✓ DG-mode compensation philosophy defined ✓ Solar/BESS effect considered where applicable

Panel

✓ Incomer sized ✓ Busbars sized ✓ Protection coordinated ✓ Capacitor-duty switching devices used ✓ Ventilation designed ✓ Reactor thermal protection included if applicable

Standards & Quality

✓ IS 16636 considered for Indian APFC panel ✓ Applicable capacitor standard considered ✓ FAT defined ✓ TPI requirement defined where applicable

Conclusion

Calculating APFC size is not difficult.

But sizing an effective APFC system requires more than one formula.

Start with:

Actual kW

Existing PF

and:

Target PF

Then calculate:

Qc = kW × (tan φ₁ − tan φ₂)

After that, check:

• Maximum demand
• Minimum demand
• Existing capacitors
• Transformer reactive requirement
• Harmonics
• VFD/UPS loads
• Capacitor voltage
• Normal vs detuned design
• Step sizes
• DG operation
• Solar/BESS interaction
• Future expansion

Do not size a capacitor bank from transformer kVA alone.

Do not assume the capacitor nameplate kVAr equals its actual output at every voltage.

Do not install a huge bank simply because future expansion may happen.

And do not chase:

PF = 1.000

without considering how the plant actually operates.

The correct APFC panel is not the panel with the highest kVAr.

It is the panel that can continuously provide approximately the reactive power the electrical system actually needs at each operating condition.

Planning a new APFC panel or replacing an existing capacitor bank?

Share your maximum/minimum kW demand, existing PF, target PF, transformer data, VFD/UPS loading, DG details and harmonic measurements with Wisdom Techno Solutions for project-specific APFC sizing and panel engineering.

Related Guides

• Normal APFC vs detuned APFC
• Load list and transformer/incomer sizing
• Panel heat dissipation and cooling
• Product page: APFC panel

Frequently Asked Questions

How do I calculate required capacitor kVAr?

Use Qc = kW × (tan φ₁ − tan φ₂), where φ₁ corresponds to the existing PF and φ₂ to the desired PF. The kW figure must be the plant's actual measured demand, not connected load and not transformer kVA. In practice the bracketed term is read from a multiplier table — for example 0.679 for 0.75 PF corrected to 0.98 — so the calculation reduces to one multiplication. Remember that this gives the fundamental-frequency requirement only; harmonics, minimum load and capacitor rated voltage still have to be assessed separately.

How much kVAr is required for 500 kW at 0.75 PF to improve to 0.98 PF?

The approximate requirement is 339 kVAr, from 500 × (tan φ₁ − tan φ₂) = 500 × 0.679. That is the fundamental compensation figure before adjustment. Before ordering, deduct any healthy existing capacitors, confirm the effective kVAr at the actual bus voltage rather than the capacitor nameplate voltage, and check minimum load so the smallest step is usable at light load. A practical bank might then be built at 350 kVAr with graded steps rather than as a single 339 kVAr block.

Can APFC size be calculated from transformer kVA?

No — transformer kVA is useful system information but it is the wrong basis for the main calculation. A 1000 kVA transformer may serve a plant whose actual maximum demand is only 450 kW, in which case a bank sized from the transformer rating would be far larger than the load requires. Transformer rating tells you nothing about the existing power factor, the minimum load or the reactive load profile, which are the three inputs that actually determine bank size and step arrangement. The transformer's own magnetizing reactive requirement is a separate item, assessed from its no-load current data.

Should APFC be sized from connected load or maximum demand?

Size it from actual or expected maximum simultaneous demand, not from total connected load. Diversity means a factory with 1200 kW connected may never exceed 700 kW at once, and calculating from the connected figure can oversize the bank by a large margin. Use utility meter data, the incomer multifunction meter, an energy-monitoring system or a load study as the source. Oversizing does not just waste money — it worsens light-load control because the steps end up too coarse.

Why is minimum plant load important?

Minimum load determines the smallest capacitor step the bank needs. If peak compensation is 300 kVAr but the night-shift requirement falls to 20–30 kVAr, a bank whose smallest step is 50 kVAr cannot regulate at all at light load — it must either supply nothing or overcompensate. The result is continuous hunting between lagging and leading PF, premature contactor wear and, depending on tariff, a leading-PF penalty. Maximum demand sets total bank capacity; minimum demand sets the first step.

Should APFC target power factor be 1.0?

Not automatically. Very aggressive unity-PF targets can increase the chance of temporary leading PF as load changes. Target PF should follow utility/project requirements and the available step resolution.

What is leading power factor?

Leading power factor occurs when the capacitive reactive power supplied by the bank exceeds the inductive reactive power the load actually needs. It is most common at light load, when the smallest available capacitor step is larger than the plant's remaining reactive demand. Consequences can include unwanted voltage rise, unstable controller switching, difficulty for generator AVRs during DG operation, and a utility penalty depending on the tariff. Avoiding it is a step-resolution problem, not a total-kVAr problem — a large bank is fine provided its first step is small enough.

Does APFC save electricity?

Only indirectly, and usually modestly. APFC reduces reactive-current flow and therefore the total line current, which lowers I²R losses in upstream cables, busbars and the transformer — a real but generally small kWh saving. It does not reduce the active power the process itself consumes: a 100 kW motor still needs 100 kW after capacitors are added. The genuine financial benefit is normally the avoided utility reactive charge or PF penalty, plus released current capacity in existing assets, so treat claims like "APFC will cut your bill by 30%" as unsupported without a site-specific study.

Does capacitor kVAr change with voltage?

Yes — capacitor reactive output varies approximately with the square of the applied voltage, per Qm = (fm/fn) × (Um/Un)² × Qn. Operating a capacitor below its rated voltage therefore gives less kVAr than the nameplate states, and operating above it gives more, along with additional dielectric stress. This is why an APFC specification should state the required effective kVAr at the actual system voltage. In detuned banks the effect matters even more, because the series reactor raises the fundamental voltage across the capacitor and the units are normally rated higher as a result.

Will a 25 kVAr 440 V capacitor give exactly 25 kVAr at 415 V?

No. Applying the voltage-squared relationship, (415/440)² is approximately 0.89, so the unit delivers roughly 22 kVAr on a 415 V bus rather than 25 kVAr. Twelve such units therefore give about 267 kVAr effective, not 300 kVAr — a shortfall of around 11%. This is one of the most frequent quotation errors, so always confirm whether a quoted kVAr figure is nameplate at the capacitor's rated voltage or effective at the actual bus voltage.

How many steps should an APFC panel have?

There is no universal number. Step count and sizes should reflect total compensation, minimum reactive demand, load variation, switching duty and controller capability.

Is 6 × 50 kVAr better than mixed steps for a 300 kVAr panel?

Not necessarily. Equal steps are simple, while mixed smaller/larger stages can provide finer control across a wider load range.

Do VFD loads require detuned APFC?

Not automatically. The decision should consider actual nonlinear loading, harmonics, transformer impedance and resonance risk.

Can an APFC panel remain connected when the plant runs on DG?

Potentially, but the compensation strategy should be coordinated with generator loading, AVR behaviour and project/OEM requirements. Full capacitor capacity may not always be appropriate during DG operation.

What is IS 16636?

IS 16636:2017 is the BIS standard titled Automatic Power Factor Correction (APFC) panels for voltage rating up to and including 1000 V. BIS shows it as reviewed in 2022.

Which IEC standard applies to LV PFC capacitor banks?

IEC 61921:2017 applies to low-voltage AC shunt capacitor banks intended for power-factor correction. It covers the capacitor bank as an assembled product — ratings, losses, service conditions and testing — and notes that where applicable such banks also comply with the relevant IEC 61439-1 and IEC 61439-2 requirements. An APFC specification is therefore clearest when it names the bank standard, the capacitor standard and the assembly standard separately rather than writing "as per IEC".

What Indian standard applies to self-healing power capacitors?

BIS currently lists IS 13340 (Part 1):2024, which adopts IEC 60831-1:2014 for self-healing AC shunt power capacitors up to 1,000 V. It covers the capacitor units themselves — ratings, tolerances, losses and tests — rather than the panel that houses them. For a complete Indian APFC specification it is normally paired with IS 16636:2017 for the APFC panel and the IEC 61439 series for the assembly requirements.

What information should I provide for APFC quotation?

Provide maximum and minimum kW, existing PF, target PF, transformer kVA and impedance, details of any existing capacitor bank, VFD/UPS loading, harmonic measurements if available, whether the plant runs on DG or has solar/BESS connected, system voltage and the project specification. Minimum kW and the harmonic picture are the two inputs most often left out, and they are precisely the ones that decide the step schedule and the normal-versus-detuned decision. State the required kVAr as effective at the system voltage so that offers using different capacitor rated voltages remain comparable.