Review of Capacitor Bank Control Practices

Ilia Voloh, Thomas Ernst – GE Grid Solutions

I. Introduction

This paper reviews capacitor bank control options available in modern capacitor bank protective
relays and common utility practices for the control of single or multiple shunt capacitor banks. The
purpose of the capacitor bank control is to switch the capacitor bank on or off in order to provide
necessary reactive power or voltage support. There are, however, important considerations when
planning and designing capacitor bank control schemes:
- Manual (local or remote) versus automatic control
- Automatic control based on time/season or measurement based automatic voltage
regulation (AVR)
- AVR based on VARs, current, voltage or power factor
- Voltage drop compensation when important loads are located remotely from the bank
measurement point
- Ability to lockout the bank until capacitors discharge to the safe level to energize bank
again
- Ability to control the number and frequency of bank operations
- Ability to control bank closing to minimize transients
- Ability to coordinate the control of multiple banks with varying closing sequences
This paper examines the options above for switched capacitor banks and illustrates advantages and
disadvantages of each option to help engineers choose the optimum approach for their application.
The paper gives useful insights for engineers, helping them to enhance their schemes based on the
available features in modern capacitor bank relays.

II. Overview of standards and regulations

Shunt capacitor banks are important for the reliable operation of the power system. The capacitor
banks support the system voltage and reduce reactive power flow through the power system. In
general, capacitor banks benefits are following [2]:
• Voltage support.
• VARs support
• Increased system capacity
• Reduced system power losses
• Reduced billing charges
There are a multitude of standards, guides and white papers that apply to the protection of shunt
capacitor banks but there are relatively few standards or guides that apply to the control of shunt
capacitor banks. Reference [2] defines Capacitor Control as “The device required to automatically
operate the switching device(s) to energize and de-energize shunt power capacitor banks.” It also
indicates that “Shunt capacitors support the system voltage and reduce the reactive current flow

1
through the power system” and are “required when acceptable system voltages cannot be maintained
by the generators and transmission system alone.” The IEEE considers capacitor control to operate
automatically and are coordinated with generator voltage and power factor controls.
The North American Electric Reliability Corporation (NERC) reliability standard VAR-001-5 “Voltage
and Reactive Control” requires transmission owners to develop a transmission voltage schedule and
to “schedule sufficient reactive resources to regulate voltage levels under normal and Contingency
conditions”. This regulation is to be accomplished through “reactive generation scheduling,
transmission line and reactive resource switching, and using controllable load.” Reactive resource
switching is not defined in the NERC standard but it likely includes the switching of shunt reactors and
capacitors as well as the control of static VAR compensators (SVC). This scheduling will include both
manual and automatic switching and is to be coordinated with the generator voltage and power factor
controls.
While IEEE and NERC differ slightly in their requirements for shunt capacitor controls, this paper will
consider both manual and automatic control of shunt capacitor banks to regulate voltage or power
factor and will discuss coordination with generator controls and with transformer tap changer settings
and controls.
III. Methods to control voltage/VARs
Shunt capacitors reduce the reactive power flow through the system and help support voltage.
Acceptable voltage levels cannot be maintained only by the generators in the transmission system –
shunt capacitor banks or other means of the reactive power are required. Mostly, power system loads
are inductive and resistive, operating at the lagging power factor. Such loads require reactive power
from the system, which means reactive power has to be transferred from generation causing system
losses and reducing power system voltage because of these losses. Figure 1 below demonstrates
effect of switching on the shunt capacitor bank.

VL
Inductive Load Q PF=lag
System VST S C
I, S M I SC
~ IC, SC Resistive Load
VST P
C PF=lead
I

Figure 1. Effect of capacitor bank switching

From this figure the benefits of the capacitor bank switching on can be seen:
• it reduces reactive power required from the system by bank capacitance C
• it reduces reactive power plus apparent power flow through the line.
• inherently, current through the line will decrease from I to IC and
• voltage at the bus VL will increase because losses were decreased.
• power factor is improved with the capacitor bank switched on.

2
 Depending on the application (transmission or distribution), location and power system parameters,
availability of current/voltage measurements, bank size, availability of other bank nearby, availability
of the load tap changer (LTC), etc., different approaches can be chosen to switch On/Off the capacitor
bank by the capacitor bank controller.
There are a few common methods available in the capacitor bank controller to control voltage and
VARs by switching capacitor bank On/Off.
• Timed control- capacitor bank will be switched On/Off depending on the time-of-the-day, day,
month.
• Voltage measurement control– capacitor bank will be switched On/Off based on the
measured voltage.
• Reactive power control – capacitor bank will be switched On/Off based on the measured VARs.
• Power factor control - capacitor bank will be switched On/Off based on the measured power
factor.
• Temperature and current control – capacitor bank will be switched On/Off based on the
measured current or temperature.
• Manual control – capacitor bank will be switched On/Off by the operator, based on needs.
We’ll review these methods to elaborate on which method is most appropriate for certain
application.
A. Time-based automatic control
Timed-based control is beneficial when the VAR demand pattern is well known during each day and
during certain seasons. It minimizes the number of switching operations of the capacitor banks and
therefore reduce switching transients, which improves power quality. An example is a large industrial
plant, having the same VAR demand each working day of the week.

Figure 2. Example of the timed control to meet peak VAR demands

3
 Figure 2 above illustrates an example of meeting MVAR demand in the area with 2 capacitor banks
based on the time-of-the-day switching. Modern capacitor bank controller can be time-synchronized
to the UTC time, providing the ability to program control actions based on the real time. Control actions
can be programmed for the entire year using month-of-year, day-of-month, second-of-day
functionality. This allows creating virtually any time-of-the-day and any season-based control actions.

B. Voltage measurement-based automatic control

Voltage measurement-based control is used when the capacitor bank is required to provide voltage
support. Typically, capacitor banks with the primary objective of voltage support are installed at the
major transmission and distribution buses serving large geographic area. Switching the capacitor bank
in results in a voltage rise at the capacitor bank location, which can be estimated by the Equation 1.
𝐼𝐶 𝑄𝐶 Equation 1
𝛥𝑉 ≈ ≈ × 100%
𝐼𝑆𝐶 𝑄𝑆𝐶
where 𝐼𝐶 and 𝑄𝐶 are capacitor bank current and MVAR, while 𝐼𝑆𝐶 and 𝑄𝑆𝐶 are system 3-
phase short circuit current and system 3-phase short circuit MVA at the bank location.
Inherently, a capacitors bank controller, using voltage measurement control, will switch the bank On
during peak loading conditions, causing increased system voltage and will switch bank Off during light
load, causing decreased system voltage. To minimize effect on power quality for the customers,
voltage change is recommended to be within 2-3% range for distribution systems and <5% for
transmission systems.

For example, if system 3-phase short circuit power is 1000 MVA, switching on the 20 MVAR capacitor
bank will result in 2% of the voltage increase at the bank location.

When a capacitor bank is located at a distribution station serving a critical customer located
downstream on a radial line, the voltage level at the customer location may be significantly lower than
the voltage level at the substation due to voltage drop along the line: VL=VST - IL x ZL. To maintain
voltage at the customer location, voltage at the supply substation has to be higher and the controller
has to take into account this voltage drop and turn On/Off the capacitor bank accordingly.

System VS VL Load
IL ZL
~ M

Controller

Figure 3. Voltage drop compensation

4
 Modern capacitor bank controllers are able to estimate voltage level at the critical customer location
by measured current and known feeder impedance, rather than at the substation bus and switch
On/Off the capacitor bank accordingly.

When multiple banks are used to control voltage, it’s important that all controllers respond to the
same voltage, which may be selected in the controller as phase-to-phase voltage or phase-to-ground
voltage, or the average of the three phase-to-phase voltages or positive sequence voltage. This will
ensure all banks are controlling voltage in unison and there is no racing between them. Considerations
to switch bank on or off in the voltage-based control mode are:

• Set the close voltage settings lower than the open voltage setting by more than the voltage
change expected from capacitor switching, plus margin.
• Coordinate between multiple banks in the area, which is achieved by proper voltage level
setting value and time delay to switch bank on or off.
• Coordinate with a load tap changer (LTC), when using this mode in the distribution substation.
Load tap changer will increase voltage but will not improve reactive power flow and power
factor at the substation. Therefore, capacitor bank should have a preference over LTC,
especially at the distribution substation, supplying industrial load - this is achieved with a
proper time delay at both controllers. Also, capacitor bank controller should use voltage from
the HV side, while LTC controller from the LV side of the transformer, regulated by the LTC.

LTC
System VS VL Load

~ M

Cap. Bank
Controller

Figure 4. Coordination between LTC and capacitor bank controller

Figure 4 above depicts voltage measurements for coordination between LTC and capacitor bank
controller. Usually, onload tap changers regulate voltage in steps of 0.625% to 2.5% with a time delay
1-3 minutes for each step operation but switching on a capacitor bank can give the same or larger
voltage increase much faster. Therefore, it is important to exclude possible counter action between
these two control devices, where switching On the capacitor bank can cause the LTC to decrease a
tap and vice versa. Frequent LTC operations are not desirable because they reduce LTC lifetime,
increase maintenance cost and cause voltage step changes at the customers. One approach would
be for the LTC controller to check capacitor bank status and increase step change delay after a
capacitor bank switching operation. Another approach would be to switch On the capacitor bank(s)
permanently during high load time and allow tap changer to fine adjust the voltage during this time.
Properly sizing the capacitor bank or dividing one large bank into multiple smaller banks can also
improve coordination with the LTC.

5
 C. Reactive power-based control
Reactive power-based automatic control directly provides all benefits of the shunt capacitor banks
such as: improved voltage, improved power factor, reduced system losses, increased system power
transfer capacity and reduced reactive power requirements from generators. This control mode can
be used on both transmission and distribution substations. Reference [2] gives details of the economic
impact of adding capacitor bank to the system.

Q Q

CLOSE PKP CLOSE PKP

T
IMI
O RL
ACT
RF
WE
PO
VAR LEVEL VAR LEVEL
TO CLOSE TO CLOSE
VAR LEVEL VAR LEVEL
TO OPEN TO OPEN
0 0 minimum expected kW
OPEN PKP P OPEN PKP demand P

a) b)

Figure 5. VAR control mode without (a) and with (b) PF supervision

Reactive power-based control can be used to regulate both VARs and power factor of a load to
reduce transmission and distribution costs. If the capacitor bank’s primary objective is to minimize
system power losses, then power factor supervision is not used, and the characteristics will look like in
Figure 5a. When the primary objective is to minimize power factor rate penalties, then power factor
supervision is used, and the characteristics will look like in Figure 5b. Larger margins between VAR
Close and VAR Open result in less frequent capacitor bank switching. The difference between VAR Open
level and VAR Close level should be the amount equal to the capacitor bank VAR rating plus a margin.

D. Power factor-based control

Although it’s looking attractive to use power factor for automatic switching to avoid penalties for the
low power factor, in general, power factor alone is not a good basis to control capacitor bank switching
[2]. This is because power factor measurement itself cannot distinguish between low load and high
load, which may result in the leading power factor due to “fixed” size of the capacitor banks. At low
load it can also cause a “pumping” condition because adding capacitor bank can easily cause a
significant change of the power factor. At high loads it may be beneficial to use power factor-based
control, which requires adding VAR or current measurement to a control mode. Therefore, the method
described in [C] above is more attractive to use both VARs and power factor together. If, however, user
has a need to automatically control a capacitor bank using power factor-based control, modern
controllers provide this functionality.

E. Temperature and current-based control

6
 Although temperature and current magnitude increases are an indication of the VAR demand, they
are not giving direct measurement of the VARs deficit at the given bus or geographic area. Therefore,
these methods are not preferred when measurements of the voltage and current are available and
when the capacitor bank controller is capable of methods A to C described above. However, it’s still
possible when direct correlation between temperature/current and voltage/VAR change is well known.

F. Manual control
Manual control, local or remote, still can be used by the operator for switching the capacitor bank
On/Off depending on system conditions. Sometimes, due to scheduled maintenance of the VARs
resources, it is known in advance that a VARs deficit is imminent and is desirable to have capacitor
bank(s) switched On in advance. With the communications available today, remote control is very
much possible, but local control may be needed as well in case of communications failures and
emergency situation. Automatic, manual and remote controls have to be interlocked not to contradict
each other which will be discussed in the section V below.

IV. Transients control

The capacitor bank controller is directly controlling switching devices, which are HV circuit breakers
or circuit switchers. Capacitor switching devices require special attention because very severe
switching transients exist during both switching on and off the shunt capacitor banks, affecting both
the capacitor bank and the adjacent system.

Energization of the shunt capacitor bank creates severe transient over-voltages, affecting insulation
of the adjacent equipment. The most severe over-voltage occurs in the phase, which is closed at the
voltage peak. Figure 6 below illustrates an example of a 230kV 26MVAR bank energization where all 3
breaker poles were closed simultaneously with phase A at the voltage peak. Over-voltage in phase A
reached 1.85pu, while phases B and C experienced very small overvoltage.

Figure 6. Overvoltage during capacitor bank energization

7
 Although these over-voltages are below utility surge protection, they will pass through step-down
transformers to lower voltage loads. To reduce these over-voltages, it is desirable to energize each
phase separately close to voltage zero-crossings, if the breaker and the capacitor bank controller are
capable to do so. Typically, capacitor bank energization transient frequencies are from 300Hz to 1kHz
[2]. Figure 7 below demonstrates the single bank energization current inrush. When a capacitor bank
is energized in the proximity of the previously energized capacitor bank, sympathetic inrush currents
frequency and magnitude will be even higher than the isolated bank energization and can be mitigated
with the addition of a small series reactor on the second bank.

Figure 7. Overvoltage during capacitor bank energization

Another important concern is de-energization of the capacitor bank. When the breaker opens,
capacitors remain charged at whatever instantaneous voltage value they were left disconnected.
Across the breaker contacts there is now the bus sinusoidal voltage and the constant voltage of the
capacitors. The worst case is when voltage is interrupted at the phase voltage peak, because
capacitors remain charged initially at this value and 2 times peak system line to ground voltage (2pu)
appears across the breaker contact. A dielectric breakdown in the gap between the breaker contact
results in the resumption of current through the breaker contacts causing a restrike. Figure 8 below
demonstrates field case capacitor bank breaker restrike. It can be seen that restrike is happening in
phases B and C which were interrupted close to the corresponding phase peak voltage. Phase A was
interrupted close to zero-crossing – no restrike occurred.

Undetected breaker restrike will cause damage to the dielectric withstand of the breaker insulation
and may eventually lead to a catastrophic breaker failure. Capacitor bank controllers should be able
to detect and alarm on each occurrence of the restrike. Reference [2] gives recommendations for
controlled switching, where, for a grounded bank each phase should be closed individually on the
phase zero-crossing and for an ungrounded bank the first 2 phases should be closed first on the phase-
to-phase voltage zero and then delaying closing the 3rd phase by 90.

8
 Figure 8. Breaker restrike during capacitor bank de-energization

It is important to remember that capacitors remain charged for a long time after bank de-
energization - attempt to re-connect the bank with capacitors having significant trapped charge may
result in even more severe transients overvoltages and increased duties on the capacitor bank breaker.
Interlocks to delay re-energization are required. This will be discussed in the section V below

V. Interlocking

The capacitor bank controller needs to switch the bank On/Off to ensure safety of the personnel and
equipment. To achieve this, the capacitor bank controller needs to have the means to provide
interlocking between close/open commands from different sources and needs to open or close the
bank at optimal time.

A. Discharge Interlock

Capacitors store energy in the form of electric fields. This necessitates precautions when energizing
and deenergizing capacitor banks. Capacitor elements are not ideal capacitors so, given enough time,
they will discharge any trapped charge following a de-energization. However, the rate of time required
to discharge the trapped charge is unpredictable and quite long. For this reason, individual capacitor
can design includes an internal discharge resistor as shown in Figure 9 which will dissipate trapped
charge to less than 50V within 5 minutes.

9
 Internal Discharge
Resistor

(a) (b)

Figure 9. Fuseless (a) and internally fused (b) capacitor cans discharge resistors

To allow the resistor time to discharge any trapped charge, a time delay is required to inhibit closing
after the bank has been switched Off. Even though a 5 minutes delay is frequently used [2] in the
industry, re-energizing capacitor sooner or later may be required and [1] gives details of the discharge
time calculations.

B. Manual/Remote/Auto Control Interlocks

Capacitor control includes On/Off commands from various sources (protective trips,
local/remote/auto commands, etc.). It is important, for safety and administrative reasons, that the
control maintain remote/local and auto/manual control rights. Local control refers to control from the
front panel of the relay or relay panel. Remote control refers to control via communications from an
on-site interface, and/or from an off-site control system (SCADA, master VAR management system,
etc.).

Auto control allows the capacitor control system’s automatic control system On/Off control of the
capacitor bank switching device. Manual plus Remote allows remote control and Manual plus Local
allows local control. In some cases, it is desirable to force the auto/manual mode to manual on a
protection operation to prevent the control system’s automatic switching scheme from closing a
faulted capacitor or where closing the bank might cause cycling. This is especially important for trips
like bus over-voltage where the protection does not operate the capacitor bank lockout device. For
safety reasons, the local off (trip) control is typically always active. When this is the case, it is also
necessary to toggle the control to Manual following a local off (see Figure 10 for example logic).

10
 Set control to Capacitor Control
“Remote” S
in “Remote”
Latch
Set control to R Capacitor Control
“Local” in “Local”
Reset dominant
Non-volatile
“Remote” set to AND
“Auto”
OR
“Local” set to AND Capacitor Control
S
“Auto” in “Auto”
Latch
AND R Capacitor Control
“Remote” set to
Reset dominant in “Manual”
“Manual”
OR Non-volatile
AND Local control
“Local” set to AND enabled
“Manual”
Remote control
AND enabled
Protection Trip
AND
“Remote” Open
OR Breaker Open
“Local” Open AND Command

“Auto” Open AND

PU = 0
DO = 5 min
AND Discharging Timer
“Remote” Close Breaker Close
OR AND
Command
“Local” Close AND

“Auto” Close AND

Figure 10. Example logic for Manual/Remote/Auto control interlocking

In the example logic above, there is always a “conscious” decision and indication about who has a
control priority, either local operator, or remote operator or controller left to control automatically per
control mode programmed. Another important consideration is controlling the number of switching
operations of the capacitor bank – the controller may include a lockout timer after each close
operation or count and then limit number of operations per day.

C. Bus and Line Loss-of-Voltage Off interlock

In many cases it is desirable to open the capacitor bank switching device when the connection point
to the system de-energizes. This prevents re-energization of the bank when the connection point re-
energizes. This is especially important when the connection point is a line or a portion of the bus that
de-energizes for a line fault and the connection point is subject to automatic reclosing. Opening the
switching device enables the Discharge Interlock logic, discussed previously to assure the capacitors
are fully discharged before re-energization. This is easily accomplished by tripping the switching device

11
with an under-voltage element set to operate on total loss of connection point voltage with a time
delay less than the line reclosing time.

VI. Optimizing multiple banks control and coordination with other Volt/VAR controls

It’s quite simple to apply a controller to automatically control a single capacitor bank at the
substation bus. It is, however, a rare case that this is the only Volt/VAR regulating means in the area.
Usually, there are many capacitor banks and other VAR regulating apparatus on the transmission, sub-
transmission and distribution substations. Besides VAR regulating means, there are transformer load
tap changers, regulating voltage at the LV buses. Therefore, Volt/VAR regulating problem becomes
multi-dimensional and very complicated.

System A 230kV System B

~ Line ~

13.8kV
69kV

Line

Figure 11. Multiple Volt/VAR resources in the power system

Another requirement is to keep the number of switching operations for both capacitor banks and
LTCs as low as possible to extend the life of switching devices. There are optimization programs
available for the system planner which allow analysis of multiple approaches to control voltage,
reactive power, losses, minimize switching and select the best one to meat all these objectives.
Therefore, P&C engineers need to program capacitor bank controllers in accordance with system
planning requirements. .

VII. Conclusions

Modern capacitor bank controllers offer many options and functionality to control capacitor banks
for voltage/VAR regulation in both transmission and distribution systems. They allow manual (local or
remote) or automatic control based on the available measurements. They also provide interlocking
means to ensure safe operation of the capacitor banks. They have the capability to control multiple
capacitor banks and execute programmed operations sequence automatically or manually.

When it comes to coordination with other devices controlling voltage and VAR, such as other
capacitor banks and load tap changers, it becomes quite complicated with automatic volts/VAR
control mode to ensure all controllers are working in unison and no switching races exist between
them. For these situations, its preferred to have a combination of the timed-based approach per known

12
VAR deficit to take care of large chunks of the VAR demand and capacitor bank automatic control for
smaller chunks of the VAR demand. Having different capacitor bank sizes helps to control voltage and
VARs more effectively.

Alternatively, Distribution Management Systems can provide a centralized Volt/VAR strategy by

taking measurements from multiple points of the system and controlling capacitor banks and LTCs
remotely. This can achieve optimum system voltage profile by regulating voltage and reactive power
flow throughout the system, thus improving power quality and minimizing power losses.

VIII.Literature:
1. IEEE Std. C37.99-2012 -Guide for the Protection of Shunt Capacitor Banks.
2. IEEE Std. 1036-2010 - IEEE Guide for the Application of Shunt Power Capacitors.
3. GE publication GEK-130995, C70 Capacitor Bank Protection and Control System - Instruction
Manual, 2017

IX. Biographies

Ilia Voloh received his Electrical Engineering degree from Ivanovo State Power University, Russia. He
is currently a senior application engineer with GE Grid Solutions in Markham Ontario, and he has been
heavily involved in the development of UR-series and 8-series of relays. His areas of interest are current
differential relaying, distance relaying, advanced algorithm and advanced communications for
protective relaying. Ilia authored and co-authored more than 40 papers presented at major North
America Protective Relaying conferences. He is a recipient of the best paper award at Georgia Institute
and Technology Protective Relaying Conference in 2012. Ilia is a senior member of the IEEE, member
of the IEEE PSRC main committee, a member of the IEC TC 95 committee and a member of the CIGRE
B5.65 working group.

Tom Ernst is a semi-retired P&C Technical Application Engineer for the GE Grid Solutions North
American Commercial team. He has been with GE since 2011 supporting the Grid Automation
Protection and Control Portfolio. Prior to joining GE, Tom has been with Minnesota Power as a
Supervising Engineer, Delta Engineering International as a Manager of Electrical Engineering, HDR
Engineering as a Manager of Electrical Engineering and Northern States Power as a Supervising
Engineer. He received his Bachelor of Science in Electrical Engineering from the University of Minnesota
in 1978 and his Master of Science in Power Systems from Michigan Technological University in 2008.
He is a long-time member of the IEEE Power and Energy Society.

13