Fuseless Capacitor Bank Protection

Tom Ernst, Minnesota Power

30 West Superior Street
Duluth, MN 55802
(218) 722-1972/(218) 720-2793 [fax]
ternst@mnpower.com

Abstract
The use of fuseless capacitor banks requires subtle changes in the protection approach from the more
traditional fused banks. This paper covers the aspects of protecting fuseless capacitor banks of various
voltage classes. Comparison of fused versus fuseless capacitor bank protection is discussed, along with
examples and insights from Minnesota Powers experiences with fuseless capacitor banks.

Introduction
There are many different VAR and voltage ratings available on power capacitors and many different
physical arrangements of power capacitor banks. However, the installations can be grouped into one of
three applications - internally fused, externally fused or fuseless. The individual power capacitors are
designed uniquely for the application intended and have their own unique failure modes. As a result,
protection systems must also be tailored to the application.
All applications of power capacitors require the same basic protection objectives, including system short
circuits between phases or to ground within the bank, and element overvoltages, caused by power system
overvoltages or by the failure of other elements within the bank. Since the failure modes differ between
capacitor types, the performance of protection schemes will also vary. In particular, fuseless capacitors
present unique protection considerations.

Capacitor Types
There are three main types of power capacitors, internally fused, externally fused and fuseless. Internally
fused capacitors were used extensively in the past when the element technology was all-paper or paperfilm. The individual can is constructed from series groups of parallel capacitor elements, each element
individually fused within the can (refer to Figure 1a). When an individual capacitor element shorts, the
fuse for that element blows, opening the element with minimal overvoltage stress on the remaining
elements.
Externally fused capacitors utilize modern all-film element technology. The individual can is constructed
from series groups of parallel capacitor elements which are designed to be operated with a common
external fuse (refer to Figure 1b). The external fuse will generally not blow for failure of an individual
element and the can will continue to operate with a reduced VAR flow and increased voltage stress across
the remaining elements within the can. The increased voltage stress usually results in cascaded failure of
the remaining elements within the can, ultimately blowing the external fuse.

a) Internally Fused

b) Externally Fused

c) Fuseless

Figure 1:Internal configuration of power capacitors

Fuseless capacitors, like externally fused capacitors, utilize modern all-film element technology and are
constructed from series groups of parallel capacitor elements. However, they usually contains more
elements than a typical fused capacitor (see Figure 1c). When an individual element shorts, the can
continues to operate with a slightly reduced VAR rating. Due to the larger number of series groups within
the can and the power wiring methods used, the overvoltage stress on the unshorted groups is smaller than
for externally fused capacitors. As a result, cascaded element failures within the same can are not as
common.

Typical Capacitor Bank Arrangements

Figure 2 shows a typical high voltage capacitor bank arrangement using externally fused capacitors (note:
the discussion of fused capacitor banks will be limited to externally fused). An element failure results in
a significant increase in voltage stress on the remaining elements within the can (and a slight increase in
voltage stress on the other cans). As a result, a cascaded failure of elements within the same can is
common, causing the can fuse to blow. A blown fuse causes a more noticeable increase in voltage stress
on the remaining cans. The bank must be removed from service if the voltage stress on the remaining cans
becomes high enough to cause cascaded can failure (10% above nameplate is a typical value).

Phase 1

Phase 2

Phase 3

Figure 2: Typical arrangement of an externally fused capacitor bank.

Figure 3 shows a typical high voltage capacitor bank arrangement using fuseless capacitors. In this case,
an element failure within a can results in a uniform increase in voltage stress on all the remaining
elements in the parallel group. As a result, multiple element failures will occur anywhere in the group, not
necessarily within the same can. Again, the bank must be removed from service if the voltage stress on
the remaining elements in the group becomes high enough to cause cascaded element failure (10% above
nameplate is a typical value).
Phase 1

PM

Phase 2

PM

Phase 3

PM

Figure 3: Typical arrangement of a fuseless capacitor bank.

Protection Objectives
The objectives of capacitor bank protection are the same, regardless of the type of capacitors used or the
physical arrangements employed. They include short circuit protection for phase and ground faults,
overvoltage protection resulting from excessively high power system voltages and overvoltage protection
resulting from element failures.

Short Circuit Protection

The failure mode for short circuits (faults) within the capacitor bank is the same for all types of capacitor
banks. Consequently, short circuit protection for fuseless capacitor banks is the same as for fused
capacitor banks and is generally provided in the form of phase and ground time-overcurrent relaying.
Where available, the relaying is generally connected to current transformers located at the capacitor bank
breaker. The pick-up of the phase relays should be set above full load with the time delay set to
coordinate with other protective devises (bus overcurrent relays, remote zone 2 impedance relays, etc.).

System Overvoltage Protection

Excessively high system voltage can cause capacitor failure, regardless of the type of capacitor. For all
types of capacitor banks, protection against overvoltages that are caused by excessively high system
voltage is generally provided by a high speed overvoltage relay connected to the substation bus voltage
transformers. This relay trips the capacitor bank breaker or vacuum interrupter before capacitor damage
can occur. The overvoltage relay pick-up should be set to correspond to a system voltage which results in
an unacceptably high (typically 10% over nameplate) can voltage.

Element Failure Related Overvoltage Protection for Fuseless Capacitor Banks

For capacitor banks having more than one series group, failure of individual elements causes the applied
voltage to increase on the remaining elements and cans. There are three common methods of detecting
can or element failure voltage differential, neutral overvoltage and neutral overcurrent. All three of
these methods can be used on fused and fuseless banks, with varied applications.
When voltage differential is used for a fuseless capacitor bank, the bottom can in each phase is a single
element protection module (PM). The voltage differential relay (87V) is connected to look at the
difference between the bus voltage and the protection module voltage (see Figure 4). As elements fail
within a series group, the phase current will increase and the voltage across the protection module will
increase relative to the bus voltage. The difference voltage will become smaller. At the same time, the
voltage will increase across the elements within the same series group as the failed elements and, to a
lesser degree, across the elements in other parallel groups. This is in contrast to fused banks where failed
elements blow fuses, resulting in reduced phase current, reduced tap voltage and an increase in the
difference voltage.

In both fused and fuseless capacitor banks, the voltage differential relay provides alarm and tripping
functions. The alarm should be set to alert maintenance personnel in advance of a trip so that damaged
equipment can be repaired during a scheduled outage. For a fused capacitor bank, this generally means
alarming for one blown fuse. On fuseless capacitor banks, the alarm point is not as obvious. Since there is
no outward indication of which cans have bad elements, alarming for one failed element may not be
useful for maintenance. If the alarm is set for more than one element failed, there is a strong likelihood
that several cans will need to be replaced (there is no logical failure mode to suggest that all of the failed
elements will be in the same can). Generally accepted practice is to alarm for some number of failed
elements that is slightly less (1 or 2) than the trip point.

Phase 1

Phase 2

Phase 3

Bus
VT

87V

Tap
VT
PM

51N

Figure 4: Voltage differential application on fuseless capacitor bank.

Tripping from the voltage differential relay should be set to occur at the point where one more failure
(element in the case of fuseless or blown fuse in the case of fused) will result in a voltage stress that
exceeds 110% of rating on the remaining elements. Note that, it is important to use nameplate or
measured capacitance for the protection modules when performing the voltage differential calculations as
a small variable in the protection module impedance results in a significant change in differential voltage.
If the voltage differential relay being used has the ability to automatically calculate the nulling factor, it is
recommended that this feature be used once the bank is known to contain no bad elements.

Neutral overvoltage relaying can also be used to detect failed elements in both fused and fuseless
capacitor banks. In the case of fused capacitor banks, the application is generally limited to ungrounded
banks. For fuseless capacitor banks, neutral overvoltage relaying can be applied on grounded wye banks
by grounding the bank through a single element protection module (see Figure 5). The neutral overvoltage
relay (59N) operates based on the voltage across the protection module. In this application, the voltage
across all three phases changes as elements fail within the series groups.

Phase 1

Phase 2

PM

Phase 3

59N

Figure 5: Neutral overvoltage application on fuseless capacitor banks

The setting criteria for alarm and trip are the same for the 59N application as in the case of the voltage
differential. However, the calculations for neutral voltage as a function of failed elements uses a wyedelta conversion of the phase impedances to calculate individual phase line currents and associated
neutral current/voltage. A disadvantage of the neutral overvoltage application is that it may respond
adversely to primary system voltage unbalances.
Neutral overcurrent relaying can be used to detect failed elements in any grounded capacitor bank (51N
Figure 4). On transmission voltage capacitor banks it is generally applied as back-up to a voltage
differential application. Since the relay operates on relatively small currents (25 - 50 amps primary), it is
generally required to use a low-ratio neutral CT (10:5 or 50:5) with primary protective gaps. The relay
can generally be set to provide the same alarm and tripping points as the voltage differential relay. As in
the case of the neutral overvoltage relay, the neutral overcurrent application may respond adversely to
primary system voltage unbalances.

Conclusions
All of the traditional methods of protecting fused capacitor banks can be applied to fuseless capacitor
banks with the same basic objectives. However, a careful understanding of the failure modes of fuseless
capacitors is required in order to correctly apply and set the protective relays. Since there is no outward
indication of failed elements within fuseless capacitor cans, maintenance and trouble shooting procedures
will be different and the relay settings should be designed taking this into account (especially with regard
to the failed element alarm).