Connections & required protections
This technical article explains the protection practices applied to shunt reactors and capacitors as well as to static var compensators (SVCs) and static compensators (STATCOMs). Dry-type and oil-immersed reactors usually use overcurrent, negative sequence, percentage differential, or impedance relays depending on some specific application.
On the other hand, the main protection for capacitors is a function of their installation within a station.
Capacitor overvoltage protection is provided by fuses supplied by manufacturers. SVCs and STATCOMs are gaining increasing popularity as a means of controlling both inductive and capacitive reactive deficiencies at the point of their connection to the power system.
The protection is a combination of protection required for both reactors and capacitors and the special requirements associated with the SVC and STATCOM design.
However, coordination with manufacturers is essential.
- Shunt reactor protection
- Shunt capacitor bank protection
- Static var compensator protection
- Static compensator (STATCOM)
Reactors are connected into a power system in either a series connection or a shunt connection. The series reactor is used to modify the system reactance, primarily to reduce the amount of short-circuit current available. The shunt reactor is used to modify the system voltage by compensating for the transmission line capacitance.
In general, the protection of reactors is very similar to that of transformers.
For both types of reactor construction, there are two more considerations that have an effect on the protection.
- Single-phase reactors, i.e. each phase is in its own tank. These are usually applied on EHV transmission lines. There is no possibility of having a phase-to-phase fault within the reactor enclosure although such a fault can occur in the bus and bushings.
- Three-phase reactors where all three windings are in the same tank. These are primarily applied at lower voltages.
The faults encountered in dry-type reactors are the following.
1. Phase-to-phase faults on the tertiary bus, resulting in high-magnitude phase current. These faults are rare since the phases of the reactors are located physically at a considerable distance from each other.
2. Phase-to-ground faults on the tertiary bus resulting in low-magnitude fault current depending on the size of the grounding transformer and resistor. These faults are also rare since the reactors are mounted on insulators or supports with standard clearances.
3. Turn-to-turn faults within the reactor bank, resulting in a very small change in phase current. Winding insulation failures may begin as tracking due to insulation deterioration which eventually will involve the entire winding.
The result is a phase-to-neutral fault which increases the current in the unfaulted phases to a maximum of √3 times normal phase current.
Fault protection for the dry-type reactor is achieved through overcurrent relays connected as shown in Figure 2 and differential relays as shown in Figure 3. This protection is the same as the overcurrent and differential relaying for generators and transformers.
Since phase-to-phase and phase-to-ground faults also produce negative sequence currents, a negative sequence relay, connected the same as the overcurrent relay, can be used. Load is not a consideration with negative sequence, but there must be enough current to operate the relay.
Differential relays can provide sensitive protection but they do not see turn-to-turn faults since the current entering a reactor with shorted turns is equal to the current leaving the reactor. Instantaneous relays are not usually applied since the only fault location that will produce enough current to operate an instantaneous relay is at the phase end of the reactor or in the bus or bushings.
These fault locations are usually protected by the bus differential relay.
The failures encountered with oil-immersed reactors are the following:
Failure № 1 – Faults resulting in large changes in the magnitude of phase current such as bushing failures, insulation failures, etc. Because of the proximity of the winding to the core and tank, phase-to-ground faults can occur, the magnitude being dependent upon the location of the fault with respect to the reactor bushing.
Failure № 2 – Turn-to-turn faults within the reactor winding, resulting in small changes in the magnitude of phase current.
Failure № 3 – Miscellaneous failures such as loss of oil or cooling.
An impedance relay can detect shorted turns since there is a significant reduction in the 60 Hz impedance of a shunt reactor under such a condition.
Figure 4 shows the connection for an impedance relay (device 21). Protection against low-level faults or mechanical failure involving the oil system is by pressure, temperature or flow devices.
One of the principal difficulties with shunt reactor protection is false relay operation during energizing or de-energizing the iron core. During these periods, DC offset with long time constants and low-frequency components of the reactor energization current cause most of the problems.
High-impedance differential relays rather than low-impedance relays are recommended if this problem occurs.
Capacitors are also installed as either a series or a shunt element in power systems. The series capacitor is used primarily to modify the transmission line reactance for stability or load flow considerations. The protection of fixed or switched shunt capacitor banks requires an understanding of the capabilities and limitations of both the capacitors and the associated switching devices.
Capacitor bank protective equipment must guard against a variety of conditions:
Condition № 1 – Overcurrents due to faults between the bus and the capacitor bank. The protection afforded is the conventional overcurrent relay applied at the breaker feeding the capacitor bank.
Condition № 2 – System surge voltages. The protection afforded is the conventional surge arresters and spark gaps.
Condition № 3 – Overcurrents due to individual capacitor unit failure. The manufacturer provides the necessary fuses to blow for an internal unit failure. The fuse link should be capable of continuously carrying 125–135 % of the rated capacitor current.
Condition № 4 – Continuous capacitor unit overvoltages.
Figure 5 shows the general arrangement for a high-voltage capacitor bank consisting of parallel units to provide the required current capability and series units to provide the desired capacitive rating.
An overvoltage can be imposed across individual capacitor units as a result of the loss of one or more units, usually by the operation of a unit fuse.
The overvoltage is the result of the increased impedance of the series section from which the faulty unit has been removed. As units are removed, the impedance of that section increases. However, since there are many sections in series the effect of the increased impedance in one section does not decrease the phase current in the same relative proportion.
The result of the slightly reduced current flowing through the more markedly increased impedance causes a higher voltage to appear across the remaining units in that section.
Static var compensators (SVCs) are devices which control the voltage at their point of connection to the power system by adjusting their susceptance to compensate for reactive power deficiencies. The basic reactive components of SVCs are shunt reactors and shunt capacitors (Figure 6).
Reactor controls are either thyristor-controlled or thyristor-switched. Capacitors are either fixed or thyristor-switched.
As the load varies, a variable voltage drop will occur in the system impedance. This impedance is mainly reactive. Assuming the generator voltage remains constant, the voltage at the load bus will vary. The voltage is a function of the reactive component of the load current, and system and transformer reactance.
A SVC can compensate for the voltage drop for load variations and maintain constant voltage by controlling the gating of thyristors in each cycle. With fixed capacitors and variable reactors, leading or lagging current can be provided to the bus and will correct the voltage drop or rise.
Sometimes, these protective functions can be provided as part of the integrated protective system supplied by the manufacturer or they can be provided by the user. In either case the settings must be coordinated between the manufacturer and the user.
Differential relays, phase and ground overcurrent relays, gas pressure, low oil level and temperature relays have all been effectively applied in accordance with standard transformer protection practice.
The transformer guide should be applied to transformers that are part of the SVC. The connection of the transformer windings determines the types of relay and their connection.
Phase fault protection
Conventional differential or time overcurrent relays are applicable. Time-overcurrent relays are usually used for backup protection. For some installations, the SVC bus is included in the protection zone of the transformer differential relays.
Ground fault protection
The voltage supplied to SVC buses may be grounded through a resistance or impedance in the main transformer or through a grounding transformer, or ungrounded. Grounding transformers are sized such that ground fault currents are limited to reduce damage, yet large enough to selectively operate ground relays.
In addition, ground fault currents on the SVC low-voltage bus should be limited to 500–1500 A to prevent thyristor valve damage.
Typical protection schemes are as follows:
- Time-overcurrent relays connected to CTs to measure the zero sequence current in the main transformer or grounding transformer.
- Time-overvoltage relays connected across the broken-delta secondary winding of a voltage transformer as shown in Figure 7. This scheme does not provide fast location of a fault since the ground fault could be anywhere in the low-voltage bus or any of its branches.
In most installations, relays connected to the bus voltage transformers are provided to protect the entire SVC system from excessive overvoltages. The capacitors in the SVC are vulnerable to overvoltage and therefore determine the relay settings.
Reactor branch protection
A thyristor-controlled or thyristor-switched reactor is usually considered a separate zone of differential and coordinates with it. Conventional reactor differential and overcurrent relays are used. In many designs, the reactor branch is connected to the SVC bus by a relatively slow motor-operated disconnect switch.
The relaying would then trip the SVC main breaker, the faulted element would then be removed and the remainder of the SVC installation returned to service.
Capacitor branch protection
Conventional overvoltage and unbalance protection is applicable.
Switching of SVC elements will introduce harmonics into the power system. Depending upon the particular design, the manufacturer must provide the necessary filters and should provide, or recommend, the protection required.
The magnitude of the harmonic voltage generated depends on the type of SVC, the SVC configuration, the system impedance and the amount of reactance switched.
The even harmonics are removed by symmetrical gating of the TCR thyristors.
This protection is normally provided as a part of the thyristor control system. Typical protection is provided for overvoltage, overcurrent, temperature and, where applicable, coolant flow and conductivity.
A static compensator (STATCOM) provides variable reactive power from lagging to leading, but with no inductors or capacitors for var generation. This is achieved by regulating the terminal voltage of the converter.
SVCs have generally proven to have lower equipment costs and lower losses. STATCOMs have been used in transmission where land constraints, audible noise or visual impact are of concern.
STATCOM can provide both reactive power absorption and production capability whereas an SVC requires individual branches for capacitors for var generation and reactors for var absorption.
Source: Power system relaying by Stanley H. Horowitz and Arun G. Phadke