Abstract

This report documents the work performed by the student on the topic Transformer protection.

Faults on a power transformer can cause an outage in power supply, destruction of a power system and the loss of profits and inconvenience to the customers. Hence suitable protection schemes are vital to ensure the safety of the power system and human personnel. Included in this report are the causes of faults involving transformers and suitable protection schemes required to protect transformers have also been presented.

Contents
Abstract ................................................................................................................................................... 1 1. 2. 3. Introduction .................................................................................................................................... 3 Requirements for power system protection................................................................................... 4 Transformer protection .................................................................................................................. 5 3.1 The need for transformer protection ............................................................................................. 5 3.2 Components of a Transformer Protection Scheme ................................................................... 6 3.2.1 Current transformers ......................................................................................................... 6

3.2.2 Application of current transformers ........................................................................................ 9 3.2.2 Voltage transformers ............................................................................................................. 10 3.2.2.1 CCVT in Power Line Communication ................................................................................... 11 3.2.2.2 Ferro resonance in CCVT ....................................................................................................... 12 4. Types of protection devices .......................................................................................................... 15 4.1 Fuses ........................................................................................................................................... 15 4.2 Automatic circuit reclosers ....................................................................................................... 16 4.3 Sectionalizers .............................................................................................................................. 16 4.4 Circuit-breakers and relays ......................................................................................................... 17 4.4.1 Gas actuated relays .............................................................................................................. 18 4.4.2 Sudden pressure Relays ....................................................................................................... 19 5. Differential Transformer protection ............................................................................................. 20 5.1 Earth Fault Relays........................................................................................................................ 22 5.2 Overcurrent relays ...................................................................................................................... 23 6. 7. Protection co-ordination............................................................................................................... 24 Switchgear (Busbar) Protection ................................................................................................. 26 7.1 The need for busbar protection .................................................................................................. 26 7.2 Types of busbar protection ......................................................................................................... 27 7.2.1 Bus differential protection ................................................................................................. 27 7.2.2 Bus differential protection with linear couplers ................................................................ 27 7.2.3 High Impedance bus differential protection ........................................................................ 28 7.2.4 Fault bus protection ............................................................................................................. 29 7.2.5 Frame leakage protection .................................................................................................. 29 7.3 Saturation detectors ................................................................................................................... 30 Conclusion ............................................................................................................................................. 31

1. Introduction
A power systems core functionality is to generate, transmit and distribute electrical energy to its customers. The aim of the power system is to supply this generated energy to the customer in a reliable and economical manor which comes at a huge task since there usually exists a compromise between these two requirements. These three sectors of generation, transmission and distribution comprises of many diverse items of equipment that are very expensive, requiring a big capital investment. It is therefore important to protect these pieces of equipment at all costs. Examples of such equipment include generators, turbines, network breakers, transmission lines or cables, protection devices and for our interest, transformers and busbars. Power system protection is important for ensuring that an electricity distribution network is capable of operating within preset requirements for safety for individual items of equipment. It is also important because it safeguards the entire system and maintains the continuity of power supply reduces damage and repair costs to a minimum and ensures the safety of human personnel. Protection systems are generally installed to clear faults and to limit any damage to electrical equipment. The special equipment that is used to detect faults is referred to as protective equipment and systems that use such equipment are protection systems.

2. Requirements for power system protection

In order for a protection system to detect and localise faults early and to ensure prompt removal of faulty equipment from the system it must possess the following qualities: Reliability- Is dependant on two factors namely dependability and security. Reliability is defined by the following equation:

% Reliability =

Dependability- It is the ability for the system to trip when called upon to do so. Dependability is defined by the following equation:
% Dependability =

Security- It is the ability of the system not to trip when it is not supposed to. Security is defined by the following equation:
% Security =

Sensitivity- To detect even the smallest fault, current or system abnormalities. Selectivity- To detect and isolate the faulty item only. Speed To operate at the minimum time to clear a fault in order to avoid damage.

Example 1 [1]: The performance of an overcurrent relay was monitored over a period of one year. It was found that the relay operated 16 times, out of which 12 were correct trips. If the relay failed to issue a trip decision on 3 occasions, calculate the dependability, security and reliability of the relay. Answer: Number of correct trips = 12 Number of desired trips = 12 + 3 = 15
% Dependability =

= =

% Security =

% Reliability =

3. Transformer protection
The rising demand of power transformers in the last few decades has led to the development of transformers of a higher efficiency and a higher working voltage, meeting todays current demands. The reliability of the transformer in power systems is deemed most important and is the first requirement in the design process and although other elements, such as cost, are of importance they cannot be compromised in terms of reliability. Transformers mainly operate under predictable, normal conditions but they should be designed in such a way as to have the ability to withstand abnormal conditions such as lightning strikes, short-circuits, flash-overs etc. Faults of a power transformer can cause an outage in power supply and loss of profits and inconvenience to the customers. We therefore need to detect incipient faults in a power transformer as early as possible so that we can take them out of service and repair them safely to improve the reliability of power systems. Through the centuries the transformer has proved to be one of the most important links in a power system. The benefit of having a simple design deems the transformer a highly reliable piece of equipment. This reliability, however, depends on numerous aspects on the operating environment of the transformer such as, an adequate design, care in erection, proper maintenance and the provision of the respective protective equipment.

3.1 The need for transformer protection

Transformer protection is needed because: It protects the transformer against the existence and effects of faults occurring on any part of the system. These faults are classified as external faults. It protects the system against the existence and effects of faults occurring internally within the transformer. These types are faults are classified as internal faults. External faults include [2]: Overloads The effect of overloads involves the overheating of the transformer and has the potential to cause permanent damage or loss of life to the unit. The cause of overload could be due to unequal load sharing of parallel transformers or the unbalanced loading of three-phase banks. Usually no protection is provided for overloads but alarm is often used to alert operating personnel of the overload. Overvoltage Overvoltages occur under two distinct conditions namely, short-term transient conditions or long-term power frequency conditions. Short- term transient overvoltages cause end-turn stresses and possible breakdown. Surge protective devices are used for protection against these transients. Power frequency overvoltages can occur due to a sudden loss of load on an isolated portion of the system that causes the voltage to rise, as a result over fluxing occurs which increases the iron losses of the transformer. Underfrequency External system short circuits

Internally, the main causes of faults in a transformer are due to: Winding and terminal faults Core faults

Tank and transformer accessory faults Onload tap changer faults Abnormal operating conditions Sustained or un-cleared external faults Overheating of the transformer due to failure of the cooling system

3.2 Components of a Transformer Protection Scheme

The traditional method of detecting faults in a transformer and protecting it is done by using current and voltage measurements. To accomplish this current transformers (CTs) and voltage transformers (VTs) are used. Along with these fundamental measurement devices network breakers and relays are used to make up the protection scheme, which collectively functions to detect a fault and to isolate it from the system.
3.2.1 Current transformers

Current and Voltage transformers are necessary to ensure isolation of the protection, control and measurement equipment from the high voltage of a power system, for providing the equipment with the appropriate values of current and voltage and to act as sensors for the relay. A current transformer is connected in series with the primary circuit and produces a scaled down replica of the primary current in its secondary circuit. For example, a CT of ratio 100/1 would produce a secondary current of 1A when 100A flows in the primary circuit. As compared to regular transformers current transformers have current source excitation where the primary winding of the current transformer is connected in series with the transmission line. The equivalent circuit model of a CT is shown in figure 1. It is assumed that the primary winding resistance and leakage reactance are neglected and the total impedance on the secondary side is the sum of relay burden, lead wire resistance and leakage impedance of secondary winding.

Figure1. Equivalent circuit of a Current transformer [1] CTs can be classified into two types namely measurement CTs and protection CTs. Measurement CTs: Have lower VA capacity than protection grade current transformers Are not expected to give linear response during large fault currents Has a accuracy range of 5-125% of normal current

Protection CTs:

CT burden has to be kept low in order to obtain linear response Performance accuracy range between normal currents to fault currents Gives linear response up to 20 times the rated current

Depending on the type of application, CTs can be classified by a further two types: 1. Wound primary type 2. Bar primary type The wound primary type (shown in figure 2) is used for the smaller currents; however it can only be applied on low fault level installations due to thermal limitations as well as structural requirements due to the presence of high magnetic forces. The bar primary type (shown in figure 3) is used for currents greater than 100A.

Figure2. Wound primary type current transformer [3]

Figure3. Bar primary type current transformer [3] ANSI/IEEE standards classify CTs into two types: 1. Class C CTs 2. Class T CTs Class C CTs: Leakage flux is very small More accurate than Class T CTs Performance can be evaluated from standard exciting curves

Class T CTs:

Performance can be evaluated by appropriate testing Wound type CT with one or more primary turns on a core Has high leakage flux

The ratio error of a current transformer is the percentage deviation in the current magnitude in the secondary from the desired value. Ratio error =
| | | | | |

When the CT is not saturated - ratio error = Where Ip and Is are the CT primary and secondary currents respectively.

Figure4. Class CT excitation curve [1] Example1.[1] A 1200/5, C400 CT with excitation curves shown above, is connected to a 3.0 ohm burden. Based on the accuracy classification, what is the maximum symmetrical fault current that may be applied to this CT without exceeding a 10% ratio error? Answer: CTratio=1200/5 Secondary resistance=0.61ohms Relay burden = 3 ohms For 20 times rated secondary current, i.e., 100A Secondary voltage = 100 x (2 + 0.61) = 361 Volts Linearity will not be lost even at higher currents (<400V). Approximate limit on secondary current is given by Imax = 400/3.61 = 110.8 A

Maximum symmetrical fault current = 110.8*(1200/5) = 26592A. Example2 [1]. Assume that secondary burden of a 300:5 class C CT is 5 . The relay setting is 2A and the CT ratio is 300/5. Using the calibration curve, calculate the primary current required to operate the relay? Answer: Secondaryburden=5ohms Secondary resistance for 300/5CT=0.15ohms Relaysetting,Is=2A Vs=2x(5+0.15)=10.3V Corresponding exciting current=0.045A Total secondary current = 2 + 0.045 = 2.045A Primary current to operate the relay = (300/5)*2.045 = 122A

Example3.[1] If a 300:5 class C CT is connected to a meter with resistance R1 = 1 ohm and secondary current in the CT is 4.5A find out the primary current, voltage developed across the meter and % ratio error. Lead wire resistance secondary resistance of a 300:5 CT = 0.15 . Answer: Ri = 1 ohm; RL = 0.02 ohm; Rs = 0.15 ohm; Is = 4.5 A Total secondary resistance RT = Ri + RL + Rs = 1.17 Secondary voltage = Is* Rs = 4.5 * 1.17 = 5.0265V = 0.03A Turns ratio N = 300/5 = 60 Ip = N (Is + IE) = 60(4.5 + 0.03)= 271.8A Voltage across meter = Is * Ri = 4.5 * 1 = 4.5V Ratio error = (IE/IS)*100 = (0.03/4.5) *100 = 0.67%

3.2.2 Application of current transformers

AC trip coils are designed to carry normal and fault currents, but it is difficult to use the same trip coils for higher current circuits. To overcome these limitations current transformers are used so that the high primary currents are transformed down to manageable levels that can be handled comfortably by protection equipment. An application where CTs are used in is fused AC trip coils.These use current transformers, which must be employed above certain limits i.e., when current rating and breaking capacity becomes excessively high.

Some basic schemes are: 1. Overcurrent In this application, the fuses bypass the AC trip coils as shown in Figure 5. Under normal conditions, the fuses carry the maximum secondary current of the CT due to the lowimpedance path. Under fault conditions, the secondary current having reached the value at which the fuse blows operates the trip coil TC to trip the circuit breaker. Characteristics of the fuses are inversely proportional to the current, so a limited degree of grading is achieved.

Figure5. CTs for overcurrent use in series trip coils [3]

2. Overcurrent and earth fault There are two methods involving the connection of CTs, the first is shown in figure 6 and the second type is shown in figure 7 which is the most economical arrangement for this protection.

Figure6. CTs for overcurrent and earth fault protection using series trip coils [3]

Figure7. Economical use of overcurrent and earth fault configuration [3]

3.2.2 Voltage transformers

A voltage transformer is connected in parallel with the power system, usually in an earthed star configuration. It produces a scaled down replica of the primary voltage in its secondary circuit. Voltage transformers are usually arranged so that a normal system voltage produces a nominal secondary voltage. There generally are two types of voltage transformers used in protection equipment: 1. Electromagnetic type (commonly referred to as a VT)

2. Capacitor type (referred to as a CVT). VTs are commonly used in relay applications such as distance relays, directional overcurrent relays which require measurement of voltages at a bus. The secondary voltage of a VT is typically standardized at 110V (ac), when the primary voltage increases the turns ratio N1:N2 also increases. This results in the transformer becoming bulky at high voltages. Hence to reduce the cost and size of the VT a capacitance potential divider is used. This results in a reduced voltage being fed to the primary of the transformer. This type of transformer is called a coupling capacitor voltage transformer (CCVT). Due to the presence of the capacitive divider, the accuracy of the CCVT is not good due to the voltage drop created. This results in the connection of a tuning inductor which compensates for this voltage drop. The equivalent circuit is shown in figure 8.

Figure8. Capacitor compensated voltage transformer [3]

3.2.2.1 CCVT in Power Line Communication

A schematic diagram of a CCVT with RF connection is shown in figure 9. CCVT is commonly used in HV/EHV systems where carrier line communication is implemented. CCVTs allows for high frequency RF signals to be coupled to power lines for communication. In the schematic shown the capacitance potential divider provides a shunt path to high frequency signals used in power line communication. A drainage reactor is added to block the path to ground for RF signals.

Figure9. Schematic diagram of a CCVT with RF connection [1]

3.2.2.2 Ferro resonance in CCVT

Ferro resonance is caused by the nonlinearity of the iron cores of the reactor and transformer. As a result a Ferro resonance suppression circuit is connected to the secondary of the CCVT. The ferro resonance suppression circuit reduces the transients in the CCVT. The differential circuit of an RLC circuit is given by:

During a fault, v(t)=0 and the above equation becomes:

This equation is expressed in standard form as:

From the above equation it can be seen that the circuit depends on the: 1. Damping coefficient 2. Natural frequency n in rad/sec 3. Point on the voltage waveform where the fault occurs The transients that are formed are known as subsistence transients. Figure 10 shows how subsistence transients can reduce the accuracy of distance relays.

Figure10. Typical CCVT subsistence transient [1] Example 4 [1] The equivalent circuit of a CCVT is shown in figure 11. The values of C1 and C2 are 0.0018F and 0.0186F respectively. Tuning inductor has an inductance of 497Hand a resistance of 4620 . Xm of the VT referred to 6.6kVside is 1M, core loss = 20 watts per

phase, VA burden =150VA per phase. The value of Cm for compensating the current drawn by Xm is equal to 3.183 109F. (a) Verify the appropriateness of choice of L and Cm. (b) Find out the nominal value of V/V2. (c) If the frequency drops from 50Hz to 47Hz, what would be the values of the ratio error and phase angle error?

Figure11. Equivalent circuit of a CCVT [1] Answer: , where = 2f and f = nominal frequency. Thus, = 496.7 H, which is equal to the given value of L. Now, Xm = 1106 . Cm has to be in parallel resonance with Xm. Therefore,

This is also the same as the selected value of Cm. Hence, the selection of both L and Cm is appropriate.

V = 11.33 6.6 =

Thus this VT is connected to a 132 kV bus. Core loss = 20 W

VA burden = 150 VA (resistive)

Xm= 106 at f = 50Hz 2fLm = 106

= 0.45910-6 j1.06410-6 + j0.9410-6 + 3.4410-6 = (3.902 j0.124)10-6 = 3.90410-6(-1.82O) Z = 256018.32 + j8135.15

=
VT =IthZ =

A (256018.32 + j8135.15)

= 6480.42(4.26o)

% ratio error = Phase angle error = 4.26

= 1.81%

4. Types of protection devices

There are basically four different types of protection devices that are commonly used in South African power systems. These include [4]: 1. 2. 3. 4. Fuses Automatic circuit reclosers Sectionalizers Circuit breakers and relays

4.1 Fuses
The functionality of a fuse operates on the principle that when the current passing through the

element of the fuse is high enough, the element would melt. The melting of this element would create a short circuit and therefore isolates the fault from other parts of the system. The advantage of this device is that it is inexpensive and it can operate very fast for high currents. The fuse is called an expulsion device due to the formation of arc when the fusible element, located within a fuse tube, melts under high current. The arc causes a rapid pressure buildup, which forces the ionized gas out of the bottom of the cutout, which helps to prevent reigniting.

Figure 12: Typical fuse used for transformer protection [7] Figure 12 shows a picture of a Bussmann Medium Voltage (E-type) Fuse. This fuse is typically used on medium voltage transformers with the general purpose of current-limiting. A general purpose current-limiting fuse is capable of interrupting all currents from the rated interrupted current down to the current that causes melting of the fusible element in 1 hour (ANSI C37.40). The fuses carry a rating that falls into categories, indicated by letters, which defines the melting-time-current characteristic of the fuse. For the above fuse, the fuse is rated as an E-type fuse. This means that the following conditions must be met: [7] 1. 100E and below - the fuse element must melt in 300 seconds at 200% to 240% of its rating (ANSI C37.46). 2. Above 100E - the fuse element must melt in 600 seconds at 220% to 264% of its rating (ANSI C37.46). Industry standards specify two other important types of expulsion fuses used in power systems: K type: fast fuse with speed ratio of 6-8

T-type/E-type: slower fuse with speed ratio 10-13

The speed ratio is the ratio of the melting current at 0.1 second to the melting current at X seconds, where X is 300 for fuse ratings below 100 amps and X is 600 for fuse ratings above 100 amps. The current rating is the level of current the fuse can safely carry for an indefinite period of time. For example, a typical K-type fuse would have ratio (using 600 seconds) of 200/30=6.7.

4.2 Automatic circuit reclosers

An automatic circuit recloser is a circuit breaker that is programmed to implement a certain number of trip recloser sequences, once this is done the auto recloser locks out in an open position. The advantage of these ARCs as compared to fuses is that they are less expensive when compared to circuit breakers but the accuracy of the device is compromised. ARCs are most commonly used on overhead transmission and distribution lines. Overhead lines face mostly temporary faults that last for only a few seconds. In the case of a wind blowing a tree branch on an overhead line, causing a phase to ground fault, the ARC will trip. Since the fault will be cleared when the wind stabilizes of blows in the opposite direction, the fault will be cleared from the line, and thus the ARC will reclose. In this care of temporary fault the ARCs are favored. The basic operation of an ACR is to Sense the presence of a fault via over-current Open to interrupt the over-current and isolate the fault Reclose to re-energize the circuit Lockout if the fault persists after 4 attempts to reclose

The first reclose is generally done instantaneously which means there is no intentional time delay embedded in the first reclose. Typical timing sequences used in industry include 0-15-30 seconds 0-30-60-90 seconds 5-45 seconds

4.3 Sectionalizers
Sectional are used for remote applications that require remote isolation. Sectionalisers are programmed to isolate the section which they control depending on the number of unsuccessful reclosres of a recloser or an upstream breaker. Sectionalizers are similar to switches. The main difference between a switch and all other devices described previously is that all the other devices can detect and interrupt a short-circuit fault current, but a switch can do neither. It is typically applied together with an ACR or a circuit breaker to isolate faulted sections of the distribution system.

The basic operation of a sectionalizer is as follows: It counts the number of operations of the backup device After a pre-set number of interruptions, it opens while the backup device is open. Then the backup device can reclose to restore service to remaining circuits.

4.4 Circuit-breakers and relays

Compared to the protection devices considered thus far, the circuit breaker and relay are the most important with respect to transformer protection. The protection relay forms the most important part of the protection scheme, as a result its capability and functionality plays an important role in how the protection scheme operates. The basic function of a relay is to respond to abnormal conditions in the power system, and control a circuit breaker which will then isolate the fault of the system. The types of relays include [4]: Unit Trip Relays This relay, also known as the master tip relay, is responsible for tripping an entire unit, (eg. a transformer). It receives its inputs from any of the other relays, which if trip causes the unit trip relay to operate. Instantaneous Relays These relays are set to trip immediately when there are very high fault currents detected on the line. The current at which it trips is set on the relay. Over Current Relays These relays can have two characteristics on which they can operate, namely a Definite Time (DT) and Inverse Definite Minimum Time (IDMT). The IDMT characteristic has an operating time that is dependant on the fault current, ie. the higher the fault current the faster the operating time, while the DT characteristic has a set trip time for any value of fault current. Earth Fault & Sensitive Earth Fault Relays These relays are connected in the CT neutral lead to detect and operate for phase-ground short-circuits only. Since current only flows in the CT neutral lead when there is a ground fault (under normal conditions phase currents are 120 out of phase and cancel out), these relays can be set to be very sensitive and have a short operating time. Directional Relays This type of relay uses both the voltage and current as its inputs. This is because the phase angle between the current and the voltage determine the direction of the current flow. If there is a current flowing that is above the over-current pickup level in the tripping direction, then the relay will operate. Distance Relays This is another type of relay that uses both the current and voltage as inputs. It basically calculates an impedance value from the current and voltage and when the impedance is below the set value, which means that there is a fault down the line, the relay will operate. Differential Relays These relays are used to protect major pieces of equipment such as transformers and bus bars. The relay compares the current that goes into the equipment to the current that comes out. If the currents on either side of the protected equipment are not equal and in phase then a fault must exist, which causes the relay to operate. Alarm Relays This type of relay gives an alarm indication of certain fault conditions such as low oil level, high temperature or Buchholz alarm on a transformer. If the alarm

is left unattended, then this relay will send a trip signal to the unit trip relay, which will trip the faulty equipment. The most important type of relay will now be discussed.
4.4.1 Gas actuated relays

As was discussed previously, some of the faults that could occur in a transformer are insulation failures and poor electrical connections. These types of faults can cause the generation of heat in the transformer. Heat developed within a transformer is of great concern since this leads to other complications internally such as decomposition of the solid insulation and oil. Decomposition causes gasses to develop in the transformer oil and will therefore rise to the top of the transformer tank. The presence of different gasses is an indication of different faults that are occurring within the transformer. The initiation of these faults is sometimes detected at an incipient level by a dissolved gas in oil analysis. The gasses accumulated at the top of the tank act as triggers for the gas actuated relays.
4.4.1.2 Buchholz Relay

As already stated, due to localized heating and decomposition of the insulation, gasses are developed in the transformer oil and rise to the top of the transformer tank. The Buchholz gas-actuated relay is used to trigger an alarm when a certain amount of gas has accumulated in the relay indicating to the operating personnel on duty that there is an internal problem in the transformer. The Buchholz relay is very important with respect to the detection of incipient faults in a transformer. The position of the Buchholz relay on the transformer is shown in figure 13.

Figure 13: Buchholz relays position on a transformer [8] Figure 13 shows a Buchholz relays position on a transformer; it can be observed that the relay is connected into the pipe leading to the conservator tank in a favourable position to detect any gas accumulation. As the gas is accumulated in the relay a float operates a switch which sounds an alarm. This principle is used to develop the modern Buchholz relay shown in figure 14.

Figure 14: Internal mechanisms of the modern Buchholz relay [4] This modern mechanism eliminates the possibility of a false alarm. Since the faults will be detected at an incipient level, prevention of sustained arching and further damage will be eliminated. The relay is adjusted to give an alarm when the gas accumulated in the relay has reached a volume which depends upon the size of the transformer. The table below gives a rough estimation of the settings used in industry. [4] This type of relay is used in conjunction with the differential protection scheme that will be discussed later. Table 1: Gas to volume to operate the alarm [4] Transformer size Up to 1MVA 1 to 10MVA Over 10MVA Pipe Diameter 2.5cm 5.0cm 7.5cm Setting Range 100-120 cc 185-215 cc 220-280 cc Normal setting 110cc 220cc 250cc

4.4.2 Sudden pressure Relays

In earlier decades, certain transformers were made without a conservator tank which rendered the Buchholz relay obsolete. The sudden pressure relay was then used. This relay operated on the principle that the rate of increase of the gas in the oil is an indication of the presence of a fault as opposed to the quantity of gas used in the Buchholz relay. It has a diaphragm which is deflected by a differential oil pressure; the diaphragm is by-passes by a hole which equalizes the pressure on the two sides of the diaphragm normally and also makes it responsive not to pressure but to rate of rise of pressure. [4]

4.4.3 Relays

A relay is a device that monitors the conditions of a circuit and issues instructions to open a circuit under unhealthy conditions. As discussed previously, the relay is triggered by current and voltage measurements obtained from CTs and VTs. The relay must then make a decision on [9]: Whether an abnormal condition exists or not. Whether the system is being endangered. Which element of the system is faulty? How quickly the fault can be removed. Whether the system can be safely re-energized or whether it must remain disconnected until manual inspection of the cause of the fault. The functionality of a relay in the protection scheme can be described by the following flow diagram:

FAULT
MEASURED BY CT/VT

BATTERY

RELAY
Figure 15: Flow diagram of the relay logic

ACTION
OPEN or CLOSE

5. Differential Transformer protection

The Buchholz relay that was mentioned earlier operates for a fault that occurs under the oil in an oil immersed transformer. There are occasions where faults and flashover occur above the transformer oil e.g. LV-HV bushings, bushing between phases, tank to earth, etc. These types of faults are dangerous and must be cleared quickly; it is therefore fundamental to provide a high speed biased differential protection scheme which will detect these faults that the Buchholz relay would not have detected. Usually earth fault relays (to be discussed later) are used in addition to the differential protection where this relay serves as backup protection. The differential relay operates on the principle of comparing currents of each winding. Since the current in the high voltage winding and the current in the low voltage winding are related to each other by the basic transformer equations the comparison of the two currents can be made and an irregularity can be detected if the measured current deviates from these relationships. From simple transformer theory it is known that the ratio of primary and secondary current depends on the number of turns in each winding of the transformer. This would make differential transformer protection relays a suitable device for the detection of

turn to turn and turn to ground faults, since a change in the turns ratio would affect the current ratio. A simplified schematic of a differential relay is shown in figure 16 to easily identify its operation.

Figure 16: Simple model of a differential protection system [3] The above differential scheme is called the current balance or circulating current scheme. The two CTs are connected in series and the secondary current circulates between them. The relay is connected across the midpoint where the voltage is theoretically nil, hence no current can pass through the relay; therefore there is no operation for faults outside the protected zone. During internal fault conditions (i.e. faults between the CTs) the relay operates, since both the CT secondary currents add up and passes through the relay. [3] The disadvantages of this simple relay are: [4] The current transformer characteristics The difference in C.T characteristics due to different ratios being required in circuits of different voltage levels, may cause a major difference in the respective secondary currents whenever through faults occur. Although these currents are of equal value at normal load, the problem arises when different types of C.Ts are used. One of the major problems with differential protection is the unequal saturation of the current transformers. This saturation gives rise to an inaccuracy in the ratios that is used to determine the imbalance and disturbance and may cause the relay to trip on a false fault. Also unequal length of current transformer secondary leads can cause a difference in VA burden between two sets of C.Ts. This will also lead to a ratio error. Ratio-change as a result of change of tapings Most transformers are fitted with an on-load ratio-change gear. When the ratio is changed by this gear, the ratio between the primary and secondary C.Ts can be made to match at one point only of the tap changing range. At any other points there will exist an imbalance in the ratios which will lead to a false fault. Magnetizing inrush-current

When a transformer is energized, the inrush current produced may attain peak values corresponding to several times the transformer full-load current and decays relatively slow. This inrush current generally flows on one side of the differentially connected relay only, which will tend to operate if some form of restraint is not provided. The above three disadvantages are overcome by using a percentage differential relay. From the first two problems mentioned above, the currents produced by the C.Ts increase with an increase of through current. As a result, the relay is now operated at a percentage of through current. This allows for a sensitive setting at low current so that the relay would not trip unnecessarily. The percentage differential relay operates on the principle that the operating coil is provided with the vector sum of the currents in the transformer windings and the restraining coil with the through current [4].

5.1 Earth Fault Relays

Ground faults are the most common type of fault to occur in a transformer, attributing for nearly 805 of all faults in a power system. As a result it is very important to provide reliable, fast and accurate protection against earth faults.

Figure 17. Earth fault protection of a power transformer [2] From the system shown in figure 17 one can note that the delta winding is protected with an over-current relay. In industrial applications zero sequence over current relays protect delta windings and ungrounded Y connections. These relays are supplied by C.Ts situated at the terminals of the power transformer as shown above. Note that this type of relay does not have an earth connection, which would have the implication of only operating for an earth fault due to the lack of an earthing through which to supply an external fault. To prevent incorrect operation on a false residual current, the relay is instantaneous and of a high impedance. For star connected windings, with the grounded star point, the faults are detected and managed by restricted earth fault relays as shown in the above figure. This differential connection provides relay current only for a winding fault to ground, but with instantaneous relays it is stable only if the relay circuit is of high impedance [4].

During the occurrence of external faults in the system, there exists the presence of third harmonics due to these faults. The presence of third harmonics would occur for earth fault and balanced three phase faults. Balanced three phase fault are very rare in a power system and attributes for less that 5% of all faults. Since the third harmonic components of the phase A,B and C behave as the zero sequence components of these phases, they add together directly instead of cancelling out like the fundamental components do. This problem is solved by using a third harmonic filter.

5.2 Over current relays

Over current relays are used for both overload and fault protection. An Inverse Definite Minimum Time Characteristic (IDMT) is used to obtain the settings of the relay. An overcurrent relay is operated from a C.T attached a circuit breaker. This type of relay is used when the cost of differential relaying cannot be justified. The Inverse Definite Minimum Time Characteristic is used to obtain the setting required for the system. These settings are time multiplier settings or time dial settings, Operating time of the relay and pickup current or multiple of pick up current. The Standard Inverse Time curve has the following relationship:

Where: top is the operating time of the relay TM is the time multiplier setting M is the Multiple of pickup current or ration I/IS (where I is the actual secondary current and IS is the relay pickup current setting) For over current relays the pickup current should be set to trip at 150% of the maximum and sufficient time delay to ensure that the proper breaker is tripped for external faults. (see example in the section protection coordination) Note that overcurrent relays provide less protection for internal transformer faults. As mentioned before overcurrent relays are usually used only as back-up protection. Overcurrent relays are usually installed on the HV and LV side of the transformer in which case the side affected by the fault will trip, or both. Another implementation of overcurrent protection is using a two stage scheme. This is made up of one IDMT over-current relay and one time-lag relay with the IDMT relay on the HV side. When the IDMT relay is operated the time lag relay is triggered which requires a time setting of say 0.4 seconds, so that it does not operate before the LV breaker trips. This as a recommendation of the grading margin which is made up as follows: Relay pickup and timing errors CT errors Breaker operating time Safety margin 100ms 100ms 100ms 100ms

If the fault persists the time-lag relay trips the HV breaker.

6. Protection co-ordination
The basic aim of a protection scheme is to isolate a fault from a system while concurrently preventing outages to as many customers as possible. There are two main important coordinating principles: 1. The protecting device must clear a fault before the protected device interrupts the circuit (fuse) or operates to lockout. 2. Outages caused by permanent faults must be restricted to the smallest section of the system for the shortest time.

Figure 18. A co-ordination system [10] From the figure above note that the substation receives power flow a high voltage transmission line and steps the voltage down. Protective devices in this illustration are located at the coordinating points. Device A is at the substation. Devices C and H are in the feeder. Device B is at the branch tap off the feeder. Device D is on the transformer primary and devices E, F, and G are service entrance fuses on the transformer secondary. All devices must be selected to carry normal load current and respond properly to a fault. With respect to device H, device C is the protected device. For a fault at point 1, device C must not open and device H must interrupt. With respect to device A, device C is the protecting device and must interrupt permanent fault current at point 2 before device A operates to lockout. Device B is also a protecting device for A and must operate similar to device C for a fault at point 3. Only device A functions when a fault occurs between A and C, such as at point 4. For a transformer fault at point 5, device D is expected to interrupt current and permit normal load current to flow in the remainder of the system. For an overload on the secondary tap of transformer at point 6, device E should interrupt the circuit, permitting power to the transformer to be continued so that customers on other secondary taps will receive service. This is the basic approach that is used to co-ordinate different devices and parts of a power system, incorporating transformer protection. In a full scale power system there are more that 2 breakers that need to be coordinated with each other. To understand the above coordination consider the following example:

The following example was done to calculate the settings of the section breaker 2 in the figure below:

Figure19. Schematic of the network used for example When performing the calculation for the section breaker 2 there are three different settings for over current and earth fault that need to be calculated. These are IDMT, High set and Rapid. The philosophy that is usually used for the IDMT plug setting is that it is usually set to 120% of the full load current (IFull load). The Rapid setting is usually the same as the IDMT plug setting, only with a faster trip time. The high set is usually an instantaneous trip at a specific fault current. The fault current used to set the plug setting is usually 80% of the line. The high set plug setting is usually a multiple of the IDMT plug setting. This will be illustrated in an example using figure 19. It is important to note that the section breaker has 4 trips to lockout (breaker remaining open). These 4 trips are comprised of 2 trips for the Rapid setting and 2 trips for the IDMT setting. The purpose of this is to detect if the fault clears (for example of the wind caused a tree branch to touch the line). If the fault does not clear after the 4 trips then lockout will occur. For the purpose of illustration the following example was used: All the relays are normal inverse curve relays The settings of the incoming breaker are over current IDMT =400A @ 0.4s NI The fault levels at the different points on the figure are as follows: o Fault current at point A=3500A o Fault current at point B=2300A o Fault current at point C=1000A o Fault current at point D=550A The loads fed after the network breaker add up to 11MVA

For the HV breaker the full load current is calculated by the following equation:
I full _ load S 20 10 6 131 A 3V 3 88 10 3

For the over current IDMT plug setting we use 120% of the full load current therefore this would be 1.2x131=157.2A160A. In the same way the grading for the LV breaker is done. For the over current high set we throw a three phase fault on the 22kV bus bar. From the given information this fault current is 3500A. We transfer this to the HV side by the following equation:
I1 n 2 V2 V 22 kV I1 2 I 2 3500 A 875 A I 2 n 1 V1 V1 88 kV

The over current high set is set in multiples of the over current IDMT plug setting, so this is set to 5 times 160A, which is 800A. The over current rapid settings would be the plug setting of the over current IDMT setting of 160A but at a faster time. For the settings of the network breaker the full load current is calculated in the same way as done above but with 11MVA being used instead of 20MVA and 22kV instead if 88kV as follows:
S 11106 288.68A 3V 3 22 103 Plug _ setting 1.2 288 346.42A 350A @ 0.4sNI I full _ load

The rapid setting is also 350A with a faster time. For the high set we throw a three phase fault on the section breaker 1. From the given information this is 2300A. So this can be set to 2450A (7x plug setting of IDMT=350A). It is important to note that the full load current is usually below 200A and if the full load current is beyond this, as in this case, the cable capacity should be checked that it can carry this capacity. If not the plug setting should be less than 200A. For the settings of the section breaker 1 and 2 the settings are done as above with a fault being thrown at section breaker 2 for the fault current of section breaker 1 and a fault being thrown beyond section breaker 2 for the calculation of fault current here.

7. Switchgear (Busbar) Protection

Busbars are one of the most critical components in a distribution network. Busbars can be located in an open switchyard; these bus bars are called open busbars. They can also be located inside a metal clad cubicle restricted within a limited enclosure. Busbars is the connecting point of a variety of elements and a number of transmission lines; hence any incorrect operation will lead to the loss of all of these elements. In case of an external fault, failure-to-trip on an internal fault or false tripping of a busbar could result in undesirable effects on the stability of the power system and sometimes it may even cause a complete blackout of the system. Therefore when designing a busbar protection system high speed reliability and stability are important factors that needs to be incorporated. [11]

7.1 The need for busbar protection

If bus protection is omitted from a system, bus faults will be cleared by backup protective zones of the lines terminating at the bus. The advantage of this protection method is its ability to produce satisfactory results in terms of the reliability of fault clearing, however this is a slow process and the fault will be allowed to cause a great amount of damage to the bus

structure and surrounding equipment, hence fast clearing is vital. Also slow clearing of faults often cause transient stability problems and may endanger human personnel working in the station. If bus protection is used there is a possibility of tripping the bus unnecessarily which leads to the outage of the connected load and possibly the creation of system stability problems. These unnecessary trips could be caused by a relay setting error, instrument transformer saturation or human error in working on relay equipment. Due to the fact that busbar faults are likely to cause extensive damage and possibly even destroy an entire station, busbar protection is preferred over other protection schemes.

7.2 Types of busbar protection

7.2.1 Bus differential protection

Differential protection has proven to be one of the most commonly used methods of providing bus protection, there are several different methods of implementing bus protection some of which will be explained. One of the major problems with differential protection is the unequal saturation of the current transformers used for current measurements around the bus. Unequal saturation is due to the large differences in the measured fault current magnitude at the different bus connection points and is also due to the residual magnetism in the current transformers. Two commonly used differential protection schemes that tries to reduce CT saturation will be discussed.
7.2.2 Bus differential protection with linear couplers

Bus differential relaying becomes very difficult due to the current transformer saturation problems. These saturation problems make it hard to predict the degree of saturation and to design appropriate counter measures. An effective solution to this problem is to eliminate CT saturation. This is done by eliminating the iron from the current transducer. The device that can accomplish this is called a linear coupler which is an air-cored mutual reactor located on a nonmagnetic toroidal core. This device has several advantages which include: 1. Saturation is eliminated in current measurements. 2. High speed performance with negligible transient response. 3. Very reliable design. 4. Linear coupler is easy to set and maintain. 5. Can be operated with secondaries open without damage. Linear coupler differential relaying makes use of a voltage differential scheme; this is due to the linear coupler having a large number of secondary turns with a linear characteristic of about 5 volts per 1000 primary ampere turns. A linear coupler bus differential scheme is shown in figure 20.

Figure20. A linear coupler bus differential protection scheme [2] The linear coupler bus protection scheme does have two disadvantages. Firstly, the sensitivity of the linear coupler is limited by the maximum external fault conditions and secondly this system requires the installation of linear couplers that can only be used for differential protection. The second disadvantage is a costly one because the solution requires the purchasing of all new CTs since all circuits on the bus has to be equipped with linear couplers if they are to be used.
7.2.3 High Impedance bus differential protection

This system uses a current differential relaying scheme with the use of overvoltage relays as shown in figure 21; this system is also designed to eliminate CT saturation. In this system bushing CTs or other conventional CTs that have low impedance secondary windings are used. To control the effects of CT saturation a substantial resistance is added to the branch of the circuit by the full-wave bridge rectifier. To improve selectivity the series L-C circuit is tuned to the system fundamental frequency which makes the relay responsive only to the fundamental component of current. The drawback of this idea is that it slows the response to some extent, but this is countered by the addition of a high speed overcurrent relay in series with a voltage limiting, metal-oxide resistor. This resistor is place in the circuit to limit the voltage that is produced when faults occur. The overcurrent shown in figure 21 provides the tripping for high magnitude bus faults; also its minimum pickup can be set relatively high to allow it to ignore external faults. When there are a large number of circuits connected to a bus this type of protection will be beneficial to use. The requirements required for the proper application of this principle include: a) b) c) d) Equal transformer ratios on all connections. Low current transformer secondary winding resistance. Adequate knee-point voltage output from the CTs. Low burden from the connecting leads.

Figure21. Bus differential protection using high impedance voltage relays [2]

7.2.4 Fault bus protection

This type of protection is well suited for applications where an insulating bus support structure completely isolated from ground is built for instance in metal-clad switchgear or other stations that can be completely housed in a metal enclosure. The basic structure of the protection scheme is shown in figure 22; all equipment contained within the enclosure is isolated from ground with the exception of the single ground connection which contains the protective relays. The overcurrent relay is responsible for controlling an auxiliary relay that trips all circuit breakers connected to the bus. It is important that circuit breaker tanks, supporting steel structures, high voltage bus conductors and all frameworks be isolated from ground by supporting insulators. This type of protection is well suited for isolated phase construction; in this case all faults are ground faults.

Figure22. Bus fault protection scheme [2]

7.2.5 Frame leakage protection

This protection scheme involves the measurement of fault currents from switchgear frame to the earth. The scheme consists of a current transformer connected between frames to earth points which energises an instantaneous ground fault relay to trip the switchgear. It is important to ensure that all metal parts of the switchgear are insulated from the earth to avoid unwanted currents to be circulated. The minimum recommended setting for this protection scheme is 30% of the minimum earth fault current of the system. Figures 23 and 24 show the frame leakage protection system.

Figure23. Requirements for frame leakage busbar protection [3]

Figure24. Schematic connections for frame leakage protection [3]

7.3 Saturation detectors

In many bus protection schemes saturation detectors are used. The current transformers used in the system feeds the differential circuit via auxiliary transformers, these auxiliary transformers then feed a typical saturation detector circuit as shown in figure 25. Making reference to figure 25, a voltage Vc is produced across the resistor R. The capacitor C is then charged to the peak value of the voltage Vc, a comparator is then used to compare the voltage with 0.5 Vc stored in the capacitor. On saturation, Vc drops below 0.5 Vc capacitor voltage, the comparator then turns on the switch across the buswires eventually the pulse widths increase.

Figure25. Saturation detectors [3]

Conclusion
From the information contained in this report, the importance of transformer protection cannot be over emphasized. Due to the requirements of a power system, reliability is of great importance and therefore protection of the transformer is necessary in order to deliver reliability. Faults on a power transformer can cause an outage in power supply, destruction of a power system and the loss of profits and inconvenience to the customers. Hence suitable protection schemes are vital to ensure the safety of the power system and human personnel.

References
[1]http://nptel.iitm.ac.in/courses/Webcourse-contents/IIT-KANPUR/powersystem/ui/Course_home-3.htm, NPTEL Online-IIT Kanpur PS, 30 August 2010 [2] P.M Anderson Power system protection,New York: McGraw-Hill, 1999 [3] L Hewitson, M Brown, B Ramesh Practical Power Systems Protection, First Edition, IDCTechnologies [4] A.R. van C. Warrington, Protective relays their Theory and Practice, Volume 1, Chapmen&HallLtd.,1968 [5] Document: 4PB2 Transformer Protection, Ms. K Pillay, University of KwaZulu-Natal [6] T. Short, Electric power distribution handbook, CRC press, 2004. [7] Cooper Bussmann, Equipment protection, 2002, Available at: www.bussmann.com, [Accessed17thSeptember2010] [8] http://www.automation2000.com/option_IB_en.html, Automation 2000, 18 September 2010 [9] Eskom Basic Protection Course, Alec Aylward, Eskom Distribution [10] McGraw-Edison Power Systems, Distribution system protection manual. [11] Mohan, S.M, Chatterjee, S, Busbar Protection A Review, IEEE Region 8 SIBIRCON, July 2010