Вы находитесь на странице: 1из 24

Busbar Protection

Problems caused by Inrush Phenomenon

The word bus is derived from the Latin word omnibus which

means common for all.

Busbars are the nerve-centres of the power system where various circuits are connected together. These are the nodes of the electrical

circuit. Figure 5.1 shows a busbar having an N, number of

incoming lines and an N2 number of outgoing lines. The protective zone, to be generated by the protective relays, is also shown. It may

be noted that under the normal power flow condition the sum of
incoming currents is equal to the sum of outgoing currents, i.e

Figure 1: Introduction to busbars

There is a large concentration of short-circuit capacity at the busbars. A fault on the busbar, causes enormous damage. When protective relays operate to isolate the busbar from the system, there is a large disruption to the loads. Busbars are located in switchyards, substations, and supply kiosks. The switchyards are well shielded from direct lightening strokes but because of their outdoor nature, are subject to the vagaries of weather. The substations are well protected in all respects and fault probability is indeed very low. Similarly, supply kiosks are totally enclosed and sealed.

The causes of faults experienced on busbars are: weakening of insulation because of ageing, corrosion because of salty water, breakdown of insulation because of overvoltages, foreign objects, and so on. For example, rodents, lizards and snakes are known to have caused busbar faults in remote unmanned substations.

Differential Protection of Busbars

Selection of CT Ratios in Case of Busbar Protection: Wrong Method
Figure 2 shows a busbar, having two incoming feeders and one outgoing feeder, being protected by a simple differential protection scheme. The currents shown are for normal load flow. Let us decide the CT ratios on the basis of maximum primary load current seen by each CT. Thus, the CTs on the incoming feeder will have CT ratios of 1000/1 A and 2000/1 A, respectively. The CT on the outgoing feeder will have a CT ratio of 3000/1 A. However, with this choice of CT ratios, it can be seen from the diagram that there is a spill current even during the healthy condition. Thus, the method of selecting CT ratio on the basis of maximum primary current seen by the feeder is not correct.

Figure 2. Wrong method of selection of CT ratios for

Selection of CT Ratios in Case of Busbar Protection: Correct Method Figure 3 shows the correct method of setting the CT ratios for the busbar differential protection. It can be seen that the CT ratios of all the CTs are equal and are based on the primary current of that feeder which carries the maximum current. Thus, all the CT

Figure 3. Correct method of selection of CT ratios for

ratios are 3000/1 A. Therefore, there is no spill current through the OC relay connected in the spill path and the scheme remains stable. We draw an important rule for the selection of CT ratios for all the CTs in a busbar differential protection, namely:

External and Internal Fault When the CT primary current, or the burden on it, is within its design limits, the CT can indeed be assumed to be more or less ideal. However, as the primary current exceeds the design value or the CT burden (output of CT in VA) becomes excessive, the CT no longer behaves in an ideal fashion then on-ideal behaviour of the CT has very serious implications for the protective schemes.

Figure 4 shows currents during an external fault beyond CTC. It can be seen that CTC, the CT on the faulted feeder, has to carry the sum of all currents fed into the fault by various feeders. Therefore, CTC sees a substantially larger primary current than either CTA or CTB. In all likelihood, CTC will therefore become saturated. We can, therefore, no longer assume that CTC will faithfully transform the fault current. For the sake of illustration, we have assumed that the secondary current of CTC is only 4 A instead of 10 A. It can be seen from Figure 4 that this results into a spill current of 6 A, causing the scheme to maloperate, i.e. lose stability on external fault.

Figure 4 : Behaviour of busbar differential scheme on external fault

In the worst case scenario, CTA and CTB continue to transform faithfully as per their nameplate CT ratio but CTC, which carries the total fault current, gets completely saturated. This clearly indicates the occurrence of an imbalance in transformed secondary currents, resulting in

substantial spill current. This situation most likely will

cause the scheme to operate. Operation of a differential scheme under external faults is, therefore, clearly a case of maloperation

Interestingly, as the fault shifts by a small distance to the left and

becomes an internal fault, still drawing the same current, the

situation dramatically changes as far as CTC is concerned. This is depicted in Figure 5, wherein it can be seen that CTC now does not

carry any fault current (assuming a single-end-fed system with

source on left hand side). Since CTA and CTB are not carrying excessive primary currents, they transform the current without too

much error. There is thus a spill current in the spill path and the
scheme operates as expected. The maloperation of the busbar differential scheme on external

faults is caused due to non-ideal behaviour of a CT carrying

excessive primary current. It will, therefore, be pertinent, at this point to take a closer look at the actual behaviour of protective

current transformers.

Figure 5 : Behaviour of busbar differential scheme on internal fault

Actual Behaviour of a Protective CT

Figure 6 shows the equivalent circuit of a current transformer referred to the secondary side. Rp and Xp are the primary winding resistance and leakage reactance, respectively, referred to the secondary side. Rs and Xs are the resistance and leakage reactance of the equivalent circuit (referred to secondary). At low values of primary current Ip and therefore Is voltage Es, to be induced by the secondary winding, which is approximately equal to (Zburden * Is) , is quite low. The working flux in the CT, which is equal to (Es/4.44 fN) is also very low. The magnetizing current requirement is, therefore, correspondingly low. Thus, the secondary current Is is substantially equal to Ip/N.

If the primary current increases, initially, the secondary current also increases proportionately. This causes the secondary induced voltage to increase as well. Increased secondary voltage can only be met with an increase in the working flux of the CT. AS the flux increases, the transformer needs to draw a higher magnetizing current. However, because of the nonlinear nature of the B-H curve for the CT, as the knee of the excitation characteristics is passed, any further increase in

flux demand causes a disproportionately large increase in the magnetizing current requirement of the CT. This is illustrated in Figure 5.7. It may also be noted that I, is no longer sinusoidal and its waveform has a prominent peak.

Figure 6 Equivalent circuit of CT.

Figure 7 Operation of the CT beyond the knee point of the B-H curve

As the primary current goes on increasing, a stage comes when the magnetizing current requirement is so large that almost all the transformed current is taken up by the CT for the sole purpose of magnetization, with the result that there is hardly any current available for the burden. When this occurs, it means that the CT is completely saturated. The secondary induced voltage and burden current waveforms of a CT, operating in saturation, are highly distorted. They, in fact, consist only of sharp pulses near the zero-crossings of the primary current. This is shown in Figure 8, where it can be seen that in order to reach the peak of the sinusoidal flux waveform, the CT is driven deep into saturation. Due to flatness of the excitation curve in this region, a very large amount of exciting current is demanded by the CT. The waveform of the exciting current becomes distorted and is, in fact, very peaky in nature.

Figure 8 CT saturation due to excessive primary current

Circuit Model of Saturated CT

The circuit model consists of a current source of value (Ip/N) feeding into a short circuit through Rp and Xp. The connections to the outside world are available through Rs and Xs.

Figure 9 : Circuit Model of Saturated CT

External Fault with One CT Saturated: Need for High Impedance Busbar Protection
Consider the external fault of 30,000 A, on the busbar shown in Figure 4. Assuming that CTC gets completely saturated while CTA and CTB continue to operate normally, Figure 10 shows the equivalent circuit as seen from the CT secondaries. Note that for simplicity, all leakage reactances have been neglected. It can be seen from Figure 10 that currents [(IA/N) I0A] and [(IB/N) I0B] sum up and the resultant current has two parallel paths available. One path is through the over-current relay and the other is through (Rs + RL) via the short representing the saturated CT magnetizing branch.

Thus, the part of fault current will flow through the over-current

relay, causing it to trip, even though the fault is external. Hence,

the OC relay needs to be restrained from tripping on external faults (with one CT completely saturated). We can easily

accomplish this by connecting a high resistance (known as the

stabilizing resistance) in series with the OC relay. The stabilizing resistance should be of such a value that under the worst case of maximum external fault and full saturation of one CT, the current through the OC relay is less than its pick-up value. Such schemes are known as high impedance busbar differential schemes.

Figure 10 Secondary equivalent circuit with one CT fully saturated during external fault,