Use of Single CT Secondary Output for Both Differential and Overcurrent Protection

If I understand the question properly it is theoretically possible that it could be done with one transformer if a high impedance sensor were connected to the output and amplified to trigger the appropriate relay.

An increase in secondary load will also produce an increase in primary current.

There may be a single CT but are you sure that one secondary is used? From past experience and testing of substations, transformers and generators and the like the CTs have had single primaries but multiple secondaries for each individual protection or metering, some had as many as eight secondaries.

Yes, single secondary core is being used for both differential and overcurrent protection relays. For metering, another secondary core is used but for protection scheme, single core is used.

As far as I know and please correct me if I'm wrong, for differential protection, the burden on CTs on both sides of the protected zone should be same.

For combined differential and overcurrent protection relays, secondary core of CT on the upstream side or line side of the generator is used. Downstream or neutral side CT is only connected to differential relay. Will this not cause imbalance in the burden on both CTs ?? How will this work?

With modern electronic relays, the relay burden in the CT circuit is small, particularly since one relay may replace several electromechanical relays, each with a burden. The CT winding resistance plus the cabling resistance CT to relay is most of the resistance.

If it is a problem, one can "cheat" by adding extra cable length on "Diff only" side to equalise resistance - the length of run from relay to CT may require that anyway because cable runs are not equal length.

There are two schemes which have been popular for differential protection of generators....

1. Biased - the two currents from the CTs are used to make the differential relay become less sensitive as through current increases, so the CT error does not cause a false trip. This is OK if generators never fail during an external fault.
2. The assumption is made that one [saturated] CT will give zero voltage [short-circuit] while the other gives saturated voltage. The relay, with equal resistance to each CT and a high resistance [usually with added external resistance] sees half the saturation voltage and has a setting greater than that. For an internal fault, the relay is not shunted by a saturated CT and receives the differential current. Practically, the relay resistance does not have to be so high for stability and can be about 1/3 of the limit value and the CTs require matched saturation voltage.

N.B. There are reasons why the overcurrent etc relays should use CTs on neutral side of gen with usual star connection- a phase to phase short in the generator would not flow through line side CTs.

Differential Protection for generators almost always uses separate CTs which are identical type each side of generator and identical loading [burden] including matching cable resistance from each CT to relay and may be closely matched as pairs for saturation voltage. This appears the case for modern multi-function digital relays which have separate inputs for differential CTs. Consequently, I do not think it is advisable to share a differential CT with other functions.

This is because generator short circuit currents are high multiples of rated current & can have high asymmetry (DC component) which demand high saturation voltages from CT. Note that a 5P20 CT must deliver 20 times rated current with 5% error into its rated burden + its own resistance, hence having a high saturation voltage. Because a differential relay is desired to operate very quickly [~1 cycle] and with small real differential currents, these CT errors are a major limit on performance.

Necessarily, generators must be the last elements to trip in a power system, so it is very important that a fast protection like differential does not operate when the fault is not in the generator e.g. a high current "through" fault.

I believe that hydro machines are commonly low speed machines with high reactance (L/R) and time-constants for DC components, risking saturation.

The number of CTs or windings on a CT and the number of relays may reflect the reliability/burdens of CTs, windings & relays and the technology used when the plant was built.

Modern digital relays can have very small CT burdens and multiple functions so that two relays & two CT sets can provide a full back-up against a single failure [except a software fault, which would disable both relays for a specific fault]. However, since differential protection must necessarily have CTs at both neutral & line sides this is bound to lead to more CTs.

For zone protection or differential protection, the PS class CTs are used.The knee point voltage of the differential protection class(PS- Protection special) class CT is much more than the KPV of protection class CT 5P20. The magnetization characteristics of the PS class CTs mounted at both sides of the equipment(transformer primary and secondary side) must have the same KPV at Ie/2 excitation current.

The equipment will be at risk if the protection special(PS) class CTs are not used for differential protection.

Read More:Importance of Knee point Voltage and its calculation

Thank you for your input. But in most cases that i have seen, they have used class CT 5P20 for differential protection also.

it is theoretically possible that it could be done with the help of one transformer if any high impedance sensor were connected to the output and amplified to trigger the appropriate relay.

For more detail please fallow the link below..

Current Transformer Questions with Answers