In the protection scheme of 20MVA, 16.5/6.9kV , Delta-Star connected transformer having an impedance of 10 ohms, bushing CTs with ratio of 300/1A are provided for REF Protection. The CTs provided on the line side have an Ith rating of 2000A. The earth fault current of the transformer is limited to 300A approximately using a resistance of 13.3 ohms. (The LV side neutral is earthed)

While calculating the knee point voltage of the CTs for REF protection of the above transformer, should the full fault current of the transformer or merely the earth fault current (300A) be considered?

Regards,
Pramod

 

Responding to Pramod's 23-Feb-08 (10:18) query:

1) What is meant by CTs? REF ony requires one CT! Please clarify!

2) Knowledge of the CT kneepoint-voltage is unecessary because the maximum REF current is only 300A.

3) Please explain the "10 Ohm" value... is it the anticipated burden of the REF relay?

4) Knowledge of the kneepoint of the 2000A CTs might be necessary if there are very large motors on the 6.9kV bus!

Regards,
Phil Corso([email protected])

 

Sir,
1.REF refers to the restricted earth fault protection of the low voltage winding of the transformer in which the ground fault happening only inside the winding of the transformer is detected by placing one current trasformer(CT) on the neutral grounding and three CTs in residual connection on the line side.

2. CTs knee point voltage need to be calculated because, in case of a through fault outside the zone covered by REF CTs, saturation of any one of the CTs on line side may lead to undesired operation of relay.

3.The transformer impedance is 10 percent which I wrongly mentioned as 10 ohms. Sorry for that.

4. The knee point voltage is calculated for 2000A CTs (provided on the line side for differential protection) using the maximum through fault current value.

 

Pramod, the REF you defined is called the partial-differential scheme (earth-faults only) in the USA. It is used for protection of a transformer’s star-connected winding when faster response is required. Following are additional comments:

5) It is more likely found in systems having solidly-earthed neutrals! Thus, earth-fault currents are substantially higher than 300A!

6) Like the typical differential scheme, the area of major concern is unusual CT-ratio mismatch, not CT-saturation. For the parameters you cited, maximum fault-current (neglecting offset and motor contribution) is about 17kA (rms) or about 8 times the CT’s rating. So, if the CT is typical of those used for protection, saturation is not a problem! The reason is that the knee-point is much higher than those used for billing or measurement purposes!

7) Presumably you are not using the secondary line-side CTs asigned to the transformer's differentil protective circuit. Is this correct?

8) Frankly, if the majority of 6.9kV loads
consist of motors or distribution transformers, I would not recommend the REA!

9) BTW, did you mean residual-connection or did you mean parallel-connection of the three line-current CTs?

Regards,
Phil Corso ([email protected])

 

Pramod, further to my earlier 24-Feb-08 (19:29) response:

10) Hopefully, you realize that since the REF scheme does not respond to earth-faults beyond the protected-zone (star-connected winding), an additional EF relay is required as backup for earth-faults in outgoing 6.9kV feeders!

Regards, Phil Corso ([email protected])

 

Sir,

6) If CTs with ratio of 300/1A were used for REF protection in place of 2000/1A CTs (i.e fault current is approximately 57 times the CT rating in this case), would a knee point check be required even if CT secodaries are arranged in a parallel (residual) connection?

7) Yes, we are using a seperate set of CTs for differential protection.

8) You would not recommend using the REF if the majority of 6.9kV loads consist of motors or distribution transformers. Is that because the starting current of motors & inrush magnetising current of transformers may cause an undesired operation of REF relay.

9) By residual connection, I meant the parallel connection of three line side CT secondaries & one neutral side CT secondary before conneting them to relay.

Regards,

Pramod

 

Pramod, re: your 26-Feb-08 (00:27) reply:

Answers 6) & 9), yes!

Answers 7) & 9), thank you for clarification!

11) If you continue with the REA scheme, then keep the 2000/1 CTs. A resistor in series with the EF relay will preclude false operation!

12) Target 10% of EF current (i.e. 10A) as trip set-point.

13) Let me know if you need recommendation for EF protection of 11kV feeders!

Regards, Phil Corso ([email protected])

 

Sir,

10) Yes, we are providing a differential relay to take care of earth and phase faults from HV side circuit breaker to LV side circuit breaker. In addition an O/C & E/F (51/51N) is also used in the downstream of LV side breaker. Further, a stand by earth fault relay is also provided on the neural of the transformer.

Regards,

Pramod

 

Okay, I have read your clarification,
but I have seen in my plant that REF is not used for distribuition transformer, i.e. 6.6 kv/415 v. What would be reason? Also please clarify, differential protection is also not given in case of distribution transformer. If we connect REF for resistance grounded neutral what would be effect of that?

 

Responding to Samsung’s 08-May-08 (12:52) query… First, let me point out that this thread started out discussing CT saturation. It morphed into a discussion about REF, but for higher voltage levels than evident in your query.

That said, I’ll address your concerns: first, Restricted Earth Fault relaying, or second, Differential Protection, is never used with the typical LV distribution transformer. Putting safety concerns about neutral-earthing aside, there are several reasons why higher levels of protection are not used:

1) Solidly-earthing the neutral of the secondary wye-winding assures that earth-fault current is high enough, i.e., about the same order of magnitude of a 3-phase or a phase-to-phase fault (please, no trivial arguments that ph-ph is only 87% of 3-ph!) Thus protection for an earth-fault relies on the same devices, i.e., breakers and fuses, installed for that purpose. If the REF method were used then every single distribution circuit, regardless of the protective device used, would require additional sensitive earth-fault detection relaying. The cost would be prohibitive!

2) Differential Protection reduces fault duration. While it can be argued that it provides more sensitive protection for the transformer’s primary winding, its advantage in secondary circuits is limited to situations where it was impossible to coordinate with downstream feeder circuits using only time-current relaying!

3) Protection methods, regardless of sensitivity can’t eliminate electrical faults, but can only reduce their duration, hence diminishing the extent of damage!

4) The single LV distribution transformer, if considered a major economic asset like electrical equipment found at higher voltage levels, would require that alternative distribution schemes be used!

Let me know if additional information is required!

Phil Corso ([email protected])

 

Sir,

Our distribution system has a 6.6kv/415v delta star transformer having a REF protection. The LV side breaker having E/F protection also. One of the outgoing motor winding E/F created and it tripped the transformer by REF. Feeder also having E/F relay. Can you give some advice?

 

Responding to Suresh’s 26-Jul-08 (02:95) query… please provide details:

1) Transformer parameters, rating and impedance.
2) Phase CT, rating and ratio.
3) Neutral CT, rating and ratio.
4) All CT secondary resistances.
5) Burden (relay) impedance!
6) Secondary cable impedance.
7) Value of stabilizing resistor, if installed!

Phil Corso ([email protected])

 

Sir,

Few queries regarding REF.

1) What is Knee-point voltage of CT's & how to calculate it?

2) Most of the time i read that REF scheme is applied for Delta-Star transformers (Am i correct?). My question is, Can we apply REF scheme for Delta- Delta transformers. If yes then how would be the connections of CT's? And how much CT's we should used?, Also where will be location of neutral CT?

3) REF relays are of two types (If i am not mistaking). a) High impedance b) Low impedance. With which conditions & where these should used?

4) If your REF is acting for the through faults, what would be the possible reasons?

SANTOSH
santosh.jagale [at] adityabirla.com

 

Santosh... have you done a search of the Control List Archives?
Regards, Phil Corso (cepsicon [at] aol.com

 

>1) What is Knee-point voltage of CT's & how to calculate it? <

Firstly, The Knee-Voltage of CT is the Voltage beond which the CT is saturated. If Knee Point Voltage of any CT is 200 and if burden on the same goes beyond this specified value then the CT becomes Saturated and it is no longer able to tranform the Current from Primary to Secondary.

Secondly, The Knee Point Voltage value of Protection Class CT is often described by the Relay Manufacturer, for Example AREVA make HORM4 relay in which they describe Vk = 50/In + If(Rct+2*Rl) Where In is Rated Current of Relay, If is Max through fault Current, Rct is Resistacne of CT Secondary Winding and Rl is Resistance of Single Lead from CT to Relay.

The Knee Point Voltage of CT varies from Relay to Relay as per their implementation for different Protection.

> 2) Most of the time i read that REF scheme is applied for Delta-Star transformers (Am i correct?). My question is, Can we apply REF scheme for Delta- Delta transformers. If yes then how would be the connections of CT's? And how much CT's we should used?, Also where will be location of neutral CT? <

Dear Santosh, please understand that REF is nothing but Differential Protection specifically employed for Inner Zone Earth Fualts. In Delta-Star Trafo, Earth Faults have Neutral circuit to flow. Thus only we provide one CT on Neutral Side to detect Earth Fault flowing through the Neutral. For Delta-Delta Trafo as no Neutraling is done so such faults shall be detected by the Differential Proection CTs. and Diff Prote will operate for Inner Zone Earth Faults.

 

Dear Hitesh,

you have mentioned above that For Delta-Delta Transformers, such faults are detected by differential CTs.I am not able understand this.If there is a single point fault in any one of the delta winding how the fault current is going to flow. Current will only flow when there is another fault point in the same winding then only fault current will flow and differential prot. will operate. pl comment.

 

I understand you are talking about REF protection of a resistance earthed star winding. You are applying REF protection because of the limited sensitivity of the differential protection to earth faults on the LV star winding. This is common in the UK.

I assume you are using high impedance REF protection. Through fault stability should be based on the maximum through fault level (I think that will be a 3ph LV fault in your case, pls check). Hence the stabilising resistor (for a current operated relay) will be calculated based on this level. The CT knee point voltage is normally based on the actual voltage setting, e.g. 2x Vs(act).

The maximum internal fault level will be low, and metrosils may not be required to limit the voltage across the relaying point during internal faults.

Care must be taken if you intend to use residually connected CT's on each phase, because of the high 3ph fault level and load, relative to the CT ratio.

As you have already stated, you would also have separate differential, LV stand-by earth fault and HV over-current protections.

You would be best to discuss this application with the relay manufacturer.

As far as earth faults and delta/delta transformers are concerned, you need to draw out the sequence networks. As you say, another earth is required on the system. Low fault levels may exist due to cable capacitance, i.e. analysis of earth faults on ungrounded systems.

 

I think for line CTs of REF protection knee point voltage is important and shall not saturate for through fault current. However for single CT in transformer neutral for REF protection (residual connection) fault current is limited by transformer NGR, hence CT knee point voltage is not so important.