Home Page News Search Contact Us Language Bar فارسی English
Protection Riddle No.38 - Restricted earth fault protection and Inrush current
As you know, the inrush current just appears in one side of transformer when we energize power transformers. Therefore transformer inrush current can be cause the malfunction of differential protection; because of unbalance detected current between primary and secondary side. On the other hand, according to technical recommendation, the applied restricted earth fault protection to one side of transformer shall be check for inrush currents too. Why? Regarding activation of restricted earth fault protection system in one side of transformer ( except auto transformers ), why shall we check the existing inrush current in unbalance zero current detection process similar the overall differential protection cases?
Author : Hamid - From: IRAN
Wed, June 17th, 2009 - 08:53
The inrush current flowing in the primary while energisation of the transformer is not of same magnitude in each phase, they are different in magnitude.  Hence there will be phase currents of different magnitude and also it is not balanced through neutral.  So there will be differential current in REF relay and hence it may operate during energisation if proper filtering/suppressing of 2nd harmonics is not done.

Author : Abdul Hameed Pathiyil - From: Indian from Qatar
Wed, June 17th, 2009 - 10:41
Please note, high impedance type restricted earth fault relays just contain one signal input and individual phase current do not insert to relay. In low impedance type equally, the function of relay is related to zero sequence calculator stage of relay; therefore the phase unbalance of inrush current cannot cause relay mal-operation. So REF must be stable against through zero sequence inrush current theoretically. 
Author : Hamid - From: IRAN
Wed, March 31st, 2010 - 18:29
Dear All,

REF protection being a current balance scheme applied to only one side of the transformer, it would appear that no second harmonic restraint is required as in the case of differential relay.
However, REF relay, usually being set very sensitive (about 10-15% of CT rating) and without any intentional time delay, it is imperative that CT secondary current balance is maintained fairly accurately for all through faults, in zone phase faults, system recovery transients (after external fault has been cleared by another relay) and transformer  ‘inrush’ currents. Of the above, transformer ‘inrush’ imposes the most severe duty on CTs and can cause saturation. The inrush current contains high magnitude of both odd and even harmonics which are not symmetrically placed across all phases (resulting in  neutral harmonic currents other than triple n.) and DC offset of long decay times of about 100 msec to several seconds with different magnitudes in each phase. If fast reclosing is employed in feeder circuits connected to the same side of the transformer, the remnant flux  in CT core can aggravate the situation further. Under these conditions, there will be mismatch in secondary currents and if suitable precautions are not taken the relay may maloperate.

The well known High Impedance REF scheme, if properly set, is quite secure under the above system conditions. Here again the relay is made immune to DC and tuned to system frequency with the further advantage that the setting voltage can be made only on the basis fundamental and therefore,  CTs of lesser knee point can be used. But the High Impedance scheme has a disadvantage that dedicated special protection class CTs are normally required. Also, voltage limiting devices are required across relays in each phase to limit excessive voltages during in-zone faults fed directly by the system.All these add to the cost.

With numerical relays, manufacturers claim that dedicated CTs are not necessary because of the processing capability of these relays. Most of the present-day numerical relays use low impedance measurement principle with the following security measures:
(a) Current bias (static or a combination of static and dynamic)
(b) Directional check (using fundamental phasor) of residual current with neutral current as polarizing signal.
(c) Coincidence of residual and neutral current.( In case of actual faults they are almost coincident.)
(d) Operation only on the fundamental component of the  differential current.

With the above security measures, especially the directional check, I understand these relays offer good security against maloperation in the above mentioned scenarios. As Mr.Hamid has mentioned, some numerical relays still use second harmonic restraint and block the relay if it is more than a certain percentage ( fixed setting). I am not sure why they do this. May be they place more onus on the security of the protection. Other reason I think would be the inability of the algorithm to make directional check in the event of too low neutral current and thus use the restraint function for security. I am not sure about this. Please post you comments.
Author : K B Santharam - From: India
Sat, April 3rd, 2010 - 10:15
As Mr. K B Santharam said, the DC off set inrush current can cause CT saturation in initial short time fault current which subjected to REF operations. I think mentioned facility in new numerical relays cannot cover mal- function of saturated CTs in protection system. 
Author : Hamid - From: Iran
Submit Your Answer

Change Language :