Directional Phase Overcurrent picking up on reverse ground faults

[charlierod]

Hello everyone,

In a MV distribution network with local gas engine generators we use directional relays due to the meshed topology of the grid.

Step up 34,5/ 0,48 kV transformers associated to generators are protected by reclosers on the HV side. These reclosers have two groups of settings running at the same time. One group is faster with lower dials intended to protect the transformer and it is adjusted in both phase and ground elements with maximum torque angle set to detect internal faults. The other group has O/C elementos with higher dials and acts as a back for devices protecting the external grid.

We have had several external line to ground faults in which phase directional O/C elements set to operate only for faults within the transfomers or associated feeders pick up and trip resulting in a loss of selectivity and a disconnection of local generation. Analyzing recloser control records we can see that the operation of the device is correct. The problem is that the contribution is mainly due to zero sequence current and the unfaulted phases do fullfil angle conditions for picking up. The device use typical 90 degree polarization.

We know we could solve the problem with a modern relay that includes the posibility to block protection elements with flags from other protection and control elements and we could for example use the ground directional elements flag to supervise the phase directional or use another type of polarization. Well, none of these solutions are possible because the recloser does not include such a functionality.

Has anyone faced this problem in his/ her applications? Any suggestions to deal with the problem? By the way, transformers do not have a diffential relay, the only protections are reclosers installed at HV.

Thanks in advance for your feedback.

Regards,

Carlos

 

[RRaghunath]

Is neutral CT of 33kV winding wired to relay? For directional Earth fault protection never use neutral CT, it should be residual current of phase CTs that should be used.
Is 33kV neutral solidly earthed?
What is the characteristic angle chosen for 67N. If I remember right, the recommended value is -45deg for a solidly earthed system. Since the current in to the faulted transformer has to come all the way from remote, the earth loop impedance of the fault is likely to be high. Thus, even a lower characteristic angle (-30) could be better. Is the 33kV distribution system with shield wire on top?
What is the polarisation quantity - neutral current or negative phase sequence current or residual voltage?
Above information will help analyse the maloperation better.

 

[charlierod]

Hi RRaghunath and thanks a lot for your reply. Here are the answers to your questions:

Is neutral CT of 33kV winding wired to relay? No. In fact neutral CT is not available.

Is 33kV neutral solidly earthed? Yes, solidly grounded.

What is the characteristic angle chosen for 67N? -45 degrees (default setting)

the 33kV distribution system with shield wire on top? Yes, two continuous ground wires which are grounded at each tower.

What is the polarisation quantity - neutral current or negative phase sequence current or residual voltage? Residual zero sequence voltage and the operation quantity is 3I0.

Just to clarify, 67N elment operation is OK. We have problems with 67 elements in non-faulted phases. If we have an AG reserve fault then 67 elements of phases B and C would pick up and if there is some delay in clearance or if the fault is close to the subestation would trip.

Here is the picture of an external SLG fault AG:

MAGNITUDE ANGLE (DEG)
VA 7,140 -9,400
VB 21,660 235,900
VC 21,120 125,100
IA 0,699 102,700
IB 0,431 102,700
IC 0,430 103,800


Regards,

Carlos

 

[RRaghunath]

Since the overcurrent elements are operating, check the directionality with a test kit. It can also be verified ith load current when the feeder is loaded. The current should be (-ve) when the current is flowing towards the line / away from transformer. This is because we want the direction to be positive when it flows towards transformer terminals.
Also, you can verify wiring to relay vis a vis the relay technical manual recommendation.

 

[charlierod]

Hi RRaghunath. Thanks for your feedback. We will check again.

Regards,

Carlos

 

[cranky108]

I'm almost sure you are using newer microprocessor recloser controls, but, with the old form 3's, to add a directional we used directional relays, in the backwards direction to short out the element that we wanted to directionalize.

 

[charlierod]

Hi cranky108,

Thanks for your comments. Yeah, that's right. Recloser control is a microprocesor based unit but not that modern so as to define user logic or at least we have not found how.

Regards,

Carlos

 

[thermionic1]

This does not directly address you problem, but SEL makes a 651R Recloser control that may work as a plug and play retrofit. You would find limitless possibilities available in a 651R for protection elements and very flexible logic.