ABB REL 511C Line Relay - Ground Distance Element

[vandal06]

I am performing a system coordination study and this relay is installed all over the client's 115 kV transmission system. My experience has been with SEL's offerings so I was immediately struck by the quadrilateral construction of the impedance elements, both for phase and ground. I'm most familiar with phase/ground mho's and quadrilateral grounds, with my experience in the latter much less than the former. SEL distance relays to my knowledge only implement phase and ground mhos and quadrilateral ground.

I have access to all the accompanying technical documentation for the ABB REL-511C relay and am somewhat familiar with the operation of the distance elements. I have reviewed the in-service settings for the relays and noticed that only 21P and 67N elements are active; ground distance have been disabled. A direct transfer trip between this terminal and the remote terminal is active. I have asked my supervisor why they would disable 21G and use 67N as primary ground protection and he thinks it's a carry over from older philosophies where ground distance relaying was not common (electromechanical units).

The quad ground distance appears to operate much differently than an SEL version. For one it is positive sequence voltage polarized whereas the SEL version has the capability for I0 or I2 polarization. I've reviewed an industry application guide that states that for long high-voltage transmission lines pilot-wire 67N is ideal and 21X is not. Does anyone have any experience with the REL-511C and can give their thoughts on why one shouldn't enable the 21X quad ground element?

 

[veritas]

Hi vandal06

It is more than 12 years ago that I worked on the REL511 and no I cannot think of an reason the EF distace is not used. Maybe local policy? or history as you say?

How long is the feeder you are working on? Any other funnies like series reactor? Though I would not expect to see one at 115kV.

Watch out for the GFC. I normally set it to trip in 5s as a general backup - care must be tken for heavily loaded feeders.

 

[DTR2011]

I don't have a lot of experience with that particular relay, however your supervisors comments are quite valid.

Consider the philosophy if it isn't broken...

Additionally, for routine testing, the procedures and test plans for Mho and 67 functions are well defined in North America. Testing the Quad characteristic is quite different from a Mho, although not all that difficult to perform.

 

[vandal06]

Hi vandal06

It is more than 12 years ago that I worked on the REL511 and no I cannot think of an reason the EF distace is not used. Maybe local policy? or history as you say?

How long is the feeder you are working on? Any other funnies like series reactor? Though I would not expect to see one at 115kV.

Watch out for the GFC. I normally set it to trip in 5s as a general backup - care must be tken for heavily loaded feeders.

Veritas, thanks for the response. The transmission circuits range from a 1 km long tap all the way to a 145 km long critical line. No series reactors, though one terminal of the long line has a shunt capacitor. I'll need to examine the GFC closer, I've not seen a similar element in the other distance protection relays I've dealt with (Schweitzers). It just looks like a duplicate of the phase and ground distance protection?

I don't have a lot of experience with that particular relay, however your supervisors comments are quite valid.

Consider the philosophy if it isn't broken...

Additionally, for routine testing, the procedures and test plans for Mho and 67 functions are well defined in North America. Testing the Quad characteristic is quite different from a Mho, although not all that difficult to perform.

Thanks smallgreek. What I forgot to mention was my supervisor said to go ahead and set the 21X elements for ground faults, but I'm of the same opinion you are. If it isn't broke...and I'm trying to come to a reasonable justification for either leaving the philosophy as is (21P and 67N) or adding 21N as my supervisor has said. The plan right now is to just go ahead and set 21N, but I'm doing my due diligence as well to make sure it's not something to be avoided.

That's a good point from a testing perspective, but I'd like to point out that the 21P function is actually a quad. This particular relay does not have any mho functions in the 21 elements.

 

[DTR2011]

I forgot to ask, what are you coordinating with on the other end(s)? Are there any issues with the communication system or back up protection that the 67N function excelled at?

 

[vandal06]

This installation is a ring-bus with multiple transmission lines heading out to generating sites with step up transformers, and the long line heads back to the big distribution sub. No line relays at the generating stations, nor 115 kV breakers for that matter. My line relays would be coordinating with transformer protection. For comms there is just a direct transfer tripping scheme that opens the local ring-bus breakers if the remote generator breakers open for any reason, and trips out both ends for faults detected by the line relay.

If we lose comms we'd be relying on local generator protection clearing line faults from that end, but that is a separate issue that's been discussed in scheme report I've already delivered.