Transmission Line Protection Diff vs POTT 3

[George556]

I work for a big utility and am working on a new relay package for two parallel transmission lines (230kV). Each line is 12 miles.

The other end of the lines are owned by another big utility.
My utility’s practice is to put two current differential relays with backup elements in each relay (L90, SEL-411L) on diverse comm channels.
The other utility has a different practice which is one diff relay on fiber and one POTT scheme on digital microwave (SEL-411L, D60).

We already have two other lines between us and the other utility’s station which have L90 diff on fiber with backup elements and SEL-311C POTT on digital microwave. This was done in 2013 before I worked at my utility.

I tried to push for two diff relays with backup elements since we have the comm channels, but the other utility wants to keep consistency with the other shared lines, L90 and 311C.

I’ve never set a POTT scheme. I feel more comfortable with two diff’s, since they are solely current based and more sensitive, but what are your thoughts? Will this be a reliable scheme and how is the 311c?

 

[RRaghunath]

The Line differential protection relays come with built-in distance protection option as well.
This can probably meet the requirement of both the parties.
Coming to POTT scheme for distance protection, it is the recommended option for short lines as the Z1 reach is small and hence linking teleprotection communication with Z2 distance element makes it more reliable.
The only difference between PUTT and POTT schemes is whether Z1 or Z2 that is linked with teleprotection communication channel.

 

[davidbeach]

Why "vs."? Why not both?

My approach is to lay down a good foundation with a pair of matched relays; the SEL-411L in my case but could be two of something else as well. Why two of the same? Because failure to trip isn't a problem the industry has to deal with. 95%+ of all misoperations are unnecessary trip rather than failures to trip. Two different relays provides twice the opportunity to create an unnecessary trip. Back in the early days of the transition from EM relays to numeric (including the solid state interlude) when the numeric were new and untried the two different approach may have been very prudent, but it doesn't seem to be today. Another reason to go with two of the same is that for every hour I get to work on that terminal I get to spend that hour on getting one relay setting right; with two relays I get 30 minutes for this one and 30 minutes for that one. In the real world where the primary common mode protection failure sits in a chair, the more time we can spend trying to get it better, the better off we will be.

So, with a matched pair of relays, the first level of settings is a fully functional step-distance protection, both phase and ground. We use phase mhos and ground quads. Ground distance is far easier to set and get coordinated than directional ground overcurrent despite an awful lot of nay-saying. Mutual coupling means ground distance isn't quite as easy to set as phase distance, but the radical changes in source impedance caused by line and transformer out cases doesn't cause nearly as much havoc as it does for ground overcurrent.

Given the availability of decent, or better, comms we then add a layer of DUTT/POTT. Given that we're using SEL relays that's done with Mirrored Bits. In addition to DUTT (bit 1) and POTT (bit 2), we also use remote drive to lockout (bit 3) and local breaker status (bit 4 plus bit 5 where applicable). All of those are reasonably forgiving of minor comm issues and can be run over reroutable paths. For POTT we use directional ground overcurrent since actual reach isn't as important getting the direction correct.

Given the availability of good, point-to-point, comms we'll add 87L on top of the rest. With 87L less forgiving of comm glitches we generally block 87L on loss of the normal channel rather than rerouting.

87L is nice, and we use it where we can, but many 2-terminal lines the truth is that for nearly all faults the zone 1 elements will get at least one of the ends tripped before any of the comm assisted schemes and that the POTT almost always gets to a trip faster than the 87L. For longer, higher voltage, lines 87L give the ability to subtract out charging current; that can be useful as sometimes charging current can be higher than the current associated with a high resistance ground fault. 87L is also better suited to lines with series capacitors, but traveling wave may be even better.