Faster clearing Time 69 kV Substation Grounding

[cuky2000]

For a new 69 kV substation with protection and control based in modern microprocessor relays, what should be the minimum safe clearing time recommended to design a ground grid?

 

[davidbeach]

Local breaker failure protection? Single or dual batteries/trip coils? What’s the overall protection philosophy? In other words, It Depends.

 

[davidbeach]

Oh, and how many contingencies do you consider to be credible?

 

[cuky2000]

Hi Dave,
Thanks for been the first to start answering this question. Here Is a background information of the issue:
The conventional philosophy selecting a clearing time is predicating a breaker failure and clearing the fault by the backup circuit breaker. It is common practice to use 0.5 seconds (30 cycles) or larger as a clearing time to design a safe ground grid at 69 kV. This parameter is used to determine the allowable step and touch potentials that are inversely proportional to the square root of the electric shot duration.

In adjacent 230 kV substation built last year, the utility accepted a clearing time of 17 cycles as an special unusual case since this is the maximum time allow by the ISO to maintain the system stability limit for stuck breaker condition.

The engineers proposed using the same clearing time of 17 cycles for the new 69 kV substation and was rejected. It should be noted that the allowable stability limit for 69 kV is 38 cycles. At difference of the 230 kV system, no fiber optic(OPGW) is used. However, similar microprocessor relays and 3 cycle circuit breakers are designed such as the backup protection device theoretically may clear any ground fault within 17 cycles.

The cost of the ground grid could be unusually expensive due to the high soil resistivity of this site and the large clearing time of 30 cycles. So the question from a grounding design prospective is: what is the minimum safe clearing time that can be obtained from a distance protection, DTT or a typically backup scheme commonly use for 69 kV applications?

 

[davidbeach]

Yep, there are definitely pressures to come up with very fast clearing to minimize ground grid costs. We're working on a number of projects, some with 115kV ground faults approaching 30kA and the clearing time questions get intense. For one project we found that speeding up the remote clearing saved enough ground grid cost to pay for adding breaker failure protection at the remote end.

The pessimistic Protection Engineer wants to consider failure of lots of breakers, say a battery failure. The project manager looks at that cost and pushes back. Ultimately the company has to decide how many contingencies to build the ground grid to contend with and then figure out which collection of those contingencies produces the worst case. That's highly likely to include breaker failure at the remote end of the strongest source.

With ground distance you pretty much know what the remote zone 2 time will be. If all of the remote zone 2 times are the same the worst will be the strongest. But with directional ground overcurrent at the remote ends the remote clearing times vary from line to line and from system configuration to system configuration; a lower current but slower clearing source can be worse than a higher current and faster clearing source.

Ultimately you're weighing the balance and determining that you'll build a ground grid for a really bad event, but not for a terribly horrible event. (Terribly horrible would be at least as bad as no local clearing, no transfer trip, and breaker failure at all remote ends.)

By the way, I very much prefer David over Dave. Thanks.

 

[powerpi]

Hello
Is your network grounded directly? Perhaps isolated? Do you have any experience of remote end Z2 tripping in case of high infeeds from other lines in case of S/S nearby faults?

Regards

 

[jwatty]

It may be worth asking for fault currents and worst case operate times at the point instantaneous protection no longer works (ie, the edge of zone 1/zone 2 step distance, and the edge of 50/51 relaying). 17 cycles may well be sufficient for close in faults where instantaneous relaying may operate, and the lowered fault currents & longer clearing times at the edge of zone 2 may not be the worst case. When I usually get this question, I'm asked for a maximum clearing time and maximum current. Both of these don't usually occur at the same location. If this doesn't help, a few inches more of rock may be cheaper than a copper plate, if the utility will allow it.