Hey! The line breaker won't close.

Here are some slides from a presentation I did a few years back on an operations problem that I helped to troubleshoot

After a few unsuccessful closes, we pulled an event record from the line relay and this is what we saw. At first glance, you see the breaker close, then we get current on one phase only, we key permissive, receive permissive back (the other end of the line was not closed at the time, so it was echo), and then we trip. Without staring too hard, it looks like an A-G fault.

But hold on just a second. If this is a fault, then why do we have full line voltage on the "faulted" phase, and reduced voltage on the unfaulted phases?

And the phasors sure don't look like fault current. The current should be lagging if there were an actual fault. At this point, Operations tried again, but using the other line breaker (this was a double bus, double breaker station).

Dang... Exactly the same thing. So whatever's going on, it probably has nothing to do with the line breakers, so it's back to the event record. We obviously have voltage and line charging current on A phase, but pretty much nothing on B or C phase. Since we get the same results using both line breakers, it's probably not the breakers (to have two identical, but independent failures happen concurrently isn't something you propose as your first bet). So what else is in the line that could possibly be open on two phases?

Aha! It's the line disconnect. The station where this was happening was the one at the bottom of the transmission line one line diagram. At 500 kV, phase distances are large enough (and the moving parts are heavy enough), that you have independent motor operators for each phase, even if there is no intention of ever operating these devices single phase. So we had A phase closed, but B and C phases open, therefore we were only ever energizing one phase of the line. Also note the position of the line CVT on the line side of the disconnect. This explains why we would only see single phase voltage, even though we were closing the breakers 3-phase.

Note that the line voltage on B and C phases is in phase with the A phase line voltage. This is capacitive coupling between the energized A phase, and the floating B and C phases.

The line tripped via the permissive scheme. The far end was open, so it keyed echo on receipt of permissive, however the operation of the local end is a bit confusing. Line charging current is capacitive, so when you look at the vector relationships between Va and Ia, they appear backwards to what you would normally expect for an A-G fault. Is this right? (yes, it is, but we need some suspense here).

The permissive scheme operated on 67G2, which is a sensitive zero sequence overcurrent element torque controlled with a negative sequence directional. The important thing here is that in a fault situation, V2 points in the opposite direction to the faulted phase voltage (it causes the voltage to drop), while in a single phase energization scenario, V2 is in phase with the phase voltage. So even though our negative sequence currents in the two scenarios are 180° apart, the negative sequence voltages are also 180° apart, and both scenarios are considered a forward fault by the relay.

For an A-G fault, here are the voltage and current phasors (on the left), the associated sequence components (in the middle), and the reassembled phasors (on the right). Note how V1a, V2a and V0a add up to zero, as do I1b, I2b and I0b, as well as I1c, I2c and I0c.

And here is the same decomposition of phasors for the single phase energization scenario. Compare the sequence phasors to those in the previous slide and note how except for V1, everything is pointing in the opposite direction.

Here, we just show the negative sequence voltage and current phasors for both scenarios, and it's easy to see that even though V2 and I2 point in the opposite direction in both cases, the angle relationship between them is unchanged.