4 kV Overhead Distribution Circuit Burned Clear - All Phases Downed Mid Span

[saladhawks]

The electric utility I work for recently experienced a 4 kV Overhead Distribution Circuit line down, where all 3 phases of 4 kV circuit burned clear mid span. Span where 4 kV circuit burned clear had double-ended construction on both sides, so there was no significant tension on this span. All 3 phases of 4 kV circuit that burned clear were 336 All Aluminum Conductor (AAC) with low voltage tree insulation.

Initiating event was a 34 kV over-build circuit dropping a single phase bare conductor onto this 4 kV under-built circuit. 4 kV circuit went through a series of reclose shots and then subsequently held with normal load current of approximately 50 Amperes per phase for 60 seconds.

After 60 seconds, 4 kV circuit relayed out and reclosed again. Circuit held, but load current was now 0 Amperes. Circuit was then cleared from SCADA due to closed circuit breaker with no current indicating an energized line down condition.

Most interestingly, the distance between initiating 34 kV circuit event and where under-built 4 kV circuit phase conductors burned clear is approximately 1,000 feet apart, separated by multiple pole spans.

I am having trouble grasping how it would be possible to burn clear 336 AAC conductors that far away from initiating event.

Any thoughts would be greatly appreciated.

 

[davidbeach]

How much of the 4kV circuit was the 336 AAC tree wire? From the 4kV source to the point the 34kV got into it what else is there? Was it a deliberately slack span or are you surmising that there was no significant tension? Was any significant amount of load tapped off the line between the burn-down location and the 34kV fault?

 

[waross]

... there was no significant tension on this span.

A suggestion. Magnetic forces. This span may have been the span with the most slack and as a result was the span that experienced the greatest displacement from the magnetic forces due to the fault currents. When the fault currents were interrupted the lines swung back together just as the circuit re-closed, lighting up another arc.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[davidbeach]

Following up on Bill's suggestion - what is the reclosing sequence for the 4kV circuit?

 

[jghrist]

You wouldn't think that momentary contact from the phases touching from magnetic forces would cause a fault on tree wire. The insulation is not good enough for continuous contact, but it is meant for momentary contact with trees (although with Ø-grd voltage not Ø-Ø). Was the phase spacing such that the conductors could have been wrapped together for a long time? Once there is a fault, one problem with covered conductors is that the arc is confined to one spot and the conductor will burn down sooner.

 

[saladhawks]

As follow-up, the 4 kV under-built circuit is all 336 AAC from substation to point at which over-built 34 kV bare conductor dropped.

Distance from substation to location where 34 kV bare conductor dropped is approximately 3,000 feet.

Distance from substation to where all 3 phases of 4 kV circuit burned clear is approximately 2,000 feet.

Reclose sequence when 34 kV over-built conductor dropped was:

Trip #1
- 34 kV bare conductor de-energized when it makes contact with under-built 4 kV circuit
- Starts as 1-phase fault
- O/C elements picking up / dropping out over the course of seconds
- Final trip is on 3-phase fault
- Total time from initial o/c element pickup to circuit breaker trip is 7.54 seconds

Reclose # 1
- 3 second delay
- Circuit breaker closed
- Starts as 1-phase fault
- Final trip is on 3-phase fault
- Significant fault current
- Total time from initial o/c element pickup to circuit breaker trip is 0.238 seconds

Reclose #2
- 15 second delay
- Circuit breaker closed
- Circuit breaker holds for approximately 51 seconds with normal load current (~ 50 Amperes per phase)
- Reclose relay resets after 30 seconds

Trip #2
- Starts as phase-to-phase fault
- Final trip is on 3-phase fault
- Total time from initial o/c element pickup to circuit breaker trip is 0.813 seconds

Reclose #3
- 3 second delay
- Circuit breaker closed
- 1-phase fault
- Total time from initial o/c element pickup to circuit breaker trip is 1.303 seconds

Reclose #4
- 15 second delay
- Circuit breaker closed
- No load current (0 Amperes across all phases)
- Circuit breaker tripped from SCADA due to likely energized line down condition

My assumption is that 4 kV circuit at location of burn down got wrapped together during initial Trip / Reclose sequence.

If so, while wrapped together for the 51 seconds with normal load current (~ 50 Amperes per phase), is it possible that enough heat was generated to burn clear the 336 AAC conductor?

 

[Mbrooke]

What about the 34kv during this period? Id imagine that can burn clear way more.

 

[saladhawks]

34 kV Over-Built Circuit was de-energized during this entire sequence of events on underbuilt 4 kV Circuit.

 

[davidbeach]

So a 34kV conductor could have been laying across the 4kV tree wire during the 50A interlude?

 

[saladhawks]

Yes, 34 kV over-built cable could have been laying across under-built 4 kV Circuit.

The catch is that where the 34 kV over-built cable was laying on under-built 4 kV was approximately 1,000 feet from where all 3-phases of 4 kV circuit burned clear.

 

[JensenDrive]

I would surmise the single phase 34kV touched one phase of 4kV, but flashes phase to phase to ground at 4kV 1000 feet away. The 34kV line provided most energy to burn down wires. Each reclose was further burn down effect.