Failed 15 kV Vacuum Circuit Breaker


Thread Starter


I would like to consult you and ask your opinion about the failure, while in service, of a POWL-VAC circuit breaker protecting the main transformer (GSU). The specs are the following:

Type: 15PV1000-63 Vacuum Circuit Breaker
3-phase, 60 Hz, 15 kV, 3000 A Continuous
Rated Short Circuit Current Cap = 48 kA
Impulse Withstand Rating = 95 kV
Instruction Books: IB-60010, IB-60013

The phase "A" damage parts are burned sliding contacts (with erosion marks), burned VI lower contact block, burned strut kit, shiny molten metals deposited on the strut kit, burned yoke, burned bell crank, base of the vacuum bottle and movable stem are also burned, soot deposits on the porcelain shell, part of the lower primary disconnect near the lower wishbone insulator was also burned. Phase B and C were not damaged.

Other observations on the phase A are bent operating pin and silver whiskers growing on the surface of the lower contact block. What do you think is the cause of the severe overheating of the said parts? And what are your recommendations (solutions and interventions) to prevent similar incidents from happening in the future?

The transformer is 230 kV (wye) / 13.8 kV (delta). It is solidly grounded through the neutral bushing H0.

The failed breaker is installed at the primary side(13.8 kV) of the transformer going to the 15 kV switchgear.

Broner... thanks for info. Now can you provide some detail:

1) Which Protective device cleared the fault?

2) Was breaker "bottle" compromised?

3) Do you have photographs of the damage?

4) Is the breaker-enclosure mounted on a 'Roll-out' frame?


Here are the answers to your query:

1. None of the transformer protective devices cleared the fault. The unit was manually shutdown by the operator. The following relays did not activate:

1.1 87T
1.2 51G
1.3 51 (for phase A, B & C)

The 87B protecting the 15 kV busbar and the 51 (TOC) relays protecting the aux. feeders (emanating from the said 15 kV bus) did not activate.

2. The vacuum bottle itself has no damage.

3. Yes, I have pictures. Still have to figure out how to attach them in this thread.

4. Yes, it can be rack out and rack in.

Thanks for bearing with me.


5) What prompted the operator to trip the CB manually? Had he/she noted smoke, arcing noise, loud hum?

6) Was damage evident to any wheel or rail?


here are the answers:

5. there was smoke coming from the 15 kV VCB cell.

6. there are no damage to the wheels and rail.


You said the neutral was solidly-grounded, but I believe you were referring to the Wye or 230kV-side of the GSU. Additional questions:

8) Is the 15kV system grounded via the Generator-neutral? If so, how... NGR, NGT, other?

9) Although no current-related relays operated, do you have any recordings of the Pre-fault, during-fault, and post-fault voltage, current, or power measurements?

10) Was the generator connected to the 15kV systyem at the time?


here's my answer to your query #7:

for main transformer 50/51 (A,B,C)

Min pick up = 7 Amps @ 1 sec
Min pick up = 3.3 Amps, instantaneous

for main transformer 51G

Min pick up = 11 Amps @ 1.5 sec

for main transformer 87T

(not quite sure about this)

Unrestrained = 28 Amps (input 1); 46.2 A (input 2)
Restrained = 2.0 Amps (input 1); 3.3 A (input 2)


8. The 15 kV busbar grounding system has current limiting fuse per phase. It also has PT per phase for ground detection (alarm).

The generator ground on the other hand consists of distribution transformer with a resistor installed at the secondary winding.

9. I have no data yet

10. Yes, the generator is on line during the incident. It is supplying house load and the grid.


one last question:

11) I presume that all amp readings you provided are based on primary currents... not CT secondary currents. If the latter, then I need CT ratios!


Here's the answer to question # 9:

Pre-fault (before a burnt odor was first noticed):

Phase A = 2001 Amps; B = 2056 Amps; C = 2076 Amps (45 MW load)

During (about 20 minutes before the unit was manually stopped, smoke was observed)

Phase A = 1834 Amps; B = 1901 Amps; C = 1914 Amps (41 MW load)

Post (after the unit was synchronized back to the grid, with the VCB already replaced)

Phase A = 1870 Amps; B = 1940 Amps; C = 1960 Amps (43 MW load)


Those are secondary currents but corrections are in order. Mea culpa. Below are the correct secondary current settings:

for 50/51 relay (A,B,C)- Basler BE1-50/51B-105
Timed Pick up = 5 Amps (M curve, time dial 3.0)
Instantaneous Pick up = 33 Amps
CTR = 4000/5

for 50/51 relay (Ground)- Basler BE1-50/51B-105
Timed Pick up = 7.7 Amps (D curve, time dial 1.7)
No instantaneous setting
CTR = 200/5

for 87T relay (three phase)- Basler BE1-87T
Restrained Pick up = 0.7 Amp (input 1); 1.155 Amp (input 2)
Unrestrained P/U = 28 Amps (input 1); 46.2 Amp (input 2)
CTR (230 kV side) = 600/5
CTR (13.8 kV side)= 4000/5


I believe your 'incident' was the classic arcing-ground fault! Such faults produce current-magnitudes well below Phase-to-Ground expected Irms values. Several facts lead to my conclusion:

A) The current-magnitude (or duration) was sufficient to produce the conditions observed!

B) The current-magnitude (or duration) was insufficient to trigger any of the current-sensing relays!

C) The current-magnitude (or duration) was insufficient to produce an alarm triggered by the Open-Corner, Delta-Connected PTs!

I recommend the following course-of-action:

o Determine adequacy of the switchgear groundING connection, whether via single, but flexible, conductor, or bus-stab! Insure that the wheel-to-rail connection is never in the return path!

o Determine the ground-return path impedance between the VCB drawout frame and the generator-neutral connection!

Let me know if you need assistance with a relatively “easy” way to perform the impedance measurement!.


PS" can you send pictures to me at:

[email protected]

I also want to alert you to another phenomenon that could result in low-magnitude ground-current.

It can occur if the resistive effect of the generator’s NGT method results in a resistor too small to fully negate the effect of a large MV system’s Ground-Capacitance!

High Ground-capacitance is present if the distribution system consists of a large number of distribution cables and/or a large number of motor and transformers.

Do you know the details of the NGT design?


Thanks so much for your insightful conclusion and valuable recommendations.

My follow up questions are:

a. how can it be determined whether the wheel-to-rail connection is in the return path or not?

b. what are the usual cause or causes of an arcing ground fault?

c. how can a low magnitude SLG fault current leads to the burn down of the lower primary of the VCB?


The generator NGT is a 42 kVA, 13.8 kV/220 V transformer. I still have to verify the size of the resistor installed parallel to the secondary winding of the NGT.

I would like to restate that the burned VCB is not the generator breaker. It is protecting the GSU transformer (installed at the primary 13.8 kV). Neither the GSU and the generator was damaged.