Ageing switchgear fails after warnings

What happened?

A GEC Type VMX high-voltage vacuum circuit breaker failed, causing electrical arcing and a fire within a substation.

Protection systems cleared the fault within 700 milliseconds (0.7 seconds), but significant damage still resulted.

The original VMX unit had suffered severe partial discharge and was replaced in 2004 with a second-hand unit.

The safety notice applies to all three forms (A, B and C) manufactured in the UK between 1985 and January 2003, including units badged GEC, GEC Alsthom, Alstom and Areva T&D.

The failure had the potential to cause serious injury or death.

Why did it happen?

Partial discharge activity within the ageing vacuum circuit breaker was the most likely cause. It progressively degraded the insulation, leading to disruptive failure with arcing and fire.

Partial discharge was recorded at 6-8 dB 24 months and again 14 months before the failure, but the warning signs were not acted upon. Relative humidity was measured at 71% and 81%.

Substation environmental conditions contributed to the failure. The mechanism is strongly linked to relative humidity above the manufacturer maximum of 80%, condensation and temperature fluctuation.

Intrusive maintenance two months before the failure detected no discharge. A single clear test result did not clear equipment with a history of partial discharge.

Outstanding legacy design modifications and Improved Maintenance Practices may not have been completed or repeated. The provenance, modification status and condition history of the second-hand unit also required verification.

What did they learn?

Identify whether GEC Type VMX switchgear is installed on, or forms part of, any substation accessed by the site. Confirm the manufacture date and check serial and work order numbers against the original equipment manufacturer Table of Potential Problems.

Review maintenance records to confirm that all applicable design modifications and Improved Maintenance Practices have been implemented and repeated where required. Where records are incomplete, contact the current OEM for guidance.

Verify that relative humidity does not exceed 80%, and control condensation and temperature fluctuation. Record relative humidity and ambient temperature at the same time as every test result.

Arrange an initial partial discharge survey and vacuum interrupter integrity review by a competent person, in line with manufacturer instructions, BS 6626:2010 and HSG 230. Establish routine, periodic monitoring within the maintenance programme.

Review risk assessments and apply access and operational restrictions until all applicable measures are implemented. Do not enter where there is an ozone smell, audible discharge or crackling, visible tracking, overheating, high humidity, condensation or water ingress.

Ask yourself or your crew

How can something like this happen here (e.g. on our site)?

What safety measures (i.e. procedures, controls/barriers) do we have in place to mitigate the risk?

How do we know the risk controls/barriers are working?

What improvements or changes should we make to the procedures, controls/barriers, or the way we work?

What happened?

A GEC Type VMX high-voltage vacuum circuit breaker failed, causing electrical arcing and a fire within a substation.
Protection systems cleared the fault within 700 milliseconds (0.7 seconds), but significant damage still resulted.
The original VMX unit had suffered severe partial discharge and was replaced in 2004 with a second-hand unit.
The safety notice applies to all three forms (A, B and C) manufactured in the UK between 1985 and January 2003, including units badged GEC, GEC Alsthom, Alstom and Areva T&D.
The failure had the potential to cause serious injury or death.
Why did it happen?

Partial discharge activity within the ageing vacuum circuit breaker was the most likely cause. It progressively degraded the insulation, leading to disruptive failure with arcing and fire.
Partial discharge was recorded at 6-8 dB 24 months and again 14 months before the failure, but the warning signs were not acted upon. Relative humidity was measured at 71% and 81%.
Substation environmental conditions contributed to the failure. The mechanism is strongly linked to relative humidity above the manufacturer maximum of 80%, condensation and temperature fluctuation.
Intrusive maintenance two months before the failure detected no discharge. A single clear test result did not clear equipment with a history of partial discharge.
Outstanding legacy design modifications and Improved Maintenance Practices may not have been completed or repeated. The provenance, modification status and condition history of the second-hand unit also required verification.
What did they learn?

Identify whether GEC Type VMX switchgear is installed on, or forms part of, any substation accessed by the site. Confirm the manufacture date and check serial and work order numbers against the original equipment manufacturer Table of Potential Problems.
Review maintenance records to confirm that all applicable design modifications and Improved Maintenance Practices have been implemented and repeated where required. Where records are incomplete, contact the current OEM for guidance.
Verify that relative humidity does not exceed 80%, and control condensation and temperature fluctuation. Record relative humidity and ambient temperature at the same time as every test result.
Arrange an initial partial discharge survey and vacuum interrupter integrity review by a competent person, in line with manufacturer instructions, BS 6626:2010 and HSG 230. Establish routine, periodic monitoring within the maintenance programme.
Review risk assessments and apply access and operational restrictions until all applicable measures are implemented. Do not enter where there is an ozone smell, audible discharge or crackling, visible tracking, overheating, high humidity, condensation or water ingress.
Ask yourself or your crew

How can something like this happen here (e.g. on our site)?
What safety measures (i.e. procedures, controls/barriers) do we have in place to mitigate the risk?
How do we know the risk controls/barriers are working?
What improvements or changes should we make to the procedures, controls/barriers, or the way we work?