earth fault

..If a high voltage greater than rated capacity is applied across capacitor, its dielectric strength will break down and eventually capacitor will explode. .. Electrolytic capacitors fail due to leakage or vaporization of the electrolyte inside. This can be caused due to heating in operation.....

GA

Another possibility is some extraneous object striking the <...capacitor...> unintentionally, thereby causing internal damage and the <...earth fault...>, perhaps?

It's a little difficult to see from here.

Thanks everyone for the reply

the earth fault was due to potential transformer failure on busbar

Then that is <...why surge capacitor exploded...>.

Surge Capacitor failure is most likely due to over voltage in the same phase.

When an earth fault occurs in one phase, the voltage of the other two phases (unfaulted phases) rise temporarily. Depending on how fast the protection isolates the fault, it is possible for the surge capacitor to fail.

Failure of capacitor in the same phase as the faulted one, I can't think of.

Normally the protection for a potential transformer, a high voltage current limiting E Class fuse, would clear a fault way way way before you would exceed the BIL rating of your surge capacitor. Surge capacitors applied on effectively ungrounded systems have a minimum rating line to line voltage, so that a ground fault would not impress high voltage on the device. It is possible your switchgear designer did not account for a fault during (unusual) ungrounded operation, and the surge capacitor voltage rating is too low.

There has been a similar question asked before, but never has the question of the power system grounding method been answered. If the bus section with the PT has no ground when the generator breaker is open, then the fault may have started as a ground fault, but had to escalate to phase to phase, as is often the case, before protective equipment cleared the fault.

Yes, it makes more sense when you switch cause and effect, GA.