We had something odd happen in a wind storm last week. We had a 78" guy strain insulator (we use 2 circuits on a pole so thus the extremely long guy strain insulators). It appears that the phase lead swung over and hit the guy strain insulator about 24-30 inches from one end. There is clear evidence of tracking over that area and we had a trip on the breaker as a result. The circuit voltage is 22.9 kV. I've been told that it was "21 kV rated" but this doesn't make sense. It seems like it was actually 21,000 lbs. rated. Is this failure mode common? What could possibly cause this to happen? What is the BIL (kv/foot) for these?

The "straight" type guy strain insulators (assuming we are talking about GSI's without any skirts) that are the fiberglass type have BIL levels of about 150 KV/foot and one minute wet withstands of about 30 KV/foot. So even one foot of GSI should have prevented a flashover (at 23 KV, 150 KV BIL and 12" phase to ground clearance are normal design ratings). The GSIs are usually rated in lbs. of tension, not KV. They provide clearance, so to properly apply them would be based on clearance rules and distances, not a KV rating.

Did the GSI fail, or just show tracking?

If nothing else is apparent, my guess would be some debris in the wind might have caused the flashover and breaker tripping. Your weather event may shed some light - was it raining, what was the wind speed, etc. and the physical surroundings - trees, debris piles, etc. may also help explain.

I have examine your post but I think there should be more comments about the topic that has been originally started so that the pool of thoughts can be acquired. Hope you understand. Regards.

Could the insulator been contaminated? That could explain surface tracking or is the insulator compromised somehow physically? The breaker tripped but is there sophisticated relaying that might reveal some answers thru a DFR or fault location? Maybe it is coincidence that there is tracking but maybe fault elsewhere?

Sorry just more questions.

Its a distribution system so no DFR, etc. The main sub (radial-like system) is getting up in age so relays right now are mostly ABB Circuit Shield or electro-mechanical. There is “monitoring“ but its 10 second sampling (PLC), no waveform level capability. Will eventually fix this but not now. Did not see any obvious signs of contamination, recent install, or obvious internal issues like voids in an epoxy insulator. like I usually see so no smoking gun. So thought I'd ask since its such an unusal incident.