Underground Distribution loops - Arrester placement

[atlbraves49]

In reference to URD (underground distribution) loops, we currently install arresters at the "open point" of the loop for our 35kV lines. However currently we do not do so for our 12kV lines. Would it be recommended to do so, and if so, why?

 

[magoo2]

You don't really need arresters at the open point for 15 kV class voltages in general. You could provide tem for 12 kV as extra insurance.

If you look at the clamping level of 9 or 10 kV arresters, even if you double that value it will provide sufficient protectice margin. At 35 kV its a different story which is why the open point arrester is used.

 

[atlbraves49]

Thanks for the reply as usual Magoo,

Can you expand a bit and/or point me to some additional resources? I will likely need to provide recommendation to a director for this within a week or so and would like to have a firm understanding on what I am recommending.

 

[stevenal]

Two arresters at each open point, one elbow arrester and one parking stand arrester. Cheap insurance against costly repairs.

 

[atlbraves49]

I have continued my research on this topic and have found great information, but still have some questions/concerns.

I've found that we don't use them currently, due to the fact the doubled wave magnitude ends up being around 60kV, and the transformer BILs are 95kV (for the 12kV system). Therefore the transformers should be protected without the use of an open point arrester.

However, I spoke with some companies who suggest that there's more to it than just BIL, and that the cable itself could suffer degradation from even 60kV (deficiencies in the cable could be exploited over time; not in one incident, but over time they could end up failing). However I can not locate any data to verify this. Thoughts?

 

[stevenal]

I agree. Who knows what the BIL is of a cable that's been in the ground treeing for 20 years, or the splices and elbows that may not have been installed quite up to spec. Cheap insurance to help keep the lights on.

 

[magoo2]

From a high level perspective, you can think of the potential damage of a lightning surge to underground cable like someone driving a nail through wood. If you tap the hammer lightly, it takes many, many shots before the nail goes through the wood. If you use a heavy swing you might be able to do it in one or two strikes.

Cable has a similar withstand to surges. It can tolerate many smaller surges and have negligible effect on its performance, but it is still degrading slightly. If you get one or two high magnitude surges, it may fail the cable.

Hopkinson of GE published a paper on this in the 1980s. I think Mashikian and one of his grad students did another paper on it a little later, maybe the early 1990s.

 

[atlbraves49]

Stevenal, is it cheap insurance if the benefit is rarely seen? The reason I ask, is that we've been NOT using elbow arresters on 12kV URD systems for probably 15 years or more, and there really hasn't been many issues. However, it's impossible to say what state the cable is in at this point. I've spoken to several contacts who say that generally arresters are not needed on 12-15k URD systems, and that most utilities do NOT use them on such systems. Then I talk to someone else who is amazed that we don't.

Having a tough time coming up with a decision on this matter. It's one thing to recommend we start using them; it's another thing to have to be able to actually justify that decision.

 

[stevenal]

Look at the cost of a single UG failure. Crew time to locate, dig, splice, restore the ground; material costs, vehicle and equipment costs, overhead. I think you can justify a few arresters on cost alone.

 

[jghrist]

What is the failure rate of elbow arresters?

 

[stevenal]

jghrist,

I have no data, but I don't believe it's very high. I'm more aware failures at the riser pole arresters, which are exposed to the weather. Point taken, though; arresters cause as well as prevent outages. An arrester failure is quickly addressed, though. Cable failures are not.

atlbraves49,

Yes. Remember risk equals impact times probability. Even low probability events can have high risk if the impact is high enough; and UG cable failures are high impact.