Surge Arrester Blowing 2

[Amex12]

I have a project involving a sub transmission line with wind turbines interconnected. Surge arresters are all along the pole line and at the base of each turbine in the switchgear with the same sizing. For some reason, the last turbine at the end of the line keeps blowing its arrester at the basement switchgear. Any reason this could be happening? It is almost exploding and leaving burnt marks. The area is prone to a lot of storms.

Thanks in advance!

 

[davidbeach]

What's the grounding scheme? AKA, in a fault study, of a correctly modeled system, what voltages do you get on the unfaulted phases at that location for a ground fault? How much fault resistance can you have and still exceed the MCOV of the arrestor?

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[Amex12]

The system is a solidly grounded wye. Not sure how to answer your question about fault resistance.

 

[magoo2]

What is the subtransmission voltage and what is the MCOV of the surge arresters?

How many times has this happened before (arresters blowing)?

 

[davidbeach]

Solidly grounded wye? Where? Just at the head end of the system and then every turbine connects with a delta winding?

Do you have a good model? If so, place a SLG (1LG) fault at the last turbine. What's the voltage to ground on the unfaulted phases? Does the collector breaker trip first leaving you with a delta system? There's probably an easy answer to your question after all of the facts come out.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[Amex12]

Voltage is 34.5 kV (L-L) and the arresters are 24.4 kV MCOV. This has happened twice before and they are elbow type arresters connected to switchgear bushings.

 

[Amex12]

@davidbeach - it is a solidly grounded wye at the autotransformer secondary that feeds the turbines. The feeder breaker has a fact acting ground switch so you don't lose your wye when the breaker opens.

I do have a good ETAP model and will try what you suggest and report back.

 

[davidbeach]

Ok, well, good luck then.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[cuky2000]

Check if the reflected traveling wave amplifies at the end of the line. There is a known phenomenon that the reflected wave at the end of a line could double the voltage potentially exceeding the energy release capability of the surge arrester.

It is possible that excessive input energy will raise the temperature of the SA blocks that may force the unit into a thermal runaway condition creating the damage described.

 

[dpc]

I'd also run a power flow and check the voltage at the end of the line with no generation. Voltages can get pretty high due to capacitance of the long runs.

 

[jghrist]

I'd also run a power flow and check the voltage at the end of the line with no generation. Voltages can get pretty high due to capacitance of the long runs.

The voltage will be higher during maximum generation because of power flow (particularly reactive flow if any) into the system causing a voltage rise. Measure the voltage to compare with the calculated voltage. The utility source voltage may be higher that modeled.

I wouldn't expect steady state voltage to be a problem since the MCOV is 22% higher than nominal Ø-N voltage.

You might need an EMTP study to determine if the joule rating of the arresters might be exceeded during a stroke. The line capacitance may be discharging into the last arrester, which would discharge first because of voltage reflections.

 

[Amex12]

During steady state, the voltage is only 106.5% of nominal as the turbines are fed from overhead lines. @cuky2000 how would I check what you are saying? @jghrist I would have to learn a program to do that. Is it something I could pick up with tutorials in the free ATP program?

Is a simple solution to just up the arrester MCOV?

 

[RRaghunath]

Arrester MCOV is recommended to be equal to LL voltage (i.e. 34.5kV in your case)if the system is ungrounded. 24.4kV is correct for solidly grounded system.
If the system can run ungrounded and if you find the SA failures are happening during that period, you may choose SA with higher MCOV. However, SA residual voltage has to be verified against the BIL to confirm that SA still provides effective protection to the equipment.

 

[Palletjack]

Are these inspected after storms? Do they only blow after a lightning event?

 

[jghrist]

@jghrist I would have to learn a program to do that. Is it something I could pick up with tutorials in the free ATP program?

You would have to learn something like ATP to do this or hire a consultant. Tutorials are hard to find for ATP and it is not easy to learn. EMTPWorks (

https://www.emtp.com/)

has more training available and support, but is expensive.

 

[pwrengrds]

Check to see if the grounding breaker's grounding point is actually attached to ground. I've seen a few that weren't attached. The connection is more in the center, underneath the breaker, if there is just a slot for the bus to go through it's not really a grounding breaker as configurated. If the site is running normally at 106.5%, consider tapping the MPT down a tap or changing the center point of the tap changer controller.

 

[cuky2000]

The best way to determine the adequacy of surge arresters is to prepare an accurate system study to value the arrester rating. However, this is not only time-consuming but also could be very expensive if contracted with an experienced organization.
A conservative rough estimate could be determined per the enclosed excerpt below. For the solution of replacing the surge arrester, consult with a serious manufacturer and suggest if a parallel surge arrester may help the suspected exceeded energy capability during a lightning strike in the region of the project.
I hope this help