Negative Sequence Directional Overcurrent 1

[anggapra]

Hi All,

At my workplace, there is an 34.5 kV subsea distribution cable that is 30 km long, connected to WIDP01 and KARA01 bus.

Both WIDP01 and KARA01, have generators connected at a voltage level of 13.8 kV (generator at KARA01 is not shown in this drawing), where the transformers used are Dy transformer.
Based on simulations and actual cases, no zero sequence current can be detected at bus WIDP01, because there is no ground reference at WIDP01 bus.
I plan to add directional negative sequence current protection to bus WIDP01 to detect line to ground fault. Based on the simulation, negative sequence current can be detected at bus WIDP01. The Negative sequence current should be I0.

I need some input, can this protection mode be applied? Because I haven't found many references regarding this type of protection mode
What are some of the potential issues that may arise that I need to consider when implementing this?
As a note, power flow can flow from bus WIDP01 to bus KARA01 and vice versa

Thanks,
Angga

 

[davidbeach]

Once the breaker at KARA02 trips there is no more fault current to be detected at WIDP01. Negative sequence overcurrent is part of every new relay with any form of overcurrent I've worked with for the past couple of decades, so just turn it on, but at WIDP01 it has to be faster than any interruption at KARA02. 3V0 detection at WIDP01 would be more appropriate.

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

 

[bacon4life]

I second David's suggestion to consider 3V0 detection.

A couple consideration for negative sequence elements:
1)You need to check coordination for LG faults on the far side of any transformers.

2)For setting the negative sequence pickup, consider the effects of single phase switching. A slight difference in timing during breakers or disconnect opening/closing will produce negative sequence current. In order to avoid tripping on negative sequence current during switching, you can either set a time delay longer than the maximum pole disagreement, or can set the pickup current higher than the maximum current produced by the phase(s) open condition. Although I do not have timing records for medium voltage switches, 115 kV air break disconnects routinely have tens of cycles of disagreement between phases.

 

[anggapra]

Thank you all for your input. I forgot to mention that there are no Wye-connected PTs at the WIDP01 bus. Therefore, I think there is no way to measure the 3V0 without purchasing a new set of PTs. The only available PTs at the WIDP01 bus are open delta-connected PTs.

I think we will purchase new Wye-connected PTs to detect 3V0. In the meantime, we will activate 67_2 since purchasing new PTs will take quite some time.

Are there any considerations that I should take into account when choosing the PTs?

 

[davidbeach]

Two bushing PTs with a primary rating of 34.5kV rather than single bushing rated 20kV.

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

 

[anggapra]

Thank you, David. Why does it need to be rated at 34.5 kV? Is it because of voltage shifting during LG fault? There were Wye-connected PTs at the WIDP01 bus, but they are only rated at 20 kV. However, during a fault at another location, the insulation failed at a voltage level of 34.5 kV. I believe this is the answer we have been looking for, as we had previously attributed the failure to ferroresonance