Temporary overvoltages and surge arresters

[lume7006]

Hello everybody,

If a transformer neutral is grounded through a resistor to limit the magnitude of the single line to grund current, on the "healthy phases", there will be some overvoltages lasting the time overcurrent protection clears the fault.

Well, my question is related to these overvoltages:
Do surge arresters need to be sized considering them?

Because, in a case like this, they would have to be lower in MCOV and TOV than when neutral is solidly grounded.
Does anybody have some recommendations about it?

Any help/comment will be appreciated!

 

[dpc]

For any resistance-grounded systems, the arresters should be rated as if the system is ungrounded.

The MCOV of the arrester for this situation must be HIGHER than a solidly-grounded system, not lower.



David Castor

http://www.cvoes.com

 

[potteryshard]

Arrestor vendor information typically illustrates this point with different rating for solidly and resistively grounded systems.

 

[lume7006]

Hello,
I thank your answers!
Any other comment will be welcome!

 

[cuky2000]

The usual range of arrester MCOV rating is:
1.05*VLG_nominal ?MCOV ? 1.10*VLL_{nominal
NOTE: For low MV and UHV, some application could select MCOV > VLL -- See first file --}



As a rule of thumb, arrester MCOV rating usually is selected as follow:

1) Ungrounded or impedance grounded: MCOV ? 105% VLL_nominal
2) Solidly grounded: MCOV ? 80% *VLL<sub>max

To determine the actual TOV without using transient analysis, It is useful to use normalized curves to determine the Coefficient of Grounding (COG) or the Earth Fault Factor (EFF) (Both are equivalent EFF=1.73*COG)</sub>

The TOV (min MCOV) on the unfaulted phase for SLG fault could be determine as follow:

MCOV_min= COG*(1.05 to 1.10* VLL_Nominal)

For additional details see the second enclosed files.

 

[cuky2000]

Here is the first file mentioned in the above post for typical MCOV rating used in various applications.