XFMR Neutral GNDg Resistor sizing

[JBUDA54]

I am working on a project where we are replacing two (2) three winding 13.8-4.16kV-480 3750kVA XFMRs with two (2) two winding 13.8-4.16kV 7500kVA XFMRs. I found a one line that shows the neutral resistors for the old 3750kVA XFMRs to be 8kV 30 ohm for 300A for 10secs. Should these resistors be replaced with the new XFMRs? If so how do I size the new XFMRs neutral resistors? Is there a IEEE std on this?

 

[JBUDA54]

I did some digging with IEEE color books and found a page out of the Brown book that shows an eqn of calculating the fault current using the symmetrical components and the Rg grounding resistor. The eqn is:

Isc = 3V/(3Rg+(R1+R2+R0) + j(X"1+X2+X0)

Isc is maximum FC in amperes
V is phase to neutral voltage
Rg is the grid resistance to earth in ohms
R1 is positive seq in ohms
R2 is negative seq in ohms
R0 is Zero seq in ohms
X"1 is positive sequence subtransient reactance in ohms
X2 is negative sequence system reactance in ohms
X0 is zero sequence system reactance in ohms

Now I have to detemine my power system's symmetrical components. Once I come up with this data am I trying to evaluate whether or not this selected Neutral Resistor will limit the ground fault current to protect the XFMR? I have the ABB XFMR drawings and do not see any SC Ratings.

 

[dpc]

First, an engineering decision must be made regarding what the desired level of ground fault current should be. If 300 A is what is desired, I don't see a need to replace the resistors if they are in good condition. You may want to review the IEEE Red Book and Green Book for discussion on various grounding methods. There is no single right answer.

As for calculating the current, for resistance grounding, just divide the line-to-neutral voltage by the resistance. Nothing else in the system is going to make much difference.



"The perfect is the enemy of the good." -- Voltaire

 

[trosepe]

You really don't need to perform a short circuit study or symetrical componant analysis to size the resistor or determine your max zero sequence current using the existing 80 ohm resistor. Regardless of the available fault duty, the zero sequence current is limited to V(l-n)/resistance of grounding resistor; in this case 2400/30=80amps. You just need to be sure your ground relay operates at below 80 amps. The only ratings a grounding resistor has are Vmax(l-n), ohms, amps and seconds. Your resistor can withstand without damage 300 amps for 10 sec. Your resistor is designed for use on an 13.8KV Wye connected system which has a phase to neutral voltage of 8KV. 8,000v/30ohms=266 amps. Since you are connecting this to a 4.16/2.4KV system you will have 80 amps as opposed to the 266amps on the 13.8KV/8KV system. No problem as long as you are sure the ground relay is set to operate at 80 amps or less and will clear the fault before about 20 seconds.

 

[racobb]

I do not understand why you would want to wait 20 seconds or even 10 seconds to trip. Anytime I have used a neutral grounding resistor, I was because I had no line to neutral loads, hence no normal neutral current. Since neutral current indicates an abnormal case, I usually use an extremely sensitive relay and trip quickly.

Don't see a reason to wait unless there is some application that is new to me. If so will you please enlighten me.

Thanks

Alan

Democracy is two wolves and a sheep deciding what to have for dinner. Liberty is a well armed sheep!
Ben Franklin

 

[trosepe]

I was not saying to set the ground relay for 10 or 20 seconds, I was saying the ground resistor he is using could withstand the 80 amps for that period of time without damage. You would of course set the ground relay to coordinate with down stream ground relaying if there is any. Ground relays on solidly grounded systems are set faster to prevent damage to the shielding on the MV cable as the zero sequence currents are nearly the same as the phase short circuit currents. In a resitance ground system and in particular this case, the zero sequence current is limited to 80 amps so there is no rush.

 

[trosepe]

The purpose of the neutral grounding resistor is not to protect the transformer but, in most cases, to limit the damage to the stator laminations of rotating equipment in the event of a phase to ground fault in the windings. The resistor will allow enough current to accuratly operate a ground relay but limits the damage to the stator so it can be more easily repaired.

 

[lz5pl]

Just an addition to trosepe -there is some typical (I am not sure it is defined in the standards) values for thermal rating of NER's - to withstand 300A for 10s. The same requirement applies in our grid (IEC world), but of course ground fault protections are set to maximum 1,5 s. Remaining time is as safety margin. Also we should keep in mind that the NER heats up from the flowing current and changes it's resistance. If it is thermally rated too close to the limited current (for example 100 A for 2 s), on the end of relay protection time delay resistance will be changed and possibly ground fault current may be limited to a lower value.

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[CoryA]

Hello JBUDA54,

There is a paper that addresses neutral-grounding-resistor selection specifically. I'll add a link to it below.

Ground-Fault Detection, Charging Current and Neutral-Grounding Resistor Selection

I would suggest, however, that if the resistors are intact, their replacement may not be necessary. An appropriate neutral-grounding-resistor monitor will verify the entire neutral-to-ground connection. A second paper detailing why neutral-grounding-resistors should be monitored can be retrieved from the IEEE website; I'll add a link to it below.

Why Neutral-Grounding Resistors Need Continuous Monitoring

Cory Anderson

http://startco.ca/

 

[dpc]

The reason to delay tripping is to allow time for downstream ground relaying to trip clear the fault. That is the main advantage of low-resistance grounding over high-resistance grounding - there is adequate current to allow selective coordination over three levels (or more) of ground overcurrent relaying. If the plan is to trip the entire system off instantly for any ground fault, you would be much better off with high resistance grounding to limit ground fault damage.



"The perfect is the enemy of the good." -- Voltaire

 

[racobb]

DPC....Thanks for the input. I would consider 30 ohms at 2400 volts to be a high resistance ground.

Alan

Democracy is two wolves and a sheep deciding what to have for dinner. Liberty is a well armed sheep!
Ben Franklin

 

[racobb]

Actually, seems I remember from somewhere in my past that usually high resistance grounding limits the l/n current to less than 10 amps which would require a 240 ohm or greater at 2400 volts. Don't know if it is a standard or not though.

Alan

Democracy is two wolves and a sheep deciding what to have for dinner. Liberty is a well armed sheep!
Ben Franklin