Zigzag grounding transformer - Strange Results when modelling in ETAP

[stevesummers]

All,

I have an 11kV site that when in island mode has a handful of EDGs that run unearthed, with an 11kV system earth provided by a zig-zag grounding transformer. the grounding transformer is rated at 131A (142ohms) for 30s.

My understanding of these units is that the grounding transformer impedance should limit the fault current to around about 131A.. in a similar manner to having a neutral earthing resistor in a step-down transformer.

I have modelled the system in ETAP using, their suggested method for creating an equivalent model for the grounding transformer... So all good so far, but when i run the fault study, the earth fault value calculated is in the order of 2.5kA, so much higher than i would expect.

I have been back and forwards with the ETAP tech support guys on this, and they are stating that although the generators are unearthed and thus have no zero sequence impedance, the positive and negative sequence do interact with the zero sequence rating of the grounding transformer making the fault current higher. i.e. so the zero sequence current is given by:

Ia0 = Ia1 = Ia2 = E/(Z1+Z2+Z0)

I also did a similar experiment with a 20MVA, z=10% YNd transformer with the same zigzag and no generators, and got a similar very high earth fault current.

So, while i can see what the ETAP guys are saying, the end resultant figures just don't look correct to me. I would expect some variation, but not a resultant fault current that is out by a factor of 20.

This obviously has a knock on effect to the protection settings, and potential damage to the generators as the earth fault current is much higher. So i am keen to make sure i understand the issue correctly,

Can anyone shed any light on the above. Are the ETAP guys correct, and i am missing something (its a been 20+ years since i did sequence currents in anger) - or is my hunch correct, and ETAP is coming up with a 'wrong' answer for some other reason?




Steve

 

[waross]

It is my understanding that if you want to limit the current through a zig-zag transformer to 131 Amps you have to add the 142 Ohm grounding resistor.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[JuanBC]

Hello,

I suggest you include a single line diagram to help us give our point of view about your problem

JBC
.......
"The more I read, the more I acquire, the more certain I am that I know nothing"

 

[stevesummers]

Hi,

see attached for a word document with some screengrabs from the power transformer case i set up. The first image is showing the calculated earth fault current and the second is showing the sequence current flows - the third is the model i have of the earthing transformer.

Waross - you may be right, i have seen both configurations before on an SLD, but never had cause to look into the specifics of them until now.

 

[JuanBC]

Hi Steve.

Look picture attached (a calculation I made in a previous project i worked on). I think you are missing the resistance; you should configure it in the "Grounding" tab of the transformer

Regards.


JBC
.......
"The more I read, the more I acquire, the more certain I am that I know nothing"

 

[stevesummers]

Hi JuanBC,

thanks - but i that is a different scenario, where there is a resistor / reactor connected to the star point of the zig-zsg. In this case we just have a directly connected zig-zag.

 

[mgtrp]

I built the same scenario into ETAP and obtained the expected results, so I suspect that you have a data entry error somewhere. Quick hand calculations also give the same answer as ETAP. Note that you have calculated the zig-zag transformer impedance based on I0, but the fault current is actually 3I0. Therefore, the 0.40pu zero-sequence impedance will give you a maximum ground fault current of approximately 387A. You need an impedance of about 1.20pu to get 131A of fault current, which is the result that you get from the formula in the ETAP paper.

ETAP is correct that the positive and negative sequence impedances of the main transformer will affect the ground fault current, although with a 20 MVA transformer at only 0.1pu impedance, the effects are negligible against the impedance of the grounding transformer. The positive and negative sequences of the grounding transformer don't figure in the calculation. You should see this when you draw out the sequence network.

 

[stevesummers]

Hi mgtrp,

thanks - Yes i can see the positive and negative sequence of the grounding tx is not relevant... my rough estimation is similar to yours... its just that ETAP is telling me something different..

I have been staring at the model i have for the last few hours and cant see anything obviously wrong.

Earthing Tx
Primary V = 11kV
Secondary V = 11kV
Rating = 1MVA

Positive %Z = 9999
Zero %Z = 120% (1.2pu)
Tap - OFF
Grounding - Primary - Star (Wye) Solid & Secondary - Delta IT (ungrounded)

Would you share the details of your etap model for the zigzag.. i.e. a screen grab or two?

 

[mgtrp]

See attached. ETAP 16.1 if it makes any difference.

 

[stevesummers]

Thanks mgtrp

A key difference i spotted, was that you had set the earthing transformer positive / negative sequence impedance to 10%, where i had set it to 9999%, and this appears to be causing the difference.

I am not sure why, as surely with a grounding transformer the pos/neg impedance should not be that relevant... but i found that adjusting it pushes the final fault current up or down significantly.

Perhaps i am missing something obvious but i can't think why this would be the case..

 

[mgtrp]

The earthing transformer positive impedance has no effect, I get the same results either way. That might be a clue as to the source of your troubles.

 

[stevesummers]

Hi mgtrp - interesting. I swapped the calculation method to ANSI and found that i got the same results as you, regardless of the Z1 value of the grounding Tx. However, when i use the IEC method, the problems start appearing..

I am not fully aware of the differences between the two short circuit methods - the main one being IEC's use of voltage factors and treatment of variable speed drives and wind turbine converters / inverters, but i don't believe they are that fundamentally different!

 

[mgtrp]

Interesting. I hope that you report this back to ETAP!

 

[prc]

I think it is a 11 kV NGT , with neutral current of 131A ( phase current=43.7A, Phase Z= 142 Ohms)
11,000/(root3 x131) = 48 Ohms (phase ohms =144 ) So KVzz= 11 ; Igzz= 131 A