Earthing System Impedance

[JBirch]

We recently conducted an earthing audit of a local substation. Part of the testing was fall of potential testing to determine overall ground Impedance. The measured value for this test was 10.2Ohms. This value is much higher than any other values previously measured in other locations. Due to the high earth Z and the high fault currents our EGVR is in excess of 40kV!! Has anyone else come across such high EGVR's/Earth impedance values?
Thankyou

 

[jghrist]

Are you assuming that all of the fault current flows through the earth? This is probably not a good assumption with a high grid resistance. A lot of the fault current will flow through the neutral/shield wire.

What is the line-to-ground voltage? The GPR cannot exceed this value.

 

[lansford]

The fall of potential method may not be accurate for substations that occupy large areas.
You should check some references to be sure that the method is reasonable for your application.
Jim

 

[peebee]

Further to lasford's post: A testing agent recently informed me that they used the "slope" method in lieu of the specified fall-of-potential method due to the physical size of the installation being tested.

 

[Zogzog]

Probally because they didnt have long enough test leads

 

[JBirch]

Test leads were run out to a distance greater than 10x the diagonal dimension of the earth grid.

Same test method as used every other time where results have always been acceptable...

Line to ground voltage of 110kV / sqrt3 (~63.51kV.

Its a relatively small switching station with no TX. Only a couple of isolators (disconnectors) and a circuit breaker.

 

[jghrist]

A grid resistance of over 10 ohms for a small station does not surprise me. The main determinant of grid resistance is area of the grid and there just isn't that much area.

Your GPR must have considered all of the current going through the grid. Most of it will return through the shield wire. This is what saves the day and makes the design practicable.

 

[Zogzog]

"Test leads were run out to a distance greater than 10x the diagonal dimension of the earth grid."

Thats long enough, have you considered soil treatment?

 

[JBirch]

The station consists of 110kV 3 phase Over Head Transmission Lines with NO Over Head Earth Wire (OHEW) or OPGW.

Hence all earth fault current must return via earth.

We performed a soil resistivity test, not exactly sure what is meant by soil treatment though.

The surrounding land is undisturbed bushland with just a transmission line track heading out in either direction from the station. The station has been there for many years so the ground has had plenty of time to 'settle'.

Current was injected down one transmission line track with the voltage readings taken 180deg in the opposite direction.

 

[NorthSlope]

As a guy who makes part of his living doing these tests - let me add two cents. Slope method, if done correctly, is just as valid as fall of potential. Ten times the major diagonal is an ideal - not a "make or break".

I have tested lots of substations, from 345 kV on down. Some come close to or exceed this 10 Ohm number. Some utilities spend a lot of money, time and effort to get below 2 Ohms - some don't.

Soil resistivity, hence substation resistance varies with soil moisture and consequently the season and recent rainfall. Yes, fault current, and step and touch potential are all affected by soil resistivity (and fault paramaters). It all depends on where the emphasis is placed - and how much money you have to throw at the problem.

 

[JBirch]

All step and touch reading were well within the accpetable limits. Its just the EGVR that we are concerned about...

Its seems very high at ~42kV with a worst case fault clearance time of 1 second.

However, as this station is not the source all fault current would be returning to the remote stations which are much larger with more substansual earthing systems.

This station is in a very remote location with little or no known underground assets in the area, ie. Gas, Water, Telecommunications, etc.