Surface Material for Substation Ground Grid 5

[Bilegan]

I am involved in a substation design for an additional facility of an industrial plant, one of the requirements of the project is to design a grounding grid of the substation. The outdoor transformer is 10/12.5MVA, 13.8kV/4.16kV, DYn1, secondary neutral grounding resistor rated at 400A for 10 seconds. The primary of the transformer is tapped from the indoor 13.8kV metal clad switchgear via 13.8kV cable bus installed above ground. There is no grounding transformer connected to the bus of the 13.8kV metal clad switchgear.

Attached is the elevation of the transformer area. There is a perimeter steel cyclone fence with steel post around the transformer area. The surrounding of the transformer area and beyond the perimeter fence is asphalted for about 0.15 m thick, beneath the asphalt is a compacted dry soil. However, the top of the transformer oil containment is filled with gravel and extended at 1.8 meter around the transformer.

For the step and touch potential substation grounding calculation, we are debating for the surface material to be used. I suggested the crushed rock should be the surface material to be used, my colleagues suggested should be the asphalt, as they are telling me that the crushed rock bottom is steel grating, and below the steel grating is an open space for transformer oil containment. Is the step and touch potential calculation is treated separately? Where can I reference the touch voltage, in the transformer or on the fence?

Can anybody please help and cite section in IEEE Std. 80 and or standards to resolve my concerns?

Thanking you in advance.




Regards,
Bilegan

Murphy's Law - If anything can go wrong..it will.

 

[jghrist]

Your ground fault sources are all local, so none of the fault current flows through the earth and there will be no step- or touch-voltages.

A 13.8 kV ground fault will return through the ground connection to the metal clad switchgear. A 4.16 kV fault will flow through the ground system and neutral resistor to the transformer neutral. See Chapter 15, "Determination of maximum grid current," in IEEE Std 80-2000.

 

[rcw retired EE]

Conenct the steel grating under the gravel and the transformer case to the ground grid. It provdies a good equipotential surface.

As jgchrist says, you won't have any appeciable current flow to or through ground to create step or touch potential issues.

 

[Bilegan]

To jghrist, thank you for the brief but concise technical pointers for the scenario of my grounding system locally.

To rcwilson, what a good pointer as well for the steel grating to be bonded to the ground grid conductors.

For clarity, I attached the plan of the substation and more information are added in addition the transformer elevation which was previously attached to my original post that might be needed to further clarify my concern for the step and touch potential substation grounding calculation, we are debating for the surface material to be used. I suggested the crushed rock should be the surface material to be used, my colleagues suggested should be the asphalt, as they are telling me that the crushed rock bottom is steel grating, and below the steel grating is an open space for transformer oil containment.

Thanks.


Regards,
Bilegan

Murphy's Law - If anything can go wrong..it will.

 

[jghrist]

As I understand it, a ground fault at 13.8 kV is limited to 800A by the NGR on the 69-13.8 kV transformer. A ground fault at 4.16 kV is limited to 400A by the NGR on the 13.8-4.16 kV transformer. I assume that the ground bus of the 13.8 kV switchgear is tied to the substation ground grid.

You do have outgoing 13.8 kV lines, so there is a possibility of some earth current flowing from the grid for a ground fault on one of lines. This will be minimal, however, because the total fault current will be no more than 800A and most will flow through the cable ground instead of through the earth.

You have minimal touch- and step-voltages to worry about. I would guess that the GPR is below the allowable touch- and step-voltage. Step- and touch-voltages will be less than the GPR.

The first thing to do would be to check allowable touch-voltage with no insulating surface. Compare this to the maximum possibly grid current of 800A times the grid resistance. This can be done easily and will probably indicate no problem, even with these very conservative assumptions.

If there is a problem, then you need to estimate how much 13.8 kV line ground fault current will flow back to the 69-13.8 kV transformer through the cable grounds and how much will flow through the earth. Only the portion flowing through the earth will cause GPR or touch- and step-voltages.

If there is no touch- and step-voltage problem with no insulating surface, then you can use either stone, asphalt, or nothing. Either stone or asphalt will do anyway probably because they both have high resistivity.

It's not clear where you are considering asphalt vs crushed rock. I'd make the decision based on something other than grounding considerations. Driving on asphalt is easier. You can't use asphalt over the oil containment pit or spilled oil might not get into the pit.

Grounding the grating around the transformer as suggested by rcwilson will prevent any touch- or step-voltage because it will be a constant potential surface at the same voltage as the transformer tank.

 

[Bilegan]

jghrist,

Yes, your understanding is correct, we tied the 13.8kV ground bus to the substation ground grid with 70mm2 green PVC insulated grounding with in two opposite location.

Initially, our contractor declined to install equipment grounding wire between 13.8kV switchgears (because 1500 meters apart and additional cost for them) as they are thinking both switchgear are tied up each in their respective separate substation ground grid and it is enough, however we pushed them to install as we cite the requirement from NEC Article 250 (effective ground current path), not to rely on the substation ground grid. Although, they will install steel wire armoured feeder cables between switchgears, our local standard in the middle east is not allowing steel wire armour as equipment ground conductor, a separate equipment grounding conductor is required, capable to carry the short time rating fault current that will fow back to the 69kV-13.8kV transformer.

To run the simulator program for substation ground grid calculation, we need to identify the surface material, upper layer material and lower layer material. We all agreed on surface material to be asphalt and the lower layer to be wet soil, however, we have different idea what material to be used on the upper layer. There my concern arises.


Regards,
Bilegan

Murphy's Law - If anything can go wrong..it will.

 

[Elmir]

With regards to selection of top material you can try running simulations with both.

Crushed rock when wet has an approximate resistance of 3000 ohms/m. Asphalt when wet has a higher resisitiviey, approximately 9000 ohms/m. In either case their purpose is the same, to act as an insulator between people and the buried ground grid. For purposes of simulation I would use the worst case, crushed rock in this case.

 

[Bilegan]

Elmir

Yes I agree with you, we can simulate both, first using crushed rock with an approximate resistance of 3000 ohm-m, second we will used asphalt with an approximate resistance of 9000 ohm-m, at the end I've got 2 set of simulation results. Assuming, both simulation results are acceptable, calculated step and touch potential voltages are lower than the allowable step and touch potential voltages.

My question, is there any section in the IEEE Std. 80, the IEEE Std. 142, nor in NEC art. 250 that will recommend or criteria to select between crushed rock or asphalt as upper layer material for substation ground grid calculation? Assuming no economics involved, we will think the ultimate safety and reliability for the operation of protection relays..etc.


Regards,
Bilegan

Murphy's Law - If anything can go wrong..it will.

 

[umrpwr]

Assuming economics are not issue which in most cases they are since there is almost always a trade off between cost and design the asphalt would be the better of the two materials. What drives the decision between the two from an engineering design perspective is the touch and step potentials. I have seen some extreme cases where it was not possible to achieve an acceptable value using crushed stone so entire substations were paved with asphalt. It is a somewhat rare case and definitely a more expensive option, but in a few cases maybe the one of the only ways to acheive the the design criteria.

 

[NAZ55]

IEEE-80 (Table 7) gives some typical surface material resistivities. As a rule of thumb 1" of asphalt is equivalent to 4" of crushed rock.

 

[jghrist]

In my last post, I advised that grid current can come from a fault on an outgoing 13.8 kV line. This is probably not the worst case if you have incoming 69 kV lines to serve the 69-13.8 kV transformer. A local 69 kV ground fault will probably give the most grid current. To analyze this situation, you would have to model the entire grounding system, including the 69 kV portion.

To run the simulator program for substation ground grid calculation, we need to identify the surface material, upper layer material and lower layer material. We all agreed on surface material to be asphalt and the lower layer to be wet soil, however, we have different idea what material to be used on the upper layer.

It sounds like the simulator program is expecting the asphalt or crushed rock as the surface material, and a two-layer soil model. The upper and lower soil layers would be determined from soil resistivity tests and do not refer to the surface material. Is there an option in the program to use a homogeneous (single-layer) soil model?

 

[Bilegan]

In our case, our local standards calls for asphalt as the surface paving in the whole transformer yard, the thick is 0.15 m and beneath the asphalt is compacted dry soil. However, each transformer, there is oil containment around, the width is 1.8 meters, therefore we cannot cover the oil containment with asphalt. In case of oil spill during the testing of oil, the oil shall be diverted to the oil containment connected to the sewer system. The civil engineers, extended the depth of the oil containment and they put steel grating 0.3m below the grade level, then they suggested to put ¾ cube gravel stone up to the grade level.

If you refer to the transformer elevation in my original post, the surface material above the oil containment is gravel stone at 1.8 meters wide, and the remaining areas are all paved with 0.15m thick asphalt. You will notice that there are 2 different material used in the transformer yard.

An example scenario, assuming electrical fault occurs and there is person holding the body of the transformer and standing on the gravel stones, If I will calculate my substation grounding step and touch potential voltage, I will consider the gravel stone as surface material used, but my colleagues keeps telling me that I cannot consider the gravel stones as the surface material because beneath the gravel stones is open space for oil containment, they are recommending the asphalt as surface material to be used in my calculation.

What do you think folks, I would appreciate if you can lead me to resolve the concern in reference of some sections of IEEE Std. 80 or IEEE Std. 142.

Thanks


Regards,
Bilegan

Murphy's Law - If anything can go wrong..it will.

 

[Bilegan]

jghrist,

Our simulator is ETAP software. Attached is the snap shot of the soil editor.

Soil editor is;

Surface Material; drop down selector are asphalt, gravel, crushed rock, clean lime stone etc. resistivity in ohm-m, and you can input the depth.

Top layer; dry soil, moist soil, wet organic soil, bed rock, resistivity in ohm-m, and you can input the depth.

Lower layer; dry soil, moist soil, wet organic soil, bed rock, resistivity in ohm-m, no input window for the depth.

As you can notice in the surface material selector drop down, I can select either asphalt or gravel to run the simulator. If you will ask me, I will select gravel stone as surface material to be used, but my colleagues keeps telling me that I cannot consider the gravel stones as the surface material because beneath the gravel stones is open space for oil containment, they are recommending the asphalt as surface material to be used in my calculation. (Refer to the transformer elevation of my previous post).


Regards,
Bilegan

Murphy's Law - If anything can go wrong..it will.

 

[jghrist]

It doesn't matter what surface material you use for the calculations around the transformer, but I fail to see why anyone would recommend using asphalt if the actual material is gravel stone.

The surface material is used in the calculation of allowable step- and touch-voltage, not in the calculation of actual step- and touch-voltage. The actual step- and touch voltages will be zero because the metal grating provides an equipotential surface (no step-voltage). The metal grating is the same voltage as the transformer tank, so there will be no touch-voltage.

You will have some step-voltage at the edge of the containment area where you could have one foot on the grate (or the concrete berm) and one foot on the asphalt around the containment pit. This is sort of a hybrid situation between step- and touch-voltage because one foot is not insulated (the concrete can be assumed to be at grid voltage) and one is. I'd make sure that you have low enough touch-voltage around the containment pit, using asphalt as the surface material. This would be a worse case than step-voltage with one foot insulated.