Touch Potentials Beyond the Substation Grid System

[RSChinn]

In reference to the GPR voltage gradients around a substation during a close-in fault and the extreme drop-off point at the edge of the station ground grid, is there a possible touch (not step) potential hazard if a metal stake was planted into the ground within say 10 feet away from the fence? Hypothetically:
1) There is a grid extension of 3 feet beyond the fence so that the metal stake is actually 7 feet away from the edge of the ground grid.
2) Case 1: The metal stake is not grounded to the station grounding system nor is it grounded at all other than being planted say 1 foot into the ground, or
3) Case 2: The metal stake is not grounded to the station grounding system, but is grounded with an independent grounding rod connection at its location in addition to being planted 1 foot into the ground.
4) Case 3: If the above metal stake is actually the end of a cyclone fence that goes 100 feet away from the substation fence, is there touch potential hazards along this fence that are considered transferred voltages even though the this fence is not connected to the substation grounding system?

Of course, there are many factors that need to be considered in order to define the magnitude of the Touch potential and whether or not it is truly a hazard. Let’s say the available fault current is 20,000 amps, clearing time is around .7 seconds, current division factor is .5, surface covering is average native soil (not substation rock). Regardless of the values of these factors, does any touch potential exist?

 

[GroovyGuy]

Hi RS,
What is the GPR target that you are trying to obtain (ie 5000V). If so, then the station resistance must be < 0.25 ohms. This is doable, but depending upon soil conditions (which can vary widely), may require a substantial amount of copper in the ground.
I think that answers to your questions above will be difficult to provide without undertaking a detailed ground study. There are a lot of variables, including the lack of crushed stone 10ft from the fence.

Lately I have been using ETAP (FEA) for doing ground calculations. The station potential rise does tend to drop off rapidly as you distance yourself from the outer most ground conductor, but I can't say for certain if there will be any touch or step issues at the 10ft mark.
Sorry,
GG

"Wish I didn't know now what I didn't know then." -- Bob Seger

 

[GroovyGuy]

Hi RS,
I ran a quick ETAP (FEA) analysis of your issue above, and as suspected the GPR roll-off is very rapid at the edges of the grid. Refer to the pics below.
The results are based on;
- a 100ft x 200ft overall substation
- 20ft x 20ft grid
- 20ft ground rods
- nominal soil resitivity
- Reultant 0.20 ohm ground resistance
- 20kA GF current
- 4000V GPR
Of course the soil resistivity plays a big role here, and you need to consider this for your circumstances. While the GPR was limited to an acceptable value here, the touch and step potentials were still excessive.

Regards,
GG

"Wish I didn't know now what I didn't know then." -- Bob Seger

 

[Marmite]

A touch potential hazard only exists if you can simultaneously physically touch two conductive structures which may be at different potentials.
Regards
Marmite

 

[RSChinn]

Hi GroovyGuy: My concern is for a homeowner that lives adjacent to a substation with the substation fence built right on top of his property line. Not only do we not have the opportunity to extend the apron of the ground grid 3 feet beyond the fence to negate an outer touch potential at the fence, but the drop-off of the GPR gradient is the most steep in his yard right at the fence line. If he puts up a tent in his yard with a metal tent post say 5 feet away from the fence, I believe the possibility of a touch potential hazard exists even if the post is not physically connected to the fence.
Also, my mentioning of the 100 foot fence that may create a transferred voltage is a concern dealing with a little league baseball diamond that has a cyclone metal fence installed from the home plate backstop to the outfield along the sidelines. If the fence at the outfield is within 10 feet of a substation fence although not physically connected to it, I believe the possibility of a touch potential hazard exists via transferred voltage all the way to the backstop.

To Marmite: Touch potential is not between two structures, but it is between a voltage level touched by the hand (or another part of the body) and a different voltage level at the feet. By touching the tent metal pole and standing on native soil during a GPR event a touch potential situation exists that will cause current to flow through the human body… and in his yard, there is no substation rock to break this current flow.

I know that power companies that evaluate the touch potentials end their hazards study at the edge of the substation grounding grid and do not go further generally. My concern is for what happens to the adjacent neighbor that is in this steep surface voltage gradient if he decides to put up structures such as a tent pole, an antenna, or even a swing set.

I’d like the thoughts of GroovyGuy, Marmite and anyone else on what I just wrote. Regardless of any calculations, can a situation exist such that touch potentials (and even step potentials) are possible and big enough to be considered hazardous. Because it only takes 125 ma to place a heart into fibrillation, I believe this is possible. Your thoughts?

 

[IrishPower]

HI RSChinn,

Short answer: Yes, I think you could in theory (and possibly in practice!) have touch potential issues at the metal stake or cyclone fence. In my own experience I have seen cases, both modelled and measured, where excessive touch potentials were present at third-party fences and metal objects which were close but not actually bonded to the substation ground grid.

Longer answer:

Whether or not the touch potentials exceed safety limits or not depends... on the GPR at the substation and local surface potentials at and near the metal stake (which will also depend on distance from the substation ground grid and the soil resistivity / soil layer structure). If the GPR is relatively low you may be lucky and not have any significant problems.

I would expect Case 1 and Case 2 to result in similar touch voltages at the metal stake, assuming that the ground rod in Case 2 is buried immediately adjacent to the metal stake. The touch potentials you could experience could be quite high if the stake is very close to the edges or corners of the substation ground grid where the surface potential gradient is steepest.

On the other hand, Case 3 could result in far worse transferred touch voltages at various points along the route of the metallic fence, particularly at the ends of the fence closest to and farthest away from the substation ground grid. Again, the transferred touch potentials will tend to be worse for longer fences extending radially away from the substation ground grid and where one end of the fence is particularly close to the ground grid.

Whether in your situation any of the cases you have outlined will result in touch potentials exceeding safety limits is almost impossible to say without either more detailed modelling or physical measurements (post-construction) - or both!

I'd suggest that you get professional assistance from an experienced grounding design engineer / consultancy with access to software that can model multiple electrodes at different voltages. Note that many grounding software packages cannot do this; however one that I am aware of that can is CDEGS (using one of the CDEGS packages which uses the MALZ module rather than the MALT module).

Hope this helps - best of luck with your investigation!

IP

 

[basilasq1]

Hi RS,

As stated by everyone else, it is impossible to state whether calculated potentials will exceed threshold limits without using a detailed grounding simulation.

However, I can share some of my experience studying similar situations:

Transfer of potential to a nearby facility fence due to a fault at the substation is very common. In most the the cases we studied, a facility fence was placed in close proximity to a substation fence and usually isolated through a 3m wide isolating fence section (wood or any other non-conductive material is very common in the industry). We used HIFREQ (one of the more advanced modules of CDEGS) and modeled both above and below ground conducting material. All situations were quite different but to summarize some of the findings:

- Transfer of potential through soil is very common even when the adjacent facility fence is isolated from substation fence. Of course, that is highly dependent on the soil composition. Some simple mitigation measures include installing rods at grid extremities to control GPR, increasing isolation distances and installing a gravel strip around the plant fence to control touch voltages. In cases where plant fences were installed over a very large area footprint, the fence and gates were never tied to the main plant or substation grids.

- Transfer of potential is very common through neutrals, shield wires, telephone wires and pipelines. A detailed Transfer of Potential analysis is required using actual data from field to assess magnitude of hazards.

- EM induction effects on fence are also very common especially during fault conditions on adjacent transmission lines and may induce touch voltage hazards on fence sections.

- It is extremely important to assess the risk factor, especially when dealing when very large fence sections. Because mitigation measures can become extremely expensive for such situations, a targeted approach to mitigate hazards should be used only for essential and high risk area. This would also include probability of failure analysis.

 

[7anoter4]

In my opinion, the fence at any well grounded point will get this ground point potential. So, if both ends are well grounded no transferred potential will take place.
If only one end is grounded and the other is not at the insulated point the transferred potential will be the difference between the ground potentials.
If the entire fence is well insulated no transferred potential will be.

 

[cuky2000]

"....My concern is for a homeowner that lives adjacent to a substation with the substation fence built right on top of his property line."

Although it is recognized the ground potential away from the substation decay with the distance from the fence, this scenario is definitely a hazard concern that need to be addressed since the allowable voltage in the substation is different than inside the house for the following reasons:
a) Substation are designed for adult usually 50 kg (110 lb)to withstand safely the electric shock, the person in the house could be children with less than 50kg.
b) Substation has crushed rock (>2000 Ω.m) but house are not built with voltage control surfacing materials (concrete slab ~50 Ω.m, wet soil with low resistivity, etc.)
c) Grounding (earthling) through metallic contact such as plumbing, gas, utility, etc. could create a zero potential while the house floor or backyard will be at higher potential do to SC at the sub(transfer potential)
d) Kitchen, bathroom or swimming pool could be another source of concern wet conditions
Unfortunately the solution of this problem is limited and most utility do not properly address those issues sampling ignoring the potential hazard and pray for the best. I see in other instance that utilities pass the problem to consulting engineers adding language in the grounding study contract.