Deep Rod Earth Grounding

[Dumbo2929]

I would like to propose some discussion to those smarter than I. I have been asked to evaluate the possibility of installing deep rod earth grounding around a new facility. Here are the details: New building 50k sq.ft. for a satellite uplink facility. The building was installed with a ring of 10' rods, 20' spacing, and a #2/0 ground loop. The resistance is several hundred ohms since the soil is quite sandy. The contractor tells me they practically pushed the rods in, as the soil was so loose. There is an old existing well, but it's resistance was poor as well; about 600ohms at 100'. I do not have soil data yet, but I wanted to ask people experience and success rate with installing deep rods at the corners of buildings, etc. No body wants to deal with chemical rods at this point.

 

[PowerSynch]

Soil resistivity data should tell you about impedance of different layers of soil. If lower impedance soil is located at 100meters below surface, your earth rods will not have much effect in reducing the impedance of your earthing system if they don't reach that dept.
Perform a soil resistivity test and on the basis of that model your earth, then you know the dept at which your earth rods can penetrate.
There are some soil enhancing material that you can purchase and treat the imediate surronding of your earth rods with. These have a great effect in reducing the soil impedance.

 

[cuky2000]

Here is a sample ground rod resistance versus depth. Actual values will depends on the resisyivity of specific

s

http://www.erico.com/newsDetail.asp?newsid=64ite

conditions.

 

[Skogsgurra]

A very interesting diagram, cuky!

I almost feel how dry the soil is down to 30 feet. You add lenth after length and it doesn't get any better. Then, you hit water table! And here we go again!

Done that. Seen that. You really have to go deep for a low earth resistance. The deeper the lower ;-)

Gunnar Englund

http://www.gke.org

 

[jghrist]

If the primary purpose of the ground system is for lightning protection, you need to look at the high frequency impedance of the ground. Long rods and long conductors are less effective at higher frequencies because of inductance. From Lightning Protection & Grounding Solutions for Communication Sites by Ken R. Rand, PolyPhaser:

There is a limitation to the length that a single
ground rod can penetrate poor conductivity soil
on its way to water table and better conductive
earth. Unless “shunted” by conductive earth, the
series inductance of the rod section in poor
conductivity soil, “chokes off” current flow to a
possibly more conductive lower section creating a
voltage drop along the more inductive top section.
The top section of the rod “breaks down” the soil.
There will be saturation and local ground potential
rise due to this breakdown.

 

[Dumbo2929]

Thanks guys. I appreciate the discussion. I have read the white paper that the graph above is from. Once I have better soil resistance data, I should be able to decide on deep rods or not.

 

[wareagle]

Why do you need to lower the ground resistance? The 10ft rods at 20 ft is generally what is done. Is lightning a problem?

 

[Dumbo2929]

The resistance of this grid is no where near the 1-3 ohms the client wants. The soil is very loos sandy. I would agree that for most substations the 20ft grid would be good. I did do one substation a few years ago that required 20ft long roads on a smaller grid size, due to rocky soil.

 

[dpc]

Not sure of the location, but you also have to get down below the frost line. You cannot take credit for any portion of the ground rod in the frost zone. When earth freezes, its resistivity goes up by orders of magnitude.

In areas such as Minnesota, 30-foot ground rods are common in substations.

 

[wareagle]

To get 1 to 3 ohms you probably need to go to chemical ground rod installations. 1 to 3 ohms is tough to get.
Why does the cleint think this is necessary?

 

[Dumbo2929]

The client is in the telcom business and that is there standard. I spoke with my technician who does a lot of Sprint cell sites and they try to get 8 ohms. I don't think my client knows any better, someone probably said 1 ohms is good.

 

[rcw retired EE]

Look at a Ufer grounding system if the building rebar is still available. The building's rebar and concrete footings provide a low resistance connection to earth. The resistivity of the concrete is realtively high, but the sheer mass of it in contact with the soil provides a much better ground than any singel or multiple ground rod.

One problem with ground rods is the small surface area of the typical 1/2" (50 mm) diameter rod. A 3.5 meter (10 foot) rod has about 0.55 square meters in contact with the earth. A typical footing (0.5 m wide x 0.5 m deep) has that much surface area in 360 mm (14") of length. With a buiding that size, you have a wonderful connection to earth, even if the earth is a little sandy.

Drilling a deep well or driving a steel pipe or pile might do the same thing by giving more surface contact area and by hitting the lower resistivity layer below the water table.

 

[waross]

I like ufer grounds, but check whether there is a moisture barrier under the concrete. With an insulating moisture barrier, the ufer is obviously not effective.
In Canada, a ufer can not be retrofitted. It must be a copper cable installed before the concrete is poured.
yours

 

[Dumbo2929]

Does anyone have any pros vs. cons for using a water well as a deep ground rod. I would be inclined to keep the two (water well and ground grid) separate. Does anyone know of any good papers on this topic? Thanks.

 

[cuky2000]

High cost compare with other grounding system is one of the drawbacks for ground well. Since you mentioned that there is one available, I would use as part of the grounding system.

Appear unusual that well has poor resistance. Is 600ohms(cm?) at 100'the resistance or resistivity? How this was measured?

Just a curiosity, does the meter is properly calibrated?
Did you compare your data with other facilities in the area?

 

[stanlsimon]

Is that well pipe a new well? Most new well casings are PVC plastic. Not a great conductor. If the well is still being used then the water table should be above the 100' invert. Check with the county well boring log to get a survey of surrounding water table depths. In that graph above I would say that the water table was at 35'. Notice the 2nd derivative at that depth.

The problem with using metal water piping as part of the grounding system is that often the metal pipes are iron or a combination of copper, brass or iron. Put together two different metals in an aqueous solution, clamp an aluminum ground clamp on the cold wet ductile iron pipe, pass a little stray current through and presto now you have blue green water. Especially if the water is softened. Galvanized iron piping is the worst, and it only takes one little fitting to cause all the copper pipes to be pitted out in a couple of years.

They are supposed to install dielectric fittings where they join dissimilar plumbing materials to prevent this from happening. Unfortunately this occurs often at the main water service entry where the electrician installs a bonding jumper around the meter and voids the dielectric fitting anyway.

If you have 600 ohms resistance I would say that the plumber did his job and put in a dielectric. I'm just speculating but that sounds like the the resistance of the electrical path through the water inside the pipe.

Someone should tell the NFPA that many new water pipes are plastic nowadays.

 

[stanlsimon]

I forgot to mention to try to get a ground in a wet location near a downspout or get the HVAC guy to pipe his condensate from an air handler down to that point. A nice steady 24/7 trickle is what you need. Under the building the ground can be very dry.

 

[mc5w]

You could also try installing a subsurface irrigation system e.g. Porous Pipe. You probably would need to add a dash of salt to the water.

Army Technical Manual 5-690 specifies a 1/0 copper ground ring at the bottom of a 10 foot deep trench and 10 foot ground rods driven to 10 feet below the bottom of the trench for lightning grounding. You could get the equivalent by connecting your existing ground rods to extensions using threadless coupling form Erico. However, the minimum diameter copper clad ground rod for use with couplings is 5/8 inch and if the soil is consolidated 3/4 inch needs to be used.

Your soil below 10 feet deep might actually be quite compacted due to the weight of the soil above and you could get a dramatic increase by extending your ground rods from 10 to 20 feet. Here is northeast Ohio, USA the top 6 feet of a ground rod driven from the surface of clay soil does not really act as a grounding electrode. Hence, some electricians drive a ground rod beneath the basement floor to access better soil.

 

[EEJaime]

I had a project where the client, a network television affiliate in Palm Springs, CA had their network engineers provide us with specifications for the signal reference ground grid which consisted of a continuous copper strap, 12"x 0.02" buried 24" below finished grade with silver soldered joints and XIT chemical ground rods at the corners. The trench was backfilled with treated soil to reduce it's impedence. This grid completely encircled the building and had lateral connections to the interior. All connections were by exothermic welds. All satellite dishes, uplink control equipment and data room equipment racks and studio data rack equipment were bonded to this grid. It was quite expensive, but in the desert sand/soil where this facility is located, they felt it was worth the cost. I believe that the test reports were that the system's resistance was less than 2 ohm's almost down to 1 Ohm. I don't know if this is the application Dumbo2929 has in his building. But it sounds like a similar facility.

This system was completely seperate from the power system grounding electrode system which utilized Ufer grounding for the UPS,generators and switchgear.