How much does the ground rod really do?
In thoery the current if there is a fault, is to travel through the rod, through the ground and back to the source.
If the the Service equipment is bonded correctly, shouldnt the current travel through the service ground back to the source??
Opinions wanted

I had a service call today, where the pole by the street was hit and completly taken out. A triplex went to my customers building, feeding a single phase service.
A second Quadplex went to the building for a three phase Delta service. When the pole was hit, it tore the triplex off the building, which completly chafed the service conductors. The quadplex pulled on the 500's in the service conduit but did not damage the conductors. Now each service has its own ct cabinet. Neither was bonded. Second there are no ground rods. Only a building steel ground bonded to one of the single phase service panels. The ground wire in the delta service was burnt from the first service disconnect through the cabinet and up the service pipe.
Now it appears to me that when the single phase service conductors chafed and where touching the conduit this caused a short. Which in turn sent the current through the delta service grounding electrode.
Now if the cabinets where bonded, the current should of went through the conduit and through the cabinet and back through the service ground.
How much would a ground rod have really done??

Depending on soil conditions and distance probably not much. The ground rod is more for earth reference and lighting disappation than for clearing a fault. The earth is a poor conductor.

Why in the NEC, does it state 250.66 that size of electroce conductor should be 1/0 cu
for service conductors over 350-600kcmil. Yet 250.66(A) states the conductor that is sole connection to the grounding electrode shall not be required to be larger than 6awg cu.
Now if you say, run the 1/0 into the building from the service head and than in the service enclosure, where bonded cgange to 6awg, that contradicts 250.64(C) where it states grounding electrode should be continuous without splice unlesss compression type connectors are used.

The grounding electrode conductor running to the water service pipe may very well have to withstand every bit of the fault current the utility transformer can deliver.

This is possible because that metal water pipe may be electrically continuous with the other structures around it. Those other structures will also be bonded to the metal pipe providing a low resistance fault path.

On the other hand a ground rod or a concrete encased electrode can only be expected to carry the current it can pass into the soil which is a poor conductor.

Lets say you have a ground rod with 25 ohms of resistance, a 120 volt fault to this ground rod will result in only 4.8 amps of current flow. (120 V / 25 O = 4.8 A)

Even if you happen to get the rods resistance down to 3 ohms (unlikely) a 120 volt fault will result in 40 amps of current on the GEC.

No it does not contradict itself.

The conductor from the service head to the panel is a 'Service Conductor' it is not the 'Grounding Electrode Conductor'

The Grounding electrode conductor starts at the bonding point in the service disconnect enclosure and ends at the grounding electrode.

If you chose to daisy chain more than one electrode together the GEC is the conductor from the bonding point to the first electrode. The conductor between electrodes is a bonding jumper forming the grounding electrode system it can have splices.

Bob

Where does it state that?

I can not say that "it states that" in any one place.

You got to jump around.



Now take a look at part of 250.50


Bonding jumpers are used to bond the electrodes together into a grounding electrode system.

Once you have made this grounding electrode system the conductor from any point in this system back to the service disconnect is the grounding electrode conductor.

Bob



[This message has been edited by iwire (edited 06-20-2004).]

I understand and have practiced evrything you said. I am trying to get specific articles. I am weel awaree of the bonding process in the service. I overkill, as to sleep at night lets say.
The code states the GEC should not be broken but continuous with exception of crimps.
This I interpret as from the ct cabinet if used, bonded to neutral continuous to disconnect or MDP,on to ground rod,to building steel and water pipe.
We both know nobody does this, but I interpret this article as stating to do so.
I originaly posted because I do not beleive that in circumstances I presented with the burnt ground, that a rod would not of prevented this, but proper bonding of the service entrance neutral or ground, to the cabinet and disconnect would have.
Agree or disagree?

Hopefully some others will jump in and offer some other suggestions, the info you want is spread out throughout Chapter III and Chapter V of Article 250.

If you have access to the NFPA handbook you can see a lot of different ways of installing a grounding electrode system.

I could run a GEC from the Service Disconnect enclosure to building steel, then on the other side of the building I could run a bonding jumper from building steel to the water service pipe.

This is all shown in the handbook.

Here is a relevant section.


Now we have to decide, are we going to talk about how you and I may do it beyond code or are we going to stick with what is required.

I will usually jump the water meter without a splice but we are not required to.

I do not believe a ground rod would have done anything for you.

Would proper bonding of one service have protected the other?

Hard to say,no reason the bonding for one service would be large enough for the other service. One service might be 100 amps and the other 600 amps.

Also these services came from separate transformers so whatever bond wire between them on the primary side may not be large enough to carry the full secondary current.

JMO hopefully others will jump in.

Bob

The ground rod is my point. I fail to see its need for service bonding. Electricity will not travel miles throught the ground with all the different resistance back to the source, when it will takes its quickest, least reisitive path. Agreed??
I think someone mentioned dissapate. It wont dissapate. Physics proves electrons travel in lets say a loop. What leaves comes back in full. Cannot see current traveling through rock dirt, fill whatever, when a system is properly bonded at the enclosure, it should travel back through the grounded service conductor.

Here is a graphic from the handbook that labels the conductors of a grounding electrode system.




A ground rod by itself can become very important if you have a primary voltage conductor fall across the bare neutral of a triplex heading toward a building.

In that case the voltage might be 7000 volts or more to ground.

A 25 ohm ground rod with a 7000 volt fault to it will result in 280 amps of current flow, (forgetting about the POCOs grounding resistance) well above the primary overcurrent protection device.

In that case the ground rod would aid in the opening of the primary overcurrent device.

Then there is some added protection from lightning strikes.

There are a few cases where a "25 ohms" ground at the service entrance can be useful. Consider a "line cross," or primary-to-secondary fault in a distribution transformer. For a 13.8KV circuit, you may expect to see ~8kV-to-ground on the low-voltage service-entrance wiring. 8000V/25 ohms = 320A, which should be able to operate some upstream 13.8kV overcurrent devices. Distribution-switching transients and indirect lightning hits can force significant current through a 25-ohm "made electrode."

No question that concrete-encased steel or significant underground bare-metal piping can beat out 25 ohms for "better" grounding electrodes, but 25 ohms may be useful in some caes. Always consider properly-sized bonding as a good means of limiting potential difference between exposed conductive electrical-enclosure surfaces.

One plug... IAEI Soares Book on Grounding—even a used copy—is well worth having for code-based grounding concepts and practicalities. [Also, the 1954 R H Kaufmann AIEE paper is excellent for detailing real, high-current tests of AC fault-current propagation that much of present codes are based on.]




[This message has been edited by Bjarney (edited 06-20-2004).]

Bjarney,
Is the resistance of 25 ohms a standard figure in the US?.
It sounds quite high, compared to a maximum Electrode + Ground Lead resistance value of 1 ohm combined, given in our Regulations.
Mind you, our TN-C-S system relies on high Fault currents to disconnect faulty circuits in under 0.4 seconds.

Trumpy, that is a good question. I have searched for an answer in earlier IEEE and AIEE Transactions {the predecessor to IEEE} and found no history, other than the nominal "25 ohms" has been used for many decades. I believe that particular benchmark/default value of resistance is listed in the 1953 National Electrical Code Handbook, and I don’t think it was new then. Mind you, it is supposed to be a absolute maximum, but there is no specific code requirement that every electrode or ground grid be tested after installation.

Nowadays in the NEC, two driven rods suffice for a grounding electrode without testing. In parts of the continent, it is a very hard number to achieve, where in other places, one driven rod tests suitably low. {Often the very desirable “Ufer ground" re-bar with concrete encasement is acceptably and consistently low regardless of weather and soil conditions.}

Other E-C.net members here have reported that some areas of Canada routinely use [nonferrous?] buried-plate electrodes as a alternate to rods.




[This message has been edited by Bjarney (edited 06-20-2004).]

Just as a point of reference, last time I checked the driven galvanized pipe that is my electrode at the service panel [from 1935?] reads about 65 ohms with the plumbing/water-pipe bond isolated, but the steel casing on the water well with static water at around 70 feet measures about 2 ohms using a conventional Biddle “3-lead”/fall-of-potential test set.

There is a 1970 paper ieee-pcic.org/archive/fagan_lee_paper.pdf that is sort of a followup to George Ufer’s earlier work.

At a new public-school building once, a contractor wanted a test on a driven rod, which measured about 22 ohms. He thought that may be ‘too close’ to 25 ohms and likely the inspector would accuse him of “fudging the numbers,” so he used a bucket of water to soak the two test electrodes. Then, he blew a gasket when the retest reading increased to about 23 ohms(!)




[This message has been edited by Bjarney (edited 06-20-2004).]

I only choose 25 ohms for lack of a better figure.

The NEC uses 25 ohms as a cut off point for requiring a second ground rod.

That said once you add a second rod there is no requirement to achieve any amount of resistance.

Looking at the replies so far in this thread, it would be very redundant for me to reply also! (yet here I am replying! ...)

The topic is really interesting, so if I can figure something non-boring, interesting and concise, and most importantly - get some time to do so, I would like to toss in a few additional comments to this thread.

As a side bar, here's a Pirate Joke for everyone's amusement:

Q: Where Do Pirates Take Their Vacations?

A: ARRRRRRRgentina!!!

<add cheezy drum fill>
Tatttttttttattttttttatttttt--- chhhhhing!

Scott35

Just to throw in a couple of alternate ideas from Merrie Olde England:

On TN-C-S supplies, which are effectively like a standard American service where the house grounding system is bonded to the incoming neutral, there is no requirement here to use a local ground rod at all.

In practical terms, there's likely to be a ground connection due to the bonding of water pipes, but as plastic supply pipes are becoming more and more common even that is changing gradually.

In many rural areas of Britain we have a TT system, where there's is no neutral-ground bond, so the ground rod and earth is part of the fault path. The resistance can be considerably more than 25 ohms:
http://www.tlc-direct.co.uk/Book/5.5.2.htm

Iwire stated ""In that case the voltage might be 7000 volts or more to ground.

A 25 ohm ground rod with a 7000 volt fault to it will result in 280 amps of current flow, (forgetting about the POCOs grounding resistance) well above the primary overcurrent protection device.""

the rod etc is more for surge and lightning,,,,someone do that math!

Grounding can be very confusing but it really is not. A ground rod is not for clearing faults on under 600 volt systems. even a 480 sustem (480/25=19.2) is not likely to clear a fault very quickly.
The grounding electrodes larger than #6 almost always go to the water service entrance and/of building steel.
The ground rod also is for lightning disapation not disapation of an electrical circuit's electron flow. While electricity takes ALL availble paths, the most current will flow on the path of least resistance. Some current will flow from a ground rod back thru the earth to the next closest grounding means ( a water pipe , another ground rod of the POCO ground on a pole). The amount of this current flow will be small in relation to the current back thru the service neutral.
SO the question does a ground rod work? Yes for what it is designed for. No for clearing faults.

Remember that the 25 ohms to ground is just that 'to ground'. What the resistance back to the grounding electrode at the transformer may be is anyones guess.

And yes the ground rod and grounding electrode will 'dissipate' lightning current, as it does not need to complete a circuit. It will also function as a ground reference.

The original question was what will it really do?
In my opinion - remember my opinion!! it is a waste of time and money in a standard service installation. For years and years, the cold water or steel sufficed. I know that with the advent of plastic pipe, the rods supplement the cold water, but the concrete encased electrode should be made to supplement the cold water, if they really want to accomplish any real purpose.

Pierre

By cs409

A rod all by itself can open the overcurrent protection at the voltage levels that the power company uses for local distribution.

It also can help dissipate a lightning strike.

But what kind of surge can it help with?

Only a surge on the neutral, it will do nothing for a surge on either of the ungrounded conductors.

I have a mind set like Pierre's, a ground rod is of little use for the services I generally deal with.

Bob

If you didn't ground the service at the building, wouldn't you risk the earth around the building, and the cement foundation, being at a different potential than the service ground?

It is my view that the purpose of the service ground is to bring the earth to the same potential as the building's bonded systems, and the ground conductor is sized so that it can withstand fault currents that may occur.

twh, I was only talking about a ground rod, most of the buildings I deal with have other much better electrodes that we are already connected to.

Bob

Okay, I handled this poorly. Let me explain:

Where I am, we don't use a ground rod when there is a water line. Last week, I grounded a service to a water line that was 100 feet away from the service.

I think that is a bit far because I once bonded two buildings together that were 200 feet apart and the voltage between the grounds was over 100 volts. That was an unusual circumstance, but once bit, twice shy.

The topic comes up here, and I am interested in the comments. I don't want to take sides, because I don't know.

nesparky said:

In my clumsy way, I'm saying I agree with him.

Then, nesparky gave an opinion:

And iwire gave a somewhat different opinion:

From the context, I didn't think iwire meant that it was of little use "for what it is designed". Perhaps a direct question would help me understand.

Bob, how can I tell when a ground rod, in addition to the water line, would be of use?

tim

In a standard service, there is usually a cold water copper pipe that grounds the service. Generally it is located fairly close to the service disconnect.
I will agree that if the coldwater pipe connection is not close to the service disconnect, that ground rods will help to keep the surrounding area in the same potential as the disconnect/system.

But as a supplement to the system in general it is not altogether very effective. That is why I would like to see that a concrete encased (footing) electrode become mandatory for all new installations. I believe that there are some jurisdictions requiring this.

Pierre

Pierre,
Not really. About 80% of the total voltage drop associated with a ground rod occurs in the first 24 to 30" around the rod. This small area would be the only area that is at the same potential as the system.
Don

No disrespect intended, because sometimes I will speak "matter of factly" too.
But you all state your "opinions" freely, which, again I have no problem, but where is the physics proof in the statements here?

ayrton I do not think anyone here is offended when questions are asked.

Which statements would you like more info about?

I will be glad to learn more about this subject.

I would like to see the 'physics proof' that one or two ground rods does much at all.

Time spent going through Soares Book on Grounding will save years of wondering-even if it's an old edition.