Ground Rod Installation

[jeff48356]

When upgrading a service panel, the code requires two ground rods. But could someone clarify the requirements? Are we supposed to place the ground rods exactly 6 feet apart? Or is it at least 6 feet apart, or no more than 6 feet apart? For the ground conductor, do we run a separate wire from the service panel to each ground rod, or one continuous wire to one rod, then onto the second one?

 

[Volta]

At least 6' apart per NEC, but better that they are at least as far apart as they are long. You can run the wire either way that you mentioned, as the particular situation warrants.

 

[infinity]

You can wire them many different ways, here are three methods, one continuous GEC from the service disconnect to the first rod and then on to the second. A separate GEC from each rod to the service disconnect or a GEC to the first rod and then a bonding jumper between the two rods. As Volta stated the rods must be a minimum of 6' apart.

 

[copper chopper]

here is how you do it. most ground rods are 8 feet long, put the first one in then put the second one the lenth of it(8') away from the first one and one 1 continous wire from the panel to the rods. if you run one wire to each your boss is going to frag your ass for wasting all that copper wire... LOL..

 

[Dennis Alwon]

I have heard the length of the rod apart as well as twice the length of the rod. Apparently the further apart the better however "The code states minimum 6 feet".

 

[petersonra]

I have heard the length of the rod apart as well as twice the length of the rod. Apparently the further apart the better however "The code states minimum 6 feet".

That kind of assumes there is any real benefit to the 2nd rod in the first place.

It is not even real clear to me that there is all that much benefit to earthing at a structure at all. It is usually earthed at the pole, what extra good is there from earthing it again 10 or 20 feet away?

 

[Volta]

That kind of assumes there is any real benefit to the 2nd rod in the first place.

It is not even real clear to me that there is all that much benefit to earthing at a structure at all. It is usually earthed at the pole, what extra good is there from earthing it again 10 or 20 feet away?

The more paths you provide for the lightning current to reach the earth, the less current has to travel through the electronics. Far to complex to accurately predict with precision, so we just have to go with 'more is better'.

To figure 'how much more is how much better' is all a cost-benefit ratio, there is no totally right or wrong answer.

 

[kwired]

When upgrading a service panel, the code requires two ground rods. But could someone clarify the requirements? Are we supposed to place the ground rods exactly 6 feet apart? Or is it at least 6 feet apart, or no more than 6 feet apart? For the ground conductor, do we run a separate wire from the service panel to each ground rod, or one continuous wire to one rod, then onto the second one?

First, code does not automatically require two ground rods, often times after going through all the options two ground rods is what is determined to be the easiest or most economical choice.

If you have qualifying structural steel or concrete encased electrode you must use them, and if you do use them they do not require supplemental electrodes such as ground rods. So if you have a CEE you can use it and nothing else. A metal water pipe does require a supplemental electrode, but again if you have a CEE or building steel you have a supplement to the water pipe and a rod is still not needed.

If you do use a rod and can prove it has resistance of 25 ohms or less then a second rod is not required.

 

[infinity]

First, code does not automatically require two ground rods, often times after going through all the options two ground rods is what is determined to be the easiest or most economical choice.

If you do use a rod and can prove it has resistance of 25 ohms or less then a second rod is not required.

One thing to note, under the 2011 the wording has been changed so that two rods are required. There is an exception for 25 ohms or less with one rod. Kind of has the same end result.

 

[kwired]

One thing to note, under the 2011 the wording has been changed so that two rods are required. There is an exception for 25 ohms or less with one rod. Kind of has the same end result.

As far as I am concerned the rules have not changed, only how they are worded is all that has changed.

If you choose not to prove 25 ohms or less then two rods is the rule, 2008 and previous or 2011 NEC.

The way it is written in 2011 may put more emphasis on the 25 ohms requirement - now an exception, but is essentially the same requirement. Before it was sometimes questioned whether or not a third rod would be reqired if the second one does not result in 25 ohms or less. I think it is a little more clear now that two is all that is needed even if 25 ohms is not achieved.

 

[cmreschke]

Ok from my past experiences two things here disturb me. I was a believer that when you drove two rods you could go panel to rod a and panel to rod b I have been violated for that in the past. The inspector and the handbook show panel to rod to rod to try to achieve less than 25 ohms if you go basic rod to panel to rod you theoretically have two separate gecs with more than 25 ohms of resistance.
My other issue is this. Why do we have to prove that we are compliant with less than 25 ohms? In America we are innocent until proven guilty. The burden of proof is in my opinion on the ahj. Simply my opinion.

 

[infinity]

Ok from my past experiences two things here disturb me. I was a believer that when you drove two rods you could go panel to rod a and panel to rod b I have been violated for that in the past. The inspector and the handbook show panel to rod to rod to try to achieve less than 25 ohms if you go basic rod to panel to rod you theoretically have two separate gecs with more than 25 ohms of resistance.
My other issue is this. Why do we have to prove that we are compliant with less than 25 ohms? In America we are innocent until proven guilty. The burden of proof is in my opinion on the ahj. Simply my opinion.

Yes, you can run two separate GEC's to two separate rods. An inspector who violates this is misinformed. 250.53(A)(2) says that the rod must be supplemented by another electrode not that the two electrodes have to be connected together with a bonding jumper.

Regarding your second thought, this is one of the reasons that the 2011 NEC changed the code wording. Now the requirement is that you install two rods or a single rod and another electrode on the list to supplement it. IMO now if you choose to try and meet the exception of 25 ohms or less then you have to prove it.

 

[cmreschke]

As you stated the ROD must be supplimented with another rod. Imo you would not be doing that going rod to panel to rod.

 

[infinity]

As you stated the ROD must be supplimented with another rod. Imo you would not be doing that going rod to panel to rod.

The NEC does not say that it must be supplemented with another rod just another electrode from the list in 250.53(A)(2). It also does not say that the supplement must be connected directly to the first rod. 250.53(A)(2)(3) says that it can be connected to the service grounded conductor. A metallic water pipe requires a supplemental electrode and that is not required to be connected directly to the water pipe either.

 

[cmreschke]

Ok all of my references are based on 2008 since that is what we are working under. Article 250.56 is my reference point. You are augmenting the additional electrode for one that doesn't meet 25 ohms to ground. If you run the gec as two separate conductors then you potentially have two gecs with roughly the same resistance to ground. Now electricity takes the path of least resistance A would it not be better to connect the two together with a bonding jumper making one electrode of less than 25 ohms than it would be to have two electrodes of equalish value?
B. Off the wall question but would you not be parralleling your gec if you run two separate gecs and violating the rules of parralleling?

 

[Volta]

Ok all of my references are based on 2008 since that is what we are working under. Article 250.56 is my reference point. You are augmenting the additional electrode for one that doesn't meet 25 ohms to ground. If you run the gec as two separate conductors then you potentially have two gecs with roughly the same resistance to ground. Now electricity takes the path of least resistance A would it not be better to connect the two together with a bonding jumper making one electrode of less than 25 ohms than it would be to have two electrodes of equalish value?
B. Off the wall question but would you not be parralleling your gec if you run two separate gecs and violating the rules of parralleling?

Electricity does not take "the path of least resistance".

Electricity takes all possible paths.

Two 25 ohm resistances in parallel will be 12.5 ohms.

Section 310.4 does not refer to grounding conductors. In 2008, only, "phase, polarity, neutral, or grounded circuit conductor".

 

[kwired]

Ok all of my references are based on 2008 since that is what we are working under. Article 250.56 is my reference point. You are augmenting the additional electrode for one that doesn't meet 25 ohms to ground. If you run the gec as two separate conductors then you potentially have two gecs with roughly the same resistance to ground. Now electricity takes the path of least resistance A would it not be better to connect the two together with a bonding jumper making one electrode of less than 25 ohms than it would be to have two electrodes of equalish value?
B. Off the wall question but would you not be parralleling your gec if you run two separate gecs and violating the rules of parralleling?

The two are connected together - either at the service equipment or where one taps to the GEC, as well as any other grounding electrodes that may be present. All the grounding electrodes regardless of what type are parallel to each other creating an overall lower lower resistance of the grounding electrode system.

 

[cmreschke]

Two 25 ohm resistances in parallel are just that, 2 25 ohm resistances. Two 25 ohm resistances in series are what we are trying to achieve, 12.5 ohms to ground. If you had 2 15 amp circuits in parallel would you then have thirty amps?

 

[ActionDave]

Two 25 ohm resistances in parallel are just that, 2 25 ohm resistances. Two 25 ohm resistances in series are what we are trying to achieve, 12.5 ohms to ground. If you had 2 15 amp circuits in parallel would you then have thirty amps?

No, it is the opposite. The more parallel paths the lower the resistance.

Ten 100 watt light bulbs in series will not trip a 15 amp circuit breaker, nor will fifty, nor will a hundred. The exact opposite is true if the bulbs were wired up in parallel.

 

[petersonra]

The more paths you provide for the lightning current to reach the earth, the less current has to travel through the electronics. Far to complex to accurately predict with precision, so we just have to go with 'more is better'.

To figure 'how much more is how much better' is all a cost-benefit ratio, there is no totally right or wrong answer.

You are assuming that the ground rod is that path. Just ain't gonna happen through a #6 wire.

In any case, in a typical residential setting you have L1, L2, and N. Only N is grounded. How does lightning that strikes L1 or L2 get to ground? How does light that impresses a voltage from L1 to L2 get to ground.

My contention is that the N-G bond does little if anything, especially given it is already earthed at the pole.