residental grounding

[tac]

I had a 200 amp residental heavy failed by the inspector because I had used a size #6 solid bare copper conductor for the cold water ground.He stated that I must use #4 as per 250-66 in the 2002 code. I explained that 250-66(A) states that when using a permitted electrode (I used a 1" copper water pipe) the conductor shall not be required to be larger than #6. He said that the code is talking about the ground rod not the water pipe in this case.

Can anyone clairify this for me?

 

[al hildenbrand]

Re: residental grounding

While 2002 NEC 250.66(A) talks about a "pipe electrode", that is not a "water pipe". The inspector is correct.

 

[gndrod]

Re: residental grounding

Hi tac, Also note that the UFER ground on new construction is a minimum #4 AWG.
rbj, Seattle

 

[bphgravity]

Re: residental grounding

While I agree with the inspector and the requirement per the NEC, I do not understand the reasoning behind it. If #6 is sufficient for satisfying 250.4 when connected to a ground rod, it should also be sufficient when connected to a water pipe.

 

[haskindm]

Re: residental grounding

A ground rod cannot and will not carry the return current in the event of a short circuit or ground fault. If we achieve a ground rod resistance of 25-ohms and have a 120-volt hot conductor shorted to the rod, it will develop 4.8 amps. Is a #6 sufficient for carrying 4.8 amps? If that same 120-volt conductor shorts to a water line, and the waterline has .01 ohms resistance (just picking a number here), it will develop 12,000 amps (if the transformer is capable of producing it) until an overcurrent device trips. This is the code's acknowledgment of the fact that a ground rod can't and won't carry enough fault current (at 120-volts) to cause an overcurrent device to open. The waterline BOND on the other hand may be called upon to carry fault current.

[ December 14, 2005, 09:36 AM: Message edited by: haskindm ]

 

[jbwhite]

Re: residental grounding

Originally posted by bphgravity:
While I agree with the inspector and the requirement per the NEC, I do not understand the reasoning behind it. If #6 is sufficient for satisfying 250.4 when connected to a ground rod, it should also be sufficient when connected to a water pipe.

I was told that it is because the resistance of the rod is such that a wire bigger than 6 is a waste, the rod limits the fault path. The resistance of a water pipe is such that the larger wire is a bennifit...

someone help me too..

 

[al hildenbrand]

Re: residental grounding

JB,

The image I use about the lower resistance of a water pipe electrode comes from a group of occupancies supplied by a common water source.

This is not in the NEC.

Many millions of occupancies in the country are supplied by community water systems that are still metal pipe.

When I, as the new service installer, connect the GEC to the water pipe, I am connecting through a 100 feet or so of conductive metal pipe to the neutrals of my neighbors services (the neighbors that are supplied by the same PoCo transformer secondary as my new service.) There are one or more parallel paths back to the source transformer that are continuous low impedance metal all the way.

 

[jbwhite]

Re: residental grounding

i agree al....
and the ground rod is just in the dirt... that is why a wire larger than 6 is not required. there is no fault current path greater than a 6 wire could carry.

 

[bphgravity]

Re: residental grounding

The grounding electrode systems is not and should not be used as an effective fault path for current. That is not its intended use. This is the job of the service grounded conductor. Grounding provides a very limited function and there is no evidence that providing a full sized Table 250.66 GEC to a water pipe will be any more effective than a #6 to a ground rod.

 

[A/A Fuel GTX]

Re: residental grounding

So am I correct in saying that the GEC is just a little something to help lower the resistance of the grounded conductor in a service and not a vital part of the actual fault clearing process?

 

[haskindm]

Re: residental grounding

The grounding electrode has NOTHING to do with clearing faults. Bonding has to do with clearing faults. That is why we BOND to the waterline and other conductive materials. The grounding electrode may help in the event that primary voltage enters the building. I saw this happen once when a tree fell on the power company primary wires and brought them into contact with the overhead secondary conductors to a house.

Look at 250.4(A)(1), which gives the purposes for grounding. It mentions, lightning, line surges, contact with higher-voltage lines, and stabilize the voltage to earth. It doesn't say ANYTHING about clearing faults.

 

[A/A Fuel GTX]

Re: residental grounding

OK.....I need some clarification in simple terms regarding the difference between bonding and grounding. I think this is perhaps the most misinterpreted topic in article 250. I appreciate those of you that are so knowledgeable on this subject because I really get confused sometimes.

 

[roger]

Re: residental grounding

M73214, study the graphic and associated commentary below.

Notice the ground rods (grounding) really have nothing to do with the fault current path (acheived by bonding) back to the source.

Effective Ground-Fault Current Path. An intentionally constructed, permanent, low-impedance path designed and intended to carry fault current from the point of a line-to-case fault on a wiring system to the grounded (neutral) at the electrical supply source, see 250.4(A)(5). Figure 250?1


Author Comment: An effective ground-fault current path is created when all electrically conductive materials that are likely to be energized are bonded together and to the grounded (neutral) at the electrical supply. Effective bonding is accomplished through the use of equipment grounding conductors, bonding jumpers, metallic raceways, connectors and couplings, metallic sheathed cable and cable fittings, and other approved devices recognized for the purpose. A ground-fault path is effective when it is properly sized so that it will safely carry the maximum ground-fault current likely to be imposed on it.

Intent: The addition of these definitions will be very helpful for the Code user to understand the performance requirements contained in Article 250. Defining the words or terms will help the users more readily understand the requirements of Article 250.

Ground-Fault. A ground-fault is an unintentional electrical connection between an ungrounded (hot) conductor and metal enclosures, raceways, equipment, or earth. Figure 250-2

Ground-Fault Current Path. An electrically conductive path from the point of a line-to-case fault on a wiring system through conductors, equipment, or the earth extending to the grounded (neutral) terminal at the electrical supply source.

FPN: The ground-fault current paths could consist of equipment grounding conductors, metallic raceways, metallic cable sheaths, electrical equipment, and other electrically conductive material such as metallic water and gas piping, steel framing members, stucco mesh, metal ducting, reinforcing steel, shields of communications cables, or the earth itself.

Author?s Comment: These definitions originated from the Usability Task Group recommendations for clearly describing what grounding and bonding is intended to accomplish.

Roger

[ December 14, 2005, 03:20 PM: Message edited by: roger ]

 

[A/A Fuel GTX]

Re: residental grounding

Thanks.....That helps tremendously. This is an awesome forum!

 

[LarryFine]

Re: residental grounding

If I may throw in a little plain ol' English:

Bonding is effectively tying conductive parts to one another to minimize the possibility of voltage differences between the parts, whether this results in the parts being at zero volts to earth or not.

Grounding is effectively tying those parts to earth, with as low a resistance as possible, to minimize the possibility of voltage differences between parts as mentioned above, and the surrounding earth.

A grounding electrode does more to keep the earth around electrical equipment at the same potential as that equipment, than it does to keep the equipment itself at the system's neutral voltage (zero, hopefully).

The main bonding jumper is where the neutral and equipment grounding conductor(s) are tied together. This is the pathway through which any line-to-ground (as in metallic equipment, conduit, etc.) fault has a low-enough resistance to trip a breaker or fuse.

Bottom line: the earth alone is never called upon to be part of that fault-current pathway. The system neutral is the point toward which "return current" attemps to flow. The neutral being grounded is why the earth seems to behave similarly to neutral.

Some say that having a grounded system causes more danger than it circumvents, because a non-grounded system theoretically reduces likelihood of shock, but it also minimized gross system over-voltages from, say, lightning or primary-to-secondary faults.

[ December 14, 2005, 09:17 PM: Message edited by: LarryFine ]

 

[suemarkp]

Re: residental grounding

Now the real question becomes, what is so special about the water pipes that you have to use table 250.66? NEC 250.104(A)(1) says you use 250.66 in most cases when bonding water piping. But other metal piping can use 250.122 along with normal equipment grounding.

So you just need to remember that structural steel and water pipes have larger bonding requirements than everything else.

 

[twistin214]

Re: residental grounding

FPN: The ground-fault current paths could consist of equipment grounding conductors, metallic raceways, metallic cable sheaths, electrical equipment, and other electrically conductive material such as metallic water and gas piping, steel framing members, stucco mesh, metal ducting, reinforcing steel, shields of communications cables, or the earth itself

[ December 21, 2005, 10:49 AM: Message edited by: twistin214 ]

 

[bphgravity]

Re: residental grounding

Originally posted by suemarkp:
...Now the real question becomes, what is so special about the water pipes that you have to use table 250.66?...

That is the real question. Mr. Ufer's study seemed to indicate the concrete-encased electrode as being the lowest impedance electrode possible, yet the NEC only requires a #4 GEC. This is because that is the size Ufer chose for his study. There really is no logic to requiring 250.66 sizes to water pipe and strucutral steel electrodes.

Other studies including those included in the NFPA 780 show ground impedance not really being a major factor for events such as lightning or other high energy surges.