Using an offset nipple between a meter socket and an outdoor service disconnect would require a bonding bushing where some eccentric KO rings are left intact. The nipple is considered part of a service raceway. Correct?

I would tend to think so, at least, that's the way I interepret it...

I interpret 250-80 requires this offset nipple to be bonded.

250-94(4) suggests bonding bushing or set-screw locknut.

Only one end of the metal "service" raceway needs to be so bonded.

If concentric KO's are encountered, bond both ends. Or at least the end that has any concentric KO's.

Glenn

I find it much less trouble to just use Pvc when possible. If a metal nipple was used here We would have to bond with a #6 solid Insulated (not taped) Green conductor. Try and buy that wire in a pinch.


Bill

That, and my PoCo doesn't want any bonding bushing or #4 bare in their meterbase...

I use PVC too...

I don't believe that the bonding locknut is acceptable with concentric knockouts that have portions of the KO's remaining. It doesn't serve any more purpose than would a standard conduit locknut in this case.

Bonding of this nipple (I assume it is not connected to a hub at one end, in the case of the meter socket above an outdoor rated panel) is required even if there is no concentric ko's. If it attached via a hub, no special bonding is required. If power companies don't want you in their sockets, and you need to bond it, either install a meyers hub on the panel, or install a grounding bushing inside your panel, or use a pvc nipple. Just make sure to bond the pvc with a nylon jumper!(ha-ha)

Rick Miell

It only needs to be bonded at one end, so there is no need to install something in the meter can that the utility doesn't want there.
Don(resqcapt19)

I was unaware that there needed to be a bond to a nipple. I have connected service to thousands of customers and have never seen any sort of bonding at all to a nipple. The only thing I was ever taught to look for is the plastic bushings that protects the wire from rubbing the edge of the nipple. Even in the city (Houston) where they have inspectors and require permits I have not seen this practice.

Also what good would it do if it were only bonded on one side(breaker box). If the meter-can has knockouts then how would there be true continuity tying the two boxes together? Seems like a waste of time to only install a bond to one side. But as was stated by someone earlier, the utility does not allow "any grounds" in there meter can. As also we do not either.

Without the nipple having a true bond between the boxes has helped me to troubleshoot some neutral problems. If the neutral connection was ever lost in the breaker box then the load would try to use the nipple to conduct. If not bonded then it begins to spark/burn around the KO where the nipple makes contact with the box. It has gotten to a point now where I look for this tell tale sign whenever the cust mentions dim lights.

glenn35,

Maybe it's a local thing. Here, the PoCo even made the manufacturers add a separate lug to the Meter Pan to accomodate this bonding wire. I don't know for sure if they'd call for bonding in all situations though. When they stopped allowing the wire to be taped green I went for PVC and haven't looked back.

Bill

Glenn,
The intent is not to bond the enclosures together with the nipple. This is on the line side of the service disconnect, so both enclosures are bonded to the grounded conductor. The purpose of the bonding the nipple or raceway is to provide a good fault path in the event of a ground fault within the nipple. This bonding requirement is more stringent here on the line side of the service OCPD because the only protective device that could clear a fault at this location is on the line side of the utility transformer.
Don(resqcapt19)

Don,

Not sure which Glenn you are mentioning.

On my post, I mentioned " Only one end of the metal "Service" raceway needs to be so bonded".

I did slip-up on the complete info about the set-screw locknut. A set-screw locknut is not allowed when concentric/eccentric KO's are envolved.

On enclosures for Service conductors, see 250-80.

Only 250-64(e) for a metal enclosure for a GEC needs to be bonded to the GEC at each end and to each enclosure the GEC is in.

Glenn

Guys,
The answer has be bandied around quite a bit here. Let's not forget that it is a service raceway, it is obviously metallic, therefore has a possibility of carrying the entire fault of what must be condidered "unlimited" buss, the fault of the entire outside conductors and network. That is the reason service conduits must have bond bushings installed at EACH end, KO's have nothing to do with it. If your AHJ won't allow bond bushings on each end, please change to PVC asap, not a bad idea altogether.

My inquiry is solely directed at the nipple, and only where concentric rings remain. The meter socket is factory bonded, so the bonding bushing, which I install only on the disconnect side is to bond the nipple to the service disconnect which we are required to bond. My observation has been that the bonding of this short section of service raceway has been ignored. Similarly, in off-peak systems where 2 meter sockets are nippled together, I rarely see the nipple bonded.
BTW,
PECO doesn't want any bonding wires in their sockets either. This poses a problem when entering into the side of the socket with remaining rings.

George,
Please cite the code section that requires each end of the service raceway to be bonded. Thanks.
Don(resqcapt19)

You're right Don, I should have posted an entire answer, sometimes when I throw my 2 cents in, I am merely responding to the question, sorry, I do that in real life too !!!!

The first problem here is 250-92. This requires that ALL non-metallic service parts be bonded together - period. If the utility with the meter in question is not allowing a ground tail to the equipment, they are in violation of the NEC. My copy of the NESC is at work, but I will look when I go in. That is 250-92 (a)(2)

250-94 gives us several problems in this configuration. It gives us several methods of bonding, but only (4) applies to this question. "Other approved devices, such as bonding-type locknuts or bushings " The paragraph below states that "Standard locknuts or bushings shall not be the sole means for the bonding required by this section." Notice that it does not say, if you use a bond bushing on one end, locknuts are OK on the other end. It plainly states that locknuts are NOT acceptable as the bonding means. I have thought for 29 years that was plain enough.

I can see, however, if you are used to smaller services like this one, that it may seem excessive, but you are leaving a HUGE weak spot in your bond to unfused conductors if you fail to use a bond bushing. Check out Soares, now available from the IAEI for entire details, too lengthy to go into in this venue.

OK ??? Friends ????

Once again, all this mess is a strong argument for PVC...

I've switched!

It appears bonding is required at one end. Source: 1999 NEC Handbook, Fig. 250.72
http://www.geocities.com/cinkerf/dwg.jpg

George,
I still don't see anything that requires both ends of a metallic service raceway to be bonded. On the line side of the service disconnect the grounded conductor is used to bond the service equipment, the equipment is not bonded by the service raceways. I always thought that bonding both ends was a code requirement because that is what our local utility requires, but I debated this with Phil Simmons (CMP5 chairman) and lost. The code only requires one end of a metallic service raceway to be bonded. If you want to bond both ends, that is fine, but the NEC does not require it.
Don(resqcapt19)

George,
A follow up question. How do you bond both ends of a service mast for an oeverhead service?
Don(resqcapt19)

Let me at it again.
Both enclosures are bonded to the neutral, and therefore, bonded together. You can't get a lower impedance path than that. The nipple now needs to be bonded to only one of the (bonded) enclosures, so in the (unlikely?) event of a ground fault inside the nipple, a low impedance path is encountered, thereby burning clear, or operating the primary fuses.

Hmmmm, unlikely ??? You guys haven't driven service trucks long enough. Bond an overhead service enclosure ??? We use weatherheads which are bonded by being threaded, no locknuts are relied on, or even used, at the weatherhead end, don't know what you guys use where you are.

Is the practical concern for the travel and/or parrallel fault path or it's overall connectability?

this IS making pvc look good...

George,
Are you telling us that you bond the grounded conductor to the weatherhead????
Don

would 250-142 allow this?

Sparky,
It is permitted by 250-142, it's just that I have never seen it done. My point is that George is telling us that both ends of a service raceway must be bonded. I am asking how he bonds both ends of the riser. If he doesn't bond both ends of a riser, then why are both ends of a raceway between 2 service enclosures required to be bonded? I still don't see any requirement to bond both ends of any service raceway in the NEC. I don't see any problems bonding both ends and it is probably a good practice, but it just isn't required by the code.
Don(resqcapt19)

Maybe someone is confusing service raceway bonding with bonding both ends of a GEC raceway. This is required to minimize the choking effect caused by the reactance of the single conductor enclosed by metallic raceways.

OK guys one moreand I think we've worn this puppy out. Yes, you can bond the service race way to the grounded conductor, it is is small enough, it is even practical.

offset nipples aside, You're telling me,that if the service conduit in this question were 200' long, 2" rigid, you feel the code says one bond bushing on one end, and locknuts on the other are sufficient ?

Same thing, now were just debating lengths.

Yes, George, that is exactaly what I'm telling you the code says.
Don(resqcapt19)

>if the service conduit in this question were 200' long, 2" rigid, you feel the code says one bond bushing on one end, and locknuts on the other are sufficient?
I see your point. In that case, I personally would bond at the end nearer to the service.

Secondly, if I thought that it might have to dissipate lightning or radio frequencies, I would bond at both ends.

If i may be the thread simpleton here...

all NEC aside, more bonding is better than less?

I agree with Don here. By the NEC, the conduit could be >200' of EMT and not require bonding on both ends.
But I think more is better.

[This message has been edited by electure (edited 08-20-2001).]

Hi, I’m new to the ECN BB. Those who know me from IAEI know I sometimes get wordy, but I feel short posts sometimes raise more questions and confusion than they solve.

Picture the following arrangement of service equipment from top to bottom:

metal weatherhead
2” steel mast
2” bolt-on hub
meter socket
2” nipple
2” bolt-on hub
service disconnect enclosure

There are 7 metallic items that must be bonded so that, if any of them become energized, the bonding system will carry enough current back to the source to clear the OCD ahead of the transformer. Since this is service equipment, the grounded service conductor (neutral) is permitted by 250-94(1) to serve as an equipment bonding conductor.

The service disconnect enclosure is bonded to the grounded service conductor through the main bonding jumper or screw. If the disconnect is “suitable for use ONLY as service equipment” there is no main bonding jumper because the enclosure is already bonded by virtue of its construction. The meter can is also bonded to the grounded service conductor by virtue of its construction.

So far so good. This leaves metal raceways, nipples, hubs and a weatherhead. Since the grounded service conductor is being used to bond the meter can and the disconnect, none of the nipples and raceways are required to be treated as equipment grounding conductors between pieces of equipment. However, 250-92 still requires all metal parts be bonded together and 250-94 says how.

250-92 doesn’t say that the metal parts in the service have to be bonded together at every fitting like a daisy chain, they just have to be effectively bonded together and this can be done star fashion through the neutral conductor as permitted by 250-94(1). They are each required to be bonded only once.

The steel mast is bonded through the threaded hub to the top of the grounded meter can and the metal weatherhead is grounded to the steel mast by its clamp. No problems yet.

This leaves the nipple between the meter and the disconnect. If it goes into the top of an outdoor disconnect, it will enter through a threaded hub and the nipple will be effective grounded (once) as required. There is no requirement to bond it a second time in the meter can, therefore two metal lock rings and a bushing will suffice at the meter can.

If, however, the nipple runs into a disconnect or any other enclosure without a hub, then a bonding bushing or OTHER approved device will be required to bond the nipple to an enclosure or to the grounded service conductor inside it. George C... this applies to your 200 foot long raceway as well.

Once again, since the nipple (or raceway) is not being treated as a grounding conductor between pieces of equipment, it is required to be bonded, but only once. Which end of the nipple or raceway should be bonded? I prefer bonding at the line end because the return path to the source of power will be shorter, but the NEC does not specify and permits bonding at either end. Other factors such as NO BONDING BUSHINGS IN METER CAN may make bonding at the load end necessary.

Fault currents should flow a closed CIRCUIT in the equipment. That is, a fault current will arrive through a legitimate conductor and then, from the fault point, travel back on the metal enclosure, raceway and grounded service conductor to the source of power. As this circuit is self contained it should have a low impedance.

Where the above statement may break down is when a large current from a lightning strike or fault currents imposed from shorts to OTHER SYSTEMS occur. These currents will not be self contained and will travel only in one direction over the metal parts of this service. In this case the same concerns would apply that are dealt with in 250-64(e) and 250-92(a)(3) regarding “transformer” or “skin” effect on a ground electrode conductor. If ALL raceway/enclosure connections are not bonded the impedance presented to this type of current might rise to some appreciable value.

The point becomes, we are required to meet the NEC, are we required to exceed it? The NEC requires the bonding system to be capable of carrying any fault LIKELY to be imposed upon it. Faults in the service conductors would be “likely” but would lightning and currents from other systems be considered likely?

This is my “take” on bonding service equipment from the 1999 NEC. If anyone can provide NEC references to disprove of any of my assertions, I would appreciate hearing from them.

Welcome, Max!

Great to have another fellow WV'er here!

No need to apologize for the lengthy posts... It's par for the course here!

Especially when one explains things so well (and in a way I can understand!).
IAEI is sometimes too far over my head for me to keep up!

ECN is a little more laid back than IAEI, but we still get into some heated topics...
We try to keep the flames down (although I'm guilty myself) and the topics interesting.

Bill's got a great thing going here... Enjoy!

-Virgil Kelly - Kelly Electric
Greenbrier County
West "By (the grace of) God" Virginia

Knock all of them out and use a larger nipple and put just one set screw locknut in meter base side...you can put it on the panel side but alot of inspectors like to see it on the meter side ..