EGC or EBJ?

[Twoskinsoneman]

any short ahead of your OCP device (at the transformer can, in the supply to the OCP, etc) would require bonding to be sized per 250.66. Any falult ahead of the secondary has no OCP to trip so you would treat it just as you do a service.

In this situation, since you already have a ground bar connected to XO for your grounding electrode connection, IMHO, you need a 5th wire to the panel which would serve as the ahead of OCP bond for the panel, and the path for the EGC to return power to the source.

You know I think this IS right. It makes since and I remember this illustration from one of MH books showing the difference. It was a service disconnect but the idea is the same. Inside the disconnect the jumpers going from the MBJ to the raceways on the load side were sized to 250.122 and the jumpers going from the same point (MBJ) to the raceways on the line side were sized to 250.66.

Oh well. You think anyone on this site changes their mind as much as I do?

I like what MH says- have no loyalties to your opinions.

 

[charlie b]

Here is an update.

I withdraw my earlier statements to the effect that we did not instruct the installer where to place the N-G bond. We did. There is a Grounding Plan included in the drawing package. It shows the N-G bond at the transformer, the N and G busses in the panel separated, and an EGC run from the N-G bond point (at the transformer) to the G bus (at the panel).

I understand the problem with my interpretation that we can size this ?ground wire thingy? per Table 250.122. The OCPD is not ahead of that wire. What I don?t understand is the sense in which this wire connects two or more portions of an EGC (from the definition of EBJ). I understand the sequence of code articles that would direct me to size it per Table 250.66, if it were to be called an EBJ. What I don?t understand is why it would be called an EBJ. I am beginning to think there is a ?hole in the code.?

I know that if a fault were to occur downstream of the panel?s MCB, that wire would serve the purpose of allowing a high enough fault current to trip that MCB. That tends to make it look to me like an EGC. I also know that if the fault occurred upstream of the MCB, the wire would serve no purpose, and it is possible that no overcurrent protection device would terminate the event. I believe that this is considered an acceptable risk, specifically because of the strict requirements of the tap rules. That is, the probability of a fault taking place upstream of the MCB is considered acceptably low, because the run is required to be for only a short distance, and because the wires are required to be physically protected against damage. So perhaps, once again, there is a ?hole in the code.?

Am I the only one confused by this situation?

 

[cpal]

I'm thinking the SBJ from your transformer to the MSB serves to function as any main bonding jumper . it will provide a path to the transformers XO in the event of a fault on the load side of the service Eq. In this case the location of the N-G connection is back at the halo bar in the vault.

If there is a fault at the swb something has to provide a reliable path back to N-G interconnecting the XO and grounding electrodes.

Thus it should be sized per 250.28 (not smaller that 250.66 and as large as 12.5% of phase conductors. )

 

[ibew441dc]

What I don?t understand is why it would be called an EBJ. I am beginning to think there is a ?hole in the code.?[/SIZE]

What should it be called?

Although a different subject, I have a similar complaint with how Short Circuit,Ground Fault, and Objectionable current are used ......http://forums.mikeholt.com/showpost.php?p=1036660&postcount=52

Am I the only one confused by this situation?[/SIZE][/FONT]

Unfortunately Yes ........You are the only one confused by this situation.......(just kidding)

 

[al hildenbrand]

I also know that if the fault occurred upstream of the MCB, the wire would serve no purpose, and it is possible that no overcurrent protection device would terminate the event.

Am I reading this to declare that the only methods will be conductive raceway?

Have the tap installed in PVC, and imagine the line side conductor faulting to the MCB enclosure.

 

[charlie b]

The ?ground wire thingy? has to be strong enough for the transformer primary protection to clear before the ?ground wire thingy? sustains damage.

That sounds reasonable. But it doesn?t always work. Here is an example:

? Transformer rating: 480-120/208V, 225 KVA
? Secondary current rating is 625 amps.
? Secondary side protection (at the panel) is 125% of 625 (781), which I will round up to 800 amps.
? Secondary conductor size selection: 3 sets of 300 MCM.

If I base the ?ground wire thingy? size per 250.122, calling it an EGC, then,
? It will be 1/0 copper, and
? Since I am using 3 conduits in parallel, 250.122(F) requires me to put a 1/0 in each of the 3 conduits.

If I base the ?ground wire thingy? size per 250.66, calling it an EBJ, then,
? 250.30(A)(2) tells me to size it per 250.102(C).
? 250.102(C) says to make it no smaller than is shown in 250.66.
? HOWEVER, 250.102(C) also says that the EBJ in each of the separate conduits need only be sized in accordance with the ungrounded conductors in each conduit.
? In my example, each conduit has a set of 300 MCM conductors.
? Thus, the EBJ in each conduit will be sized at #2 copper.

Summary: If it is an EGC, it will be 1/0 in each conduit. If it is an EBJ, it will be #2 in each conduit.

Conclusion: It is required to be bigger, if I call it an EGC.

That brings us back to the "hole in the code" conjecture. :roll:

 

[Smart $]

I am beginning to think there is a ?hole in the code.?

I've been thinking that on this very subject for years :grin:

 

[al hildenbrand]

Charlie,
I think the difficulty lies with 250.122(F).

When the "ground wire thingy" is no longer in the parallel conduits, it still only has to be 1/0 copper. Nothing prevents the splicing of the three 1/0s into just one 1/0.

However, the three parallel #2s must, when they leave the parallel conduits, be left separate, or be spliced to a 2/0 copper to match the "Equivalent Area for Parallel Conductors," Table 250.66

Interestingly:
#2 = 66.36 kcmil
3 x 66.36 kcmil = 199.08 kcmil

1/0 = 105.5 kcmil

2/0 = 133.1 kcmil

IMHO, the sizing of the EGC in 250.122(F) parallel installations is overly robust.

 

[augie47]

The "hole in the Code" in regard to multiple feeds in instersting.

Some random thoughts:
Ovelooking that for a moment, if you have a single set of conductors from the transformer secondary to a panel, and a fault occurs ahead of the MCB, the protection you will have will be from the transformer primary protection that would trip if the secondary fault is sufficient enough taking the turns ratio into account, Yes?
If that is the case, the larger your MBJ, the more lilely you are to cause this to occur. So, in cases without parallel conduits, treating it as 250.66 would make sense. Correct ?

 

[Smart $]

The "hole in the Code" in regard to multiple feeds in instersting.

Some random thoughts:
Ovelooking that for a moment, if you have a single set of conductors from the transformer secondary to a panel, and a fault occurs ahead of the MCB, the protection you will have will be from the transformer primary protection that would trip if the secondary fault is sufficient enough taking the turns ratio into account, Yes?
If that is the case, the larger your MBJ, the more lilely you are to cause this to occur. So, in cases without parallel conduits, treating it as 250.66 would make sense. Correct ?

Yes and Yes.

But at the same time, why can we not call metallic conduit between transformer and first disconnect the EBJ and a grounding wire a "supplementary" grounding conductor sized to 250.122? Can we technically use the conduit as the EBJ and eliminate the grounding wire altogether? If so, why must we even call the grounding wire an EBJ?

 

[charlie b]

Overlooking that for a moment, if you have a single set of conductors from the transformer secondary to a panel, and a fault occurs ahead of the MCB, the protection you will have will be from the transformer primary protection that would trip if the secondary fault is sufficient enough taking the turns ratio into account, Yes?

Maybe, maybe not. It is possible that nothing will trip, and the fault will continue until something burns up. Again, by virtue of the existence of the tap rules, that is considered to be an acceptable risk.

If that is the case, the larger your MBJ, the more likely you are to cause this to occur. So, in cases without parallel conduits, treating it as 250.66 would make sense. Correct ?

I am not addressing the Main Bonding Jumper, if that is relevant to your question. I am talking about the “5th wire” between the transformer and the panel.

Regarding a fault that occurs upstream of the MCB in the panel, the only fault that would bring that wire into play would be a contact between an ungrounded conductor and that specific wire. If nothing upstream trips, as is possible, then the larger that wire the more likely it is that that will not be the thing that burns up first.

I think that the one and only (useful) purpose of this wire is to carry current from a fault that takes place downstream of the MCB in the panel. Since the N and G bars are not connected within the panel, the only way to complete the fault circuit, and to force the breaker to trip, is via this wire. Such wires that perform such functions are called “EGCs" and are sized per 250.122.

 

[charlie b]

However, the three parallel #2s must, when they leave the parallel conduits, be left separate, or be spliced to a 2/0 copper to match the "Equivalent Area for Parallel Conductors," Table 250.66

That's a good point, Al. Once my three conduits make their way from the transformer to the panel (just a few feet away, of course), how the wire(s) in question is (are) connected to the ground bar is certainly not described in a typical engineering design document. I can see that it matters very much whether they are connected separately or spliced together before being landed within the panel.

 

[charlie b]

IMHO, the sizing of the EGC in 250.122(F) parallel installations is overly robust.

This, at least, does make sense to me. The EGC in each parallel conduit is full size because a fault within a conduit would not only be fed directly from the source (i.e., via that conduit) It would also be fed via the other conduits, starting at the source and running through the point downstream at which the parallel runs are tied back together. It is a "full size fault," so it needs a full size EGC.

 

[al hildenbrand]

This, at least, does make sense to me.

Again, I intimate a conductive raceway chauvinism.

Also, the rules we're looking at essentially don't have length limitations. While these rules apply to your SDS tap (and the tap rules), they also apply more generally. I'm not just thinking of the short distance between your transformer secondary and your MCB.

The EGC in each parallel conduit is full size because a fault within a conduit would not only be fed directly from the source (i.e., via that conduit) it would also be fed via the other conduits, starting at the source and running through the point downstream at which the parallel runs are tied back together. It is a "full size fault," so it needs a full size EGC.

A line to ground fault is going to be fed this way (except for the cases of one end coming loose from its lug) regardless of how the "ground wire thingy" is sized. 250.66 is large enough, parallel or not, to be the "full fault size".