- Location
- New Jersey
- Occupation
- Journeyman Electrician
If the yellow line is the current flow then yes but it may only be for an instant. The OCPD on the secondary should open.
SSBJ clarification
- Thread starter Pinnie
- Start date
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
The quote you were responding to was predicated on "a bond in more than one location". That means a bond in the secondary system in more than one location, not the fact that you have both an MBJ (for the primary system) and a SBJ (for the secondary system). The diagram has only one bond on the secondary system.
So you are correct that in your diagram the current would primarily only flow in the right half of the squiggly yellow line, to the SBJ at the transformer, to XO, to complete the circuit. Very little current would be expected in the left half of the squiggly yellow line. No current would flow past the ground bar in the service where the EGC connects to the GEC, so you should remove that little bit of yellow squiggle to the meter. But some current would flow to that EGC/GEC junction, along the GEC for the service to the GES, along the GEC for the transformer, and back to the transformer X0.
However, that path on the primary side EGC, the service GEC and the transformer GEC is in parallel with the fairly short path of the SBJ itself. So the current would divide in proportion to the resistances. If all those conductors are the same size, and the roundabout path is 50', while the SBJ is 6", you'd get 1/100 of the current taking the roundabout path.
Cheers, Wayne
Last edited:
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
To illustrate the parallel paths, the purple path below is the intended path for ground fault current at the panel on the right, and the orange loop is an alternate segment that is parallel with the SBJ, and so would carry a little current.
In practice, is the diagram accurate about the GEC going directly to XO, or could it be landed on the same bar as the SSBJ is connected to? If the latter, the orange path would only be in parallel with a very short piece of terminal bar, rather than the SBJ, and so the current on the orange path would be even smaller.
Cheers, Wayne
Gotcha I was drawing this and I got it right. It’s honestly not a bad thing that there’s a parallel path through the GEC in case someone undersizes the SBJ or a similar issue. Are GECs allowed to be a ground fault path or do we think that will change. Couldn’t that potentially energize gas pipes or create an arc somewhere in the building causing a fire? That’s probably why Mike says it should be a short of a wire as possible
Attachments
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
This example illustrates that GECs can end up as part of an additional ground fault path for installations in compliance with the NEC. And 250.121(A) exception can allow the same conductor to act as both a GEC and an EGC, so certainly such a conductor can be part of the intentional ground fault path due to its role as EGC.
So I think the question you want to ask is can cutting a GEC that is not also acting as an EGC ever eliminate the sole ground fault path (in an otherwise compliant installation) and let a ground fault occur without tripping a breaker? And I think the answer is that only if the ground fault is to the GES (a grounding electrode or one of the bonding jumpers). And even then, in an example like in the diagram where there are multiple GECs for multiple systems, it would probably require cutting all of them.
Cheers, Wayne
I guess I try to separate GECs and EGC in my conceptualization because they serve two distinctly different functions while both being called “ground”. I never really thought about until now that GEC also can carry fault current even though that’s not their intended design per se.
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
With a fault to the EGC at the far right panelboard the intended path is to the secondary neutral, you want a low impedance path so that high enough current will flow to cause overcurrent protection to operate. With no system bonding jumper that secondary system has no neutral to ground reference and essentially is an ungrounded system though this kind of system is not permitted to be ungrounded by NEC.
By installing the SBJ we have a low impedance path to the neutral of the supply that connects to all non current carrying metal components out in the system. Depending on where the fault occurs once you get out in the branch circuits and/or additional feeders you may have more than one path for fault current to flow. We still design to have a low resistance EGC so that regardless what other incidental paths there may be there is always that one path we supposedly can count on and it likely carries a majority of the fault current in most cases.
The EGC of the transformer primary is only going to carry incidental current that found an alternate path to this conductor. Fault current is not going past the SBJ and on to the service grounded conductor - that is not a part of the separately derived system, current is trying to get back to some component in the secondary whether it be the neutral or one of the other two phase conductors, the bond to the primary neutral is there to prevent having voltage between the EGC of each system - they bond together to create one EGC network. Fault currents carried on this network are only intended to be very short in duration - long enough to cause the overcurrent device to respond and open the ungrounded conductor that has faulted.
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
You mean MBJ (for the service), not SBJ (for the SDS). The orange path in the diagram I marked up and that you quoted shows that some small amount of fault current will go past the SBJ because the orange path is in parallel with the SBJ.
Still looking for confirmation that the transformer side connections shown in the original diagram are fairly typical when the SBJ at the transformer: that the GEC goes directly to XO, rather than depend on the transformer SBJ for its connection to XO. As in the latter configuration, the orange path is no longer in parallel with the SBJ and so would carry almost no secondary fault current.
Cheers, Wayne
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- Semi-retired engineer
I am having a hard time figuring out how the ssbj contributes to safety. There is no OCPD to trip. Unless you are trying to make it trip the primary side OCPD when there is a fault on the secondary side.
Its purpose is to create a reliable fault current path for all circuits on the secondary side. In the diagram there is no bond past the transformer so no ground fault would trip any breaker without it.
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
I meant MBJ, but I see what you are getting at. There is path from SBJ to MBJ via EGC and GEC conductors to the XO of the transformer that is parallel to the path from SBJ and XO. The latter should carry the higher amount of current as it likely is less resistance, but current will take all possible paths with respect to how much resistance is in each path.
IMO a better way to do this is use one conductor between service and the separately derived system sized to meet both ECG and GEC requirements. If you only have one conductor between those two points it won't carry any current if a fault occurs where shown in that drawing, that fault current is seeking to get back to the transformer secondary and has no reason for going to the service grounded or ungrounded conductors, at least until something else abnormal would occur to make them become a potential path to any of the secondary conductors.
are you saying have your GEC at one point to avoid that? I think that’s best.
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- Semi-retired engineer
This is only because you are allowed to make the sbj and gec at the transformer instead of the first disconnecting means.
So are you saying you don’t understand the role of it when you do bond at your first disconnecting means?
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- Semi-retired engineer
What OCPD does it trip then?
The service Over Current Protection means? Is there a fuse on the utility side?
If you bond at the first means of disconnect, rather than your transformer, are you required to have an SSBJ? Your service drop when you have a pole transformer doesn’t have a ground so I’m thinking no?
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
You need to run a GEC from the separately derived system to the premises GES, which can be to the neutral bus in the service equipment.
You also must have an EGC with the transformer primary conductors.
Nothing wrong IMO particularly if the transformer primary circuit originates in the service equipment, to simply run one conductor sized to which ever one of the two items above requires the larger conductor and have that conductor serve both purposes. If you run the two conductors mentioned above they are effectively parallel to one another anyway.
I agree. I think issues actually arise with having two points of contact to earth. Is there an actual reason we do have to run a GE both at the utility transformer and the service disconnect? I understand having a backup in case for whatever reason one fails but we could have one single point where we run our GEC to. Just go to where we bond.
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
Utility company electrode at the transformer is not under the scope of NEC.
A majority of utilities have MGN distribution systems. The potential stray voltages that can come with that are there whether you have a grounding electrode at your service disconnect or not.
IMO the only advantage you have for installing an electrode(s) at the service disconnect is you have that one spot that connects near the service entry to potentially direct lightning surges to earth. This being incoming surges on the service conductors from lightning strikes elsewhere in the distribution. A direct hit to your premises or even to close proximity is still likely going to have damages and the electrical system GES isn't likely going to protect much here - that is more of what lightning protection systems are about but they are a little more complex