GEC, EGC, AND SERVICE EQUIPMENT

[mcdow44]

I'm a Temporary Services Electrician in a Shipyard. We work to our Industrial Process Instruction along with the NEC. A typical setup for us would be running 5 sets of Single 8's (4000A) from the local 480VAC 3 ph ungrounded service point to our 4000A JBox to multiple 800A switches that would provide power for our load centers which could service blast equipment, etc. We install a GEC at our furthest upstream service equipment. Our definition for service equipment is the temporary service equipment downstream of the service point and includes a service disconnect and the very fist overcurrent protection device for the temporary distribution system. Our JBox does not fit this definition. Typically, I've seen a cable tied in to the service point ground to our JBox and then to our switch. Should this cable be sized as an EGC to be able to trip the 4000A setting at the service point or should this be sized as a GEC with a separate GEC at the 1st switch in line? I've seen it both ways. Also, with electricity wanting to go back to source, I'm curious how this applies on an ungrounded 3 phase system. I have seen poor crimps at the cable lugs where the cabling actually heated up and pulled out of its crimp. It didn't hit anything but if it hit the case of the metal Jbox and the Jbox only had a GEC, would it be able to trip the 4000A setting and would the size of the cable be able to handle it? We typically use 3/0 AWG for our GEC.

 

[Dennis Alwon]

I am having a bit of difficulty following you but in general if the wire you are talking about has ocpd ahead of it then it is not service equipment and the conductor is then an egc not a gec. If the conductors have no ocpd ahead of it then you need a gec. 3/0 is as large as you will ever need for a gec.

 

[charlie b]

I am also confused. I am aware that shipboard power distribution systems are ungrounded. But you are talking about a shore-based system. I don't see anything in your description that would invoke the requirements in NEC 250.20 and thereby force you to use a grounded system. But if your system really is ungrounded, then why are you mentioning a Grounding Electrode Conductor (GEC)?

I will also mention, just in passing, that I do not know what you mean by the phrase "Single 8."

But what confuses (and concerns) me the most is your description of the setup. What it sounds like to me is you start with a utility-owned transformer (else you could not be calling this a "service point"). You then bring 4000 amps worth of wire to a junction box, not yet having passed through an overcurrent device. You then split out the one set of service conductors to five separate sets of conductors, each rated 800 amps. These conductors are brought to five separate 800 amp switches, and these are the first overcurrent devices downstream of the service point. So you are protecting the transformer's secondary windings and the secondary conductors between the transformer and the J-box with a set of 5 OCPDs. Without knowing more about the transformer's ratings, the conductor sizes, the distances involved, the method of routing the conductors (i.e., the degree of physical protection against damage), and whether this entire installation is inside or outside of a building, I cannot offer any opinions regarding whether this is in compliance with NEC requirements.

Can you shed more light on the situation?

Welcome to the forum.

 

[augie47]

Likewise I'm not 100% sure I follow your post but Dennis' guide is what I use as a the basic rule, however, if we are addressing bonding of service conductor enclosures note that 250.92 and 250.102(C) are involved and your required bonding may exceed the 3/0 GEC limit.

(posted before I read Charlieb's post.. agree more detail would be helpful)

 

[mcdow44]

I'll try to make things a little clearer. We tie our cables into the local utility company's service points. We call them mounds. Typically they're along the pier or in pockets of a drydock. We call for an outage and at the substation they rack out the breaker. We tie into the A, B, and C bussing. There is no neutral, but there is a ground point that we tie into the shipyard grid. We mainly deal with Delta 3 phase. Single 8's are 800kcm type w cable. We triplex the a, b, and c phase cables together. Our services can run inside buildings, outside buildings, and shipboard. Everything we do is temporary services with flexible cable. Our load centers typically have a transformer so we can distribute 110/220 for temporary lighting and tools. We run a GEC off of any separately derived system like our load centers and the equipment downstream of that with an egc. There has been a little bit of a debate as to whether our JBox downstream of the utility company and before our service disconnects is just a continuation of the utility companies service or if we treat it as service equipment and run a GEC. Some projects have 2000AMP switches that simplify the setup, however, there aren't enough to go around. The single 8 cabling is good for 800 amps. I forgot to mention, some of our tie in points to the utility system there is local control. But my question pertains more to the previous setup I described.

 

[augie47]

To confirm my understanding:
You have 3 phase 480 Delta phase conductors from an unrgrounded system extending from a service point to (thru) a metallic j box and then to service equipment. Correct ?

 

[mcdow44]

You are correct augie 47. I just got a call back from the engineering tech. They're saying to run a GEC to the JBOX. Can a fault on a delta 3 phase system return to source? The system is designed more to keep things running if things go down and less for personnel safety.

 

[charlie b]

Can a fault on a delta 3 phase system return to source?

If the system is ungrounded, then a fault from one phase to ground would not establish a return path to the source. That is the basic reason for using an ungrounded system in the first place. It gives you time to discover, locate, and repair the first fault, before a second one can occur.

I just got a call back from the engineering tech. They're saying to run a GEC to the JBOX.

From where, and connected to what at each of its ends. I repeat my confusion: The installation of a GEC is all about connecting some specific point within the distribution system to planet Earth. In most of the systems I deal with, one end of the GEC connects to the center tap point of a WYE secondary, and the other end connects to a ground rod that is pounded into the planet. But for an ungrounded system, you don't want any part of the system connected to dirt. So I don't understand why anyone would suggest installing a GEC. And since one end of the GEC is attached to a ground rod (or some other grounding electrode), where do you attach the other end? If it is to one of the phases or to a center tap point, then you no longer have an ungrounded system. If it is to the external metal case of the J-box, then it will serve no purpose whatsoever.

 

[augie47]

In my opinion, 250.92, 250.96 & 250.102 would require you to bond all metal raceways, enclosures, etc, with a equipment bonding jumper sized per 250.102(C). At the service gear a grounding electrode and conductor would be required per 250.30(B)

 

[electrofelon]

In my opinion, 250.92, 250.96 & 250.102 would require you to bond all metal raceways, enclosures, etc, with a equipment bonding jumper sized per 250.102(C). At the service gear a grounding electrode and conductor would be required per 250.30(B)

I agree. You still need a GES and GEC for an ungrounded system. Even if you are not connecting the system to dirt, you still have to connect non current carrying metal parts to dirt.

There has been a little bit of a debate as to whether our JBox downstream of the utility company and before our service disconnects is just a continuation of the utility companies service or if we treat it as service equipment and run a GEC.

I dont think it is service equipment but it is a service raceway and would need to be bonded per NEC "service bonding" requirements.

But what confuses (and concerns) me the most is your description of the setup. What it sounds like to me is you start with a utility-owned transformer (else you could not be calling this a "service point"). You then bring 4000 amps worth of wire to a junction box, not yet having passed through an overcurrent device. You then split out the one set of service conductors to five separate sets of conductors, each rated 800 amps. These conductors are brought to five separate 800 amp switches, and these are the first overcurrent devices downstream of the service point. So you are protecting the transformer's secondary windings and the secondary conductors between the transformer and the J-box with a set of 5 OCPDs.

There is no requirement to protect the windings of the utility transformer. This sounds like it could be fine and a standard 230.40 Exception 2 setup (assuming the service disconnects are grouped).

 

[mcdow44]

The GEC goes from the case of the JBOX to the shipyard grounding system at the closest point. I'm not too sure exactly why this is. If the case is energized and it's bonded to ground, there still will be 480 V present and if a person makes a connection to ground, they're still going to get 480 V but milliamps due to our higher resistance. Am I correct on this?

Sent from my LGL15G using Tapatalk

 

[GoldDigger]

If there is no other ground on the 480V side connecting a 480V, then a connection from a hot lead to the box will not feed any current to the box except capacitive leakage. If the 480 is ground referenced and the box is not grounded, the box will be energized, but there will be no high risk unless something actually grounded is nearby.

 

[mcdow44]

So, on an Delta 3 phase 480VAC ungrounded system if 1 phase shorts to the metal jbox and the jbox is grounded via a gec, than that phase should read 0 or a very low voltage to ground and the other 2 phases would still read 480V phase to phase and higher than 277V phase to ground? If that's the case, than a person touching the case and making a path to ground, would they feel a shock? What would happen in the same situation if the jbox wasn't bonded to ground?
I have another question and maybe I should put it in another part of the forum, but I will ask here. We transform 3 phase power down to 120/240 single phase grounded power. Is this truly single phase power that is separated 180 degrees apart or is it just two of the 3phase and they are 120 degrees apart? I know there is a reference to ground now and each line will read 120V to ground or 240V line to line.

 

[GoldDigger]

The output of the stepdown transformer will be single phase 120-0-120 since it is s single center tapped winding.
That will be true whether the primary is fed L-L or L-N. The voltage across the primary is single phase and all secondary windings on that transformer will have the same phase with respect to the primary.
It gets more complicated for s three phase transformer because of the opportunity to wire windings as either wye or delta.

 

[mcdow44]

I'm curious as to why it's called single phase. The two phases will still be separated 120 degrees. I thought true single phase that the two phases or lines were separated 180 degrees. Will a single phase motor work just as efficiently on a single phase system that's stepped down from 2 phases of a 3 phase system?

 

[GoldDigger]

I'm curious as to why it's called single phase. The two phases will still be separated 120 degrees. I thought true single phase that the two phases or lines were separated 180 degrees. Will a single phase motor work just as efficiently on a single phase system that's stepped down from 2 phases of a 3 phase system?

The secondary hots will be 180 degrees out of phase. Period.
When only one voltage is present between input terminals of the transformer it is only one phase, since the different L-N phase voltages have no effect on the operation of the transformer.
If instead you run two hots from a three phase wye to a panel you will technically have two phases present, part of 208Y/120, but you can still only drive single phase loads

 

[ActionDave]

....Will a single phase motor work just as efficiently on a single phase system that's stepped down from 2 phases of a 3 phase system?

Motors will work just fine, they only care about voltage and frequency.