750kva, 480/277 entrance conductors.

[fastline]

In a bit of a debate as this is getting interesting. I have personally inspected this property, which has a 750kva xfmr, serving a building, with a single set of 500mil Al conductors. Not even close to being able to handle the full grunt of that xfmr. Kind of odd if you ask me. Total disconnect ampacity is 460A over 3 disconnects.

Because all this switch gear is getting moved and power has to be cut anyway, I am highly encouraging a repair to get fully ampacity terminated to first OCPDs, which has been agreed. However, there is now the debate of Copper vs Al, sizing, etc. Generally, that short run for me between xfmr and OCPD should never be a compromise, but others are trying to 'factor' it. The reality is that xfmr can handle serious overload, which can be an advantage, and expensive later if you blow up the service wires!

I am recommending 3 sets of 500Al or 2 sets of 750Cu. Obviously you can play with that a bit but I generally size around 75c for rated xfmr so there is headroom for OL. I realized all the tables, factoring, etc. I'd just like to get ideas here.

I have never seen someone take the 4+ conductor hit on a service, but this one only has a single 4" conduit so I am scratching on the best play here. I sort of think someone got shorted here. WTF would you only run a single conduit and conductor set for that size?

 

[nhee2]

what is the load?

 

[winnie]

IMHO the rated current of the transformer is not relevant. That transformer will be able to vaporize any conductor set you can fit into a 4 inch pipe. So you ignore what the transformer can do and focus on the supplied load.

Does the calculated load exceed the ampacity of the existing conductors? Do you have any peak demand data that might tell you the actual current going through the wires?

Is there a reason to believe that load will increase?

If there is reason to believe that more of the transformer capacity is going to be used, then absolutely increase the capacity of those conductors. But 'The transformer can supply more.' isn't a good reason.

The mismatch is a bit puzzling, why have 260kVA of conductor connected to a 750kVA transformer. But if the load served is less that what the conductors can handle, then it really isn't a problem.

 

[electrofelon]

I agree with Jon. I think you may be overthinking this a bit. It's very common to have multiple service disconnects add up to much more than the ampacity of the serving conductors. Think about it this way, how long has this been installed and working fine without problems?. It's kind of funny how sometimes we look at something (I admit I do it too) and am like "oh my this is no good, undersized we gotta upgrade this" yet it's been plugging along seamlessly decades before I was in diapers. Maybe load is being added in your situation, I don't know. You said the equipment is getting replaced, so what are you putting in for disconnects, multiple or a single?

Seems like you are going overboard on the conductors. 3 sets of 500AL gives you 735A. Can you get demand data from the utility?

 

[fastline]

I should clarify a few things. I was only sharing information regarding the ampacity of the disconnects. The building does not have any major loads right 'now', but 'now' is the time to upgrade because all the service wire has to be removed anyway, and the concrete cut out because those disconnects are in a small building that is getting taken down, so they have to get moved to a wall about 20ft away.

The building is 60k sf, and very likely something will be installed that has serious demand. If this is not corrected now, then you just wait and it will cause a serious disruption to a business and retrench the concrete and add more conductors.

Thus why I am recommending to make an upgrade because now is the time. Even if that upgrade ends up only being a single set of 1000mcm.

 

[jim dungar]

it will cause a serious disruption to a business and retrench the concrete and add more conductors.

You can add the conduits now, but you don't have to put any conductors in them yet.

 

[Elect117]

I am with you Fastline. While it isn't a code issue if the load is less than the ampacity of everything and the conductors are protected, I dislike it as well.

I would (Like Jim just said right before I finished typing lol) add the required infrastructure but also increase the supply side bonding jumper or GEC size to match the conductors you would use for transformer FLA.

I would double check you don't need to do the same for the primary side. Like needing an extra conduit or if they do add load, needing to upgrade that wire size.

I would advise them to upgrade the wire to transformer FLA or to the max OCPD rating. Since you feed multiple, I would advise the sum of the OCPDs. Electricians are not always going to check the main service wire size off the transformer and perform calcs before installing new loads on subpanels in a industrial / commercial establishment. Even if they should.

But, again, not a code issue until they put in more load than the rating of the conductor.

I would also inquire if they are breaking the tap rule 240.21(C). They should terminate in a single over current device or have multiple runs off the transformer. Not taps off the secondary run. If it is actually a service then where is the utility meter? There is a difference on protecting service conductors vs feeders.

 

[fastline]

You can add the conduits now, but you don't have to put any conductors in them yet.

My thinking is to minimize future disruption, disconnection, and inspection. Shutting off the xfmr triggers some things here. Otherwise, I certainly agree with this. My recommendation was to install a couple more disconnects in the rack, terminate them to the entrance conductors, leave fuses out, and tag them out.

If there are other ways to turn up ampacity safely, I am all ears. Doing hot work is never ideal, but I realize it's done every day by the POCO stabbing Polaris lugs.

OR---- Just a thought or idea, what about bumping up the service conductors as discussed, but installing polaris lugs with spare ports in the trough? Like either way, something is going to be considered hot work. Someone would have to do the hot work to punch the trough for any future disconnects and tap into the hot lugs.

 

[winnie]

Ok, so this upgrade is being sold on the basis of anticipated future load. IMHO that is totally reasonable. The existing transformer is certainly sized to handle additional future loading.

I'm more on the side of @jim dungar suggestion: add conduit and space to make the future upgrade easy, rather than paying for the bulk of the upgrade now.

You mentioned one issue: a single 4" conduit as a limiting factor. If you want to squeeze the maximum number of amps through this single conduit, rather than spending the money to get additional conduit run, then I'd go with copper conductors, and parallel them to have 9 'current carrying conductors'. For example 3x350 kcmil per phase, 2x4/0 for the neutral, use the conduit as EGC if this is a feeder rather than a service.

 

[electrofelon]

.

I would also inquire if they are breaking the tap rule 240.21(C). They should terminate in a single over current device or have multiple runs off the transformer. Not taps off the secondary run.

Sounds like this is service conductors we are talking about so article 240 tap rules would not apply.

 

[electrofelon]

Whatever you do will probably be wrong . This is one of those tough situations where on one hand you want to provide adequate electrical capacity for what might come down the line, but (presumably, maybe this isn't an issue and the person writing the checks said I don't care what it costs) you also don't want to waste money on something that will never be used. I would probably agree with Jim's approach, just try to provide some spare conduits, and not go crazy installing conductors and equipment for some giant service.

 

[hillbilly1]

Sounds like the service conductors were installed by the utility. They are not governed by the NEC. They size by their projected load, not maximum possible load. 4/0 Al is used by the poco here for 400 amp services. I installed a 1200 amp service, and the poco engineer wanted to run parallel 4/0 to it. They had extreme voltage drop with the 4/0 ran to the existing 400 amp service, so the engineer came back with 3-sets of 350 Al.

 

[fastline]

OK, more explaining guys. Building is getting serious remodeling, with intent to use as an industrial facility with serious machinery. Transformer is currently set about 30ft from side of main building, but someone built a bastard little shed on the side of the building, then terminated the main disconnects in there. That shed is coming down, which would leave the disconnects and trough sitting in the middle of nowhere.

the xfmr is pad mount. Everything underground. Concrete will have to get trenched and conduit redone anyway to extend the disconnects to the outside wall of the main building.

I identified this 'power issue' years ago but there was no reason to chase it. Now there is reason to either install the proper infrastructure and/or conductors, etc.

My thought is to examine how much it might suck to cut a little more concrete and get one more conduit into that transformer. Run all the new service conductors into a trough outside the building, then have multiple disconnects off the top of the trough. Multiple disconnects means you can shut down the building in zones and otherwise keep things operating.

And yes, this is a a service, not a feeder.

 

[electrofelon]

OK, more explaining guys. Building is getting serious remodeling, with intent to use as an industrial facility with serious machinery. Transformer is currently set about 30ft from side of main building, but someone built a bastard little shed on the side of the building, then terminated the main disconnects in there. That shed is coming down, which would leave the disconnects and trough sitting in the middle of nowhere.

the xfmr is pad mount. Everything underground. Concrete will have to get trenched and conduit redone anyway to extend the disconnects to the outside wall of the main building.

I identified this 'power issue' years ago but there was no reason to chase it. Now there is reason to either install the proper infrastructure and/or conductors, etc.

My thought is to examine how much it might suck to cut a little more concrete and get one more conduit into that transformer. Run all the new service conductors into a trough outside the building, then have multiple disconnects off the top of the trough. Multiple disconnects means you can shut down the building in zones and otherwise keep things operating.

And yes, this is a a service, not a feeder.

Is metering at the spades of the transformer so you don't have to worry about it?

Seems to me the big decision is whether to stick with that original 4-in or you want to cut and trench all the way back to the transformer for more haz. You'd have to run the numbers on what the max you can get in that 4 inches and if you and your client would be content with that. I think you'll be right around 700 amps max.

 

[fastline]

Is metering at the spades of the transformer so you don't have to worry about it?

Seems to me the big decision is whether to stick with that original 4-in or you want to cut and trench all the way back to the transformer for more haz. You'd have to run the numbers on what the max you can get in that 4 inches and if you and your client would be content with that. I think you'll be right around 700 amps max.

That is correct, meter on the xfmr. I also agree with you that though it might be 'best' to get more pipes in the party, it may not be practical, in which deciding on the best play in using that single 4" will be required. Not only looking at conductor count and 'allowable' pipe fill, but practicality in wrestling those conductors. 750Cu ain't no fun! I take Al all day, every day.

But I was going to ask you about your previous comment that 3 sets of 500Al gets to 735A. How did you arrive at that value? On derate, you can use the 90c ampacity. Just curious.

 

[electrofelon]

That is correct, meter on the xfmr. I also agree with you that though it might be 'best' to get more pipes in the party, it may not be practical, in which deciding on the best play in using that single 4" will be required. Not only looking at conductor count and 'allowable' pipe fill, but practicality in wrestling those conductors. 750Cu ain't no fun! I take Al all day, every day.

But I was going to ask you about your previous comment that 3 sets of 500Al gets to 735A. How did you arrive at that value? On derate, you can use the 90c ampacity. Just curious.

Oops, 3 x 500s won't fit, even with the reduced neutral, not even close! The ampacity would be 350x3x.7=735. I think best you can do with aluminum is three sets of 350 with a reduced neutral, = 588. Looks like two sets of 600 aluminum gets you a little higher, 616. Running with copper you can get almost up to 700.

 

[kwired]

I didn't read every post so apologize if this was mentioned - but if the existing conductors had proper overcurrent protection at the service equipment and has never tripped (with a reasonable amount of time in usage) they are not undersized for the currently connected loads.

 

[Elect117]

Sounds like this is service conductors we are talking about so article 240 tap rules would not apply.

I couldn't tell from the OP if he was mixing it up.

 

[don_resqcapt19]

I should clarify a few things. I was only sharing information regarding the ampacity of the disconnects. The building does not have any major loads right 'now', but 'now' is the time to upgrade because all the service wire has to be removed anyway, and the concrete cut out because those disconnects are in a small building that is getting taken down, so they have to get moved to a wall about 20ft away.

The building is 60k sf, and very likely something will be installed that has serious demand. If this is not corrected now, then you just wait and it will cause a serious disruption to a business and retrench the concrete and add more conductors.

Thus why I am recommending to make an upgrade because now is the time. Even if that upgrade ends up only being a single set of 1000mcm.

That is a design and cost issue, not a safety or code issue...only the depth of the pocketbook can determine what is to be installed now.

I would never recommend 1000 kcmil...you don't gain much for the amount of additional copper. 380 amps for 500, and 545 for 1000.

Assuming 4" RMC, you could use ten 350 kcmil copper, assuming the neutral load is small, and have an ampacity of 735 amps.
That being said, Jim's idea in post 6 is the best for this application.

 

[fastline]

That is a design and cost issue, not a safety or code issue...only the depth of the pocketbook can determine what is to be installed now.

I would never recommend 1000 kcmil...you don't gain much for the amount of additional copper. 380 amps for 500, and 545 for 1000.

Assuming 4" RMC, you could use ten 350 kcmil copper, assuming the neutral load is small, and have an ampacity of 735 amps.
That being said, Jim's idea in post 6 is the best for this application.

Certainly agree. Not a safety issue right now. Only point I am bringing is since this entrance has to be torn out and redone anyway due to construction, it is a good time to address the issue......which is only the question of why have a 750kva sitting there if you can't use it's full capacity?

In which I was just looking for possible ideas to sort of "future proof" the build. Adding an additional open conduit was mentioned, but I think that is more the issue than conductors since we don't yet know how difficult that might get. The actual trench work is really not an issue as competent construction crews are onsite right now. It's more about making entrance into that transformer. I've never had to add a pipe, BUT one of the reasons I need to see it is many times I've seen spare pipes preterminated and dead headed so you can just hook on and go.