Max wire size?

[ptrip]

I have a situation where I'm pulling wire 1,150 feet through an existing tunnel system. The tunnels are full of existing utilities ... so it's not going to be an easy/straight shot.

The load is a 600A 480v/3ph service (not main service entrance, but between main switchgear and distribution panelboard).

I try to limit the size of wire in my designs to 500kcmil. Given the load and distance, I'm going to either go with 2 sets of 600's or 3 sets of 400's.

The 600A safety switch I'm installing just before the distribution panelboard only comes with 2 lugs per phase and only accepts up to 500kcmil conductors. I suppose I could specify the 800A switch, which would provide 3 lugs and up to 750kcmil cable ... with 600A fuses. Which still leaves me with the 2 sets of huge vs 3 sets of large question.

I want to provide the safety switch for ... well ... safety :roll:. This way the conductors can be de-energized for maintenance work. AND ... these feeders are temporarily re-feeding this distribution panelboard until the new system is up and running further down the tunnel. So then the switch ends up operating as an easy location to transfer between "old board" and "new board".

Which way should I go? (or is there another option I'm not seeing?)

Thanks!

 

[charlie b]

I am missing something here, Pam. You can deenergize the panel, and the feeder feeding it, with the breaker on the upstream switchgear. There is no code requirement, and no good reason I can think of, to put a local disconnect upstream of the panel. Does the panel have accomodations for the two 600's or the three 400's?

Regarding the choice of two huge or three large, there is a piece of information you left out, with regard to the future install. What do you mean by "futher down the tunnel"? Will the final configuration be longer than the present 1150 feet, because the final panel is further down the tunnel? Or is 1150 the longest that this feeder will be?

Two more things come to mind. First, a pair of 500's should keep the voltage drop below 3%. Is that good enough? Secondly, how will the conductors be routed in the tunnel (e.g., in new conduit or in exisiting tray).

 

[K8MHZ]

In a power plant I helped build, oversized conductors were common. When the conductor was larger than the lug, we used 'pin adapters'. One end was a pin terminal that would fit in the OCPD terminal and the other was a lug large enough to accommodate the conductor.

 

[fridaymean]

You could use bus duct...don't know if is a possibility or not...

 

[ptrip]

I am missing something here, Pam. You can deenergize the panel, and the feeder feeding it, with the breaker on the upstream switchgear. There is no code requirement, and no good reason I can think of, to put a local disconnect upstream of the panel. Does the panel have accomodations for the two 600's or the three 400's?

Regarding the choice of two huge or three large, there is a piece of information you left out, with regard to the future install. What do you mean by "futher down the tunnel"? Will the final configuration be longer than the present 1150 feet, because the final panel is further down the tunnel? Or is 1150 the longest that this feeder will be?

Two more things come to mind. First, a pair of 500's should keep the voltage drop below 3%. Is that good enough? Secondly, how will the conductors be routed in the tunnel (e.g., in new conduit or in exisiting tray).

A little background ... The gear I'm feeding is currently the "service entrance" switchgear for half of a high school campus. Feeds to the whole campus are being redone to come from a new single switchgear entrance already installed (as half the campus is being "re"constructed). Site work requires that the feed to this gear be removed before the new distribution network is set up ... hence the need to temporarily feed it from the new gear location. This temporary feed can then be used to permanently feed the panel distribution for this particular area of the school (which will be located approx 100 feet down the tunnel).

My feeling is that the standard maintenance man who's getting into the gear is more likely to throw a switch that's local than to walk almost 1200' to the upstream disconnect. I know, rules and regs and commons sense say to make the walk ... but we're talking about humans and the "need" for convenience. There's also the difference of "required" and "good design". As is mentioned commonly here "Required" is minimum ... which isn't always best. Oh ... and technically the two locations are in different buildings ... served by the common tunnel.

I don't know about the existing lugs ... I'm not qualified to open it! :grin: I do know that, from exist'g dwgs, it's a 700A bus in a Bulldog switchgear. I also know that this feeds half of a high school campus ... and the whole campus only had a max usage of 520A on a bill last year.

The new feeders will be run in new galvanized conduits.

I am using Mike Holt's voltage drop calculator ... that's how I came up with the possible sizes. Did I do something wrong there?


Clear as mud?

 

[ptrip]

In a power plant I helped build, oversized conductors were common. When the conductor was larger than the lug, we used 'pin adapters'. One end was a pin terminal that would fit in the OCPD terminal and the other was a lug large enough to accommodate the conductor.

That's why I'm here ... I didn't know that existed! Voltage drop doesn't tend to be a terrible issue in my projects ... this is a first for me!

You could use bus duct...don't know if is a possibility or not...

The nature of the tunnels is nasty ... the bends and dodges that would be required would make it cost prohibitive.

 

[augie47]

My calculator shows you could load a pair of 500s to 560 amps and still limit the VD to 3%, so I'd be seriously thinking 500's.
I've not seen a non-fuse or fusible disconnect where changing the lugs is much of a problem.
Off the top of the head, were it my project, I'd be running 500's and change lugs if needed.

 

[nakulak]

it may be inconvenient, but possibly cheaper to step up the voltage and step back down for the run. that's quite a bit of cable.

 

[ptrip]

it may be inconvenient, but possibly cheaper to step up the voltage and step back down for the run. that's quite a bit of cable.

That's defintely worth some thought, as running through this tunnel will be a serious challenge. Initially ... it's a clean dry and full height tunnel ... but once you look closer at it it's full of water pipes, heating pipes, tech conduits, etc, etc and etc. Every little bit I can save, the better.

 

[tkb]

Is this a steam pipe tunnel, and did you consider the ambient temp in the tunnel?
These tunnels are always pretty hot.

 

[ptrip]

Is this a steam pipe tunnel, and did you consider the ambient temp in the tunnel?
These tunnels are always pretty hot.

Thank you for the reminder! Not sure why ... but these tunnels seemed uncharacteristically comfortable!

 

[W6SJK]

Is wire bending space at the lugs or phase-to-phase clearance ever a problem with reducing pins? Seems to me I ran into that problem years ago...

 

[ohmhead]

Well most breakers have that information on them you look at your gear and it will tell you on that breaker how many parallel conductors and what size wire it can take its on the breaker itself . You can see it on the front of that breaker or get you old catalog out on that type and size breaker it will tell you .


Theres a label on it and it will give you the lug catalog number type for 2 conductor or 3 conductor or 4 conductors from the factory .


Your lugs in that breaker can be a simple change and no pins are needed check your lugs first !


If you dont want or can not change out your lugs on that breaker for what ever reason then they make reducen pins as stated and they will fit fine they come in different types if you going with parallel conductors make sure you get the OFFSET type pins do not get the straight type.

The offset will fit better into your breaker when you use paralleled conductors .
And they make 90 deg pins 45 deg pins 30 deg pins for space issues .

By the way that length is kinda normal on most of our work.

 

[winnie]

The nature of the tunnels is nasty ... the bends and dodges that would be required would make it cost prohibitive.

How many bends, dodges, and _pull points_ will the conduit you are planning require?

Would a cable wiring method be better?

-Jon

 

[ptrip]

How many bends, dodges, and _pull points_ will the conduit you are planning require?

Would a cable wiring method be better?

-Jon

Let's see.

As I have it drawn (flat on paper with no obstructions), there are 9 horizontal bends. There are probably 6 more horizontal bends minimum, to get out of the building the switchgear room is in.

Vertical bends are more ambiguous. A lot depends on which route they take through the tunnel (high, low, left wall, right wall) ... and I will leave that up to the electrician. There are eletrical panels and transformers (that serve existing buildings which are to be demo'd), phone boxes, pump controllers and other miscellaneous "stuff" along the way.

If they run the conduits low on the wall as I presume they will, there will be 12 bends going up and over transformers and panels alone (if they can't go under/behind). Add 4-5 elevation change in the switchgear building. Add another 20% for unforeseen. You up to over 20 vertical bends.

So ... short answer ... probably about 40 bends. Maybe more, maybe less. Some may be combined.

Pullboxes ... <shrugs shoulders> ... "as needed" ... don't mean to be flippant, but I don't design to that detail.

Cable wiring? Please explain! Open to all suggestions here ...

 

[augie47]

Might be worth discussing MC. The disadvantage is it's 8 lbs/ft.
Your's would be a bit "special" as standard 500 MC only has a #2 equipment ground.
It obviously eliminates a lot of bend/pull box problems.

 

[winnie]

Let's see.

[...]
over 20 vertical bends.

So ... short answer ... probably about 40 bends. Maybe more, maybe less. Some may be combined.

Remember that you are permitted a maximum of 360 degrees of bends in conduit between 'pull points'. Each bend acts as a pulling force multiplier, so in a sense the pulling force scales exponentially with the number of bends. (I'm glossing over many details, but the key is that too many bends make conduit pulls difficult).

What you are describing suggests that the conductors would need to be pulled numerous times at different locations in the tunnel.

Cable wiring? Please explain! Open to all suggestions here ...

Instead of using conductors in conduit, using cable assemblies such as MC or its more robust cousins, or using something like tray cable in a suitable tray. Something where you don't have to pull conductors into a conduit.

-Jon

(Note: I have no experience with this sort of wiring, so I can't do more than speculate and throw ideas out at you.)