This may be the wrong forum, but I thought there might be an electrician out there savvy in commercial issues.
At my work, we are building a device (military contract) that gets a 225 amp three-phase feed. The feed leaves a circuit panel on the wall with the main breaker and travels to our device. In this case the device is a 6-foot tall computer cabinet from Black Box Industries. Inside that is a Square D panel board.
The original plan was to plumb conduit from the wall mounted box up to a drop box in the ceiling, then use flexible SO cord into the cabinet. The issue is that the SO cord diameter is substantial enough that the necessary hole diameter for the bushing is cutting into things we don’t want to cut into (door flanges and other structural items).
So my question is this: Is it feasible to plumb to the ceiling drop box as described above, then drop to our cabinet with two smaller pieces of flexible conduit (or three pieces if necessary) and split the wires between the conduits. If we do not run SO cable, then we could pull 4/0 direct from the wall panel to our internal Square D panel. The only goofy part is that when we leave the ceiling drop box, two wires would be in one conduit and three would be in the other.
Although unconventional, is there a code problem? Is there an electrical/phasing/EMI issue with splitting the three-phase? If this is acceptable to do, is there any protocol on which wires should be included in which conduit?
Thanks gang.
Daen
Replies
Why not mount a junction box on the surface of your cabinet. Have the cord that runs to the ceiling terminate there. Then out the back of the box feed the wiring into your cabinet. Not the best looking but to code.
We have "real estate" issues all around the Cabinet. We can't really expand the external dimensions with a junction box. Adding one would also cause interferences with access doors on the Black Box. In addition, the Black Box can be opened up by anyone - it is not considered and electrical enclosure itself - so we would still need to plumb conduit from the added external junction box into the internal panel board.
If we can plumb from this suggested junction box into our Square D panelboard with multiple parallel conduits, then it isn't much of a stretch to just stub the conduits out the top of the black box and then run flex conduit from there to the ceiling box... I assume there is an off-the-shelf method for joining flex conduit to rigid?
Thanks, Daen
Edited 6/15/2009 6:33 pm by daen
I don't see why you don't just make the run in conduit.
But using 2 conduits for your five wires [2 and 3] is plain wrong. It mainly has to do with some magical electrodynamic forces.
~Peter
Thanks Peter, I thought I had read something somewhere that suggested splitting things up was bad juju from a physics point of view...
The problem with a single conduit is that a hole big enough for it cuts through structural cabinet items. 1.5" conduit will fit nicely. 2.0" conduit requires a lot more fussing, but could work. 2.5" conduit just won't fit.
If anyone has more details on why/how splitting things up causes problems I would like to hear.
Thanks, Daen
think about using high temp motor lead wire for the feed,, like Type SRML. You can downsize the wire which will fit in your pipe, I think #4 will work just fine...but the wire is rated for 200º C and is still limited by the temp ratings of the lugs in the panelsor go find an electric shop that will bend an offset in the bigger pipe or use flexible metallic conduit....Greenfield.....
.
.. . . . . . . .
How about trying a different shape for the drop say more of a rectangle more like a tray perhaps
huh ?.
.
.. . . . . . . .
Well it may be feasible to use a fabricated form of conduit to work past the diameter being excessive. A flange would connect it the Magic black box
Here area a couple of CAD images showing what I have. The red circles in the detail image show the area I have to punch my conduit through. The cut-away shot shows some of the flange detail that is a concern.
Keep in mind that I have an HVAC register about six inches above the cabinet top (don't ask) so adding boxes on top is tight. Also, the cabinet has the top panel that is removable so a box that overhangs that panel will keep it from being accessable for maintenance.
Thanks again.
Daen
OK, let me ask another question. What is the smallest conduit allowable to feed this 225 amps?
The conduit feeding the wall panel is 2.5" per the panel table on the building plans. 5 conductors of #4/0 are used to feed the wall panel inside that conduit. I could run 2.5" conduit right up to the entrance into my Black Box Cabinet and then use some sort of reducing bushing. I could even go back up to 2.5 once inside the box and past all the flanges and structure. So the pull length at a reduced diameter could be as short as 6 inches. However, that reduction would be in the middle of a run of about 10 feet.
Thoughts?
Thanks,
Daen
Again why not transition into another shape
A different shape is feasible. I have limited space overhead to fit it in. It would have to be wide enough to handle the flexible conduit feeding it. Perhaps feed it from the end... Any rules of thumb for bending the wiring through a 90?
Then I would have to have a mating box on the inside of the cabinet to transition back to conduit. That would be a bit tricky to get but doable.
Ah, one more in-house fabricated part. Toss it on the to-do pile.
Thanks for the suggestion. I'll chew on that one and and see what holes I can shoot in it.
Daen
You mentioned you're using 4/0 cable - for a 225 amp circuit, I think you should be able to use 3/0 THHN/THWN, and five of them will fit in a 2" EMT conduit. In some cases you can run two sets of wires in parallel in two conduits - it's done when the larger wires and/or conduits may be difficult to install. However, you may not want to do that as the NEC limits that type of installation to a minimum size of 1/0; in this case two sets of 1/0 would carry 150 amps each for a total of 300 amps. This is oversized for your situation, and each set would still require a 2" EMT conduit so you don't gain anything over the single set of 3/0.
I suggested the use of #4/0 only because that is what they are feeding the wall mounted main panel with. And my box is the only things the panel feeds. So I was just carrying the wire size through.
But glad to know #3/0 in 2" would do the job.
As for what we are feeding... the Black Box is just a gateway to the rest of the device. The Square D panelboard inside the Black box breaks the feed into multiple subcircuits with computer controlled circuit breakers. We don't need 225 amps today, but wiring for tomorrow. We know they'll ask for it even if they don't know it yet...
Thanks for the information.
Daen
I have found a fitting that looks like it will take care of most of my ills. A 2" rigid Offset Nipple seems to fit in the space provided and the offset gets around the flanges. It isn't perfect, but will require very minimal removal of cabinet material.
I found some reading that discussed the total cross sectional area of the conductor can only occupy 60% of the cross sectional area of a Rigid Nipple. I assume the cross section of the conductor includes the diameter of the insulation, not just the copper. But I can not find any information on this nipple. It is off the shelf at Lowes. Manufacturer is Gampak (India). One of the earlier posts indicated the 3/0 will fit in 2" conduit. I just wanted to confirm that 4/0 won't work in 2".
Google tells me that a THHN 4/0 nominal OD is 0.66". That equates to 0.34" square inches of cross section per wire. 5 of them would be 1.71" total cross section of conductors.
The ID of the nipple is 2". This equates to 3.14 square inches of cross section. 60% of that is 1.88". This is bigger than the conductor cross section. However, I don't know if the nipple maintains the 3.14" cross section through the offset sweep. Can anyone confirm the minimum cross sectional area of a 2" offset nipple?
Thanks,
Daen
And what do you think of this idea... sort of a variation of the very first of this thread.
Run 4/0 conductors (quantity 4 THHN, three phases and neutral) in the 2" conduit and run the ground conductor in a separate conduit? Anybody see any code or electrical engineering issues with just separating the ground into its own parallel conduit?
I attached a screen shot of how the offset nipple will work in the cabinet.
Thanks,
Daen
I don't think you want to run the ground wire separately. There are a couple other options: if the conduit run is entirely metallic (EMT or rigid conduit) the conduit itself can generally be used as the ground so a separate wire isn't needed. Secondly, if you do run a ground wire it generally can be smaller than the current carrying conductors. For a 225 amp circuit, a #4 copper ground wire should be adequate.Also, in some cases the neutral wire can be downsized as well, if the current is balanced across the three phases, but there's more to it than that so I wouldn't do it in this case.As with any advice you receive over the Internet, this advice is worth exactly what you paid for it. :)
Stuart,
Thanks for the suggestion of using the conduit as the ground conductor... I hadn't thought of that one. That just may be the ticket for this circus...
We have tried to balance the loads, but not in any scientific way. So I wouldn't want to pursue the reduced neutral method. Hey, we're mechanical engineers just pretending to be electrical engineers!
Daen
I've worked with those type of enclosures but they were usually fed from the bottom or we bent the pipe the fit it in what little space there was.I still think you should think about changing the type of wire and then you could use a smaller pipe.But what is the load that needs so much power in a little box, or do you just need a bunch of separate low amperage circuits ?.
.
.. . . . . . . .
Your post raises way too many issues that require the 'professional judgement' of a real electrician. You need to bring one in on the job.
Yes, there are several issues raised that the code addresses.
For the general level of this site ... I would assume that any job involving 3-phase .... even the connecting of shop equipment .... ought to be done by a real electrician. Any job involving installing a subpanel ought to involve an electrician. Any 'separate structure' ought to involve an electrician.
Agreed with the last post in general. You may have a bit of a problem with the Inspector regarding the use of the conduit as your ground AHJ will be able to illuminate this issue for you. Use this as a resource as well. Wow now that we are able to see the Magic box in a different light an offset is probably the solution. The inspector may not have an issue with the cross section ares for that little sliver . Let us know what comes next.
And yes, free advice might just cost you more than you paid.
FYI... The building / facility that my box is being installed in will alread have its final before we even begin. So the building inspector probably won't have a hand in this.
But it is a government installation and they have their own posse to make sure things are just so.
As far as the design requiring an electrical engineer... I don't want this to sound like it is a fly-by-night operation. The building electrical was designed by an EE for the architectural firm. The sub panel in our Black Box is the same architecture/design that has been used multiple other times by us for similar installations for the govt. So I am comfortable and confident in the engineering.
The little change that is turning into this big issue is that we are feeding this from the top instead of the bottom which has been the tradition. Typically these installations are on a raised computer floor with ample access below. This one is on solid concrete and a feed across the floor would create an unacceptable trip hazard. We have also changed cabinet vendors with the side affect of this reduction in available area for the feed. All those changes were driven by other considerations and only now has this issue come to light as a consequence.
Since there seems to be issue with running the ground in a separate parallel conduit (can anyone expand on why that is unacceptable) and with using the conduit as the grounding conductor. Do you think I can fit a smaller grounding conductor into the 2" conduit with the other 4 #4/0 conductors? Looks like for a 300 amp load, the recommended grounding conducror is a #4. So do you think quantity 4 of #4/0 and quantity 1 of #4 in a 2" conduit is feasible? Total run length is in the neighborhood of 10 feet or less.
Thanks yet again. I look forward to your additional feedback.
Daen