Parallel or not???

Status
Not open for further replies.
david luchini said:
Two sources are supplying four cabinet sections, each section with it's own main circuit breaker. I don't think any transfer equipment would be required.
Click to expand...

I must be missing something.

The OP said

The 4/0 conductors will terminate on main lugs at each separate section of the distribution cabinet.


????

Jap
 
david said:
The 4/0 conductors would still be a tap from the 400 amp supply breaker, the load end breaker would provide the single overcurrent device for the 4/0 tap.
Click to expand...

Strathead said:
Did the OP state that the run was less than 25 feet? Otherwise what tap rule would allow it?
Click to expand...

The title of the thread brought into question parallel or not. I answered the title in that it is a tap and not parallel according to what was described.

I never said the installation was compliant I only said that the tap rules applied to it.

The op indicated he was open to any comments on the installation as your comment is helpful, I would hope the OP would read over the feeder tap section in article 240. All the other comments on paralleling two sets to a closer point then taping the 400 amp feeder would be appreciated by the OP.

The op also did not state what size equipment grounds where run with the 4/0 taps are there two #3 AWG or did someone run #6 AWG from the 400 amp supply distribution
 
petersonra said:
he didn't state one way or the other.



he also didn't say that the conductors terminated at an OCPD as is also required for tap conductors.
Click to expand...

mollydodger said:
I was mistaken on my reply to you. The conductors are to terminate at four separate breakers which supply the four separate sections of the distribution cabinet.
Click to expand...

Though the OP didn?t state the size of the load size overcurrent protection he did state there where breakers
 
Since he's talking about a single 400 amp feed that will service (2) distribution panels, I'm assuming the Mains in the distribution panels are probably 200 amps or less. If that happens to be the case there wont be 4 lugs per phase on the Mains to be able to run the parallel 4/0's to the first main in distribution panel A then jumper parallel 4/0's to the 2nd main in distribuiton panel B.

If the Mains in the distribution panels do have (4) lugs per phase where you could bring the parallel feeds to the first one and jumper a parallel feed to the 2nd, I'd think the Mains would have to be at least 400 Amp and the required feeder would have to be even larger,so I doubt that's the case.

I think If it were me, I'd take the 400 amp parallel 4/0 feeder to a juction box with a Power Distribuition block of some sort close to the distribuiton panels, then tap from that point to the distribution panels main breakers like others have mentioned.


JAP>
 
jap said:
I'm still trying to figure out how your running from (2) separate sources of supply to a "Common Distribution" cabinet without going through some type of Transfer switch.
Not sure we're getting the complete picture.

JAP>
Click to expand...

jap said:
Since he's talking about a single 400 amp feed that will service (2) distribution panels, I'm assuming the Mains in the distribution panels are probably 200 amps or less. If that happens to be the case there wont be 4 lugs per phase on the Mains to be able to run the parallel 4/0's to the first main in distribution panel A then jumper parallel 4/0's to the 2nd main in distribuiton panel B.

If the Mains in the distribution panels do have (4) lugs per phase where you could bring the parallel feeds to the first one and jumper a parallel feed to the 2nd, I'd think the Mains would have to be at least 400 Amp and the required feeder would have to be even larger,so I doubt that's the case.

I think If it were me, I'd take the 400 amp parallel 4/0 feeder to a juction box with a Power Distribuition block of some sort close to the distribuiton panels, then tap from that point to the distribution panels main breakers like others have mentioned.


JAP>
Click to expand...
I think I understand what he has and will give you an example that maybe is not the same thing but is similar to what he has.

One service consisting of two 400 amp service disconnecting means.

supplied by each 400 amp service disconnect is two pieces of equipment requiring a 200 amp supply, supposedly with a 200 amp main breaker in each piece of equipment.

What is in question is whether or not he can feed each final 200 amp equipment with a feeder tap - and he should be able to if he follows the feeder tap rules. Best solution I have seen so far is to put a junction box near the load end and actually make a 400 amp feeder instead of long feeder taps that likely do not meet tap rules, and then make a shorter, compliant tap to each 200 amp piece of equipment.
 
kwired said:
I think I understand what he has and will give you an example that maybe is not the same thing but is similar to what he has.

One service consisting of two 400 amp service disconnecting means.

supplied by each 400 amp service disconnect is two pieces of equipment requiring a 200 amp supply, supposedly with a 200 amp main breaker in each piece of equipment.

What is in question is whether or not he can feed each final 200 amp equipment with a feeder tap - and he should be able to if he follows the feeder tap rules. Best solution I have seen so far is to put a junction box near the load end and actually make a 400 amp feeder instead of long feeder taps that likely do not meet tap rules, and then make a shorter, compliant tap to each 200 amp piece of equipment.
Click to expand...

That was my suggestion above also, but you cant simply bring the parallel 400 amp feeder to a junction box and join the wires.
You'd have to land the feeder on a power distribuition block or some other way to make the end of the feeder a common point.

Otherwise you'd be simply installing a long illegal tap from the primary 400 amp feeder breaker since its beyon the length required of tap conductors.

Leaving out the 2nd 400 amp feeder and only visualizing 1 setup would be like taking a parallel 400 amp feed from a 400 amp breaker in a main distribution panel to feed (2) 200 amp main breaker panels 100 or so feet away.

JAP>
 
Wow such discussion about taps and such, when we have no idea what his system really looks like.

from the OP:
(2) 400 breakers are feeding (4) distribution panels.
this is very common in data centers, where the UPS output goes to a distribution panel and 400A feeds are taken to remote receptacle panels (RPP) which usually have 200A main breakers.

(2) 4/0 feeds are run from the breakers to the distribution panels which are mounted directly next to each other.
this can be interpreted as being; (1) parallel feed of 4/0 is run from each 400A breaker to (2) MB panels which are mounted into an enclosure which holds all four panels.

The key unaswered question is "How are the line side of the 4 panels connected, 2 groups of 2?"
 
jap said:
That was my suggestion above also, but you cant simply bring the parallel 400 amp feeder to a junction box and join the wires.
You'd have to land the feeder on a power distribuition block or some other way to make the end of the feeder a common point.

Otherwise you'd be simply installing a long illegal tap from the primary 400 amp feeder breaker since its beyon the length required of tap conductors.

Leaving out the 2nd 400 amp feeder and only visualizing 1 setup would be like taking a parallel 400 amp feed from a 400 amp breaker in a main distribution panel to feed (2) 200 amp main breaker panels 100 or so feet away.

JAP>
Click to expand...
I guess I didn't describe in much detail, but yes you must tie the load ends of the conductors before you have a 400 amp parallel set. I thought when I said "actually make a 400 amp feeder instead of long feeder taps" that I covered that, just without any details, plus the details were already mentioned in other preceding posts.
 
Paralell or not

Paralell or not

To all who replied thank you! As it turned out once I actually got to the project, which is a data center, there were four PDU's (power distribution units) available. Each PDU had a 400 amp, 3 phase, 120/208 volt spare breaker. We ran a single conduit from each PDU to the RPP which contains four separate distribution panels. Each distribution panel has it's own 400 amp, 3 phase 120/208 volt main breaker. We pulled in 4X 500KCMIL copper conductors in each conduit from the individual PDU's to the RPP and terminated them on the breakers.

If we had been provided with two 400 amp breakers at two individual PDU's and run 2 X 4/0 per phase from them to the RPP they would have "split off" in the RPP and terminated at each of the four 400 amp breakers in the RPP. I don't think that would have been compliant. No tap rules can be applied and the conductors would not have been paralleled since they could not terminate together at the RPP end.
 
Since this is feeding PDU's in a data center is it possible that the loads will be mostly non-linear? If so, your 500 kcmil would be a little small for 400 amps. 4 CCC's= 80%, 430 amps * 80% = 344 amps next standard size 350 amps.
 
infinity said:
Since this is feeding PDU's in a data center is it possible that the loads will be mostly non-linear? If so, your 500 kcmil would be a little small for 400 amps. 4 CCC's= 80%, 430 amps * 80% = 344 amps next standard size 350 amps.
Click to expand...

Good point.

Do you know if there is any quantitative or formal definition of a non-linear load?

In otherwords, if I have a graph of the waveform of the load current, or details of the amplitude/frequency composition of harmonics that produce the waveform, how do I tell from this information if it is nonlinear ENOUGH to be concerned with neutral currents in an application such as this?

Certainly the load doesn't need to be a mathematically pure sine wave, to qualify as a linear load. I often see the metric of percent total harmonic distortion (%THD), and perhaps there is a cutoff value of this figure that designates a "basically linear" load from a load with a lot of harmonics.
 
One candidate metric would be the distortion power factor. Essentially it is the portion of the calculated PF which cannot be accounted for by a simple phase shift. I would propose that a distortion PF lower than .5 would indicate a non-linear load.
 
Status
Not open for further replies.
Top