How come u can't run a single phase in each conduit in parallel feeds?

[winnie]

In this isolated phase situation, if we imagine that phase A and phase B are fully loaded, and phase C is not loaded at all, then the N conductors will carry full phase current.

But now phase C isn't carrying any current.

The only point I am making is that in the if you have HHHN HHHN HHHN HHHN or HHHH HHHH HHHH NNNN you have the _same_ total number of 'current carrying conductors' (per NEC counting for derating) but in the isolated phase situation you've concentrated 4 CCCs into each of 3 conduits rather than having 3 in each of 4.

-Jon

 

[wwhitney]

The only point I am making is that in the if you have HHHN HHHN HHHN HHHN or HHHH HHHH HHHH NNNN you have the _same_ total number of 'current carrying conductors' (per NEC counting for derating) but in the isolated phase situation you've concentrated 4 CCCs into each of 3 conduits rather than having 3 in each of 4.

And that's not correct physics-wise, so to the extent that the NEC language doesn't differentiate (I don't think it does), it's wrong.

For CCCs what matters is how many conductors in a single conduit can simultaneously carry their maximum current. Having one conduit with no current in it doesn't particularly help with the heat dissipation on the conductors in the other conduits.

For HHHN in one conduit, that's 3 CCCs as normal. But NNNN in one conduit is 4 CCCCs. So the mixed phase parallel installation has 4 conduits of 3 CCCs each, and the isolated phase parallel installation has 4 conduits of 4 CCCs each. An 80% derating factor would apply to the neutral conduit just as much as it would to the other 3 conduits.

Or at least it should physics-wise, and if the NEC language doesn't require that, it's wrong.

Cheers, Wayne

 

[LarryFine]

The only point I am making is that in the if you have HHHN HHHN HHHN HHHN or HHHH HHHH HHHH NNNN you have the _same_ total number of 'current carrying conductors' (per NEC counting for derating) but in the isolated phase situation you've concentrated 4 CCCs into each of 3 conduits rather than having 3 in each of 4.

That's correct, unless all of those conductors were contained within the same metallic raceway.

That's also why cutting slots between KOs makes the current behave as if there was only one large hole.

 

[wwhitney]

Basically 310.15(E)(1) currently says

"A neutral conductor that carries only the unbalanced current from other conductors of the same circuit shall not be required to be counted when applying the provisions of 310.15(C)(1)"

And the physics requires that it should read:

"A neutral conductor that carries only the unbalanced current from other conductors of the same circuit within the same raceway or cable shall not be required to be counted when applying the provisions of 310.15(C)(1)."

So switching from mixed phase to isolated phase does increase the total number of CCCs.

Cheers, Wayne

 

[synchro]

Depending on the load, 4 paralleled neutral conductors may not be needed, and the 80% derating would not apply with 3 conductors or less.
I agree with Wayne that isolated neutrals should be counted as current carrying conductors. That's because no balancing of currents is occurring between the paralleled neutral conductors, as would happen when all of the circuit conductors are included within the same conduit. Jon's example in post #21 also makes that evident.

 

[winnie]

I partially agree on the 'counting of the neutral' in an isolated phase situation.

The 'balancing of current' inside the conduit does not decrease heating. Rather the balance of currents is what reduces the current actually flowing in the neutral wires, and thus the heat produced by the neutral wires.

In the case of an isolated phase install, if the phase currents were completely balanced, no current would be flowing on the neutral.

The total heating produced by the conductors in distributed or isolated phasing should be the same. If we were doing this as a 'duct bank' with 4 conduits, both the distributed and the isolated phase approach would dump the same amount of heat into the surrounding soil. (distributed ABCNG ABCNG ABCNG ABCNG, isolated: AAAAG BBBBG CCCCG NNNNG)

I agree that the AAAAG conduit has more CCC than the ABCNG conduit, and that could force derating. So I absolutely agree that the conductors in the phase conduits would need to increase in size because there are more CCCs in those conduits.

Where I don't agree is if the NNNNG conduit has any CCCs in it at all, and if so what that means for conductor sizing.

-Jon

 

[wwhitney]

The 'balancing of current' inside the conduit does not decrease heating.

It decreases heating by decreasing current, I'm not clear on what distinction you are trying to make.

Where I don't agree is if the NNNNG conduit has any CCCs in it at all, and if so what that means for conductor sizing.

NEC says it has 0 CCCs, but NEC is wrong. It should be 4 CCCs. Prudent practice requires treating it as such, despite NEC error.

In an isolated phase installation there is no reason to treat the Neutral conduit differently from the Phase conduits in this regard (assuming equal number of conductors).

Cheers, Wayne

 

[LarryFine]

Where I don't agree is if the NNNNG conduit has any CCCs in it at all, and if so what that means for conductor sizing.

Then why run the neutrals at all?

 

[don_resqcapt19]

I partially agree on the 'counting of the neutral' in an isolated phase situation.

The 'balancing of current' inside the conduit does not decrease heating. Rather the balance of currents is what reduces the current actually flowing in the neutral wires, and thus the heat produced by the neutral wires.

In the case of an isolated phase install, if the phase currents were completely balanced, no current would be flowing on the neutral.

The total heating produced by the conductors in distributed or isolated phasing should be the same. If we were doing this as a 'duct bank' with 4 conduits, both the distributed and the isolated phase approach would dump the same amount of heat into the surrounding soil. (distributed ABCNG ABCNG ABCNG ABCNG, isolated: AAAAG BBBBG CCCCG NNNNG)

I agree that the AAAAG conduit has more CCC than the ABCNG conduit, and that could force derating. So I absolutely agree that the conductors in the phase conduits would need to increase in size because there are more CCCs in those conduits.

Where I don't agree is if the NNNNG conduit has any CCCs in it at all, and if so what that means for conductor sizing.

-Jon

I thought there was a 3-5% increase in losses with an isolated phase system as compared to a traditional system as a result of the physical separation between the phase conductors?

 

[winnie]

I thought there was a 3-5% increase in losses with an isolated phase system as compared to a traditional system as a result of the physical separation between the phase conductors?

Hmm. I'd expect an increase in reactive impedance, but not an increase in losses...unless there is significant circulating current in the EGCs.

I've been thinking more on the whole CCC thing, and I am going to have to change my position.

As I understand the reasoning behind not counting a 'neutral' as a CCC, what this means is that in a set of M conductors (including the neutral), the maximum heating case is M-1 conductors carrying full current. Say you have ABCN, which A, B, and C each carrying 20A; the neutral carries 0. Turn off the current in C, and now the neutral is carrying 20A; still 3 conductors carrying 20A. Of course things are probably not perfectly balanced, with all 4 conductors carrying current; (Say A 20, B 20, C 10, and N 10) but in this case the net heating is lower than the balanced full current case. So it is reasonable to count ABCN as _three_ CCC since only 3 of the 4 will carrying current in the worst case heating scenario.

In the isolated phase situation (if properly installed!), if any of the N conductors are carrying current, they _all_ are carrying the same current. If (because of loading) the neutral happens to carry full current, then you have all the conductors in that conduit carrying full current at the same time.

So going back to my point about the underground duct bank; when you are considering the heat production of _all_ of the conductors taken together, the neutral shouldn't count as a CCC.

But when you are considering the individual conduits you have to consider only the conductors in that conduit. And here the N conductors should be counted as full CCC.

Jon

 

[synchro]

I thought there was a 3-5% increase in losses with an isolated phase system as compared to a traditional system as a result of the physical separation between the phase conductors?

The wider separation between conductors will increase the inductive reactance and therefore add some additional voltage drop. There will likely be some losses from currents induced in any earth, concrete, rebar, etc., by the magnetic fields surrounding the conductors, but that would be hard to estimate and I'm not sure how significant it would be.

I see that Jon just posted the same comment about reactance.

 

[wwhitney]

As I understand the reasoning behind not counting a 'neutral' as a CCC, what this means is that in a set of M conductors (including the neutral), the maximum heating case is M-1 conductors carrying full current.

Precisely.

Cheers, Wayne

 

[jaggedben]

As I understand the reasoning behind not counting a 'neutral' as a CCC, what this means is that in a set of M conductors (including the neutral), the maximum heating case is M-1 conductors carrying full current. ...

Jon

This only works if M is greater than 2, and is not 3 in a wye. Which is to say, among common wiring scenarios with a neutral, the formula only works half the time.

HN - nope
HHN in split phase - check
HHN in wye - nope
HHHN - check

Arguably better to just memorize the chart than to use a formula.

 

[winnie]

This only works if M is greater than 2, and is not 3 in a wye. Which is to say, among common wiring scenarios with a neutral, the formula only works half the time.

HN - nope
HHN in split phase - check
HHN in wye - nope
HHHN - check

Arguably better to just memorize the chart than to use a formula.

Agreed.

I should have stated more clearly that I was specifically speaking of those situations where we are permitted to not count the neutral as a CCC.

 

[Carultch]

It would also be much easier to install ct's

The fact that it would be much easier to install CT's is PRECISELY WHY you are generally not allowed to isolate the phases, unless you qualify for the conditions of isolated phase installations. It is good for a CT to capture a magnetic field as that is what it is designed to do; but it becomes a problem when you have ferrous raceways that you don't intend to magnetically energize. Even a component as seemingly insignificant as a steel locknut is enough to disqualify you from a code-compliant isolated phase installation.

CT's measure the cumulative magnetic field (closed loop integral B dot dL) along a closed path that wraps around a group of current carrying conductors, and use that to infer the current that causes this magnetic field. An application of Ampere's law. For a group of conductors that carry all phases, the magneto-motive force around the closed path adds up to zero, which is what you want it to be for a ferrous conduit. You want to avoid inducing magnetic fields in the conduit, especially those of AC that will cause heat generation as a consequence of a time-varying magnetic field.

 

[retirede]

I point that I don’t believe has been mentioned yet:
The OP asked the question in the context of parallel feeds. All of the good info provided applies any time phases are isolated - not just in the case of paralleled conductors.

 

[don_resqcapt19]

Hmm. I'd expect an increase in reactive impedance, but not an increase in losses...unless there is significant circulating current in the EGCs.

I've been thinking more on the whole CCC thing, and I am going to have to change my position.

As I understand the reasoning behind not counting a 'neutral' as a CCC, what this means is that in a set of M conductors (including the neutral), the maximum heating case is M-1 conductors carrying full current. Say you have ABCN, which A, B, and C each carrying 20A; the neutral carries 0. Turn off the current in C, and now the neutral is carrying 20A; still 3 conductors carrying 20A. Of course things are probably not perfectly balanced, with all 4 conductors carrying current; (Say A 20, B 20, C 10, and N 10) but in this case the net heating is lower than the balanced full current case. So it is reasonable to count ABCN as _three_ CCC since only 3 of the 4 will carrying current in the worst case heating scenario.

In the isolated phase situation (if properly installed!), if any of the N conductors are carrying current, they _all_ are carrying the same current. If (because of loading) the neutral happens to carry full current, then you have all the conductors in that conduit carrying full current at the same time.

So going back to my point about the underground duct bank; when you are considering the heat production of _all_ of the conductors taken together, the neutral shouldn't count as a CCC.

But when you are considering the individual conduits you have to consider only the conductors in that conduit. And here the N conductors should be counted as full CCC.

Jon

I don't understand how the impedance can be higher without an increase in losses?

 

[wwhitney]

I don't understand how the impedance can be higher without an increase in losses?

What kind of losses?

If you are just referring to I²R losses, that depends only on R, not Z, to my understanding. So if the inductance/capacitance changes, but R remains the same, you have the same I²R loss (or less if I ends up going down due to a higher Z).

But if by losses you mean voltage drop, then that does depend on Z, so you would get more voltage drop if Z gets larger.

Cheers, Wayne

 

[wwhitney]

If you are just referring to I²R losses, that depends only on R, not Z, to my understanding. So if the inductance/capacitance changes, but R remains the same, you have the same I²R loss (or less if I ends up going down due to a higher Z).

Although if the load is a fixed power load, the increased Z will lower V and hence increase I, increasing I²R loss.

Cheers, Wayne

 

[don_resqcapt19]

What kind of losses?

If you are just referring to I²R losses, that depends only on R, not Z, to my understanding. So if the inductance/capacitance changes, but R remains the same, you have the same I²R loss (or less if I ends up going down due to a higher Z).

But if by losses you mean voltage drop, then that does depend on Z, so you would get more voltage drop if Z gets larger.

Cheers, Wayne

And voltage drop are losses that cost the owner money if the installation is on the load side of the meter.