Neutral as CCC

[George Stolz]

You don't typically have high leg to neutral loads on a 4 wire delta system. The line to neutral loads are across a single transformer winding, exactly like they are on a single phase center tapped transformer. The two lines across the center tapped winding of the 4 wire delta system are on the opposite ends of a single sine wave (often called 180° our of phase) and the neutral load cancels just like on a single phase system.

I still don't get it. If I hookup a three phase load to a delta system, it behaves as a three phase load because it's a three phase source. If the B phase is part of the source, electrons will want to seek it, otherwise three phase loads would single phase. So why isn't some current on your single phase loads looking for B as much as A and C?

High leg is irrelevant; it doesn't matter which phases I use of a wye, the neutral current will equal the lines when they are equal to each other. Grow the b-phase away from the neutral point (increasing the voltage of that phase), why would it matter to the neutral current? It's still fighting to get to the third phase it doesn't have ready access to.

What am I missing?

 

[Smart $]

I still don't get it. If I hookup a three phase load to a delta system, it behaves as a three phase load because it's a three phase source. If the B phase is part of the source, electrons will want to seek it, otherwise three phase loads would single phase. So why isn't some current on your single phase loads looking for B as much as A and C?

High leg is irrelevant; it doesn't matter which phases I use of a wye, the neutral current will equal the lines when they are equal to each other. Grow the b-phase away from the neutral point (increasing the voltage of that phase), why would it matter to the neutral current? It's still fighting to get to the third phase it doesn't have ready access to.

What am I missing?

The flaw is in your conceptual thinking (i.e. theory). Current leaving A seeks paths to return to A through the wiring, loads, and connections to other voltage points, and the windings which transform the voltage. Current leaving A does not seek paths to return to either B, C, or N. Those are just connections which provide the voltage and points along the path(s) to return to A.

 

[George Stolz]

That makes no sense. I would have a one-wire "circuit". :huh:

 

[GoldDigger]

That being said, if my analogy is close to correct, why would a delta system be immune? I haven't had opportunity to witness how a delta with A and C loaded appears on a neutral, but I had always thought it did the same thing because it is a three phase transformer, not a single phase. So help me out.

Well, first you have to be thinking of a high leg delta in order to have an NEC defined neutral in the first place.
Now look hard at the A-N-C winding on the transformer.
If none of your loads connect to B, then the A-B and B-C windings will not play any role in the analysis.
For three phase loads, there will be no current in N, since the loads just connect to A, B and C.
For single phase loads on A-B and B-C there will be no current in N either. (Current from B may try to go either way around the triangle to get back to the other end of winding in question, but none of it will try to go through N.
That leaves just single phase loads on A-N-C. That looks just like a standard 120/240 single phase three wire, so any net current drawn through N will reduce the current through either A or C from the maximum value. Net heating effect from those loads will be no more than you would get from a maximum 240V load connected only to A and C.

For any given load applied to the delta, either no current flows though N at all or the net current in N only happens because you have reduced the current in either A or C by the same amount as the current through N.
QED.

 

[ActionDave]

I still don't get it. If I hookup a three phase load to a delta system, it behaves as a three phase load because it's a three phase source. If the B phase is part of the source, electrons will want to seek it, otherwise three phase loads would single phase. So why isn't some current on your single phase loads looking for B as much as A and C?

High leg is irrelevant; it doesn't matter which phases I use of a wye, the neutral current will equal the lines when they are equal to each other. Grow the b-phase away from the neutral point (increasing the voltage of that phase), why would it matter to the neutral current? It's still fighting to get to the third phase it doesn't have ready access to.

What am I missing?

The neutral on a high leg transformer meets the NEC definition of a neutral, but electrons can't read so they only care about a true neutral like the one at the centre of wye transformer.

 

[Smart $]

That makes no sense. I would have a one-wire "circuit". :huh:

All circuits are one or more loops. The simplest loop is one wire connected end to end. You won't get much use out of it until you interrupt that loop with a source and a load. The wires being interrupted by source, load, switch, or another device doesn't change the fact it is a loop.

If you used wires with all the same color insulation and didn't label or otherwise identify any connection points or wires connected thereto, you could have a one-wire circuit, per se. We just use various means of identification to track conductors and connections.

 

[George Stolz]

Current from B may try to go either way around the triangle to get back to the other end of winding in question, but none of it will try to go through N.

That makes sense. Thanks!

 

[George Stolz]

All circuits are one or more loops. The simplest loop is one wire connected end to end. You won't get much use out of it until you interrupt that loop with a source and a load. The wires being interrupted by source, load, switch, or another device doesn't change the fact it is a loop.

If you used wires with all the same color insulation and didn't label or otherwise identify any connection points or wires connected thereto, you could have a one-wire circuit, per se. We just use various means of identification to track conductors and connections.

Smart, if I have one conductor leaving A, connected to a load and returning to A, the voltage the load sees is zero. Your analogy isn't making sense to me.

 

[mjmike]

Probably the best design practice is to assume the neutral is a CCC if it is ran with the phase conductors and connected. It always has the potential to carry current. The only time it becomes an issue is with a 3-phase 4-wire circuit such as for a panel feeder. At that point, you have 4CCC so you derate by 80% utilizing the 90 degree column. The simplest way to do this, is to just write the revised amperages in the 90 degree column making your own "derated" column as follows starting with #12: 24A, 32, 44, 60, 76, 92, 104, 116, 136, 156, 180, 208, 232, 256, 280, 304, 344 (ending with 500 kcmil). So if you only have 3 CCC or less use the 60/75 degree column, if 4 CCC use the derated column you created.

You will also see, that when using the 60 degree column for circuits 100A and less (per NEC 110.14.C), the conductor size does not change when derating to 80%.

The ground size does not need to change due to a larger phase conductor for derating purposes. So for VD for a derated circuit, your starting point is the derated conductor size with the base ground then upsize from there.

 

[ActionDave]

Probably the best design practice is to assume the neutral is a CCC if it is ran with the phase conductors and connected. It always has the potential to carry current. The only time it becomes an issue is with a 3-phase 4-wire circuit such as for a panel feeder......

Why would you say this? All you end up with is bigger wire and more expense for no good reason.

 

[mjmike]

Why would you say this? All you end up with is bigger wire and more expense for no good reason.

Let me clarify, this would be based on a Y system. Not a 3-phase 4-wire delta.

 

[Smart $]

Smart, if I have one conductor leaving A, connected to a load and returning to A, the voltage the load sees is zero. Your analogy isn't making sense to me.

Your assessment is correct... but likened to the one-wire end-to-end circuit. You have to include a power source (i.e. connect to B, C, or N)... then the current can end up back at A through the winding(s).

For the following, current pathway indicated with -> is through the transformer winding between connection points.

Technically, if you connect a single load to high-leg delta transformer secondary A-N, the current path is A->load->N->A... and A->load->N->C->B->A... but only if it is the only load (or all connected loads are A-N). Once you start adding loads to other connection points, especially C-N, the current will only go A->load->N->A because current can only flow in one direction at a time on the other part and full windings.

On a wye system, the current of an A-N connected load just goes A->load->N->A, because each winding is a "branch" on its own. If you have say an A-B connected load, then it goes A->load->B->N->A.

 

[infinity]

Probably the best design practice is to assume the neutral is a CCC if it is ran with the phase conductors and connected. It always has the potential to carry current. The only time it becomes an issue is with a 3-phase 4-wire circuit such as for a panel feeder. At that point, you have 4CCC so you derate by 80% utilizing the 90 degree column.

Isn't it safe to say that in a large majority of installations the neutral in your example would not be counted as a CCC?

 

[Carultch]

Why would you say this? All you end up with is bigger wire and more expense for no good reason.

Future-proofing the feeder, in case the nature of the connected loads changes over the duration of occupancy. You might start with all linear loads today, and then ten years later they connect loads with significant harmonics, yet still the same overall KVA.

Maybe this isn't the best example, but suppose you start using all incandescent lighting today. Then the owner changes it out for fluorescent lighting, several years later. The owner takes full advantage of the now available power capacity, and connects a lot of computer and server loads.

It is unlikely that you'd start with built-in incandescent lighting today, if you know there is talk about upgrading to more energy efficient technology. That is why I say that this may not be the best example.