Announcement

Collapse
No announcement yet.

Strategy for sizing the neutral wire in a PV system

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Strategy for sizing the neutral wire in a PV system

    What is a good strategy for sizing the neutral wire in a PV system? Especially when the system size and feeder length require significantly large sizes.

    As an example, consider 160A worth of inverter (200A of breaker) at 277/480V. Due to distance, I have to upsize from #3/0 to 500 kcmil on the phase conductors, and correspondingly ground from #6 to #1. I know that neutral cannot be smaller than ground.

    Does it do anything for me, if I increase my neutral to 500 kcmil?
    Or would it be better if I left my neutral at #3/0?

    Out of ignorance, I usually make it equal to the phase/line wires. But is this overkill?

    What if I have a lot of voltage drop and distance to accommodate? Voltage isn't going to drop through the neutral, when you have a balanced inverter setup. It's going to drop through the phase/line wires.

    How much current does a neutral wire carry, when the inverter current is balanced across all the phase/line conductors?

  • #2
    ...Out of ignorance, I usually make it equal to the phase/line wires. But is this overkill?

    What if I have a lot of voltage drop and distance to accommodate? Voltage isn't going to drop through the neutral, when you have a balanced inverter setup. It's going to drop through the phase/line wires.....
    Depends on your loads. A single house is much harder to balance loads, if you turn on a toaster oven, how likely is it going to be balanced out by a hair dryer? Don't put the toaster and microwave on the same phase !
    So, I'd be inclined to play it safe, and keep the neutral that same size as the 2 hot lines (L1 & L2) to allow for un-even loads. And your inverter / transformer needs to accommodate a full load on L1, with nothing on L2.

    A big office building will be much more balanced because of so many random loads on different phases.

    How much current does a neutral wire carry, when the inverter current is balanced across all the phase/line conductors?
    None. But that is highly unlikely. A load on a single phase will see voltage drop of round trip, L1 going there, and N coming back. So you have to upsize your wires.
    Last edited by Mike90250; 01-13-2015, 01:32 AM.
    Powerfab top of pole PV mount (2) | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister

    Comment


    • #3
      Originally posted by Mike90250 View Post
      Depends on your loads. A single house is much harder to balance loads, if you turn on a toaster oven, how likely is it going to be balanced out by a hair dryer? Don't put the toaster and microwave on the same phase !
      So, I'd be inclined to play it safe, and keep the neutral that same size as the 2 hot lines (L1 & L2) to allow for un-even loads. And your inverter / transformer needs to accommodate a full load on L1, with nothing on L2.

      A big office building will be much more balanced because of so many random loads on different phases.
      I'm specifically talking about grid-tied solar interconnection circuits from the inverter to the main panel. I don't think I'm seeing how it depends on loads, given that the solar interconnect wiring is a completely different branch circuit.

      Is this because one leg might be 110 Volts to neutral, while the other leg is 125 Volts to neutral, due to imbalanced loads?


      Originally posted by Mike90250 View Post
      None. But that is highly unlikely. A load on a single phase will see voltage drop of round trip, L1 going there, and N coming back. So you have to upsize your wires.
      My inverters are 240V, and therefore connected from L1 to L2. The current travels on L1 going out, and back in on L2. L1 and L2 make the circuit loop, and the neutral is really just there for "effective grounding", harmonics, and voltage measurement purposes.

      Comment


      • #4
        Originally posted by carultch View Post
        I'm specifically talking about grid-tied solar interconnection circuits from the inverter to the main panel. I don't think I'm seeing how it depends on loads, given that the solar interconnect wiring is a completely different branch circuit.
        Yes and no...

        Because you have the grid tie, it's mostly independent. You can send a lot of current out on one leg to the grid, while the other leg is being heavily used by the toaster oven, hair dryer, and window-AC unit.
        If you didn't have the grid providing that effectively infinite spot to send power, you'd definitely need to plan on having a completely unbalanced power draw on just one leg.
        But since it is grid-tied, I think in theory you're right - that it should be pretty well balanced and there shouldn't be much current on that neutral.

        Frankly, how to size it correctly is out of my area of expertise - but I'd probably go with same size as the L1 and L2, or see if the inverter manufacturer can provide guidance.

        Comment


        • #5
          Originally posted by foo1bar View Post
          Yes and no...

          Because you have the grid tie, it's mostly independent. You can send a lot of current out on one leg to the grid, while the other leg is being heavily used by the toaster oven, hair dryer, and window-AC unit.
          If you didn't have the grid providing that effectively infinite spot to send power, you'd definitely need to plan on having a completely unbalanced power draw on just one leg.
          But since it is grid-tied, I think in theory you're right - that it should be pretty well balanced and there shouldn't be much current on that neutral.

          Frankly, how to size it correctly is out of my area of expertise - but I'd probably go with same size as the L1 and L2, or see if the inverter manufacturer can provide guidance.
          I've never seen a single ampere of neutral current on our monitoring systems. Except those that monitor building load, where I'd expect neutral current.

          Most inverter manufacturers seem to match their neutral lugs to the phase terminal lugs.

          I feel like it is overkill to match the neutral with the phase wiring. When am I ever going to see anywhere near 200A on the neutral? The most I would ever imagine in this case, would be about 20A...if that.

          Comment


          • #6
            My bad, I didn't spot this was in a grid tie forum section.
            So, with the GT system, there is virtually no current in the neutral, as long as the grid is healthy. When the grid gets unhealthy, the inverter shuts off, most likely before a breaker can trip.
            Powerfab top of pole PV mount (2) | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
            || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
            || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

            solar: http://tinyurl.com/LMR-Solar
            gen: http://tinyurl.com/LMR-Lister

            Comment


            • #7
              Originally posted by Mike90250 View Post
              My bad, I didn't spot this was in a grid tie forum section.
              So, with the GT system, there is virtually no current in the neutral, as long as the grid is healthy. When the grid gets unhealthy, the inverter shuts off, most likely before a breaker can trip.
              Mike

              You may be correct that if all is balanced there would usually be no neutral current but in my experience if there is a harmonic generating device (like the inverter) there is a pretty good chance of some level of harmonic current distortion (THDi) on the system which could cause the current not to cancel out but become additive on the neutral wire. You can perform a quick check using a DC clamp on ammeter but if the system is dynamic you may not see it during a single test.

              I hate to have someone spend the extra money on bigger wire but to be safe I would size the neutral wire that same as the L1 and L2 wires. And for a 3 phase system maybe even increase the neutral wire due to the potential 3rd return current being added. A lot of places uses a neutral wire twice as large as the other 3 phase wires between the Wye secondary of a transformer and panel.

              ***** So forget what I mentioned above that is underlined. I just went back and checked some drawings and pictures of a 1MW PV system I had visited and the Neutral wire was smaller than the L wires. So the OP might get away with that #3/0 in his system if it meets the local code requirements.
              Last edited by SunEagle; 01-13-2015, 11:55 AM. Reason: added last sentence

              Comment


              • #8
                Thanks for the help. You seem to understand the question and the issue the most.

                Originally posted by SunEagle View Post
                There is a pretty good chance of some level of harmonic current distortion (THDi) on the system which could cause the current not to cancel out but become additive on the neutral wire.
                Understood.

                Most inverter datasheets specify <3% THD. Assuming worst case that this all adds up among the three phases, I wouldn't expect more than 9% of the phase current on the neutral. Given 160 continuous amps on the phase conductors, should I expect not more than 14.4A of 180 Hz harmonics on the neutral? Am I interpreting what "total harmonic distortion" means correctly? 14 Amps is nothing compared to 160 Amps.

                Originally posted by SunEagle View Post
                I hate to have someone spend the extra money on bigger wire but to be safe I would size the neutral wire that same as the L1 and L2 wires.
                That's what I do out of ignorance. And I feel it is overkill. I'm trying to understand if I should be doing a better job of value engineering my feeders, particularly the long ones.

                In a situation where wiring is local and voltage drop doesn't matter, I'll make all wires the same size (excluding EGC & GEC).

                What about in a situation where voltage drop is significant? The local size of the line wires would be #3/0, but suppose they have to be increased to 500. Does my neutral also have to increase in size? It's not going to melt or burn the insulation, if it remains a #3/0. Worst case is that it would nuisance trip and inverter voltage relay.

                Originally posted by SunEagle View Post
                And for a 3 phase system maybe even increase the neutral wire due to the potential 3rd return current being added. A lot of places uses a neutral wire twice as large as the other 3 phase wires between the Wye secondary of a transformer and panel.
                Anyway to quantify whether I'd need to do this?
                Should I be counting a neutral as a CCC, for bundling ampacity adjustments?

                Originally posted by SunEagle View Post
                I just went back and checked some drawings and pictures of a 1MW PV system I had visited and the Neutral wire was smaller than the L wires. So the OP might get away with that #3/0 in his system if it meets the local code requirements.
                Do you have an example you can share with me? You can hide the private info on the drawing as needed.
                I'd like to see the conclusion for wire sizes, given the input data, so that I can understand what they did.

                The NEC just contains recommendations, rather than requirements on voltage drop. And there isn't an equation to determine neutral voltage drop.

                What local requirements should I expect?

                Comment


                • #9
                  Originally posted by carultch View Post
                  Thanks for the help. You seem to understand the question and the issue the most.



                  Understood.

                  Most inverter datasheets specify <3% THD. Assuming worst case that this all adds up among the three phases, I wouldn't expect more than 9% of the phase current on the neutral. Given 160 continuous amps on the phase conductors, should I expect not more than 14.4A of 180 Hz harmonics on the neutral? Am I interpreting what "total harmonic distortion" means correctly? 14 Amps is nothing compared to 160 Amps.

                  No your interpretation is incorrect. The THD or total harmonic distortion for either Voltage or Current is the ratio of the root-sum-square value of the magnitude of each harmonic (2nd to 50th) divided by the root-mean-square of the magnitude of the fundamental harmonic (1st). Even in the worst case on a 3 phase system the Neutral would not see 3 times the RMS current on the Load conductors. So for an inverter with a < 3% THD rating the Neutral would only see <3% of that 160amps or < 4.8amps.

                  That's what I do out of ignorance. And I feel it is overkill. I'm trying to understand if I should be doing a better job of value engineering my feeders, particularly the long ones.

                  In a situation where wiring is local and voltage drop doesn't matter, I'll make all wires the same size (excluding EGC & GEC).

                  What about in a situation where voltage drop is significant? The local size of the line wires would be #3/0, but suppose they have to be increased to 500. Does my neutral also have to increase in size? It's not going to melt or burn the insulation, if it remains a #3/0. Worst case is that it would nuisance trip and inverter voltage relay.

                  I went over the NEC and like you could not find any particular statement concerning the size of the Neutral wire when the CCC are required to be increased due to voltage drop. Now this may be the case for a 3 phase system with only 3 phase loads but with any home circuit you have multiple 1 phase loads so IMO the Neutral wire should be sized the same as the L1 & L2 wire because of the voltage drop is for the complete round trip circuit which includes the Neutral wire. The example I mentioned in my previous post was for wires between an inverter and Y transformer for 3 phase system. I am still not sure why the Neutral is smaller than the other 3 phase conductors but it may due to the Load (inverter) being 3 phase. I am still looking into this a little more.


                  Anyway to quantify whether I'd need to do this?
                  Should I be counting a neutral as a CCC, for bundling ampacity adjustments?

                  No the neutral wire is not considered to be a CCC which does not require you to de-rate all the wires because you are not exceeding the Three CCC in a raceway



                  Do you have an example you can share with me? You can hide the private info on the drawing as needed.
                  I'd like to see the conclusion for wire sizes, given the input data, so that I can understand what they did.

                  The example I mentioned was designed by another outfit so I am not sure how they did their wire size calculation. I got involved with the project due to the inverters shutting down multiple times a day so I was given a copy of the wiring diagram.


                  The NEC just contains recommendations, rather than requirements on voltage drop. And there isn't an equation to determine neutral voltage drop. To be honest I can't find an equation to determine the neutral voltage drop but I would say that under section 210.19 for Branch Circuits all of the conductors need to be sized to prevent a VD exceeding 3%. So to be safe I would size your neutral the same as the phase wires.

                  What local requirements should I expect? The Local requirements usually involve your City and State codes. While the US follows the NEC everyone does not follow the latest revision and some places actually have a more stringent code to follow. You need to check with someone in your area to see what codes must be met.
                  See my answers above in Bold.

                  I hope I haven't caused too much confusion with my posts. I certainly need to do more research on the matter.

                  Comment


                  • #10
                    The NEC allows you to use a smaller neutral compared to the phase conductors as long as you can justify it with a load calculation.
                    For a typical 120/240V three wire (120-0-120) system the neutral wire needs to be sized the same as the phase wires because you might connect and turn on a set of loads that all ran from the same phase wire. But if you have mostly or only 240v loads, you can reduce the size of the neutral. If you have only 240V loads you can even not run a neutral at all.
                    Most GTIs for connection to a 120/240 system only output current on the phase wires. The connection to the neutral is there only for voltage balance monitoring as required for UL approval. So you could justify using only the minimum wire size (i.e. 14ga) for the neutral. Whether your inspector would allow that without an electrical contractor or PE providing documentation to back that up is the real question.
                    In the case of three phase power, e.g. 208Y/120, the typical inverter or micro inverter will connect to two of the three phase wires and again will not put any current on the neutral. When you use several inverters to try to balance the load among the three phase wires, you still will not be putting any current into the neutral. The in and out on the two phase wires will balance out for each inverter.
                    If you somehow end up with inverters (such as some hybrid inverters) which provide a 120V output, then you will have neutral current to worry about.
                    SunnyBoy 3000 US, 18 BP Solar 175B panels.

                    Comment


                    • #11
                      Suneagle, Inetdog, Mike, and Fubar...

                      Thank you all very much for discussing this concept with me, and helping me to understand this concept. I'm still not at a good conclusion, but maybe I will be someday.

                      Comment


                      • #12
                        Originally posted by SunEagle View Post
                        See my answers above in Bold.

                        I hope I haven't caused too much confusion with my posts. I certainly need to do more research on the matter.
                        No problem. I understand most of your responses to the extent I was expecting to understand them.


                        Originally posted by SunEagle View Post
                        Anyway to quantify whether I'd need to do this?
                        Should I be counting a neutral as a CCC, for bundling ampacity adjustments?

                        No the neutral wire is not considered to be a CCC which does not require you to de-rate all the wires because you are not exceeding the Three CCC in a raceway.
                        So in the case of a 3-phase/4-wire circuit for a 3-phase PV inverter, the 4th wire is not counted as a CCC for bundling ampacity adjustments? Am I understanding your response correctly? The neutral simply wouldn't count, because its current is negligible compared to the others.

                        My scope of work is not limited to single phase, and I hope you weren't assuming that.

                        I understand that when the neutral is an integral part of forming the round trip path, that it does count as a CCC. One example would be line-to-neutral circuits, which have no other option than for current to flow through neutral. In that case, I have no reason to try to value engineer its size.


                        Originally posted by SunEagle View Post
                        The example I mentioned in my previous post was for wires between an inverter and Y transformer for 3 phase system. I am still not sure why the Neutral is smaller than the other 3 phase conductors but it may due to the Load (inverter) being 3 phase. I am still looking into this a little more.
                        Single phase inverters on a 277/480V wye system are connected line to neutral. The aggergate of a group of them might not perform uniformly. It could be that all the A-phase inverters are at full power, while the B and C phase inverters are sourced by a shaded array. And thus the neutral should be expected to carry a significant amount of current.

                        For three phase inverters on any wye or delta system, it usually is harmonics on the neutral, and that's about it. Transformerless inverters usually can't do this, since they only use neutral for voltage measurement.

                        Some are built in a manner where the interior of the inverter is an inverting unit for each phase, such as the Fronius IG Plus inverters. They have a mix mode setting that allows current imbalance to improve performance, in contrast from their balance mode. Fronius IG Plus delta 208V inverters wouldn't put the imbalance on the neutral. The WYE 277 version in this family, by contrast would put the imbalance on the neutral.

                        Comment


                        • #13
                          Neutral = L1 = L2 for single phase system. If you have sized your system for 100 amps, all three are th esized for 100 amps at whatever distance you up sized.
                          MSEE, PE

                          Comment


                          • #14
                            Originally posted by Sunking View Post
                            Neutral = L1 = L2 for single phase system. If you have sized your system for 100 amps, all three are th esized for 100 amps at whatever distance you up sized.
                            I thought single phase Line-to-Line 240 volt inverters put all their current on the black and red wires.

                            When 100A travels outward on the black wire, that same 100A travels simultaneously inward on the red wire. At any cross section, the current adds up to zero from those two wires alone.

                            When does it ever travel on the neutral?

                            Comment


                            • #15
                              Originally posted by carultch View Post
                              I thought single phase Line-to-Line 240 volt inverters put all their current on the black and red wires.

                              When 100A travels outward on the black wire, that same 100A travels simultaneously inward on the red wire. At any cross section, the current adds up to zero from those two wires alone.


                              When does it ever travel on the neutral?
                              When you use a 120V appliance in north america ( "split phase" household system. )
                              Powerfab top of pole PV mount (2) | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
                              || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
                              || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

                              solar: http://tinyurl.com/LMR-Solar
                              gen: http://tinyurl.com/LMR-Lister

                              Comment

                              Working...
                              X