Combined Ground and Neutral

Hey Installers,
I have a question that maybe be remedial. I often see ground and neutral on the same (and only) bus bar in a main panel. An electrician once told me that was fine...as far as code is concerned. I guess I'm wondering what you all do about this situation, especially in regards to NEC 250.24(C) requirements for bonding neutral and GEC in service disconnects. Should I be installing a second bus bar and separating Ground from Neutral and installing a bonding jumper, or does having both G and N installed on the same bar count as that bond? Thanks for any thoughts/help.
p.s.
I am on the 2008 code cycle..if that matters at all.

Comments

  • Ethan Brush
    Ethan Brush Solar Expert Posts: 235 ✭✭
    Aenergies wrote: »
    Hey Installers,
    I have a question that maybe be remedial. I often see ground and neutral on the same (and only) bus bar in a main panel. An electrician once told me that was fine...as far as code is concerned. I guess I'm wondering what you all do about this situation, especially in regards to NEC 250.24(C) requirements for bonding neutral and GEC in service disconnects. Should I be installing a second bus bar and separating Ground from Neutral and installing a bonding jumper, or does having both G and N installed on the same bar count as that bond? Thanks for any thoughts/help.
    p.s.
    I am on the 2008 code cycle..if that matters at all.

    The short answer is yes it is fine to land the neutrals and grounds to the same bus bar in a service disconnect or at the first disconnecting means of a separately derived system. If you really look at the wording though you will notice that the "correct" way is to have a separate neutral and ground bar and connect them with a main bonding jumper. However the former method is near completely universally accepted.
  • SolarPowered
    SolarPowered Solar Expert Posts: 626 ✭✭✭
    Neutral and ground are redundant, makes the system safer. Neutral is a reference wire when ground can't be referenced. Its good to maintain seperation for this reference. Neutrals on one side of the grounding bus, Ground conductors on the other side of the grounding bus, if there isn't a secondary grounding bus to make this seperation.
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Neutral and ground are redundant, makes the system safer. Neutral is a reference wire when ground can't be referenced. Its good to maintain seperation for this reference. Neutrals on one side of the grounding bus, Ground conductors on the other side of the grounding bus, if there isn't a secondary grounding bus to make this seperation.

    You can also usually put two bare copper ground (EGC) wires into one screw terminal on a ground bus but are only allowed one neutral under a single screw on the neutral bus. So the ground bus can be smaller.

    The whole question of what to do about bonding in a supply side PV disconnect keeps going back and forth. The NEC is very clear that this is not a service disconnect, but safety considerations tend to argue that you should make a ground to neutral bond at the PV disconnect anyway.
    Once you have passed the main disconnect, subpanels and other equipment must keep the two sets of wires separate and insulated from each other.

    The big difference between neutral and ground (more properly Equipment Grounding Conductor or EGC) is that under normal circumstances the neutral will be carrying load current while the EGC will only carry current in the event of a fault from a hot conductor to ground.
    SMA SB 3000, old BP panels.
  • solar_dave
    solar_dave Solar Expert Posts: 2,397 ✭✭✭✭
    inetdog wrote: »

    You can also usually put two bare copper ground (EGC) wires into one screw terminal on a ground bus but are only allowed one neutral under a single screw on the neutral bus. So the ground bus can be smaller.

    The whole question of what to do about bonding in a supply side PV disconnect keeps going back and forth. The NEC is very clear that this is not a service disconnect, but safety considerations tend to argue that you should make a ground to neutral bond at the PV disconnect anyway.
    Once you have passed the main disconnect, subpanels and other equipment must keep the two sets of wires separate and insulated from each other.

    The big difference between neutral and ground (more properly Equipment Grounding Conductor or EGC) is that under normal circumstances the neutral will be carrying load current while the EGC will only carry current in the event of a fault from a hot conductor to ground.

    Two ground on the same screw would not pass the inspector here, I was required to get a longer ground buss bar.
  • SolarPowered
    SolarPowered Solar Expert Posts: 626 ✭✭✭
    The 2 bonds to EGC ground is controversial mostly because inspectors incorrectly interpret the language.

    When EGC grounding is a requirement for solar racking you can technically bond up to 2 EGC together.

    The whole point is to meet the requirements of NEC 2014 rules on rapid shut down.

    If you read the requirements of circuit isolation for rapid shut down EGC bonding isn't a requirement for micro inverter systems, since micro inverters meet I.G compliance not requiring a GEC since the DC side circuits are isolated at the point of micro inverters, and micro inverters meet all requirements for rapid shut down.

    To say the least some jurisdictions enforce EGC rack ground bonding as the excuse is for lightning strikes, which that arguement is void. If lightning struck a solar array the frequency that lightning travels would melt down most if not all conductors including GEC, EGC.

    The point of proper EGC installation is to maintain circuit isolation at potential hazardous points of disconnecting means.
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    solar_dave wrote: »

    Two ground on the same screw would not pass the inspector here, I was required to get a longer ground buss bar.

    Sounds like a local ammendment or shirt pocket rule making.
    How many wires are allowed under one screw (and what sizes) is a matter of the UL listing conditions of the particular bar. Most allow for two.
    But the NEC specifically adds a one conductor per screw condition for the neutral bar only. I guess they worry about inadvertently interrupting the neutral continuity for a circuit other than the one you are working on (which can have really serious consequences for a Muliti Wire Branch Circuit (MWBC), also called an Edison circuit).
    SMA SB 3000, old BP panels.
  • Ethan Brush
    Ethan Brush Solar Expert Posts: 235 ✭✭
    inetdog wrote: »

    You can also usually put two bare copper ground (EGC) wires into one screw terminal on a ground bus but are only allowed one neutral under a single screw on the neutral bus. So the ground bus can be smaller.

    I agree. The NEC specifically states only one grounded conductor (neutral) per hole. The number of EGC's is a UL listing issue and then NEC 110.3(B) would require you to follow the listing and instructions.
    The whole question of what to do about bonding in a supply side PV disconnect keeps going back and forth. The NEC is very clear that this is not a service disconnect, but safety considerations tend to argue that you should make a ground to neutral bond at the PV disconnect anyway.

    I disagree, in fact a careful look at the NEC shows that it is definitely a service disconnect and requires a main bonding jumper. I left my codebook at work so I cant give specific references right now beyond what I remember, but look at 230.40 Exception 5 which states that states that an additional set of service-entrance conductors connected to the supply side of the normal service disconnecting means can supply Solar PV systems. Then look at 230.71, and the definitions and it will be clear. I actually really hate that the NEC uses the term "supply side connection" and think they should eliminate it. All it does is confuse things. If you think about it, a "line side tap connection" is really just another set of service entrance conductors with another service disconnecting means and this installation is already allowed and very common. The only thing the current wording gives you is the allowance to have a 7th service disconnect (6 is normally the max) and the allowance to not have to group the "PV disconnect" with the other "normal" disconnects. So they should just delete the supply side connection wording, and add another exception to the disconnect grouping requirement (230.71) and it would really make everything so much simpler. This silly supply side connection thing confuses people in terms of bonding and if its a service disconnects and all sorts of semantic questions about when it is a "PV supply side connection" or when it is just another set of service entrance conductors. Further we run into utilities doing silly things lot not allowing a class 320 meter socket to be used for a supply side connection because they dont want "taps" in their meter socket.

    To say the least some jurisdictions enforce EGC rack ground bonding as the excuse is for lightning strikes, which that arguement is void. If lightning struck a solar array the frequency that lightning travels would melt down most if not all conductors including GEC, EGC.

    I am not exactly sure what you mean by "EGC rack ground bonding". Bonding metallic components of PV systems allows GFP systems to work and have better "coverage", will lower the potential difference if there are two faults on different system conductors, and provides an equipotential plane from effects such as capacitance and high impedance faults. I dont see how bonding the racking to dirt does much other than continue the equipotential plane to encompass the earth (which at the very low currents required to shock a human is valid and can work despite the very high resistance of dirt).
    The point of proper EGC installation is to maintain circuit isolation at potential hazardous points of disconnecting means.

    I dont understand what you are saying, can you elaborate?
  • verdigo
    verdigo Solar Expert Posts: 428 ✭✭
    I don't mean to hijack the thread but it is relevant to my system. My house is old and the original load center is one where the grounds and neutrals are all connected to the same bus. I am still in the throes of the inspection process and the inspector seems to want disconnects everywhere, not to mention a couple of sub panels. Am I right in my understanding that this single bus bar that has grounds and neutrals should be the only neutral ground bond in the system?
  • SolarPowered
    SolarPowered Solar Expert Posts: 626 ✭✭✭


    I am not exactly sure what you mean by "EGC rack ground bonding". Bonding metallic components of PV systems allows GFP systems to work and have better "coverage", will lower the potential difference if there are two faults on different system conductors, and provides an equipotential plane from effects such as capacitance and high impedance faults. I dont see how bonding the racking to dirt does much other than continue the equipotential plane to encompass the earth (which at the very low currents required to shock a human is valid and can work despite the very high resistance of dirt).
    I dont understand what you are saying, can you elaborate?


    Ahh yes the many controversies that plague contractors, and inspectors alike....... The revisions/amendments between 2008, 2011, and 2014 NEC.......

    Bonding metallic components of pv systems does allow for a good GFP. I agree with that according to NEC 690.43 Exposed non-current carrying metal parts of PV module frames, electrical equipment, and conductor enclosures shall be grounded in accordance with 250.134 or 250.136(A), regardless of voltage

    The purpose of GFP systems is to have a means of safety protection and ground fault for mechanical systems that have the potential to "float", or are "floating".
    Now I don't want to run on all day on NEC code, so I will send the links to the enphase micro inverter literature on rapid shut down, so you can get an understanding.
    Code has evolved significantly between 2008 and 2014, most jurisdictions don't recognize 2008 anymore.

    Micro inverters meet NEC requirement 690.35

    Listed in Enphase spec sheet: The DC circuit meets the requirements for ungrounded PV arrays in NEC 690.35. Equipment ground is provided in the Engage Cable. No additional GEC or ground (EGC) is required. Ground fault protection (GFP) is integrated into the microinverter.

    From the body of the micro inverter, which meets this requirement, a single rack which mounts the panel and inverter in one plane are all integrated, (as long as using UL listed I.G parts) DC circuit isolation is met at the point of micro inverter and the GFP, EGC trunk cable and micro inverter internals and body meet the requirement of NEC 690.43 (as long as the components are I.G), and is identified in NEC 690.35. This makes NEC 250.134 and/or NEC 250.136 redundant and hypocritical of NEC 690.35, and to refer back to NEC 690.43. The code was writen to allow either or option, not to apply the contingency of all the above, thats how I read the language, and most inpsectors have agreed to this, but still have me install on some not all projects the solid egc bare to racking incase of lightning strike, which really pertains to nothing of NEC 250.134, or NEC 250.136, of NEC 690.43.

    Since micro inverters meet all the requirements for rapid shut down, disconnects are not a requirement. All functioning parts of the system from both DC and AC side are isolated, and EGC bonded.


    http://enphase.com/global/files/Enph...Compliance.pdf

  • Ethan Brush
    Ethan Brush Solar Expert Posts: 235 ✭✭
    verdigo wrote: »
    I don't mean to hijack the thread but it is relevant to my system. My house is old and the original load center is one where the grounds and neutrals are all connected to the same bus. I am still in the throes of the inspection process and the inspector seems to want disconnects everywhere, not to mention a couple of sub panels. Am I right in my understanding that this single bus bar that has grounds and neutrals should be the only neutral ground bond in the system?

    Maybe. From your signature I assume you are grid tied? The 120/240VAC system is required to be grounded so yes you would have a N-G bond in the AC panel fed by your inverter and the grounding elecrode conductor(s) would land there too. Many people ground the negative battery terminal too. I personally dont recommend it - it may be against code, cause problems with the GFP of the CC, and IMO it makes the DC system more dangerous. Others will disagree. Your shouldnt need a ton of disconnects - Im not sure where he is requiring them, although they are nice for servicing and troubleshooting. Also what do you mean by he wants sub panels?
  • Ethan Brush
    Ethan Brush Solar Expert Posts: 235 ✭✭


    Ahh yes the many controversies that plague contractors, and inspectors alike....... The revisions/amendments between 2008, 2011, and 2014 NEC.......

    Bonding metallic components of pv systems does allow for a good GFP. I agree with that according to NEC 690.43 Exposed non-current carrying metal parts of PV module frames, electrical equipment, and conductor enclosures shall be grounded in accordance with 250.134 or 250.136(A), regardless of voltage

    The purpose of GFP systems is to have a means of safety protection and ground fault for mechanical systems that have the potential to "float", or are "floating".
    Now I don't want to run on all day on NEC code, so I will send the links to the enphase micro inverter literature on rapid shut down, so you can get an understanding.
    Code has evolved significantly between 2008 and 2014, most jurisdictions don't recognize 2008 anymore.

    Micro inverters meet NEC requirement 690.35

    Listed in Enphase spec sheet: The DC circuit meets the requirements for ungrounded PV arrays in NEC 690.35. Equipment ground is provided in the Engage Cable. No additional GEC or ground (EGC) is required. Ground fault protection (GFP) is integrated into the microinverter.

    From the body of the micro inverter, which meets this requirement, a single rack which mounts the panel and inverter in one plane are all integrated, (as long as using UL listed I.G parts) DC circuit isolation is met at the point of micro inverter and the GFP, EGC trunk cable and micro inverter internals and body meet the requirement of NEC 690.43 (as long as the components are I.G), and is identified in NEC 690.35. This makes NEC 250.134 and/or NEC 250.136 redundant and hypocritical of NEC 690.35, and to refer back to NEC 690.43. The code was writen to allow either or option, not to apply the contingency of all the above, thats how I read the language, and most inpsectors have agreed to this, but still have me install on some not all projects the solid egc bare to racking incase of lightning strike, which really pertains to nothing of NEC 250.134, or NEC 250.136, of NEC 690.43.

    Since micro inverters meet all the requirements for rapid shut down, disconnects are not a requirement. All functioning parts of the system from both DC and AC side are isolated, and EGC bonded.


    http://enphase.com/global/files/Enph...Compliance.pdf

    I dont disagree with you on the theory and codes of microinverters. I think you are confusing and mixing up some terms though.
    If you read the requirements of circuit isolation for rapid shut down EGC bonding isn't a requirement for micro inverter systems, since micro inverters meet I.G compliance not requiring a GEC since the DC side circuits are isolated at the point of micro inverters, and micro inverters meet all requirements for rapid shut down.

    I dont understand what you mean by "EGC bonding" and how rapid shutdown has anything to do with that or isolation between the DC and AC sides. It is the anti islanding protocols built into a GTI that make it shut down if the grid goes down and whether this can also meet the 690.12 requirements just depends on the location of the inverter. Put any inverter whether it be transformer based, transformerless, grounded, ungrounded right next to the array and it too will meet the rapid shutdown requirements. An isolated inverter can operate the array grounded or ungrounded. A non isolated inverter can only operate the array as ungrounded because grounding one of the PV circuit conductors would cause a short every half cycle between that ground and the AC system ground (so it really IS grounded through the inverter, its just not grounded again).
    To say the least some jurisdictions enforce EGC rack ground bonding as the excuse is for lightning strikes, which that arguement is void. If lightning struck a solar array the frequency that lightning travels would melt down most if not all conductors including GEC, EGC.

    I agree and dont see how this would offer any protection from lightning. It is almost comical to imagine a lightning strike hitting your racking and thinking that that ground wire will do anything - News flash you have already been hit. actual lighting protection would have air terminals ABOVE the array to create a zone of protection underneath it. It also worth noting that almost every large flat roofed building as many roof top HVAC units and they never are bonded directly to a ground rod.
  • Legacy928
    Legacy928 Registered Users Posts: 1
    Per National Electric Code (NEC), neutrals can only be grounded at the main panel. Grounding Neutrals in sub-panels is not permitted as it can cause premature failure of ground-fault (GFCI) receptacles and longer response time for GFCI receptacles to trip. 
    Solar may be different (being Direct Current) but once the service has been converted to 120V AC, neutrals are not permitted to be grounded on 120V systems. Any evidence to the contrary would be appreciated.   
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    edited May 2022 #14
    Welcome to the forum Legacy928.

    I am the first to admit I am not a code person... But I have some questions about your terms.

    In North America, we have 120/240 VAC split phase 60 Hz power, in general, to residential buildings.

    That is, more or less, L1 + L2 (the two "outer", transformer taps) of a pole mounted transformer, and a center tap. We get 240 VAC between L1 and L2, and From L1 to CT or L2 to CT, we get 120 VAC.

    Inside the home, there should be one CT to Earth Ground bond, typically in the main electrical panel. Which makes the CT the "ground bonded" Neutral lead and we have L1/L2/Neutral.

    Since, by definition, all 120 VAC inside the home is the result of L1+N or L2+N branch circuits--Then all 120 VAC have ground bonded Neutral plus a "power/hot" lead.

    Yes, you have "issues" if there are multiple N+G bonds in the home. One inside the Main panel, a second at a subpanel... Then you have parallel current paths. Part of the current goes through the White/Neutral cable (as designed). And another part of the current flows through the Greenwire/conduit/bare copper wire between the main and sub panel. And that is wrong for several reasons.
    • Safety Grounds (green/bare copper/metal conduit) is not supposed to carry load power
    • If you run current through two different conductors (such as L1 and Ground of some sort) through separate holes in an electrical box/panel, the two paths create eddy currents in the box sheet metal and can actually cause the box sheet metal to overheat (transformer like coupling from L1/G to box sheet metal). For AC power, all L1/L2/N power leads, per circuit must enter/exit the same single hole punched in the panel side.
    And there is the other issue... GFI Breakers and Outlets "measure the current" passing through their Lx/N leads. The total current should add up to zero amps (i.e., as AC current "goes out" L1, the AC return current most flow back on N). 

    If you have L1 current going into the GFI Current loop, and some current going through N, and other current flowing through parallel ground current path  outside the current sense loop in the GFI (say N+G bond in panel => GFI => N+G bond at load) this "looks like" an electrical fault... Say a person getting electrocuted (10-25+milliAmps) by grabbing L1 and stepping in a puddle. And trips the GFI.

    Hence the reason for the "single" N+G bond in the main panel, and no other N+G bonds elsewhere in the home (no subpanels, no loads, etc.).

    Now--There is a bit of a "lie" here... If for example, you have 5 homes connected to one pole transformer. The pole transformer has N+G at the pole (local ground base of pole). And each of the homes has N+G in their main panels. So the reality, is there can be a 1/2 dozen N+G bonds on the single Center Tap/Neutral for that pole transformer.

    And things get even more complex when (for example) recreation vehicles get involved... For example an RV may have
    • Shore Power: N+G made (somewhere) at the trailer park. No N+G bonding the trailer (L1/L2/N/Ground if 120/240 VAC. Or L1/N+Ground if 120 VAC). Neutral is ground bonded at shore power. RV AC panel is "sub panel"
    • Internal AC inverter: When RV is running on internal RV AC Inverter Power. There is typically an N+G bond made--Either inside the inverter, or in the RVs AC panel (RV Panel is now "Main Panel").
    • Internal AC Genset: Again, N+G may be made at inverter, or it may be made at AC Main panel (N+G bond made at main panel) or the RV panel is a Sub Panel (N+G at genset). And if using "portable" generators or AC inverters--They may have a GFI outlet themselves. Smaller gensets and inverters may have "floating AC outlets". Larger (3,500 Watts or larger???) are typically from the factory with N+G bonding as a default.
    • Also with AC inverters--There are two major classes. There are PSW/TSW (pure/true sine wave) type that galvantically isolated AC output (floating). Those types of AC inverter will work with N+G grounding, and (typically) negative grounding at the battery bus.
    • And there are MSW (modified square/sine wave) output AC inverters--Those typically are not isolated outputs. And if you do a N+G ground bond on the MSW inverter output, and have negative grounded DC battery bus (typical for vehicles and trailers)--That creates a short across the MSW inverter and will let out the "magic smoke".
    So, depending on lots of issues--Such as where the AC Transfer switch is located. Does the inverter or genset have an internal N+G bond or not, and the N+G be lifted or grounded as needed, etc...

    And there are always the questions of protection against lightning (surge suppressors, grounding, etc.)...

    And when you get into boats--There are new levels of grounding complexity.

    https://www.kp44.org/ftp/MarineGroundingSystems.pdf

    -Bill

    PS: I should add that AC inverters and Gensets may be (typically) 120 VAC only, or 120/240 VAC split phase.
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • solar_dave
    solar_dave Solar Expert Posts: 2,397 ✭✭✭✭
    When I did my shop building the inspector required I fix the bus bar in the mini-split disconnect to have a separate screw for each ground wire (base unit and 3 heads plus the incoming wire) only one was doubled up. That and the outdated covers on the outdoor receptacles and a non green ground screw were the items I got flagged for.  Pretty damn picky I would have to say.  

    BTW the building was initially wired by a licensed electrician that I had a falling out with, he left me with a GFI circuit that would not set and claimed I must have hit the wiring with a drywall screw.  NOT, i pulled each outlet starting at the end of the circuit till i found the ground wire touching the hot on about the 3rd outlet. Simple fix but a PITA guy but he refused to fix it.