Ground conductor patterns

The NEC states that a continuous ground conductor has to be used to ground the panels in a string. I understand the intent is so that when disconnecting a single panel, you don't interrupt the ground to the rest of the string. I've drawn two diagrams below that may meet that requirement. Can I get some feedback as to how the following two grounding patterns are interpereted by the NEC? Can either be used?

groundoptions.jpg


Thanks,

Pete B

Comments

  • stevend
    stevend Solar Expert Posts: 34
    Re: Ground conductor patterns

    In A, the bottom row's ground wire has to be connected to the ground wire coming from the top row. The temptation would be to do that in the ground lug that's screwed to the bottom-left panel. That would violate NEC I think - although if you remove that panel you could keep the wires in the lug and remove the lug from the panel. Alternatively you'd connect them separately from the panels - I'd use a split bolt. Would both be acceptable approaches?
    -Steve
  • pbartko
    pbartko Solar Expert Posts: 37 ✭✭
    Re: Ground conductor patterns

    thanks Steve,

    In case A, I intended to use a split bolt to connect the upper string's ground conductor to the lower string's ground ground conductor. In that case, you can freely disconnect any panel without disrupting the ground connection.

    Pete B
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Ground conductor patterns

    See if you can find an NEC book in the library (if you cannot justify purchasing on)... Generally, the NEC is pretty clear on requirements.

    I believe that (A) would not be allowed and (B) is allowed. The difference being is that code (as I last looked at it years ago) requires a solid, unbroken ground wire from the array to the grounding point (ground rod, etc.). You are not allowed to splice a ground wire in mid run...

    If you ran two ground wires... One from the top array to ground, and a second run from the bottom array to ground--that would be legal.

    I always wondered if somebody was getting a little strict that splicing ground wires is so bad--yet splicing every other electrical cable is fine (as long as done per code). :confused::roll:

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • pbartko
    pbartko Solar Expert Posts: 37 ✭✭
    Re: Ground conductor patterns

    When using a mounting rail as your grounding conductor, are you not inherently having a broken conductor? The rail sections are bonded to each other and the other rails in the system. Isn't this the same as having copper ground conductors bonded to each other?

    Thanks,

    Pete B
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Ground conductor patterns

    Very few systems are allowed by NEC to use mounting hardware also as a grounding source.

    The NEC theory, over the decades, has been that they do not want mounting hardware as the safety ground connection. Even when we had a dozen bolts/screws holding things together--we still needed the "silly" green wire jumper to chassis ground (such as mount a computer power supply into a metal chassis).

    The thinking is that they want the removal of a safety ground to be "obvious"--having ground through mounting hardware does not provide this "obvious" disconnection warning to the person working on the equipment. For computers, we had to put the green wire down first, put a nut on it, then we could put other local ground connections on top of the green wire nut... Does not make electrical sense--but if you follow code--it is pretty much a "get out of jail card" for a manufacturer. If we did not follow code requirements--we would be left to hiring lawyers to prove that we were not being reckless regarding safety.

    I am not in the solar business--but if the rail/panel company has approved their mounting clips as grounding clips--more power to them!

    I still do not believe that this will affect the NEC requirements for an un-spliced ground wire from the array/rail bonding mount to earth ground of the home/installation.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: Ground conductor patterns

    2008 NEC specifically allows devices that ground the modules to the rails like the Wiley Weeb clips or Unirac grounding clips.

    I don't use the grounding clips on a lot of installs though because I do a lot of Sunpower installs and they specifically don't allow them - they also provide all the grounding lugs. When I do it like that I always ground like in B. I don't know that there's anything wrong with A though, except I know some people (inspectors) would have to be argued with about whether that was continuous or not.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Ground conductor patterns
    newenergy wrote: »
    2008 NEC specifically allows devices that ground the modules to the rails like the Wiley Weeb clips or Unirac grounding clips.

    I don't use the grounding clips on a lot of installs though because I do a lot of Sunpower installs and they specifically don't allow them - they also provide all the grounding lugs. When I do it like that I always ground like in B. I don't know that there's anything wrong with A though, except I know some people (inspectors) would have to be argued with about whether that was continuous or not.

    i'll agree here with you that a is not continuous. continuous means no breaks. now i do feel that a would be fine in my book if a good heavy duty connection is made that won't come apart such as a split bolt or 2. 2 would help account for any physical strains encountered while maintaining a solid connection and thus imitating a continuous wire both electrically and physically. what whiles intends or actually allows, who knows? i do remember long ago suggesting a continuous ground wire without any breaks in it would be better and i don't know if whiles put that into the rules before or after i said that. i'm not that overly concerned to look it up, but if somebody else does then chime in with when it was implemented. i should stop going into overkill ideas or he might incorporate them into the rules.:cry:
  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: Ground conductor patterns

    I'm don't think A violates the NEC, but like you mentioned it may or may not violate some article John Wiles wrote and inspectors go to his seminars.

    The thing is, you pretty much have to run bare on the roof and you pretty much can't run bare through the conduit - so something is spliced. I would argue that continuous is not the same thing as unspliced. Continuous in this sense just means that it's all grounded everywhere and if you remove a module you don't break grounding for the rest of the system.

    The part about not breaking the ground, I believe, comes from 250.64(D)(1) (about grounding electrode conductors) which I don't have time to quote (maybe later), may or may not apply, and is generally interpreted as allowing taps, but not splices on the GEC.
  • pbartko
    pbartko Solar Expert Posts: 37 ✭✭
    Re: Ground conductor patterns

    All,

    I contacted John Wiles last night to help clarify some aspects regarding grounding. Below is his response to my questions (further below). He warns however, that his answer is not definitive and one needs to read many parts of the code book to understand all the rules:


    Peter:

    You appear to be confusing a grounding electrode conductor (GEC) with
    an equipment-grounding conductor (EGC). See Art 250 and Art 690 for
    the differences in size, routing and continuous or non continuous
    nature of each.

    The NEC requires that the rack (all pieces) and the module frames be
    effectively grounded. Any combination of series and parallel
    connections that does this throughout the array can be used.

    There is no requirement for a single, unspliced conductor to do this
    and multiple conductors can be used as long as one conductor gets to
    the combining box and then continues as the dc equipment-grounding
    conductor to the dc disconnect and then the the inverter.

    Properly installed tin-plated copper lay in lugs with surface
    preparation is the way to do this. See NEC PV Suggested Practices
    Manual V1.9 Appendix G on our web site below. IAEI News articles are
    also on the web site which may have better info that the HP articles.

    Grounding electrode conductors (GEC) are not usually found on the
    array unless you are meeting NEC 690.14(D) in the 2008 NEC. Array
    grounding is done by equipment-grounding conductors (EGC) as
    described above.

    A dc GEC must originate at the inverter and be routed as described in
    690.47(C) or either the 2005 or 2008 NEC.



    >John,
    >
    >I am in the process of installing a PV system on my roof. Some
    >questions have come up regarding grounding the DC PV array. I am
    >hoping that you can clarify the NEC code. I've read several of your
    >articles in HomePower Magazine but am still not sure of the proper
    >rules to follow.
    >
    >The array consists of a single string of 16 PV panels mounted to
    >four Unirac rails (2 rows of 8 panels, 2 rails per row). A DC
    >combiner box (Soladeck) is mounted to the roof to convert the panel
    >wires to THHN and incorporate a fuse and lightning protection. From
    >the DC combiner box, the conductors run in conduit down to the DC
    >Disconnect/Inverter located below in the garage.
    >
    >I have the option of either bonding the rails to ground and use
    >piercing bonds such as a WEEBs to bond the panels to the rails or I
    >can bond the panels directly to a GEC (6 AWG) bare copper conductor
    >using lay in lugs.
    >
    >Question 1: Using the mounting rails as a GEC, can the rails be
    >linked in parallel with an individual bonding jumper that connects
    >each rail to the Ground Lugs in the combiner box, or do each of the
    >rails need to be bonded in series with each other with a single
    >conductor going to the ground lug in the combiner box?
    >
    >Question 2: Using a copper conductor as a GEC, a single conductor
    >will be used to bond all the panels in series. Must the rails also
    >be grounded to this same conductor in series? If so, this will
    >increase the conductor length by requiring it to run alongside each
    >of the 4 rails and the panel ends where the grounding attachment
    >points are located.
    >
    >Question 3: Should the GEC also be bonded to the Service Entrance
    >Ground Electrode? If it is, will this create a ground loop as the
    >ground lug in the combiner box will be connected to the DC
    >Disconnect/Inverter Ground. DC inverter ground is internally
    >connected to Inverter AC ground which is connected through the main
    >panel to the Service Entrance Ground Electrode.
    >
    >Thanks for your help, I look forward to your Code Corner in
    >HomePower Magazine each issue.
    >
    >Pete B
    >
    >
    >
    >


    --
    If I can provide further information, please do not hesitate to call,
    e-mail or fax me. See our web site below more PV/NEC information and
    a schedule for future PV/NEC presentations throughout the country.

    John

    John C. Wiles, Program Manager
    Southwest Technology Development Institute, New Mexico State University
    575-646-6105 575-646-3841 (FAX)
    http://www.nmsu.edu/~tdi/Photovoltaics/Codes-Stds/Codes-Stds.html
    SWTDI/NMSU
    3705 Research Drive
    Box 30001/MSC 3 SOLAR
    Las Cruces, NM 88003-8001
  • solarix
    solarix Solar Expert Posts: 713 ✭✭
    Re: Ground conductor patterns

    My thinking is that grounding is important enough and difficult enough to do on aluminum surfaces for 25 years, that I believe some redundancy is in order. On my installs, I will typically run the required grounding conductor in a loop. Like example B, except bringing the end on around back to the starting point and "splice" it in at the lug. I understand that the primary reason for the grounding is to bleed off static charge, but I still do not like a convoluted path for any high currents to follow and much prefer the loop pattern as it gives a much more direct path for lightning to follow.
    I also like using Weeb washers even if others don't and we put them on every other module so each module gets grounded on each rail at least once. These practices do not cost much more and help me to sleep well during lightning storms. Very few of my inspectors actually even get up on the roof, and have never had any comments from them so would appreciate any from you all.
  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: Ground conductor patterns

    Solarix - you put a lug on every module and use the weebs? Or just lugs on rails? If I use the weebs, I just put lugs on the rails and it's my impression that the weebs make an excellent ground to the rails and the rail obviously can carry whatever current the EGC could carry.
  • stevend
    stevend Solar Expert Posts: 34
    Re: Ground conductor patterns

    Solarix, For the solar array we do either the lugs without redundant paths or WEEBs between panels and rail in which case we usually have redundancy through multiple rails. Down at the inverter/charge controller where it's my choice I opt for redundancy where possible. For example, I've seen some installers ground only the DC side of Outback inverters or the AC side, which is okay according to Outback. I do both under the philosophy of "the more paths to ground, the better" where it makes sense. I've had one experience where a wire melted during an installation due to a short and fortunately I was there. I want to be able to sleep at night knowing that won't happen when I'm not there.
    -Steve
  • solarix
    solarix Solar Expert Posts: 713 ✭✭
    Re: Ground conductor patterns

    Yes, I put a lug on each module and use weebs. There is no such thing as an excellent ground when it comes to exposed aluminum.
  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: Ground conductor patterns
    solarix wrote: »
    Yes, I put a lug on each module and use weebs. There is no such thing as an excellent ground when it comes to exposed aluminum.

    That's a big problem then because it seems like most of the grounding cables as well as a lot of the conductors used by the utility companies are exposed aluminum.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Ground conductor patterns

    the grounds coming down the poles here are copper, but al will work if some precautions are taken and it's sized right. i personally don't like al either.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Ground conductor patterns

    From what I understand, alum gets an oxide layer on the outer skin only, but copper will slowly oxodize all the way through?
    Powerfab top of pole PV mount | 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 ,

  • solarix
    solarix Solar Expert Posts: 713 ✭✭
    Re: Ground conductor patterns

    The problem with aluminum is it oxidizes very fast and that causes an insulating layer on its surface. The good news is it prevents further oxidation - bad news is it makes electrical contacts iffy.
    What you want to do is make "gas tight" contacts which don't oxidize. The lugs or weeb washers or whatever have to dig in through the oxidation to the aluminum and be tight enough to keep out air. The practical solution is to use no-alox compound on the joint as well.
    All this goes for other kinds of metal contacts too (except gold), just a lot more so for aluminum
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Ground conductor patterns

    wouldn't apply to stainless steel. i wonder why they don't use that more for connections?:confused:
  • dwh
    dwh Solar Expert Posts: 1,341 ✭✭✭
    Re: Ground conductor patterns
    niel wrote: »
    wouldn't apply to stainless steel. i wonder why they don't use that more for connections?:confused:

    Maybe this?

    http://wiki.answers.com/Q/What_is_the_electrical_conductivity_difference_between_copper_and_stainless_steel
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Ground conductor patterns

    i know it won't have as low of a resistance as copper, but that should not matter as much for aluminum also has a higher resistance than copper. the connection will stay untarnished, as in galvanic reactions, with stainless. to overcome the extra resistance one could just use a larger connecting area, aka larger connector.
    brass is also acceptable for connections to copper and it has a higher resistance too, but it will have little or no galvanic reactivity. brass, like copper, is considered soft, but is not as bad as copper and this can create problems in making secure connections in some instances.
    what you are overlooking is that a good conductivity may be compromised later on due to corrosion or the lack of physical strength and rigidness. high strength, lower conductive connections can be made to encompass a larger connection area to overcome resistive differences and are less likely to be compromised later on through physical weaknesses or galvanic reactions.
    again, what's wrong with using stainless?