Announcement

Collapse
No announcement yet.

Planning a 8.58 kW grid tie system

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

  • #16
    Originally posted by wienerdog View Post

    That is how I showed it in the drawings submitted. Engineering at power company or the AHJ didn't say anything about it. I quickly discussed it with him and submitted drawings when I applied for the permit. Would you just leave the ground rod out? Or ground the rack with the ground rod and skip the #10 from the inverter to the rack? It does seem like a waste pulling a EGC from inverter to the rack along with the DC lines.

    Why do you think it is a bad idea?
    In my limited experience inspectors might have slightly different opinion than engineers or simply interpreting things differently. If possible- try to discuss this with the inspection department.

    What follows is my opinion only: in case of lightning strike at panel's site EGC will bring charge into the house as the rod at the panels won't be able to fully neutralize it. What is worse is that charge would go through the equipment grounding path along the way before reaching house grounding rod bringing all equipment grounds up to tens of thousands of volts. If I followed recent discussions on the subject correctly two approaches were suggested to address this:

    1. bind grounding rods at panel's site, barn and the house with dedicated #6 copper conductor effectively making GES (ground electrode system) and connect EGC to the house ground at single point only - at the house MSP where it is connected now. In case of a strike that #6 wire will conduct the charge between rods leaving EGC out of this. To protect inverters and possibly other equipment it was suggested to use TVSS on all DC and AC wires coming into the barn and the house.

    2. consider all 3 sites as separate systems and run only DC & L1, L2 & N wires between them. Inside each site have its own grounding rod and local EGC. This solution requires TVSS installed on all inputs - DC at the barn and L1,L2 & N at the house with as short as possible path to the local grounding rod. As I understand this TVSS together with electric wires between sites would clamp the lightning charge to the ground along its path.

    Comment


    • #17
      Originally posted by wienerdog View Post
      .....Unfortunately the snapnrack rails are only 162" long and I am going 4 tall with 39" panels so no room for that. They use one clamp for the panel below and above. Will just have to clean when needed.......
      You are not going to be happy after a couple years of manual clearing, in the snow. And do you have space in front for ALL the snow to pile up ? At least run the upper row of panels on it's own string, you might get some power when snow covers the lower row

      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


      • #18
        Originally posted by max2k View Post

        In my limited experience inspectors might have slightly different opinion than engineers or simply interpreting things differently. If possible- try to discuss this with the inspection department.

        What follows is my opinion only: in case of lightning strike at panel's site EGC will bring charge into the house as the rod at the panels won't be able to fully neutralize it. What is worse is that charge would go through the equipment grounding path along the way before reaching house grounding rod bringing all equipment grounds up to tens of thousands of volts. If I followed recent discussions on the subject correctly two approaches were suggested to address this:

        1. bind grounding rods at panel's site, barn and the house with dedicated #6 copper conductor effectively making GES (ground electrode system) and connect EGC to the house ground at single point only - at the house MSP where it is connected now. In case of a strike that #6 wire will conduct the charge between rods leaving EGC out of this. To protect inverters and possibly other equipment it was suggested to use TVSS on all DC and AC wires coming into the barn and the house.

        2. consider all 3 sites as separate systems and run only DC & L1, L2 & N wires between them. Inside each site have its own grounding rod and local EGC. This solution requires TVSS installed on all inputs - DC at the barn and L1,L2 & N at the house with as short as possible path to the local grounding rod. As I understand this TVSS together with electric wires between sites would clamp the lightning charge to the ground along its path.
        #2 seems like a very bad idea. The #10 conductor from the inverter to the rack is needed for the DC isolation detection. If you have a DC short to the rack you don't won't to rely on the earth ground to be in the detection path. 250.32 calls for building or structures to have a grounding electrode just as the barn will have a grounding electrode system. Lighting will always take the shortest path to ground. I agree you might get a potential for a short amount of time between house and rack if a strike was to occur at the rack but does it really matter on the EGC?

        The reason for the EGC is more of a bonding issue. You don't want to rely on earth ground to trip protection circuits if one of the ungrounded conductors shorts to earth.

        Comment


        • #19
          Originally posted by Mike90250 View Post

          You are not going to be happy after a couple years of manual clearing, in the snow. And do you have space in front for ALL the snow to pile up ? At least run the upper row of panels on it's own string, you might get some power when snow covers the lower row
          You must not like snow. I don't mind it. I have to go out for the chickens in the winter anyway.

          Comment


          • #20
            Originally posted by wienerdog View Post

            #2 seems like a very bad idea. The #10 conductor from the inverter to the rack is needed for the DC isolation detection. If you have a DC short to the rack you don't won't to rely on the earth ground to be in the detection path. 250.32 calls for building or structures to have a grounding electrode just as the barn will have a grounding electrode system. Lighting will always take the shortest path to ground. I agree you might get a potential for a short amount of time between house and rack if a strike was to occur at the rack but does it really matter on the EGC?

            The reason for the EGC is more of a bonding issue. You don't want to rely on earth ground to trip protection circuits if one of the ungrounded conductors shorts to earth.
            This is all academic as your local inspection department might have very particular requirements which you better meet anyways.

            your panels are floating so 'short' to the locally grounded rack is not dangerous and inverter's detection would still trip. Your assumption about grounding rod ability to 'ground' the strike is wrong- the strike will 'lift' its site many thousands volts along with the rod as the rod impedance is not that low to conduct hundreds of kA and still keep potential within hundreds of volts. EGC in this case becomes 'the shortest path' to the rods at other sites still at ground potential. Take a look at min 8 of the video:
            https://www.youtube.com/watch?v=YuDqXFvRv94 While the video shows 1 building with 2 grounding electrodes I think it directly applies to your case as well.

            OTOH your AHJ most likely just needs this to be done in certain way and that is the only right way regardless of anyone's opinion.

            Comment


            • #21
              Originally posted by max2k View Post


              OTOH your AHJ most likely just needs this to be done in certain way and that is the only right way regardless of anyone's opinion.
              Agree. Thanks for the video I will check it out tonight when I have some time to digest it.

              Comment


              • #22
                Originally posted by wienerdog View Post

                #2 seems like a very bad idea. The #10 conductor from the inverter to the rack is needed for the DC isolation detection. If you have a DC short to the rack you don't won't to rely on the earth ground to be in the detection path.
                Most grid tie inverters sold today rely on an RCD to detect ground faults, not a fused connection to earth.
                CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

                Comment


                • #23
                  Originally posted by sensij View Post

                  Most grid tie inverters sold today rely on an RCD to detect ground faults, not a fused connection to earth.
                  Resistor Capacitor Diode? Not sure where a fused connection to earth came in.

                  Comment


                  • #24
                    Originally posted by wienerdog View Post

                    Resistor Capacitor Diode? Not sure where a fused connection to earth came in.
                    Residual Current Device: https://www.solaredge.com/sites/defa..._fault_rcd.pdf

                    Comment


                    • #25
                      Okay now I am confused. Watching max2k video linked above helped and makes sense. So I believe running the EGC from the inverter to array is a bad idea. I know I am not going to protect against a direct lightning strike and that would probably be pretty rare but the video mentions "splash" hits which seems doable for this area. The house does have a lightning protection system with braided cable going to several grounds rods at the corners of the house.

                      Looking at some other threads Sunking gives good advice here https://www.solarpaneltalk.com/forum...108#post341108 but the more I think about how I am doing things it seems even more wrong.

                      So my AHJ says my area is on NEC 2008. My house has the 200 amp feed and I am installing a new sub panel in the pole barn that is 140 feet away from the house. AHJ says he doesn't care if I run 3 wire with bonded panel and ground rod or 4 wire even though 2008 requires 4 wire. I figured I would do 4 wire since newer code has to be better. The barn and house have no other metal connections beside a copper Ethernet line that I was going to also run. I don't mind switching to fiber if it changes things.

                      SunKing mentions that the only time you should run the EGC between structures is when there is something like metal conduit or pipe between the two structures (or maybe my copper Ethernet?). The video referenced says to only make ground reference at one point otherwise you can cause issues like max2k mentions. You want all the items to rise to the same voltage at the same time as lightning spreads across the ground. But if you use 3 wire to the barn and then bond neutral to a new ground rod aren't you doing exactly that? What is the difference between what the video references and the 3 wire between the two structures and then the neutral is bonded to a grounding rod at both points?

                      Comment


                      • #26
                        This is what I was planning on doing. House and pole barn are about 120 ft apart. Array is another 150 feet on the other side of the barn.
                        Copy of Solar SLD (1).jpg
                        Last edited by wienerdog; 09-11-2017, 10:41 PM.

                        Comment


                        • #27
                          Originally posted by wienerdog View Post
                          Okay now I am confused. Watching max2k video linked above helped and makes sense. So I believe running the EGC from the inverter to array is a bad idea. I know I am not going to protect against a direct lightning strike and that would probably be pretty rare but the video mentions "splash" hits which seems doable for this area. The house does have a lightning protection system with braided cable going to several grounds rods at the corners of the house.

                          Looking at some other threads Sunking gives good advice here https://www.solarpaneltalk.com/forum...108#post341108 but the more I think about how I am doing things it seems even more wrong.

                          So my AHJ says my area is on NEC 2008. My house has the 200 amp feed and I am installing a new sub panel in the pole barn that is 140 feet away from the house. AHJ says he doesn't care if I run 3 wire with bonded panel and ground rod or 4 wire even though 2008 requires 4 wire. I figured I would do 4 wire since newer code has to be better. The barn and house have no other metal connections beside a copper Ethernet line that I was going to also run. I don't mind switching to fiber if it changes things.

                          SunKing mentions that the only time you should run the EGC between structures is when there is something like metal conduit or pipe between the two structures (or maybe my copper Ethernet?). The video referenced says to only make ground reference at one point otherwise you can cause issues like max2k mentions. You want all the items to rise to the same voltage at the same time as lightning spreads across the ground. But if you use 3 wire to the barn and then bond neutral to a new ground rod aren't you doing exactly that? What is the difference between what the video references and the 3 wire between the two structures and then the neutral is bonded to a grounding rod at both points?
                          I believe the takeaway from all this is not to mix EGC with lightning protection. EGC should serve only one purpose to provide grounding path for AC faults and be connected to the ground only at single point. Lightning protection should be done by separate wire(s) connecting the site's grounding rods together. For example, the neutral bonded to more than single grounding rod is OK as long as it is not used to bond equipment to the ground (which it shouldn't). With separate 'local' EGC at each site and wire going from one site grounding rod to another in case of a strike EGCs together with local grounding rod will provide equipotential local environment so someone standing inside a barn and touching some metal enclosure won't get electrocuted by lightning strike near the panels. Wire connecting the sites will 'lift' the barn along with its rod/GES.

                          Your drawing is almost correct in this sense- except you shouldn't run the wire connecting grounding rods through your equipment as it is on the drawing. You should connect grounding rods between themselves and then connect local EGCs to their respective local grounding rods. It sounds the same as your picture but it will be very different during the strike as every piece of wire and equipment itself has non 0 impedance and will generate thousands of volts spikes during the strike. Like the following: Grounding_3.png




                          I still want to emphasize local AHJ authority over this. OTOH if you do something which complies with their requirements + some extra protection they most likely will be fine with that. Please connect your Ethernet to your router using some surge protection device which would let the charge to the ground (EGC in this case) instead of into your router and everywhere else from there. The SPD better be installed at the place where wires enter the building and not next to the router.
                          Last edited by max2k; 09-12-2017, 12:22 AM.

                          Comment


                          • #28
                            I say yes on the the fiber, you don't want a lightning surge to be funneled into your routers and computer gear.

                            Look at the strike protection on your house, and see if you can find a way to duplicate the scheme at the solar array, to keep the strike energy off the power grid. You might loose a couple panels if you got a direct hit, and likely the GT inverter
                            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


                            • #29
                              Originally posted by max2k View Post



                              I still want to emphasize local AHJ authority over this. OTOH if you do something which complies with their requirements + some extra protection they most likely will be fine with that. Please connect your Ethernet to your router using some surge protection device which would let the charge to the ground (EGC in this case) instead of into your router and everywhere else from there. The SPD better be installed at the place where wires enter the building and not next to the router.
                              Maybe I am thinking about this too much now but my AHJ said he will take it either way for the sub-panel at the barn 4 wire from the house non bonded between neutral and EGC at sub-panel in barn or 3 wire with new EGS bonded to neutral at the barn even though that isn't code for NEC 2008. He said either way is fine.

                              I see your diagram and that makes perfect sense but then it seems like the 4 wire method is more dangerous because you have a neutral that goes back to the house that isn't bonded to the local EGS. It is bonded 150 ft away. On a strike out in the yard, as the voltage flows through the earth you get a rise at the barn but the neutral is bonded at a different place in the system. Doesn't this make the neutral susceptible to arching from it to the local ground? The plastic in the sub-panel separating EGC and the neutral isn't much. Seems like I would see a big arc there?

                              With 3 wire pulled to the barn then the new point for local EGC is bonded to neutral as your drawing showed at the barn. That keeps the neutral between ground rods of the two structures the same and then all the equipment in the barn has the local EGC going to it's ground rod so there is no difference in voltage potential besides the neutral which is continuous with no gap.

                              Does that make sense? It seems to me the 3 wire is a safer method when structures are that far away?

                              Comment


                              • #30
                                Originally posted by wienerdog View Post

                                Maybe I am thinking about this too much now but my AHJ said he will take it either way for the sub-panel at the barn 4 wire from the house non bonded between neutral and EGC at sub-panel in barn or 3 wire with new EGS bonded to neutral at the barn even though that isn't code for NEC 2008. He said either way is fine.

                                I see your diagram and that makes perfect sense but then it seems like the 4 wire method is more dangerous because you have a neutral that goes back to the house that isn't bonded to the local EGS. It is bonded 150 ft away. On a strike out in the yard, as the voltage flows through the earth you get a rise at the barn but the neutral is bonded at a different place in the system. Doesn't this make the neutral susceptible to arching from it to the local ground? The plastic in the sub-panel separating EGC and the neutral isn't much. Seems like I would see a big arc there?

                                With 3 wire pulled to the barn then the new point for local EGC is bonded to neutral as your drawing showed at the barn. That keeps the neutral between ground rods of the two structures the same and then all the equipment in the barn has the local EGC going to it's ground rod so there is no difference in voltage potential besides the neutral which is continuous with no gap.

                                Does that make sense? It seems to me the 3 wire is a safer method when structures are that far away?
                                Neutral is supposed to conduct unbalanced AC currents between L1 & L2- you need it to run to your loads (house) for that reason and I think it should be connected to your house ground only with 4 wire connection (ground bonding wire, L1, L2, N).

                                Since you have dedicated thick wire running between sites bonding grounding electrodes that wire will 'lift' the house ground as well during strike so both ends of your neutral (as well as L1 & L2 or DC wires) will 'jump' in potential during strike but they will be under much less relative voltage difference so I don't expect arching. It was still suggested to use TVSS on all power wires entering a building to clamp those jumps to the local ground even if the ground is not where it usually is during strike to shield house loads (or barn loads/inverter) from very short transient spikes which might still happen due to imbalance between wire impedances (few microseconds tops). Wiring those TVSS properly is equally important as having them at all: lightning strike is very short event and as such only short connections with low inductance work otherwise long wires will look like open circuits during those short times. Found a doc illustrating this: http://www.lepstech.com/en/pdf/Transient-source.pdf
                                Last edited by max2k; 09-12-2017, 01:14 PM.

                                Comment

                                Working...
                                X