Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

PorkChopsMmmPorkChopsMmm Solar Expert Posts: 189 ✭✭✭✭✭
I am putting up a new pole mount near my power shed. I will ground the panels and every metal piece of the array but I am wondering where I should ground the array -- with a new ground rod at the base of the pole mount or run a ground wire, along with my cables from the solar panels back to my power shed and ground them with the rest of my equipment. Thoughts?

The physical distance is only 30 feet or so but with adding in the height of the pole, etc., it is ~50 feet. Pole mount is large -- 10 feet high and will hold 12 235 watt panels.
«1

Comments

  • BB.BB. Super Moderators, Administrators Posts: 28,186 admin
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    For lightning, the array+mount+pole should be grounded with short direct run (no sharp bends) to a ground rod next to the concrete pad.

    For electrical ground, there should be a ground cable run from the pole ground rod back to the main ground rod at the home/panel/battery shed.

    There should be no solar DC Ground connection to any ground at the array (other than if you are using Midnite or similar surge suppressors, the suppressors themselves being grounded to local "lightning ground" or main ground rod--depending on where the suppressors are installed and which is closest).

    A real quick answer to a very complex question.

    You can argue that the ground cable from the pole rod to the main rod at the house is not a great idea (two ground rods connected together can pickup the voltage gradient in the earth from a nearby strike)--And I see that point. However, I am (in my old age) tending to multiple ground rods tied together (the tie together providing the grounded needed to keep an electrical short circuit from energizing metal/and to provide a return path to trip breakers/fuses).

    Your thoughts on how this applies to your situation (I am trying to start a conversation, not end one).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • gww1gww1 Solar Expert Posts: 963 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    Bill
    I have a two hundred foot run from pv to house/shed. I had decided to put ground at the pv seperate from my house ground. Pv frames, railing and combiner box. No pos or neg grounding from combiner box or pv leads. Will the breakers in the combiner box work correctly if I don't tie the rod at the pv to the house rod?
    Thanks
    gww
  • vtmapsvtmaps Solar Expert Posts: 3,738 ✭✭✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    gww1 wrote: »
    I have a two hundred foot run from pv to house/shed. I had decided to put ground at the pv seperate from my house ground. Pv frames, railing and combiner box. No pos or neg grounding from combiner box or pv leads. Will the breakers in the combiner box work correctly if I don't tie the rod at the pv to the house rod?

    1) You must connect the ground rods to each other.

    2) You should either bond the battery negative to ground, or use a GFP device.

    If you don't do #2 you have a floating ground and you need circuit breakers in both positive and negative PV wires in the combiner box.
    if you don't do #1 you have a greatly increased chance of lightning damage because there may be 1000's of volts between the two ground rods during a lightning event.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • gww1gww1 Solar Expert Posts: 963 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    vtmaps
    I really feal bad about taking over this guys thread as I have pages and pages writen to me and I still have trouble understanding.
    Here goes anyway. My negative pv lead will end up in a outback {psd2?} dc sub panel which is grounded through the mounting plate along with the inverter. I believe the installation manual shows a ground from the batteries negative going to this same box. (I had heard that NEC had dissalowed negative battery grounds?) Most post have indicated that bringing the pv gound to the house jeperdizes the house ac wiring.
    2) You should either bond the battery negative to ground, or use a GFP device.

    Does the above wiring take care of me without tying the lightning pv ground to the electrical house ground? I will not be inspected but hope to protect my equipment and be safe.

    SIDE NOTE: I remember an old post where solar guppy said no ground at pv.

    Thanks
    gww
  • vtmapsvtmaps Solar Expert Posts: 3,738 ✭✭✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    gww1 wrote: »
    vtmaps wrote:
    You should either bond the battery negative to ground, or use a GFP device.
    Does the above wiring take care of me without tying the lightning pv ground to the electrical house ground? I will not be inspected but hope to protect my equipment and be safe.

    No. Not at all. When lightning strikes in your neighborhood, your two ground rods may have 1000's of volts between them. That voltage will try to get from one rod to the other through your wiring and electronics.

    In my opinion the best connection between the rods is a thick, bare copper wire in the trench, but I think NEC wants an insulated conductor in conduit. It is better protected from mechanical trauma in conduit. And it is very important that the connection be maintained.
    gww1 wrote: »
    I had heard that NEC had dissalowed negative battery grounds?

    What the NEC wants now is to use a GFP which is itself a bond between battery negative and ground. The problem is that when the GFP trips it breaks the bond. Many smart folks recommend to use a real solid (not GFP) bond between battery negative and ground.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • PorkChopsMmmPorkChopsMmm Solar Expert Posts: 189 ✭✭✭✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    BB. wrote: »
    For electrical ground, there should be a ground cable run from the pole ground rod back to the main ground rod at the home/panel/battery shed.

    Thanks! I didn't think of doing that. I have heard it debated here (ask 10 people on here about grounding and you get 15 different answers) but I think this would be good, especially because they are so close together. I was going to buy solid copper wire for the lightening grounding for the array/panels/mount but hoped to used twisted copper wire for the ground in the conduit with the +/- array wiring. See any problem with that?

    BB. wrote: »
    Your thoughts on how this applies to your situation (I am trying to start a conversation, not end one).

    -Bill
    I appreciate the help, Bill. I need to do some searching to see how people with tilt-able/rotate-able arrays do their grounding. E.g. can solid copper wire take the stresses of being moved back and forth? I think my twisted copper #2 wiring can take it, but not sure if the solid copper is less malleable.


    BTW no worries about discussing other grounding situations in this thread. When I search for past threads I always find the discussions helpful.
  • gww1gww1 Solar Expert Posts: 963 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    Vtmaps

    Let me first say that I am not trying to be aurgumentive I am just slow on this subject. I don't care how it is wired I just want to do the best I can.
    From bill
    You can argue that the ground cable from the pole rod to the main rod at the house is not a great idea (two ground rods connected together can pickup the voltage gradient in the earth from a nearby strike)--And I see that point.
    from vtmap
    No. Not at all. When lightning strikes in your neighborhood, your two ground rods may have 1000's of volts between them. That voltage will try to get from one rod to the other through your wiring and electronics.

    All I have is a "GED education", are these quotes not saying the same thing, one connected and one not connected?

    Thanks
    gww
  • BB.BB. Super Moderators, Administrators Posts: 28,186 admin
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    gww1 wrote: »
    Bill
    I have a two hundred foot run from pv to house/shed. I had decided to put ground at the pv separate from my house ground. Pv frames, railing and combiner box. No pos or neg grounding from combiner box or pv leads. Will the breakers in the combiner box work correctly if I don't tie the rod at the pv to the house rod?

    To answer the question asked--The combiner box breakers/fuses will work correctly if the two ground rods are not tied together. The over current protective devices (OCPD) are to prevent too much current from (back flowing) through and internally shorted array.

    What won't work correctly is if there is (for example) if you have AC power from the home running to the local building/mounting framework and there is a sort from AC hot to the local metal structure. The structure can be energized with respect to earth (120 or 240 VAC) and there is not low enough resistance through the earth back to the home to trip a protective breaker.

    Similar could happen if there is a short on the DC + lead to local metal--Except that shorted solar panel outputs do not "surge" enough to pop a breaker/fuse--The local metal will still be energized and there is a shock hazard. If thee was a ground wire bond between the local and the remote ground rods--At least the metal will not be much above local ground if there is a short--reducing shock hazard and also "notify" the owner there is a fault of some sort because the array voltage is near zero volts (current will still flow because the breakers will not trip).

    How often do shorts happen--Not often. But the idea is to have built in safety (earth ground rod bonding) for safety... If you have >~30 volts and wet ground/grass/etc... Some kid or adult grabs the metal structure and gets a shock/electrocuted if there is ever a failure of insulation.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • CariboocootCariboocoot Banned Posts: 17,615 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    Ground for electrical safety is not the same as ground for lightning protection. As far as panels are concerned, the latter issue is more prevalent than the former.

    Two ground rods 50 feet apart are not the same as two ground rods 200 feet apart. The first should be treated as a "single point" grounding issue because the resistance of the actual earth over that distance is low, allowing for conductance. The second is like asking if you should connect your house's ground to your neighbour's ground rod too. Exactly how many feet of distance shifts it from one to the other is not easy to say.

    Grounds do absolutely nothing for the normal operation of the system. You can put it in wrong or leave it out entirely and never notice a thing. It's when something fails that the ground is needed, and then even a "correct" ground connection may not do the job. It's about reducing risks, not eliminating them.

    There is indeed a great deal of discussion and disagreement about how best to do this. Some of that is caused by the differences in actual installations (soil type, distances, likelihood of strikes) and some about evaluation of what is the most likely failure and resulting problem. So we render opinions.

    My opinion is that PorkChops should connect his two ground rods with buried copper, which I believe follows NEC spec for multiple rods. And that gww with his 200' distance between rods doesn't have to worry about it. See? Two conflicting opinions from the same person. :roll:
  • PorkChopsMmmPorkChopsMmm Solar Expert Posts: 189 ✭✭✭✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    Awesome. Thanks guys.
  • Coach DadCoach Dad Solar Expert Posts: 134 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    Ground for electrical safety is not the same as ground for lightning protection. As far as panels are concerned, the latter issue is more prevalent than the former.

    Two ground rods 50 feet apart are not the same as two ground rods 200 feet apart. The first should be treated as a "single point" grounding issue because the resistance of the actual earth over that distance is low, allowing for conductance. The second is like asking if you should connect your house's ground to your neighbour's ground rod too. Exactly how many feet of distance shifts it from one to the other is not easy to say.

    Grounds do absolutely nothing for the normal operation of the system. You can put it in wrong or leave it out entirely and never notice a thing. It's when something fails that the ground is needed, and then even a "correct" ground connection may not do the job. It's about reducing risks, not eliminating them.

    There is indeed a great deal of discussion and disagreement about how best to do this. Some of that is caused by the differences in actual installations (soil type, distances, likelihood of strikes) and some about evaluation of what is the most likely failure and resulting problem. So we render opinions.

    My opinion is that PorkChops should connect his two ground rods with buried copper, which I believe follows NEC spec for multiple rods. And that gww with his 200' distance between rods doesn't have to worry about it. See? Two conflicting opinions from the same person. :roll:

    Coot- After reading all of this... I have a question regarding my system...
    I have 2 pole mounts (25 feet apart) and only 1 ground rod for the 2 poles. Both poles ground wires run to the same ground rod. AND I also have a ground rod at the power shed.
    So my question is, should the 2 pole mounts be using 2 seperate ground rods or is it OK to run the wires from each pole to a single rod???
    I can chnage it pretty easily if needed..
  • BB.BB. Super Moderators, Administrators Posts: 28,186 admin
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    You want lightning grounds to be short and no sharp bends (something like 18" radius bends or larger radius?).

    And that would mean that the ground rod(s) should be near the foundation of each pole.

    Roughly, I would use ~8' as the distance as being significant regarding grounding. So, if the two rods would be 8' or more apart, then I would use separate rods for each. And, i would use a ~6 awg cable between each ground rod (including back to the home).

    As always, read all you can about an issue--There is no one "right answer" -- However, there are many "wrong answers".
    BB. wrote: »
    A couple threads about Lightning:

    Off Grid Grounding Technique?
    Another Question, this time about Lightning

    Note, the above are discussions, not a do A, B, and C--and you will be "safe". There probably is no such thing with lightning. Several different techniques are discussed--and a few of those posters even have experience with lightning. :cool:

    And our host's consolidated FAQ page:

    www.windsun.com
    Lightning Protection for PV Systems

    From other past posts here, Windsun (admin/owner of NAWS), he said that most of lighting induced failures he saw were in the Inverters' AC output section.

    Towards the end of this thread is a very nice discussion of proper generator grounding.

    -Bill


    -Bill "In my humble opinion" B.
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • CariboocootCariboocoot Banned Posts: 17,615 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    Coach Dad wrote: »
    Coot- After reading all of this... I have a question regarding my system...
    I have 2 pole mounts (25 feet apart) and only 1 ground rod for the 2 poles. Both poles ground wires run to the same ground rod. AND I also have a ground rod at the power shed.
    So my question is, should the 2 pole mounts be using 2 seperate ground rods or is it OK to run the wires from each pole to a single rod???
    I can chnage it pretty easily if needed..

    Your two poles are only 25 feet apart so having them on the same rod is a good idea.

    Here's what the issue is (and I hope I don't leave anything out):

    When something goes wrong the ground conductor becomes energized. The purpose of the grounding rod is to sink that energy to Earth, rendering the Voltage potential through a separate path (you) near nil and harmless. At the same time it should allow current to go high enough to trip any circuit protection and turn off the power at the source. System is shut down and safe.

    Now if you have two ground rods close together but not coupled properly, it is possible for the energy to go into the Earth on one rod and come back up out at the other. So instead of all the power being dissipated in the ground some of it 'escapes' and actually energizes other parts of the system - exactly the opposite of what you want to happen. But if the two rods are tied together properly they act as one rod, each being able to dissipate energy from either input to the ground.

    As the distance between the rods increases, so does the resistance of the ground itself decreasing the likelihood that some of the energy will be available to travel back up the other rod and cause a problem. This is where it gets to be a guess about how far apart it is safe to have rods that aren't connected; not all soil is the same nor does it stay the same as moisture level varies.

    So as a rule the single point ground for everything is the desired option, but sometimes it's not practical to dig 200' of trench and lay in #4 bare copper between the two (or whatever values may be involved).

    Note that sometimes "single point ground" involves multiple rods connected together (usually in a star pattern but not always) but with all the wiring from the system connected at one; the additional rods are to increase the connection to Earth from that point on. One of the biggest problems with grounding is getting sufficient soil contact to make it effective; dry, sandy, rocky ground is difficult to make a connection to. Moist loam is easy.
  • gww1gww1 Solar Expert Posts: 963 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    From cariboocoot
    My opinion is that PorkChops should connect his two ground rods with buried copper, which I believe follows NEC spec for multiple rods. And that gww with his 200' distance between rods doesn't have to worry about it. See? Two conflicting opinions from the same person.

    Thank you, I was thinking that I was going to not connect the grounds when over several days bill had tried to help me understand this issue. I had decided to go this way but everytime I read about grounding I wonder if I was understanding or making a mistake. You couldn't have made it any clearer and I feel a bit better.
    From porkchopman
    BTW no worries about discussing other grounding situations in this thread. When I search for past threads I always find the discussions helpful.

    Thank you for being so gratuitous.

    Lastly, If I did not use ground clips/fittings and insted just used stainless screws and star washers and wrapped the wire around the screw would this be considered a sharp turn of the wire or just a continuance of the strait line? Even though pv is in a field would most advise against running grounding wire in this fassion?

    Again I just want it to work but no inspection. I will save money where it works but would like it to work.

    thanks
    gww
  • CariboocootCariboocoot Banned Posts: 17,615 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    I'm not keen on wrapping a wire around a bolt or screw and then tightening to make a ground connection. It works on plugs and outlets because they are designed for it. An ordinary bolt isn't as there's nothing to lock the wire in place around the edges; it could pop out from under. Crimp a tinned lug on. NEC wants ground connections/splices to be "permanent" meaning you can't pull them apart with a good tug or disassemble them (like a bug) - which is a little weird considering the number of allowed connections you can take apart (like metal conduit being approved as ground conductor).
  • gww1gww1 Solar Expert Posts: 963 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    coot

    Point taken,thanks

    Also thanks bill and vtmaps.

    gww
  • YehoshuaAgapaoYehoshuaAgapao Solar Expert Posts: 280 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    BB. wrote: »
    What won't work correctly is if there is (for example) if you have AC power from the home running to the local building/mounting framework and there is a sort from AC hot to the local metal structure. The structure can be energized with respect to earth (120 or 240 VAC) and there is not low enough resistance through the earth back to the home to trip a protective breaker.
    -Bill

    I think the NEC code has a minimum impedance requirement for ground rods. Think its is why solar installers in AZ always use two bonded ground rods because a single rod almost never meets code in AZ's dry ground. I don't know if this code requirement covers the above scenario though.

    Wouldn't the grounding built into any AC receptacles or light fixtures be connected to house ground anyway and any boxes meant for AC be bonded to the same ground going to the receptacles so any shorts there would go to house ground anyway?
  • BB.BB. Super Moderators, Administrators Posts: 28,186 admin
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    I think the NEC code has a minimum impedance requirement for ground rods. Think its is why solar installers in AZ always use two bonded ground rods because a single rod almost never meets code in AZ's dry ground. I don't know if this code requirement covers the above scenario though.

    I believe the "usual" Ground Rod maximum resistance to ground is 25 ohms or so... At 120 VAC, that only carries ~4.8 amps. Not near enough current to trip a 15 amp branch breaker/fuse.

    So--for electrical bonding, you really need a ~6 AWG wire between each "independent" ground rod/system that share a common power source (such as 120 VAC from a home to an out building/pump shed.
    Wouldn't the grounding built into any AC receptacles or light fixtures be connected to house ground anyway and any boxes meant for AC be bonded to the same ground going to the receptacles so any shorts there would go to house ground anyway?

    Yes, green wire grounds all eventually go back to the same ground rod/metal water pipe in North American Code...

    However, if you do not have a Neutral to Earth Ground (really to ground rod) connection. If there is a short between "hot" and "ground", the ground will remain energized (with respect to other "ground" rods) unless there is a Neutral to Ground Rod bond or a GFI breaker in the circuit. More than enough current to shock/kill somebody (only need >0.005 amps to stop a heart), but not enough current to trip a breaker (less than 5-10 amps into the typical AC ground rod that meets code).

    I think this was the answer to your question (as I understood it).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • NorthGuyNorthGuy Solar Expert Posts: 1,925 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    BB. wrote: »
    So--for electrical bonding, you really need a ~6 AWG wire between each "independent" ground rod/system that share a common power source (such as 120 VAC from a home to an out building/pump shed.

    If neutral is bonded to the ground rods in each of these buildings, wouldn't the neutral wire work as an electrical bonding between the ground rods?
  • BB.BB. Super Moderators, Administrators Posts: 28,186 admin
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    Neutral to Earth Bonding should only be done in one location (i.e., the main service panel... Or can be different by switching neutral bond location with a an AC transfer switch--for example).

    If you have multi-point Neutral+Earth bonds, and a green wire ground, both the neutral and AC grounds will be sharing current flow (including conduit, panel sheet metal, etc.). This is not a good thing (safety grounds are not sized to carry sustained loads and are intended to be a safety backup in case there is a hot to earth fault).

    However--Effectively, the AC Neutral is already multi-point grounded in a normal installation... You have the transformer neutral grounded at the pole, and at each of the 1-5+ homes sharing that transformer.

    So, how can I with a straight face say that Neutral shall be bonded to earth ground only in one location at a home. However--One could argue that the utility feed does not carry a safety ground wire with it... So there is no chance of having current flow shared between neutral and safety ground.

    But--One could also argue that earth ground (i.e., "dirt") is a very poor conductor and should never be used as a "true" safety ground (lightning ground yes, over current/short circuit no). One article I read decades ago reported that a person found out that their home was miss-wired (a Hot was bonded to earth, rather than utility neutral--as I recall). The customer noticed because he saw melting snow/steam around the ground rod after a winter snow. This was an example of a "short" that energized the local ground and did not cause any other "noticeable" problems).

    So--If you were looking to save some money. Yes, you could drop carrying a "green wire/6 awg" ground between the house and a distant out building. And bond the neutral at both the home and out building to earth ground--It would emulate the Pole to Home utility bonding right outside by my house). It would not be my preference and I would worry about bringing lightning energy into the AC home wiring--But it could be done (to code after the main panel? I am not so sure). If I was really paranoid about grounding/lightning, I would consider using an isolation transformer between the home and the out buidling--That would allow Neutral+Earth Bonding without the circulating current/and less of a lightning injection issue).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • vtmapsvtmaps Solar Expert Posts: 3,738 ✭✭✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    BB. wrote: »
    If I was really paranoid about grounding/lightning, I would consider using an isolation transformer between the home and the out buidling--That would allow Neutral+Earth Bonding without the circulating current/and less of a lightning injection issue).

    Regarding the lightning injection issue... Isn't the protection limited to the voltage rating of the transformer? I would guess that when the two ground rods have 1000's of volts between them there will be arcing between the transformer primary and secondary.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • BB.BB. Super Moderators, Administrators Posts: 28,186 admin
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    Yep... But the transformer should be rated and tested during mfg. for >2,545 VDC (1,800 VAC)--And that is the tested isolation--Actual isolation should even be greater. That is a fair amount of voltage--Add good quality Midnite surge suppressors on the input/output of the transformer, then you have a good chance of reducing the chances of a surge making back into the home.

    Of course, there are the losses of the isolation transformer--Which adds to the "terrible" ~50% efficiency of an off grid power system (from solar panels thru controller, into the battery bank, and out the AC output of inverter). No free lunch.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • PorkChopsMmmPorkChopsMmm Solar Expert Posts: 189 ✭✭✭✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    OK, one last question. What gauge copper wire should I use to ground the panels and array? Doing some searching I have seen a variety of responses. I bought 6 gauge to wire the 2 grounding rods I have -- the one at the base of the pole array and to the existing grounding rod ~35 feet away with my solar equipment. Do I also need 6 gauge on the back of the panels or can I go thinner?
  • YehoshuaAgapaoYehoshuaAgapao Solar Expert Posts: 280 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    BB. wrote: »
    I believe the "usual" Ground Rod maximum resistance to ground is 25 ohms or so... At 120 VAC, that only carries ~4.8 amps. Not near enough current to trip a 15 amp branch breaker/fuse.

    So--for electrical bonding, you really need a ~6 AWG wire between each "independent" ground rod/system that share a common power source (such as 120 VAC from a home to an out building/pump shed.



    Yes, green wire grounds all eventually go back to the same ground rod/metal water pipe in North American Code...

    However, if you do not have a Neutral to Earth Ground (really to ground rod) connection. If there is a short between "hot" and "ground", the ground will remain energized (with respect to other "ground" rods) unless there is a Neutral to Ground Rod bond or a GFI breaker in the circuit. More than enough current to shock/kill somebody (only need >0.005 amps to stop a heart), but not enough current to trip a breaker (less than 5-10 amps into the typical AC ground rod that meets code).

    I think this was the answer to your question (as I understood it).

    -Bill

    I currently have 4 earth ground points - House Ufer (AC), PV (DC), Inverter/Charge controllers (mostly DC), and Generator (AC) The PV and DC ground points look properly bonded to each other but are isolated from the AC grounds as James believed AC and DC should be kept separated even in grounding. The house ground/ufer and generator are barely connected - 70ft of #10 plus opposite sides of generator frame (hots and neutral are #6 except for a short length of SOOW cable to fit the generator's 240V receptacle). The generator's neutral is floating / not bonded to ground.

    So to properly fix my grounding (asymptomatic currently, including lightning arrestors and surge protectors - so won't know until it doesn't work)?:

    Easy way -
    Connect the ground bars in the PDP together with a #6 - this provides a decent bandwidth grounding path between the house ground and the PC & PV ground rods and also a low-bandwidth grounding path (70ft of #10 plus generator frame - ground rod's wire is bonded on opposite side of generator frame as the receptacle's built-in equipment-ground bond) to the generator's ground rod

    Correct way (80 feet of digging):
    • Run a #6 bare copper wire underground between the generator's ground rod and the PV ground rod
    • (PV ground rod and DC ground rod are already bonded with the PV's #6 bare ground wire unbroken)
    • Run a #6 bare copper wire between the DC ground rod and the ufer in the main panel (or at least its ground bar)
    • (Maybe also) connect the 2 ground bars in the PDP with a #6 to connect the ground points via normal ground wiring
    • (Maybe) Replace the #10 ground wire from the generator's receptacle that connects to the top/AC ground bar in the PDP with a #6
  • NorthGuyNorthGuy Solar Expert Posts: 1,925 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    Easy way -
    Connect the ground bars in the PDP together with a #6 - this provides a decent bandwidth grounding path between the house ground and the PC & PV ground rods and also a low-bandwidth grounding path (70ft of #10 plus generator frame - ground rod's wire is bonded on opposite side of generator frame as the receptacle's built-in equipment-ground bond) to the generator's ground rod

    I wouldn't have any doubts choosing this one. Besides, they're probably already connected through the PDP case.
  • BB.BB. Super Moderators, Administrators Posts: 28,186 admin
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    I am still a little confused about the questions--But I will take a shot.
    I currently have 4 earth ground points - House Ufer (AC), PV (DC), Inverter/Charge controllers (mostly DC), and Generator (AC) The PV and DC ground points look properly bonded to each other but are isolated from the AC grounds as James believed AC and DC should be kept separated even in grounding. The house ground/ufer and generator are barely connected - 70ft of #10 plus opposite sides of generator frame (hots and neutral are #6 except for a short length of SOOW cable to fit the generator's 240V receptacle). The generator's neutral is floating / not bonded to ground.

    Again, back to the reason for grounds--Lightning and Electrical bonding (to trip fuses/breakers in the event of a short of hot to ground).

    The first three grounds: House Ufer (AC), PV (DC), Inverter/Charge controllers (mostly DC)

    Assuming that all these grounds are in the same building, I would tie all of the ground rods together with 6 AWG wire. If you want to test before bonding--Get some 40 watt 120 VAC bulbs and some wire (any type) and tie all the grounds together with a 40 watt filament bulb in between each connection point. See if the bulb glows, if there is any voltage (more than a few volts) across the bulbs. Assuming there is no substantial voltages present, then it is "safe" to attach 6 awg wire between each ground. If you do see voltage (your DC bank voltage, 120 VAC, etc.) across any of the bulbs, hunt down the reason for the voltage before making the ground connections--You have a Hot/Neutral swap and/or a short between hot and one of the grounds somewhere.

    Regarding not connecting AC and DC ground together... In a normal system, I don't agree with this. In normal operation, there is no current flow, so why not connect. If there is a short from DC or AC hot to a piece of bonded metal (chassis of appliance, inverter, etc.), then you want the current through the chassis/green wire ground to pop the fuse/breaker for the power source.

    The problem with DC shorts is that the available current is something like 4-10x the typical AC fault current (i.e., 2,400 watts of 120 VAC power is 20 amps nominal, same thing in 12 VDC is 200 amps). So anything power with heavy DC cabling/current (such as AC inverter), needs a heavy ground bond back fomr the inverter to the common ground point (specifically, the electrical path from the DC chassis, wire, ground rod, to DC Battery Negative bus, to negative DC battery bank terminal) needs to be heavy enough to carry the current. 200 amp DC circuit is probably about the maximum limit for a 6 awg wire (fusing current for 6 awg cable is ~600 amps).

    What can be a problem is sometimes folks have a DC return to green wire / earth ground bond at a DC appliance and possibly an open return cable (bad connection). The DC current flow can find the green wire ground back eventually to the battery bank and all will work... But, the typical AC Green Wire safety ground is only 14 AWG with is good for 15 amps nominally. So, if the DC Return has failed, or the small diameter green wire path has a lower resistance, it is possible for the DC current to over heat the green wire safety ground.

    That is the reason that mixing DC and AC grounds together can be an issue (over current on safety grounds). But you still want the common safety grounds because you do not want the DC safety ground at a different voltage than the AC safety ground if it is all in the same building.

    For lightning, you want a short/straight path from the grounding points (such as solar panels, wind turbine towers, etc.) to a local ground rod (less than 10' of run from base of structure to ground rod). You want lightning grounds on the outside of the building. I recommendation would be to never bring a lightning ground cable through the middle of the home--Lightning wants to go to the "edges" of the building and bring the cable into the building then out to a water pipe/earth ground to lead to the lightning energy jumping from the cable to other items inside the home.
    So to properly fix my grounding (asymptomatic currently, including lightning arrestors and surge protectors - so won't know until it doesn't work)?:

    Ain't that always true. :p
    Easy way -
    Connect the ground bars in the PDP together with a #6 - this provides a decent bandwidth grounding path between the house ground and the PC & PV ground rods and also a low-bandwidth grounding path (70ft of #10 plus generator frame - ground rod's wire is bonded on opposite side of generator frame as the receptacle's built-in equipment-ground bond) to the generator's ground rod

    Sorry--Not sure what "PDP" means...

    If you have real issues with lightning (more than likely you will be eventually hit)--You might want to look at true braided lightning ground cable. It will be much better at shunting lightning currents around to where it should go (will not be cheap).

    If you are bonding between a PV Array Frame ground to AC Main panel ground rods--Then 6 AWG is fine--You are just bonding the ground systems together so that if there is a short from Hot (AC or DC) to the framework, then the 6 awg cable to pop a breaker/fuse somewhere--or at least keep the structure from becoming "hot" with respect to ground (PV + to frame fault).

    Lightning energy will probably not pass much more than 20' or so down a 6 AWG cable--So the 6 AWG is not really doing anything between grounding systems for lightning. Burying a bare 6 AWG wire certainly is not hurting though--and is giving more area for the lightning energy to dissipate locally into the earth.
    Correct way (80 feet of digging):
    • Run a #6 bare copper wire underground between the generator's ground rod and the PV ground rod
    • (PV ground rod and DC ground rod are already bonded with the PV's #6 bare ground wire unbroken)
    • Run a #6 bare copper wire between the DC ground rod and the ufer in the main panel (or at least its ground bar)
    • (Maybe also) connect the 2 ground bars in the PDP with a #6 to connect the ground points via normal ground wiring
    • (Maybe) Replace the #10 ground wire from the generator's receptacle that connects to the top/AC ground bar in the PDP with a #6

    If you don't want to dig/bury the 80 feet of 6 awg cable (and I would bury some sort of plastic conduit or ABS large diameter pipe to allow easier pulling of cable next time)--This is a case were I would consider at the generator:
    • Generator frame to generator earth ground rod.
    • Bonding Generator Neutral to generator earth ground rod.
    • Do not bother with burying 6 AWG from generator to main ground rod.

    You will have the AC Safety ground (any shorts at the genset will find the neutral return--And you have a neutral to earth bond at the AC main panel in your home). And you will have the earth lightning bond ground local too.

    I don't think there is anything wrong with laying the 6 awg ground from generator ground rod to home AC Main Ground Rod. But I would be hard pressed to say it is worth digging an 80' trench unless you were doing this anyway (say for upgrade to larger genset eventually).

    These are my thoughts--And I welcome discussions/corrections in my suggested grounding. This is an ongoing evolution in my thinking too--And I may be very well wrong in some of my suggestions. Also, I do not completely understand this particular installation/wiring, so I may have misunderstood a point or two in the post.

    By the way, do you have Roof/High Point lightning rods/ground system (i.e., a 6+' metal pole tied to earth next to the genset)? Keeping the lighting from striking "critical equipment"/home structure in the first place is not a bad idea.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • CariboocootCariboocoot Banned Posts: 17,615 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?

    "PDP" - Power Distribution Panel. Not all systems have them, but it's what most would call the main service panel.

    No need to keep AC and DC grounding separate. Keep in mind DC grounding is different if the DC GFCI is used.

    There is a need to reduce the number of grounding points to one if possible/practical. Do not confuse "grounding points" with "grounding rods" or "grounding connections". The physical and the electrical do not always line up on terminology.

    Only thing I disagree with Bill on here is having an N-G bond at the generator; there probably already is one at the PDP, and one is enough (even if a separate ground rod is used at the gen).

    You probably don't want to lay 80' of #6 between the two ground rods. You will have that run on neutral and ground feed (in the cable) so electrically the ground would be at the PDP; the 'extra' rod would serve to keep the gen frame safe to touch (we hope).

    Don't think of the electrical safety ground as part of a lightning arresting system, even though arrestors are connected to it and the hot lines. That is for protecting the power system from induced high Voltage spikes, not 'controlling' lightning. This is why some of us contradict the NEC rule about running the panel/frame ground in to the structure and tying it to the ground rod there. Best case the rod is outside, the panel frame/mount ground goes down to it there, and the ESG goes outside to connect to it.

    Just my opinion; feel free to disagree.
  • YehoshuaAgapaoYehoshuaAgapao Solar Expert Posts: 280 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    "PDP" - Power Distribution Panel. Not all systems have them, but it's what most would call the main service panel.
    Midnite solar calls their wire/breaker/bus boxes E-Panels. My PDP doesn't have a N-G bond. That is in the main service panel. My system is grid-tied, i'm guessing in an off-grid system, the N-G bond would be in the PDP (Xantrex) or E-Panel (Mindnite Solar).
    No need to keep AC and DC grounding separate. Keep in mind DC grounding is different if the DC GFCI is used.
    BB wrote:
    Regarding not connecting AC and DC ground together... In a normal system, I don't agree with this. In normal operation, there is no current flow, so why not connect. If there is a short from DC or AC hot to a piece of bonded metal (chassis of appliance, inverter, etc.), then you want the current through the chassis/green wire ground to pop the fuse/breaker for the power source.

    The problem with DC shorts is that the available current is something like 4-10x the typical AC fault current (i.e., 2,400 watts of 120 VAC power is 20 amps nominal, same thing in 12 VDC is 200 amps). So anything power with heavy DC cabling/current (such as AC inverter), needs a heavy ground bond back fomr the inverter to the common ground point (specifically, the electrical path from the DC chassis, wire, ground rod, to DC Battery Negative bus, to negative DC battery bank terminal) needs to be heavy enough to carry the current. 200 amp DC circuit is probably about the maximum limit for a 6 awg wire (fusing current for 6 awg cable is ~600 amps).

    I'm using the built-in GFCI (it is a 1-amp fuse) in the xantrex charge controller. Not sure if it is PV-only, or PV and DC. There is no transformer in the charge controller so there is no isolation. There is no bonding to ground on the non-PV DC side (batteries) other than whats built into the charge controller. Equipment ground fairly extensive though, but only #10 in the charge controllers (including the GFCI one) becuase James saw the wiring situation to the ground terminal hellish (and southwire THHN is very stiff), but everything else grounding except the generator receptacle ground is #6. The battery boxes are double-walled HDPE (high density polyethylene) plastic (Lifetime makes great tables and chairs, also HDPE) so there is no equipment ground that is applicable to the battery boxes.

    I'm guessing the thing to do is to connect a bare #6 from the DC ground rod to the house ufer. I'm thinking James would know how to get to the ufer I think the ufer is connected directly to the ground bus in the main service panel but not sure. Maybe it would be easier just to connect the ground buses together in the PDP and let that be the linking pathway and deal with the ufer when I expand, which will require a panel upgrade line-side (200 to 225amp, leave main breaker at 200A) or supply side (main panel has zero room between the meter and the main breaker) and it can be done right then.
    What can be a problem is sometimes folks have a DC return to green wire / earth ground bond at a DC appliance and possibly an open return cable (bad connection). The DC current flow can find the green wire ground back eventually to the battery bank and all will work... But, the typical AC Green Wire safety ground is only 14 AWG with is good for 15 amps nominally. So, if the DC Return has failed, or the small diameter green wire path has a lower resistance, it is possible for the DC current to over heat the green wire safety ground.

    That is the reason that mixing DC and AC grounds together can be an issue (over current on safety grounds). But you still want the common safety grounds because you do not want the DC safety ground at a different voltage than the AC safety ground if it is all in the same building.

    I don't have any DC appliances connected directly to the batteries other than inverter and charge controllers, so I'm guessing this is not an issue.
    There is a need to reduce the number of grounding points to one if possible/practical. Do not confuse "grounding points" with "grounding rods" or "grounding connections". The physical and the electrical do not always line up on terminology.

    Only thing I disagree with Bill on here is having an N-G bond at the generator; there probably already is one at the PDP, and one is enough (even if a separate ground rod is used at the gen).

    You probably don't want to lay 80' of #6 between the two ground rods. You will have that run on neutral and ground feed (in the cable) so electrically the ground would be at the PDP; the 'extra' rod would serve to keep the gen frame safe to touch (we hope).
    BB wrote:
    If you don't want to dig/bury the 80 feet of 6 awg cable (and I would bury some sort of plastic conduit or ABS large diameter pipe to allow easier pulling of cable next time)--This is a case were I would consider at the generator:

    Generator frame to generator earth ground rod.
    Bonding Generator Neutral to generator earth ground rod.
    Do not bother with burying 6 AWG from generator to main ground rod.

    The generator neutral is not switched by the inverter (there is only one neutral going to the inverter, the AC-out). The ATS (automatic transfer switch) in the inverter only switches the hots. The generator's neutral goes into the PDP's neutral bus, which is connected to the main panel (has the N-G bond to a ufer) and the subpanel's neutral busses.

    There is a factory-built-in frame-to-ground that goes out the receptacle's ground pin plus what James drilled into the frame and tied to a dedicated ground rod. There is a #10 going from the ground bar in the PDP to the generator receptacle (hots and neutral are #6), so I'm guessing this would provide some rod-to-rod linkage though the generator frame is a part of that connection (House ground->20 ft of #6->PDP ground bus->80 ft of #10->generator frame->5 feet of #6->ground rod). I'm guessing what is there is probably sufficient then?
    Don't think of the electrical safety ground as part of a lightning arresting system, even though arrestors are connected to it and the hot lines. That is for protecting the power system from induced high Voltage spikes, not 'controlling' lightning. This is why some of us contradict the NEC rule about running the panel/frame ground in to the structure and tying it to the ground rod there. Best case the rod is outside, the panel frame/mount ground goes down to it there, and the ESG goes outside to connect to it.
    BB wrote:
    If you have real issues with lightning (more than likely you will be eventually hit)--You might want to look at true braided lightning ground cable. It will be much better at shunting lightning currents around to where it should go (will not be cheap).

    If you are bonding between a PV Array Frame ground to AC Main panel ground rods--Then 6 AWG is fine--You are just bonding the ground systems together so that if there is a short from Hot (AC or DC) to the framework, then the 6 awg cable to pop a breaker/fuse somewhere--or at least keep the structure from becoming "hot" with respect to ground (PV + to frame fault).

    Lightning energy will probably not pass much more than 20' or so down a 6 AWG cable--So the 6 AWG is not really doing anything between grounding systems for lightning. Burying a bare 6 AWG wire certainly is not hurting though--and is giving more area for the lightning energy to dissipate locally into the earth.

    The PV grounding seems to be thought of as both lighting safety and wiring-fault safety at the same time. Probably an idealistic fantasy that doesn't work.

    I don't think there is much that can protect from direct lightning strikes. A direct lighting strike at the PV array will probably destroy 2-4 PV panels at least. I don't think the puny 12 guage wire coming out of those Midnite Solar lightning arrestors will stop millions of volts at tens of thousands of amps. They will protect from 'hot' lightning though - a residual 100-amp current abouts that lasts for around 10 seconds - 'hot' lightning is the cause of most of the lightning fires.

    Was more thinking nearby lightning strikes, mainly grid-side or anywhere near the east side of the house where all the conduits are) or coronal mass ejections ('solar flares'; type 3 EMP). Only faraday cages protect from type 1 or 2 EMP - and not practical. Nuclear weapons produce all 3 types of EMP, but nearyby lightning and CME is type 3 only.

    More practically, power bricks for computers and electronics don't have the greatest of life expectancy in my house (1-2 years for the cheap ones, 2-4 years for laptop bricks) and my refrigerator died at only 5-6 years of age. Don't want my inverter dying right after the warranty runs out.
    BB wrote:
    For lightning, you want a short/straight path from the grounding points (such as solar panels, wind turbine towers, etc.) to a local ground rod (less than 10' of run from base of structure to ground rod). You want lightning grounds on the outside of the building. I recommendation would be to never bring a lightning ground cable through the middle of the home--Lightning wants to go to the "edges" of the building and bring the cable into the building then out to a water pipe/earth ground to lead to the lightning energy jumping from the cable to other items inside the home.

    I guess I should fix the PV ground wiring routing then. James brought that into the attic after it touches the last rail and then it comes out the side gable into an electrical box, passing through a ground bar that combines the grounds from the combiner boxes (the box also combines PV conduits), and then goes down to the ground rod. Or maybe I can just have a new ground rod driven on the other side of the patio (west side of house, very short path to ground from the back row on the patio and the ground is more moist there from watering stuff) and plug a #6 into the west side of one of those rails (probably can do all 4 patio rails conveniently, existing #6 is on east side of the rails)
  • BB.BB. Super Moderators, Administrators Posts: 28,186 admin
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    Midnite solar calls their wire/breaker/bus boxes E-Panels. My PDP doesn't have a N-G bond. That is in the main service panel. My system is grid-tied, i'm guessing in an off-grid system, the N-G bond would be in the PDP (Xantrex) or E-Panel (Mindnite Solar).

    It probably does not matter too much... Just a convenient place--One location only. If you have a "stock" main AC panel, many will default with the Neutral+Earth already bonded.
    I'm using the built-in GFCI (it is a 1-amp fuse) in the xantrex charge controller. Not sure if it is PV-only, or PV and DC. There is no transformer in the charge controller so there is no isolation. There is no bonding to ground on the non-PV DC side (batteries) other than whats built into the charge controller. Equipment ground fairly extensive though, but only #10 in the charge controllers (including the GFCI one) because James saw the wiring situation to the ground terminal hellish (and southwire THHN is very stiff), but everything else grounding except the generator receptacle ground is #6. The battery boxes are double-walled HDPE (high density polyethylene) plastic (Lifetime makes great tables and chairs, also HDPE) so there is no equipment ground that is applicable to the battery boxes.

    Yes, the 1 amp fuse inside the charge controller is the only DC Return to Earth Ground Rod connection... I really hate that--The fuse blows if there is any short (solar PV, battery bus, AC inverter bus, any DC Hot to Earth short) will blow the fuse (and signal to the XW controller to stop supplying charging current). If you put a #6 AWG or heavier ground cable between Battery Bus Negative and the Main Ground rod (or to DC ground rod with a 6 AWG to main AC ground rod) this will bypass the 1 amp fuse and (usually) prevent the fuse from blowing (and the charge controller will never stop supplying charging current).
    I'm guessing the thing to do is to connect a bare #6 from the DC ground rod to the house ufer. I'm thinking James would know how to get to the ufer I think the ufer is connected directly to the ground bus in the main service panel but not sure. Maybe it would be easier just to connect the ground buses together in the PDP and let that be the linking pathway and deal with the ufer when I expand, which will require a panel upgrade line-side (200 to 225amp, leave main breaker at 200A) or supply side (main panel has zero room between the meter and the main breaker) and it can be done right then.

    Code today is much more ground happy (at least in our area)... In the old days, the Neutral was connected to the nearest cold water pipe (and natural gas pipe was floating, as well as the hot water pipes, etc..

    Today, ground wire to cold water pipe (at entrance to home). 8+ foot ground rod driven into earth near panel (or near cold water pipe input). Natural gas pipe+Hot water pipe+cold water pipe all bonded with 6 AWG wire at/around water heater.

    Your DC ground rod and safety ground (green wire ground) is almost useless right now. Adding a bond from DC Rod to House Ground (UFER, cold water pipe, AC Neutral Bus, bonded AC main ground rod, etc.) will support a path back to the AC ground so that any AC short to DC green wire ground will pop the AC power breaker/fuse.

    If you jumper the 1 amp DC GFI detection fuse, then you will also provide a path back to the AC mains for any AC short to DC return. This 1 amp fuse/breaker in the DC GFI kills the whole reasoning for safety grounds and bonded returns (such as Grounded Bonded Neutrals and Bonded DC Returns).
    ...
    There is a factory-built-in frame-to-ground that goes out the receptacle's ground pin plus what James drilled into the frame and tied to a dedicated ground rod. There is a #10 going from the ground bar in the PDP to the generator receptacle (hots and neutral are #6), so I'm guessing this would provide some rod-to-rod linkage though the generator frame is a part of that connection (House ground->20 ft of #6->PDP ground bus->80 ft of #10->generator frame->5 feet of #6->ground rod). I'm guessing what is there is probably sufficient then?

    If I understand correctly, then yes, this should be OK.
    The PV grounding seems to be thought of as both lighting safety and wiring-fault safety at the same time. Probably an idealistic fantasy that doesn't work.

    There is PV Frame/Mounting Grounding. And if you follow the "Lightning Grounding" suggestions (short, to ground rod, outside of home)--It will work as a green wire fault safety--As long as there is a connection from the Lightning Ground Rod to the AC Main Ground rod (cold water pipe, UFER ground, etc.).
    I don't think there is much that can protect from direct lightning strikes. A direct lighting strike at the PV array will probably destroy 2-4 PV panels at least. I don't think the puny 12 gauge wire coming out of those Midnite Solar lightning arrestors will stop millions of volts at tens of thousands of amps. They will protect from 'hot' lightning though - a residual 100-amp current abouts that lasts for around 10 seconds - 'hot' lightning is the cause of most of the lightning fires.

    They actually may do pretty well... It appears that the DC side of a battery system is pretty rugged. My guess is the battery bank plus DC Return to Earth Bonding works together because the batteries act like a giant capacitor/chemical buffer/spark gap between + and - leads.

    Years ago, the founder of this forum ("Windsun") said that most failures he saw from lightning were the Inverter's AC output failures. So--Surge Suppressors on the AC main panel/inverter output are where they are probably most needed. Your panels and DC side may do pretty well at surviving a lightning storm (direct hits--I don't know).
    Was more thinking nearby lightning strikes, mainly grid-side or anywhere near the east side of the house where all the conduits are) or coronal mass ejections ('solar flares'; type 3 EMP). Only faraday cages protect from type 1 or 2 EMP - and not practical. Nuclear weapons produce all 3 types of EMP, but nearyby lightning and CME is type 3 only.

    CME's are typically quite low frequency and it would only affect power networks on the order of miles (or longer)--Not an off grid power system (of course, if you have Grid Power--There is an issue where the two join).

    Even then for an off grid system--An EMP would probably have to be "close enough" to ruin your day anyway (for a "home sized" off grid system):

    Making a PV system survivable in EMP/CME conditions?

    More practically, power bricks for computers and electronics don't have the greatest of life expectancy in my house (1-2 years for the cheap ones, 2-4 years for laptop bricks) and my refrigerator died at only 5-6 years of age. Don't want my inverter dying right after the warranty runs out.

    We got a rabbit a couple years ago--My power cords for the power bricks--Have a bunch of repairs and tape now. He seems to mostly like the DC side. :roll:

    Any idea why the "bad luck" with your power supplies/refrigerator? AC surges from near by lightning strikes... Could also be line voltages running high or low. Even a bad neutral (corroded neutral block--Mine was damaged by a water leak from rain) can cause issues (120 to neutral will go 240 to "neutral" if the neutral connection fails at the wiring block.). If you see some lights "brightening" when a refrigerator or other heavy 120 VAC load starts, could be neutral connection starting to fail somewhere.
    I guess I should fix the PV ground wiring routing then. James brought that into the attic after it touches the last rail and then it comes out the side gable into an electrical box, passing through a ground bar that combines the grounds from the combiner boxes (the box also combines PV conduits), and then goes down to the ground rod. Or maybe I can just have a new ground rod driven on the other side of the patio (west side of house, very short path to ground from the back row on the patio and the ground is more moist there from watering stuff) and plug a #6 into the west side of one of those rails (probably can do all 4 patio rails conveniently, existing #6 is on east side of the rails)

    Just to be clear--You are talking about PV Frame grounds and NOT PV negative wiring--Both PV plus and minus should carry through without any local connections to earth/safety ground.

    Adding a second ground on the opposite side of the array should be a good thing. The current wants to flow as far apart as possible (opposite corners of the house, etc.) because of the common magnetic fields (same current flow directing) tend to "repel" current flow from each other.

    If you want to add a second 6 awg that goes from the corner of the grounded array directly out side the all and to the local lightning ground rod--Probably a good thing. If the combiner boxes are inside the house--may not be worth the trouble to move them outside the home.

    Note--trying the best I can here. And being clear about why I recommend something so you understand the suggestion and can confirm if it is correct for your home or not--I am not there, and I am not a lightning control expert by any means.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • YehoshuaAgapaoYehoshuaAgapao Solar Expert Posts: 280 ✭✭
    Re: Quick ? -- Ground new pole mount array at pole or at power shed 50 feet away?
    BB. wrote: »
    It probably does not matter too much... Just a convenient place--One location only. If you have a "stock" main AC panel, many will default with the Neutral+Earth already bonded.

    Yes, the 1 amp fuse inside the charge controller is the only DC Return to Earth Ground Rod connection... I really hate that--The fuse blows if there is any short (solar PV, battery bus, AC inverter bus, any DC Hot to Earth short) will blow the fuse (and signal to the XW controller to stop supplying charging current). If you put a #6 AWG or heavier ground cable between Battery Bus Negative and the Main Ground rod (or to DC ground rod with a 6 AWG to main AC ground rod) this will bypass the 1 amp fuse and (usually) prevent the fuse from blowing (and the charge controller will never stop supplying charging current).

    Ok won't connect anything between battery negative bus to ground bus (currently nothing connected)
    BB. wrote: »
    Your DC ground rod and safety ground (green wire ground) is almost useless right now. Adding a bond from DC Rod to House Ground (UFER, cold water pipe, AC Neutral Bus, bonded AC main ground rod, etc.) will support a path back to the AC ground so that any AC short to DC green wire ground will pop the AC power breaker/fuse.

    Seems like a secondary safety. Probably just do it the easy way - connect the AC (connects to house ground) and DC ground bars (connects to DC/PV ground rods) with a short length of #6. The AC ground bar has a #6 going to the main panel (has the ufer) and a #6 going to the subpanel. The DC ground bus has a #6 going (insulated to the juntion box on other side of wall, then switches to bare) to the DC/PV ground bars.
    BB. wrote: »
    If you jumper the 1 amp DC GFI detection fuse, then you will also provide a path back to the AC mains for any AC short to DC return. This 1 amp fuse/breaker in the DC GFI kills the whole reasoning for safety grounds and bonded returns (such as Grounded Bonded Neutrals and Bonded DC Returns).

    I think I'll just leave the GFCI alone. I'm guessing the remaining charge controllers (#2,3,4) will be told over the network to quit charging if the GFCI blows on #1.
    BB. wrote: »
    There is PV Frame/Mounting Grounding. And if you follow the "Lightning Grounding" suggestions (short, to ground rod, outside of home)--It will work as a green wire fault safety--As long as there is a connection from the Lightning Ground Rod to the AC Main Ground rod (cold water pipe, UFER ground, etc.).

    I know PV grounding is strictly equipment grounding. There is no bonding of any DC or PV negative or positive to ground other than what is built into the charge controller.
    BB. wrote: »
    They actually may do pretty well... It appears that the DC side of a battery system is pretty rugged. My guess is the battery bank plus DC Return to Earth Bonding works together because the batteries act like a giant capacitor/chemical buffer/spark gap between + and - leads.

    Years ago, the founder of this forum ("Windsun") said that most failures he saw from lightning were the Inverter's AC output failures. So--Surge Suppressors on the AC main panel/inverter output are where they are probably most needed. Your panels and DC side may do pretty well at surviving a lightning storm (direct hits--I don't know).

    CME's are typically quite low frequency and it would only affect power networks on the order of miles (or longer)--Not an off grid power system (of course, if you have Grid Power--There is an issue where the two join).

    Even then for an off grid system--An EMP would probably have to be "close enough" to ruin your day anyway (for a "home sized" off grid system):

    Making a PV system survivable in EMP/CME conditions?

    Think i'm ok for lightning arrestors? Got 2 in the main panel, 1st all on L1, 2nd all on L2. They are in the main panel because there is no room on the PDP for them. Got 1 on the neutral bus in the PDP. The netural bus has netural connections going to the inverter AC-out, generator, subpanel neutral bus, and main service panel neutral bus - lightning could induce currents on the neutral wire. Got 1 on the subpanel (sharing L1 and L2). Got 2 on the generator, one in the PDP, other in the generator shed (80 feet of wire in between). On DC, got 1 in each of the combiner boxes (4 total) - these are mostly to protect the charge controllers. Got 2 on the battery DC - one on positive and one on negative. There is a direct path from the charge controllers to the inverter without going through the batteries first (though the batteries exert their influence). The battery bus has 2 big battery lugs - both being used - the battery bus is being used to parallel the 2 battery strings. It also has 4 conventional lugs - all being used - one for each charge controller. And there are 3 big lugs for each inverter breaker (currently only using one, will eventually use two). Basically everything DC comes together on the battery bus bars.
    BB. wrote: »
    We got a rabbit a couple years ago--My power cords for the power bricks--Have a bunch of repairs and tape now. He seems to mostly like the DC side. :roll:

    Any idea why the "bad luck" with your power supplies/refrigerator? AC surges from near by lightning strikes... Could also be line voltages running high or low. Even a bad neutral (corroded neutral block--Mine was damaged by a water leak from rain) can cause issues (120 to neutral will go 240 to "neutral" if the neutral connection fails at the wiring block.). If you see some lights "brightening" when a refrigerator or other heavy 120 VAC load starts, could be neutral connection starting to fail somewhere.

    Don't know. Did put a start capacitor in the AC in 2010 but not for electrical problem reasons. AC Guy sold it to me saying that since it is over 10 years old, a start capacitor will delay its eventual failure. No light flickering problems other than when the generator exercises and the inverter switches to it to exercise it under load. I had the inverter disconnect from the grid without the grid going out in the late evening twice - once in March and once in April. Hasn't occurred since. There were no faults logged though.

    I do get a lot of overfrequency anti-islanding faults. In April and may was getting one or two every day. Getting about 2-4 per week in July. Fault always clears when the 300 second reconnect timer runs out. Could be measurement inaccuracy - maybe a badly distorted wavelength being measured as multiple. Can't really tell. Grid voltage is usually steady between 122 and 124V (once in a while it gets to 125V for a short time). The AC running reduces grid voltage (measured by inverter AC1) by about 1-1.5V (more 1.5 at night) but is otherwise stable.

    The only thing I see that is wierd, is when AC1 (grid) is connected and the generator is running and AC2 (gen) is connected, AC2 and AC-out voltage are steady but AC1 (Grid) starts fluctuating between 119 and 124V in about 15 second intervals. This is only when AC1 is disconnected but the grid is still up.

    The other issue is that the L2 current measured at AC-out (subpane loads) is sometimes reads too low - Volt*Amps < Watts, like a >100% power factor. There is no negating opposite on L1 though (AC-out).
    BB. wrote: »
    Just to be clear--You are talking about PV Frame grounds and NOT PV negative wiring--Both PV plus and minus should carry through without any local connections to earth/safety ground.

    Yeah, its equipment ground. PV negative/positive aren't bonded to anything other than what's built into the charge controller.
    BB. wrote: »
    Adding a second ground on the opposite side of the array should be a good thing. The current wants to flow as far apart as possible (opposite corners of the house, etc.) because of the common magnetic fields (same current flow directing) tend to "repel" current flow from each other.

    If you want to add a second 6 awg that goes from the corner of the grounded array directly out side the all and to the local lightning ground rod--Probably a good thing. If the combiner boxes are inside the house--may not be worth the trouble to move them outside the home.

    I might just do that after reading about lightning being encouraged to jump to other things if its grounding path goes inside the building.

    My combiners are outside, mounted on the gable end under the shading of the eave. Most of the PV wire distance is pre-combiner and run in flex metal conduit in the attic (all 4 patio strings share a 1.5" conduit, 5 non-patio strings each in their own 0.75" conduit). Strings are tied together with the default #10 MC4 connector cable. Gets spliced to #6 when it enters the attic. #6 all the way to the combiner boxes. Longest wire run is about 80 feet to the patio. Shortest is 20 feet. #6 to minimize voltage drop and the conduits are probably buried under some of that 22" of blown-in fiberglass insulation that was added early this year (no heat dissipation). None of the PV ground goes to the combiner boxes (but the bare copper PV ground still runs through the attic free-air. Effectively equipment ground only coming from the combiner boxes, but its still #6 because of the lightning arrestors on them.

    The PV equipment ground (#6 bare copper) also goes into the attic and comes out into a separate junction box to go through a ground bar that has the combiner box equipment grounds connected to it, and then makes it way down to the ground rod. The PV ground has about 30 feet between its last rail bond (WEEBLug) and the ground rod. It is bonded to 10 mounting rails alltogether - 4 on patio, 2 on south roof, 4 on east/west roof. If I bonded the 4 patio rails to a ground rod just off of the patio, it would only be 10 feet of bare copper wire from the last rail to the ground rod. And the ground off of the patio is much more moist from irragation and hose waste than the east side of the house and the water in AZ is very hard.
«1
Sign In or Register to comment.