PV Array Grounding/System Grounding

James
James Solar Expert Posts: 250 ✭✭
Hello Folks,

We all know that grounding seems to be a rather difficult area to address and determine proper technique.
Without dealing with every minute detail, in general, I am looking for opinions regarding the following...

System setup:
Pole mounted array with negative lead grounded to ground rod next to pole.
Interior system componets (inverter, controller, etc.) bonded to battery negative lead and equipment ground with separete system ground rod at home structure.

The question;
Should the array ground rod be electrically connected (with cable lead of same gauge) to the other, system ground rod at house structure?
Should structure rod be eliminated?

Is individual rods ok?
Will two rods increase or decrease chances of lightning strike reaching componets?

??????????????????????
My first inclination is that the rods should be connected, but I have read so many articles and "Code Corners" about grounding, i'm somewhat confused. Any ideas/suggestions?
Thanks
«1

Comments

  • crewzer
    crewzer Registered Users, Solar Expert Posts: 1,832 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    I think John Wiles was pretty helpful in the article below: ground the system at the house, ground at the array framing at the array location, and connect the two grounds. See: http://www.nmsu.edu/~tdi/pdf-resources/cc94.pdf.

    I also find the following guide to be useful: http://www2.fsec.ucf.edu/en/education/cont_ed/pv_handbook/Inspector_NEC.pdf

    HTH,
    Jim / crewzer
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    i'll put it in a nutshell. yes, they need interconnected with at least the same gauge wire used in your ground system. i'm not posting this just so you don't have to read what crewzer posted either for you must follow the nec standards. i like to say ideally go beyond the nec and go at least 2 gauges larger and bury it minimumly 1-2 feet below the surface. this will take into account excess corroding of the wire being it's more subjected to moisture and it takes into account possible nicks to the wire. the interconnecting wire will add to the area of the ground system as well, so if you had doubts of your soil or the rod itself then this gives that little extra. more rods and wire under the ground in conjunction with the rods make dissipating lightning strikes or the near strikes better, thus improving the odds of equipment survival. though you are concerning yourself with the equipment survival you do understand that it could mean your survival too by lessening you being hit by the strike physically and could help to prevent a fire. there are no guarantees when it comes to lightning, but increasing your odds should make you feel better too.
    ps-don't solder the grounds.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: PV Array Grounding/System Grounding
    niel wrote:
    ps-don't solder the grounds.

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

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

  • Roderick
    Roderick Solar Expert Posts: 253 ✭✭
    Re: PV Array Grounding/System Grounding
    mike90045 wrote:
    niel wrote:
    ps-don't solder the grounds.

    Why ??
    I think the NEC definition of "continuous" ground means something that won't come off.  I'm not an authority, but think soldering counts as non-permanent, like wire nuts or terminal blocks.  Mechanical crimp I *THINK* is ok, but I ran a single uninterrupted conductor all the way down so there would be no possible objection.
    I stand corrected, below.  Thanks, Neil!
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    the reason is it could explode molten solder during even a modest strike and is not considered a good connection by itself. it does not have to be a continuous run of wire(as in no breaks or connectors), but it has to have good solid connections.
  • DLwindsun
    DLwindsun Administrators Posts: 34 admin
    Re: PV Array Grounding/System Grounding

    In my experience the grounding system itself does not seem to help during lightning strikes to save your equipment. It may provide a better path to ground to prevent a fire. The only systems that I've seen that actually work to prevent lightning strikes are air terminal systems, the same stuff you see on hospitals and large commercial buildings. You can purchase the equipment for a medium sized solar system for around $500, and It's very easy to install.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    got a link to this stuff as i've seen nothing that prevents a strike? nevermind, as it is nothing more than a lightning rod and they do not prevent a strike either.
  • DLwindsun
    DLwindsun Administrators Posts: 34 admin
    Re: PV Array Grounding/System Grounding

    You're right niel there is nothing out there to prevent a lightning strike I should have worded that differently. Air terminals simply provide a path to ground from different points on a building, just having a grounded system itself will not help do that.
    http://www.lightning.org is a good source for info on lightning protection, there is many more sites out there but this is a good place to start.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    it has also been argued that a lightning rod could attract the lightning to the area of the array that may not have been hit otherwise. i have done much research into the area of lightning protection and it is amazing the steps i've read that were considered by others to be minimal and we aren't talking the nec either. i wish i could remember the site i read this for my home that they suggested a rod of at least 20-30 feet into the ground. i have enough metal in the air from my ham radio antennas that makes my whole place one big lightning rod. you might say, that should make my neighbors feel better that i'd most likely get the hit, but being their places are in the proximity of the strike hitting me, it would give an emp strong enough to them that they might as well have been struck directly instead of me. even bad connections or wiring can amplify this as i've lit up incandescent light bulbs with my 100w radio in a place i once lived in with bad wiring. emps could far exceed what my radio did. it's a double edged sword as it can go either way with nothing for sure no matter the extent you take it too. the nec states the minimum they deem good protection and i will not go ungrounded as i've seen an antenna of mine in the past get its ground accidentally disconnected and the first storm that came by had a lightning strike to that antenna. i found that disconnect and it was never hit again, but years later it did blow off of my roof and hit the electric lines 100 feet away. 4th of july.
  • DLwindsun
    DLwindsun Administrators Posts: 34 admin
    Re: PV Array Grounding/System Grounding

    Well this is what I have seen over the years... we have an area just north of Flagstaff on the mountain with probably 50-75 solar systems. Most of these systems were installed about 15-20 years ago. Around 8-10% of these system have lightning protection systems installed including air terminals, not one of these systems has been hit by lightning in 15-20 years. In that same area we do about 20 service calls every year to repair systems hit by lightning.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: PV Array Grounding/System Grounding

    Here are some interesting thoughts (plus a linked white paper) on grounding and the NEC...

    http://www.electrical-contractor.net/ESF/Lightning_News/Standard_780_Removal.htm

    No black and white answers, but lots of interesting questions.

    I guess I should thank my lucky stars that I am not in a heavy lightening area.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    that's some of the stuff i was talking about bill, but it wasn't whiles' site that i read all of it. i'd put some really deep rods in if i didn't have lots of bedrock below my home. i do have a cage arrangement though and 7 ground rods all intertied with buried #6. i, no doubtedly, will have internal damages by the emp of a strike and would be surprised if i came out unscathed inside by one. the neighbors won't have as much of a chance as i would in my view, but lightning isn't all that predictable or tame to say for sure on the outcome of a direct strike on my home. complicating the odds is an underground stream that flows under my home near the surface. one of the rods does contact it directly in the front.
    dlwindsun,
    you don't suppose the lightning was being pulled into that general area by those with the lightning rods do you? possibility?
  • DLwindsun
    DLwindsun Administrators Posts: 34 admin
    Re: PV Array Grounding/System Grounding

    I do not think the lightning rods are attracting the lightning to the other homes.

    With all the lightning protection we have tried and seen nothing seems to have made a difference so far except air terminals. Most of the air terminals we see installed are on three story houses with metal roofs with lots of point edges, what is odd is that the systems that seem to get hit are the one story homes with a standard A frame design. For all we know it could just be total bad luck.

    Surge protection/Lightning arrestors do not seem to help prevent damage to solar equipment. Intricate grounding systems do not seem to help prevent damage to solar equipment. Lightning is a big problem for us here in northern Arizona and I would love to find something that actually makes a difference.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    if you find something, let me know too. right now my antennas are lightning rods and part of the lightning will enter my home through the antenna cables. luckilly, i'm not at the top of the hill, but that doesn't make me feel all that much better or exempt of a strike as i've seen it hit lower in the valley.
    dlw,
    the pvs are generally the opposite of lightning rods as lightning rods dissipate much static buildup at their points aimed skyward. the vast flat area pvs present, do buildup a great deal of static and make the possibility of a strike increased even when properly grounded.
  • DLwindsun
    DLwindsun Administrators Posts: 34 admin
    Re: PV Array Grounding/System Grounding

    What's interesting to me is that most of the damage that occurs to the solar systems comes from lightning strikes at least a mile away, rare to see a direct hit on a system.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    curious as to what seems to get hit, the controller? what you are seeing is the result of an emp event due to the lightning even that far away.
  • DLwindsun
    DLwindsun Administrators Posts: 34 admin
    Re: PV Array Grounding/System Grounding

    Almost always the inverter takes the most damage, if the inverter takes a hit the charge control normally also gets hit about 60% of the time, it does not seem to effect anything else.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    that's a real curiousity there with the inverter getting it most of the time. pickup of the emp must be going through the 115vac side of the inverter. do you recall if the ac side was grounded in these cases and if not could you please pay closer attention making a note of it? i have to know if they failed to ground the ac side at all or properly. also if there were lightning devices in place.
  • System2
    System2 Posts: 6,290 admin
    Re: PV Array Grounding/System Grounding

    Summer before last, we had a strike that really did a lot of electrical and electronic damage. Pretty well blew out everything in the place BUT the PV's and batteries. WOW! Now it's a big worry whenever theres a lightening storm and I run for the shed, pulling switches and plugs and hoping for the best.
    It is a very serious worry that in the blink of an eye, thousands of dollars and years of work could be destroyed, gone in a flash and puff of smoke.
    Wayne
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    i hear ya wayne. i did converse with dlwindsun by pm and the observed failures were grounded both for ac and dc sides with protection devices in place with no particular brand or model suffering this fate as the more of a brand/model out there the higher the incidence of it, putting the percentages pretty much the same for all. any protection ideas guys?
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: PV Array Grounding/System Grounding

    I would start with the rules for working with radios and reducing EMI (electromagnetic interference).

    As always, these are suggestions based on my experience in other fields... Not working with lightning myself... Your mileage may vary. For entertainment value only. :mrgreen:

    First, run all pairs of wires (especially those from solar panels and wind turbines) as twisted pairs--probably at least a few twists per foot. Never run the wire pairs separately around ground rods/guy wires and such (don't run the + on one side and the - on the other)... Prevent transformer differential coupling of energy from lightning surges through grounded metal and also to prevent the creation of a loop antenna picking up EMP from nearby strikes.

    Don't wrap (spiral) your solar panel/turbine lead pairs around grounded metal structures that may experience lightning induced currents. Don't wrap lightning ground wires around (or twist with) your +/- wires (or three phase AC wires from wind turbine). If you have excess wire/cable pairs, wrap them in "figure 8's", not in simple loops. This will help prevent common mode coupling of energy from either high current flow from direct strikes or nearby EMP coupling.

    When looking at possible sources of damage, generally, differential inputs tend to be more sensitive to over current/voltage surges (like the input to a light bulb--too much voltage/current will pop the filament). However, most devices are much less sensitive to common mode voltage/current (you can take a 6 volt light bulb, tie the two wires together, and connect them to one side of a 15,000 transformer, and the other to a sheetmetal ground plane--as long as you don't arc from the filament to the ground plane through the insulator (glass), everything is fine.

    So, if you have worked hard to prevent differential and common mode coupling to your cabling, you should now have less energy to deal get rid of with surge suppressors and such.

    Things that may help prevent some lightning damage...

    Solar Panels with aluminum frames (frames must be electrically connected at each corner through welding or good rivets). Aluminum "circle" will tend create a counter rotating current which (in theory) will counteract an EMP field.

    Twisting (or at least bundle tying) all +/- leads together. Even if you have a (for example) 10 panel series string, instead of simply running the wire in a "U" pattern, at the end panel, run the return wire back along the first wires (twisted or bundle tied) back to the initial connection--even if the total run of wire would be longer (more resistance/cost).

    Avoid long parallel runs of you solar panel cables with your ground wire. If they must come close together, then cross at 90 degree angles. Never bundle you lightning ground wire and your solar/turbine wires together (keep the energy from the lightning from coupling with the voltage sensitive equipment).

    If you have tilting/tracking solar panels or a "tilt" tower for a wind turbine, place a good piece of braided ground cable across the hinges/pivots. This may not make any difference in electrical damage to your chargers/inverters, but it can prevent the welding of your bearings by a lightning strike and preventing your tracker/tilt feature from working.

    Place (good high frequency) capacitors (and or MOV's or other voltage limiting devices) between your +/- connections right near the input to the solar converter--will help prevent large differential voltages from reaching the inputs.

    Use external metallic shielding for all of your wire pairs. Grounded, at least, at one end... Since very few people can afford braided cable for their solar/wind projects... a possible method to reduce common mode energy from lightning strikes would be to insure that, at least 10-20 feet of metallic conduit is used from the input of the solar controller out to the array. Ground at least one end of the conduit (right at the solar charger's input). You have the inductance of the cable reducing the current/voltage spike, and the capacitance of the conduit shunting energy from the wire pairs)... I used this technique (10'-20' of shielded cables then the rest unshielded as a distributed filter) to pass EMI tests on large voice mail systems.

    Also, when doing CE susceptibility testing (not lighting but simulation of radio transmitters, florescent fixtures lamp starting, brushed motors, ESD tests, etc.), I found that pretty much everything you do to reduce electro magnetic emissions, also, easily helped us to pass these tests too.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    yes i forgot it is being radiated and much of that is good advice though over some peoples ability to understand what you mean. it was revealed that the inverters are being zapped moreso than the controllers, so that means entry via the 110vac wiring. protection devices were being used so that means movs were used. most likely more of them should've been used as it is not just the hot and neutral that should have one. one from hot to ground and one from neutral to ground as well making 3 movs being used in a proper protective configuration. i like gas discharge tubes myself, but they are destroyed after one event making multiple events catastrophic. movs or gas discharge tubes can be and should be also placed on the input to a controller in the same manor as that of the inverter output. movs and capacitors and even fuses have been known to explode under the high potentials posed by lightning so keep the placements of these parts with this in mind.
    metal conduit is a good protective ground covering for the wires if you can maintain good electrical connections for its entire length and i recommend grounding only one end due to ground loops. shielding and grounding is about all that can be really effective in fighting emp. btw people, lightning rods are ineffective against emp. your arguement of opposing fields being setup in the frames may not be accurate as this depends on the attack angle of the emp. the pvs are angled towards the sky and lightning will come from the sky to the ground with the emp radiating in all directions around the bolt. this can mean it can be an angle that would become additive on the front and back of the frames making the problem twice as potent as would've normally been seen. this means one could be more susceptible from a distant strike in one direction as opposed to another in a different direction.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: PV Array Grounding/System Grounding
    BB wrote:

    Solar Panels with aluminum frames (frames must be electrically connected at each corner through welding or good rivets). Aluminum "circle" will tend create a counter rotating current which (in theory) will counteract an EMP field.

    Can you list a source for this version of EMP theory ?
    a) EMP is rotating
    b) A aluminum "circle" does anything.

    Do any panel mfgs allow for welding their frames ? (gas or arc)


    What used to be in the ARRL handbook, was at the entry point of a building, to make several loops in the cables, before they penetrate the wall. The loops, a foot or 2 in diameter, seemed to act as an inductor to a lighting strike, and prevent the main arc from going into the building itself. It's likely there would still be enough induced current to fry electronics, but it's supposed to keep most of the plasma outdoors.

    What's interesting to me is that most of the damage that occurs to the solar systems comes from lightning strikes at least a mile away, rare to see a direct hit on a system.

    A mile away ? That must be quite a strike to induce that much current. I could believe it being conducted along the city AC mains, into inverters, and frying them there, but induced current from a strike a mile away seems unlikely in my book. If ordinary household electronics last thru strikes a mile away, big beefy inverters, known to have lots of rooftop wires attached, should not be such delicate creatures.
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

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

  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: PV Array Grounding/System Grounding

    The Aluminum frame was a theory on my part... I could very well be all wet on that part.

    As I was typing about the aluminum frames, I was also wondering if placing a ground screen (metallic mesh or even corrugated roofing) under the solar panels would be a help.

    I don't know much about EMP (spectrum, field details, etc.)... But from EMI testing conducted radiation from power lines, data cables, etc., was not much of a problem above a couple hundred MHz. Above those frequencies, almost all of the radiated EMI issues were the result of slot antenna type leaks (... And those were resolved by providing grounding points (metal fingers, screws, metallic gasketing, etc.) at a minimum of a 1/4 wave length (and closer) of the frequencies we were trying to kill (from a few hundred MHz to over GHz frequencies). Our testing was only done on ranges at 10-30 meters with horizontal and vertical polarization--the antennas could be raised and lowered about 3 meters (as I remember).

    When laying out interface cables for testing, we would lay them up in figure eights to help reduce the chances of failing tests. I have forgotten the exact requirements but we were required to put 10-20 feet of cable on each I/O port (at a minimum).

    Interesting about looping the cables prior to entering a building to help reduce plasma entry. My suggestions were intended to help reduce the effects of nearby strikes.

    I had looked into gas discharge tubes before but the testing (that I had read years ago) seemed to indicate that there were not very useful when compared to other devices (MOVs, spark gaps, silicon devices, etc.).

    I guess a direct strike has a 100 to something like a 1,000 amps of current. Shunting the current away (using lightning rods, spark gaps, MOVs and such) to ground and limiting the pulse with inductance (as opposed to fusing/breakers which seem to not be fast enough or able to truly block the voltage--just would arc instead) seem to be about the best one can do (short of placing everything in a metal box and well shielded cables--but without wires, power distribution/solar is not very useful).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    bill,
    i don't think the screening would help on the pv backs. emp is a perplexing problem that goes beyond the norm of rf shielding, but it is along those lines. note i semi-did agree with not using the gas discharge tubes because of their one time usage, but on the otherhand they do allow an exellent responce time at lower voltages. as you know many rf cables are protected via gas discharge. i did conclude before that it's shielding and grounding that works against emp with all other things being iffy in their protections.
    mike,
    i never said emp is rotating. i said it is radiating like a circle from the bolt with the bolt being in the center. it is a circle that gets bigger around with time. light also radiates in a circle from the bolt and if you can see it, you can not only get the emp, you already got it. whether it is strong enough to do anything is another matter, but it is there. as with the diminished light from a far away bolt, so is the emp diminished in strength. emp 1 mile away is very strong to some items under the right circumstances and yes it is possible that enough energy travelled the utility lines to blow out something as again that is only a wire and they can act like antennas and/or rf cables.
    in general, rf type reductions by looping the wire does work, but limitedly at certain frequencies. remember the lightning is giving off every frequency in the spectrum above dc with the amplitude diminishing as you go up. it will be additive at many frequencies higher up and reasonant even at some of the lower ones. remember it is only just a coil of wire so it could act like the coil being coupled from another coil in a transformer as it knows not the source of this strong radiation it picks up.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: PV Array Grounding/System Grounding

    A reason that I was wondering if a ground screen behind solar panels would work/help--generally, if you can keep your physical dimensions to less than a 1/4 wavelength--you have a much less effective radiator or antenna...

    So, if the general experience that cables don't conduct very well above 200 MHz and wavelength(meters=300/freq-in-MHz

    Wave Length = 300/200MHz = 1.5 meters

    1/4 wave length = 1.5M * 1/4 = 0.375 meters or ~ 15 inches...

    So, if you have an EMP event, and a ground screen within 15 inches (or less) of a solar panel, it would seem that it would be difficult to inductively couple into your cabling (same thing if your cabling is routed over a ground plane/grounded structure).

    If there is capacitive coupling to the panel--that would be a common mode event--generally most devices are much less sensitive to common mode voltage/current (plus, the ground screen would act like a capacitor plate too--and should reduce the voltage rise on the solar panel--I would think.

    The above discussion shows one reason I am not so sure that I would ground (for example) the negative lead on the input of a solar array (or wind generator)... The one grounded lead--if it works well, stays near zero volts, while the other (positive in this example) lead increases in voltage in an uncontrolled manner.

    In general, it is not recommended to convert a balanced/differential circuit into a single ended circuit by grounding one lead (example +/- output un-ground referenced output of a transformer vs that same transformer with one lead grounded is now a ground and signal lead). Generally balanced (and shielded and balanced) is much less susceptible to noise reception.

    But since you still have energy to dissipate--doing things like MOV from both +/- leads to ground at the input of a solar controller would seem to be a good idea.

    As an example of a commercial grid tied inverter--from what I can see/read, it appears that the solar input to a Xantrex GT 3.x inverter is a balanced (non-ground referenced) input... However, they do have a negative ground (positive ground option available too) in the inverter through a 1 amp fuse... If there is a ground short on the Solar +/- inputs that blows the fuse--then the inverter detects that the fuse is blown--but there does not appear to be a hard ground (as defined by NEC) on the solar panel inputs. So, other than for detecting failed insulation (or possibly surges from lightning) in your solar array, I am not sure that there is anything specific to protect (in terms of shunting energy) against lighting strikes.

    It appears that the solar panel ground through a 1 amp fuse is just to detect major solar panel current leakage to ground.

    Of course, each vendor will have their own solutions (as long they comply with code).

    Lastly--for true "Faraday" type shielding--a ground connection is not required at all. Just the simple solid metal ball (or screen) is enough to prevent any major energy leakage between the inside and outside of the sphere (up to the point at which you have enough energy to actually punch a hole in the metal shield).

    But, of course, we live life in the real world and even if we can put our solar charger/inverters/batteries in a metal box--there are still those pesky wires that need to go to the outside world.

    I have read several article by HAM radio guys--they have the worst of it--antennas that need to be in open air and isolated wrt ground connected to expensive radios in the house... And the few articles I have read--the authors sometimes had exactly oposite solutions for the same problem.

    Niel, as a HAM--have you seen any good authors/articles that people who live in lightning country seem to agree with?

    One of the most difficult problems I have seen with Solar vs HAM solutions is that Solar Panels are DC and HAM's deal with AC (RF). With DC it is very difficult to stop current/voltage surges as you need a solid copper connection from the panels to the chargers/batteries/inverter inputs. With AC, a good transformer and/or capacitive coupling can block DC, common mode energy and limit the amount of energy transfered pretty easily.

    DLWindsun reports from his/her experiences:
    Almost always the inverter takes the most damage, if the inverter takes a hit the charge control normally also gets hit about 60% of the time, it does not seem to effect anything else.

    Have you seen the problem with inverter damage being on the DC input side or the AC output side?

    Also, with the Charge Controllers--have you seen any difference between a simple on/off / PWM controller vs a MPPT type converter (like MX 60 or Xantrex GT series)?

    In theory, a MPPT controller actually changes the DC to AC and back to DC again... For standard switch mode power supplies (like used in computers), it is very easy to isolate the AC / DC stage with a transformer and opto isolators (provides 2,000 or more of DC isolation between AC/DC sections of the supplies).

    I would guess that a Grid Tie type inverter would use similar technologies (isolation transformer and opto-isolators)--whereas a MX 60 MPPT charge controller could make use of these types of protection--might not if there was no NEC or other design requirements to provide input to output isolation.

    It would be interesting if there was any real data on what the failures are and if there are any advantages seen between differing design techniques.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: PV Array Grounding/System Grounding

    bill,
    i'm thinking the screen ground may work up to the point you mention in 1/4 wavelenths (more accurately i believe this is 1/8th wavelenth, but don't quote me), but only in one direction. we don't have that totally shielded box with it only in the rear and we certainly aren't going to block any part of the sun in our hopes of the shield working. reminds me of a saying one of my former bosses used to say: "nothing difficult is ever easy". that's why we do everything by the kiss method, "keep it simple stupid", then we hope for the best.
    as to grounding one and not the other, well we can't short them out or it's useless so one should go to ground. now the other can still be linked to ground, but say at a higher voltage and/or frequency leaving the lead open under the operating parameters of the device in question with some leadway. that leadway is very often enough to blow sensitive equipment and we didn't even get into the reaction time of said protective devices. like i have said to not do it to the one lead i have seen the result of and i won't go without something going to ground if i can help it. balance i don't worry about as it will never be totally balanced no matter what you do.
    now don't get me wrong here as i am not against protection devices as every little bit helps. i am assuming that the inverters that had no controller damage was encroached via the 115vac side. i'll also venture that some that inverters sent the destroying blow to the controllers. now some of the controllers were destroyed without the inverter being hit and that i attribute to long wire lengths that act as pickups. we also don't know the percentage of these controllers that took out the inverters, but we know it is probable that some did. i can't totally discount the areas between the inverter and the controller, but the odds are lessened as this is usually much shorter in length.
    as to articles i can't say i recall very many as they were sparce, but don't think the rf ac is easier to deal with. it is not easy to isolate the leads to the point that emp couldn't jump it or in layman's terms, arc. voltages with transmitted rf can easily exceed those of the utility voltages going into your house, especially in cases of impedance mismatching so this makes protection of the circuits for reception along the same antenna and cables terribly difficult. a few volts can blow those receiver circuits, so what are the chances of protection from an emp event? same as with the discussion here, you do little things and hope for the best. the best solution is isolation for either and that's not always possible. disconnecting the antenna and putting that end to ground cures most of the emp events from destroying a radio. more are saved by disconnecting from the utility power source if used. very few radios are ever destroyed when these 2 things are done and we can say the same for solar equipment, but when you do that it's making it inoperable and you may not be there to do it in time or at all. i've said enough for now as i'm no typist it has taken a long time to type this book. :roll:
  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: PV Array Grounding/System Grounding

    Niel,

    You are correct, that I should make clear that it is LESS THAN 1/4 WAVE LENGTH... 1/8 wave length is certainly much safer. 15" was 1/4 wave length at 200 MHz (in air)--and 200 MHz which was just near the effective radiator for random I/O cables out the back of a computer. 7" or less would be much better.

    I would be interested to hear more about specific details of items destroyed/saved from near by lightning hits. Problem, of course, is that house wiring has changed so much over 6 decades (from old knob and tube, through conduit, and into ROMEX) that it would probably be difficult to generalize.

    Regarding typing--I still thank God to this day that I took that high school typing class (on manual typewriters) back 35 years ago. Who knew?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • DLwindsun
    DLwindsun Administrators Posts: 34 admin
    Re: PV Array Grounding/System Grounding
    A mile away ? That must be quite a strike to induce that much current. I could believe it being conducted along the city AC mains, into inverters, and frying them there, but induced current from a strike a mile away seems unlikely in my book. If ordinary household electronics last thru strikes a mile away, big beefy inverters, known to have lots of rooftop wires attached, should not be such delicate creatures.

    I have to go with what I've seen and heard in the field, it may not be exactly 1 mile away we have no way to test the distance but most of the strikes that cause damage are very long distances away.
    Have you seen the problem with inverter damage being on the DC input side or the AC output side?

    The damage most of the time is on the ac side of the inverters but on closer strikes it can blow the FETS.
    FYI, the average cost to repair an inverter struck by lightning is around $300-800. In some cases the damage is so severe it's not worth repairing the inverter.(Use your home owners insurance on all your solar equipment, make sure they will cover it)
    Also, with the Charge Controllers--have you seen any difference between a simple on/off / PWM controller vs a MPPT type converter (like MX 60 or Xantrex GT series)?

    Lightning seems to effect the cheaper PWM controls more often than the MPPT controls. It is actually pretty rare to see a MPPT control take lightning damage.

    By Xantrex GT series I'm assuming you mean grid tie inverter?
    I've seen maybe 1-2 grid tie inverters take lightning damage it's very rare. Maybe we've been lucky.

  • BB.
    BB. Super Moderators, Administrators Posts: 33,422 admin
    Re: PV Array Grounding/System Grounding

    Regarding how far away... The old 5 second rule between the flash and the thunder (I am sure you know--but for others).

    Also, it appears that successive lightning strikes seem to be around 2-3 miles apart:

    http://www.usatoday.com/weather/wlight1.htm
    The time between seeing a lightning flash and hearing the thunder it produces is a rough guide to how far away the lightning was. Normally, thunder can be heard up to 10 miles from the lightning that makes it. Lightning heats the air around it to as much as 60,000 degrees, producing sound waves by the quick expansion of the heated air. Since light travels at 186,000 miles per second, you see the lightning the instant it flashes. But sound, including thunder, travels about a mile in five seconds near the ground. If 15 seconds elapse between seeing a lightning bolt and hearing its thunder, the lightning was about three miles away. Lightning closer than about three miles away is a warning to take shelter immediately. Successive lightning strikes are often two to three miles apart. If the first stroke is three miles away, the next one could hit you.

    For damage from strikes so far apart--there would be two methods of injecting enough energy to cause damage... One would be long wires up in the air (solar panels on second stories, wires not wrapped close together but strung out in large loops)... Or, would be from ground currents (say panels are 100's from home, panels grounded--and say negative ground at panels--then 100's away the charge controller negative possibly grounded too--causes a large surge between the + and - leads blowing the controller's input).

    It is interesting to hear that Grid Tied inverters and the MPPT solar chargers (like MX 60) seem to be less damaged by lightning... My theory is that the these can use (and I am sure that Grid Tie would be required by UL/NEC requirements) isolation transformers with the ability to withstand >2,000 volts from AC to DC (common mode)... With that high of holdoff--it is much easier to use MOV's/etc. to shunt the energy to ground before damage can occur).

    Also, the MPPT and Grid Tie input stages (on the MX 60--a high end example--and Xantrex GT series) are able with withstand much higher differential voltages (MX 60 <150 vdc, the GT <600 vdc)--again, will better withstand lower energy strikes and easier to (using surge suppressors) to dump energy away before obvious damage occurs).

    Regarding the inverters... Are you seeing issues with Mod Sine or the "True" Sine Wave types... The Mod Sine Wave generally do not have real isolation transformers (heavy, expensive)... From the repair costs--it sounds like those that are getting repaired are the true sine wave types (of course--who would repair a $30 300 watt mod sine wave inverter)...

    Are the inverters that are getting toasted more often (or less often) behind the less expensive PWM type charge controllers (my thesis being that lightning strikes have an easier time getting through a PWM type controller).

    It is interesting that a Grid Tie inverter seems that it may be more resistant to lightning and that (again a thesis point) that true sine inverters with good isolation transformers (designed and, hopefully listed, with UL/NRTL agencies) have similar functional blocks (solar panel - switchmode power supply with isolation transformer - grid tie; vs; solar panel - mppt charge controller/switchmode ps -- battery -- another switchmode ps - grid/local ac distribution). Is having a battery bank grounded (or less than ideally grounded, or sharing a ground with other functions) causing issues?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset