Ungrounded vs. grounded in the developing world

alexjalexj Registered Users Posts: 25 ✭✭

I would like to share my experience with ungrounded, small, battery based systems in the developing world and get input from others. When I say small I mean under 1kW PV, max inverter size 1200VA. These systems are 12 or 24V DC with AC at 220V. No DC or AC ground. Always mounted on a metallic pole mount with no additional equipment ground. I previously worked in the US and upon arrival in Peru was quite concerned that few small offgrid systems were being grounded, but there is ample evidence that these systems are working just fine.

After working here for two years, I haven’t seen any issues that I can directly link to the fact that these types of systems are ungrounded - no issues with lighting (other than a direct strike to a pole mount, but everyone was fine), no issues with transient overvoltages, no electrical accidents, and no problems with the system functioning improperly due to being ungrounded. This is not to say that I am completely convinced that this is the right way to be doing things, thus I would like to discuss the logic that I see behind not grounding a system and hear the opinions of others on the subject.

Reasons to not ground a system in this context:

· In rural Peru it would be extremely rare to see metallic conduit or boxes – everything is run in plastic. Therefore an equipment grounding system wouldn’t be effective.

· If a system is ungrounded, it requires a short between both wires in a circuit to have a fault, rather than just one wire to ground. Thus, it is extremely difficult for a user to be shocked in the case of a fault – unless they were completing the circuit between the two legs – as there is no return path through ground.

· The inverter itself is current limited and relatively sensitive to shorts. If there is a short between the two legs, the inverter should quickly identify it and shut down.

· It keeps costs down.


Any input on the following would be much appreciated:

1. Setting aside electrical codes, what reasons technically can you think of for grounding a system like this on either the AC or DC side?

2. Are there any other advantages to not grounding a system that you are aware of?

3.  Once again setting aside electrical codes, with a small inverter – which cannot supply sufficient current to exceed the ampacity of the wiring on any circuit on the AC side – is it necessary to install OCPDs on each AC output circuit? In my thinking, an inverter will automatically shutdown in the case of a short or in the case of any electrical accident as quickly or almost as quickly as an OCPD with a delay.

4. I have read quite a few discussions regarding lightning and grounding module frames. I am not trying to reignite this discussion, but it seems like with with a metallic pole mount sunk into the ground that it should not necessarily be a significant lightning attraction, but nor would it or a proper grounding system likely protect the other system equipment in the case of a direct strike. Therefore a ground rode and grounding system for the pole mount likely doesn't have any a advantages. Thoughts?

Thanks for your input!

Comments

  • BB.BB. Super Moderators, Administrators Posts: 31,443 admin
    Grounding a power system can be divided into 3 or 4 different reasons--Sometimes the reasons can be in direct conflict with each other.

    Reason 1. To allow the use of single pole breakers/fuses in wiring over current protection schemes. For smaller systems, say 14 AWG and 15 Amp maximum current and you use 14 AWG throughout the AC system, a single fuse/breaker will prevent overcurrent from overheating the wiring.

    One failure mode can be if you have, for example a 100 Amp AC line, 14 awg and 15 amp breakers to branch circuits (one pole only), and a floating/isolated power system (transformer output). If there is a short between the 100 amp bus and ground, then the other side of the circuit is now "hot" with respect to ground. And there is no breaker/fuse on the "return" side of the wiring, so the 14 AWG wiring on the unfused side will now have 100 amps of current available.

    When you ground reference (bond neutral to ground), The short of 100 amp bus to ground will pop fuse/breaker. And the "neutral" side of the wiring is always at ground potential, so even if you have a short from neutral to ground, there is no short circuit current flow and no breaker/fuse is needed.

    For truly floating power system... There needs to be "something" that ensures the floating output is alway floating. For a test bench, the isolation transformer is tested/certified once a month that it is still isolated output. For a ship (at least in olden days), they had floating AC power output and two lights... One from Hot 1 to ground, and a second from Hot 2 to ground. When the system is "floating", both lights are dimily lit. If there is a short from Hotx to ground, then that light glows brightly and the other goes out (fault). Ships use floating power system to avoid current flowing through hull (and prop bearing, to ocean, etc.) to prevent electrolysis (rust/erosion of metal) and avoid current flow through bearings (damage to races).

    You can get around the safety issue of floating vs grounded neutral systems by using GFI breakers and outlets. A floating power system should use double pole breakers--If there is a short circuit somewhere that pop Hot 1 breaker, then both Hot 1 and Hot 2 are turned off and the branch circuit is (relatively) safe for an end user to touch (avoid electrocution). With a GFI, since they trip if Hot 1 current does not equal Hot 2 current flow (or Hot vs Neutral), the GFI trips and protects the end user.

    Reason 2. As touched above, avoid electrolysis. Probably more important with DC power vs AC. Grounding the system makes a choice vs what is eroded (sacrificial anodes vs actual structure, equipment, etc.). For example, telephone systems are positive grounded systems (48 volts nominal) and steel natural gas piping is insulated and charged in the ground. At the house a isolating coupling is used to avoid "shorting out" the cathodic protection system.

    Reason 3. There are 120 VAC systems that need to be ground referenced for proper operation. For example, the automatic spark ignition on a natural gas stove uses the bonded neutral to help detect the current flow through the spark ignitor and the flame (as I recall, the gas flame actually rectifies current flow). A reversed Neutral/Hot wiring to your stove and the spark ignition system becomes unreliable.

    Another are some standard fluorescent tube fixtures need the fixture grounded to start reliably (especially in cold locations). The grounded fixture/reflector assembly help start the arc in the tube.

    Reason 4. Lightning... Obviously here, the idea is to get the lightning energy to ground as quick as possible over path(s) designed to shunt this energy.  Metal conduit, ground bonding a neutral, etc. all can help.

    Note that Lightning ground bonding is an "Radio Frequency" current and the physical wiring connections can affect current flow (sharp corners can cause lighting to "leave" the ground conductor to find another path because of the higher impedance of the right angle turn--Vs a nice rounded corner--like 18" radius bends). Also in a building, the lightning tends to go from the center of a building to the outside walls (skin effect do to RF nature of lightning current flow).

    When distributing AC power from a single source (like from main panel in home to well pump, and/or to outbuildings/garage/etc.) you have "earth ground" and a bonded ground. Earth grounding can be upwards of 25 Ohms from ground rod to earth. Great for lightning grounding...Not so good for safety grounding. So I highly suggest that you run (for example) local ground rods for your remote structures/buildings/solar array framework/etc for lighting control. And ground bond all the remote ground rods together with 6 AWG cable--The 6 AWG cable carries "fault current" and will trip the breaker/fuse back at "home panel" (i.e., touch hot to local ground rode at 120 VAC, and you get less than 5 amps of current flow. With 6 AWG from outbuilding ground rod back to main panel ground rod at home (ground bonded neutral/green wire/ground rod/cold water pipe, etc.) you now have enough current to trip the branch circuit breaker in the main panel (there are issues where long wire runs have too much resistance to trip--Hundreds of feet of 14 AWG cable and 15 amp breaker may not pull >15-30+ amps for a quick trip).

    Anyway, Small systems tend not to be grounded (less than 1,500 Watt AC inverter, genset, etc.)... And larger systems (3.5 kW and larger) seem to come "pre wired" for Neutral/Green wire grounding in the energy source (inside the AC inverter/genset/etc.).

    Many times, these "defaults" can cause issues when you have multiple power sources. For example, you have your main panel always ground bonded between safety ground and neutral. Your 3.5 kWatt genset will probably need to have the neutral/green wire bond lifted inside the genset or you will have "issues". You have a parallel current path for Neutral and Green wire cables (green wire/safety ground should never have current flow in normal operation). Because the genset and the main panel (of course, we are talking about North American code with 120/240 VAC split phase systems) have white/neutral to green/ground wire, this parallel current flow will cause GFI protected circuits (such as the genset output) to fault. Lifting the neutral/ground bond in the genset prevents the problem.

    And there are even more issues with RVs and Boats which can connect to a ground bonded "shore power" source. And can also have gensets/AC inverters for power when they are away from shore power connections. In RVs and Boats, there will (or at least should be) a switched Neutral/Chassis ground bond (off when shore powered, on when on the road and on board genset/inverter providing power).

    I hope this makes sense.... Details matter with grounding (your energy needs, code, local practices, etc.). Have not talked about other ground issues like static discharge for structures (a "floating structure" like antenna tower without grounding can build up static charge--Nominally ~100-300 volts per meter natural earth electric field to 10,000s of volts when a thunderstorm cell goes overhead. Cars, planes, etc. can build up static charge from airflow, rubber tires, etc..

    Grounding and ground plane for best practices antenna design (radio, other RF applications).

    Does this help (or confuse)?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • BB.BB. Super Moderators, Administrators Posts: 31,443 admin
    Some links if you are interested:
    Re: Working Thread for Solar Beginner Post/FAQ

    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 FAQ:

    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
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Dave AngeliniDave Angelini Solar Expert Posts: 5,807 ✭✭✭✭✭
    Nice Job Bill !
    I am curious, the OP said,

    "Reasons to not ground a system in this context:

    · In rural Peru it would be extremely rare to see metallic conduit or boxes – everything is run in plastic. Therefore an equipment grounding system wouldn’t be effective."

    Why would you say this?


    "we go where power lines don't" Sierra Mountains near Mariposa/Yosemite CA
     http://members.sti.net/offgridsolar/
    E-mail [email protected]

  • EstragonEstragon Registered Users Posts: 4,494 ✭✭✭✭✭
    Ungrounded (floating with respect to earth ground) low voltage DC systems are common (vehicles, boats, etc.).  Voltage isn't considered to be as much of an issue at 12-24v, so less need for safety grounding.  Batteries can put out big currents though, and DC can sustain arcs (fire risk), making proper OCP essential.  Higher voltage series strings of panels are a different matter.

    Higher voltage AC can also float, but then there's a risk a conductor at significant voltage with respect to ground chafes or otherwise contacts metal equipment chassis.  In a grounded system, a ground wire completes the circuit, and breaker opens.

    Those of us of a certain age may remember a time when most N.A. AC house wiring was ungrounded.  Lots of plugs had the round ground pin removed so they'd fit in a two prong outlet.  Like almost everyone else, I survived.  Grounded (and later, polarized) plugs are safer though, which is why they're generally now required.

    There are some applications (eg audio/video) where grounding can be a significant issue.

    Higher voltage = higher electrocution risk.  In the UK, higher voltage AC generally has a switch/breaker at point of use.

    A metal pole sunk in the ground is probably ok for lightning.  Nothing is certain with lightning, put I don't see where separate ground wire adds much lightning protection.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • BB.BB. Super Moderators, Administrators Posts: 31,443 admin
    The local ground rod + Surge Suppressors can help:

    http://www.midnitesolar.com/products.php?menuItem=products&productCat_ID=23&productCatName=Surge%20Protection%20Devices

    If you click on one of the devices above, Midnite has some white papers and videos talking about their design goals/device operation.

    Our host sells them here:

    https://www.solar-electric.com/search/?q=surge

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • alexjalexj Registered Users Posts: 25 ✭✭
    edited August 2019 #7
    Hi Bill,

    As always your input is much appreciated. I think what you posted helps, but I am more than anything curious about the specific situation that is common here. I think that I could do a better job of detailing a common offgrid install or building here:

    Description of the building: Four room rural building built out of adobe with a corrugated metal roof. No prexisting electrical system.
    Loads: AC load needs - lighting, laptop, printer, TV, projector, DVD player, radio, and cell phone charging. Varying hours of usage (assume system is properly designed).
    Modules: 2 x 325W in series.
    Array: Single pole mount
    Charge controller: MPPT
    Inverter: 24V 800W pure sine wave
    Batteries: 2x12V 240Ah in series
    Protection: Batteries located in wooden battery box. CC, inverter, terminal blocks, breakers etc in large metal enclosure.
    Wires: All wires sized properly. Run in PVC conduit with PVC junction boxes/fixtures externally on wall surface as this is the local custom.
    Grounding: None- no AC or DC grounding system. No equipment grounding.
    OCPD: On incoming solar, between battery and charge controller, between battery and inverter. No OCPD on AC circuits.

    Just to reiterate my questions from above - any input on the following would be much appreciated:

    1. Setting aside electrical codes, what reasons technically can you think of for grounding a system like this on either the AC or DC side?

    2. Are there any other advantages to not grounding a system like this that you are aware of?

    3.  Once again setting aside electrical codes, with a small inverter – which cannot supply sufficient current to exceed the ampacity of the wiring on any circuit on the AC side – do you think it is necessary to install OCPDs on each AC output circuit? One single OCPD for all AC output? None? In my thinking, an inverter will automatically shutdown in the case of a short or in the case of any electrical accident as quickly or almost as quickly as an OCPD with a delay.

    4. I have read quite a few discussions regarding lightning and grounding module frames. I am not trying to reignite this discussion, but it seems like with with a metallic pole mount sunk into the ground that it should not necessarily be a significant lightning attraction, but nor would it or a proper grounding system likely protect the other system equipment in the case of a direct strike. Therefore a ground rod and grounding system for the pole mount likely doesn't have any a advantages. Thoughts?

    Thanks!



  • mike95490mike95490 Solar Expert Posts: 9,247 ✭✭✭✭✭
    alexj said
    but it seems like with with a metallic pole mount sunk into the ground that it should not necessarily be a significant lightning attraction, but nor would it or a proper grounding system likely protect the other system equipment in the case of a direct strike. Therefore a ground rod and grounding system for the pole mount likely doesn't have any a advantages. Thoughts?

    A golfers umbrella has enough metal to ruin many a day on the greens.   Don't fool yourself.
    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: 31,443 admin
    edited August 2019 #9
    Alex,

    I will try some quick/short answers here... The battery negative bus should be connected via green wire to the metal electrical enclosure, any electrical equipment chassis ground (metal cases). This so that if there is a short inside the box, it will trip the fuse/breaker from the battery bank and not "energize" the box. 24 volts is generally considered non-shocking, but it is possible with sweaty/wet hands to get a shock if touching 24 volts. Even 12 volts is nasty if shorted with jewelry/rings/etc. -- Battery connections should be protected against people or tools form contacting (think arc welder).

    Note that the battery ground can simply be the negative battery cable "bonded" to your metal equipment enclosure).

    Most all PSW inverters are fully isolated floating outputs (as always, check the manual, almost always is not the same as all). Most (not all?) utility power is ground referenced at the pole transformer... If you had Hot/Neutral for utility power, any switches/breakers would be installed in the "Hot" lead.

    In your case, you could just "float" both leads from the inverter AC out to your loads. No circuit breaker is needed (usually CB are much slower than anything the inverter can do). If you wanted "safer", you could use GFI breaker or outlet to your AC loads. If you have two circuits, one for installed lighting, the other for your plug in loads, you would only need to install the GFI on the plug in loads. Assumes installed lighting would not have accidental contact with humans. If you have portable lighting, I would suggest two GFI outlets (or breakers). One to lights, second to balance of loads. That way, a GFI fault on the loads does not plunge the building into darkness by turning off loads with single GFI for everything.

    Using a ground rod (or buried copper ground plate) connected to the battery negative post is helpful in lightning prone area for safety (you don't want lightning getting into the building with people). No "functional" need to drive a ground rod (DC and AC will work just fine for you here).

    If you dry/rocky/bedrock in area--Ground Rods are pretty near useless for lightning control. People have tried all sorts of methods... Chemicals, and even wetting the ground with bucket/tank of water, etc.

    You don't really need an OCPD from solar array to charge controller, unless you have 3 or more parallel solar strings (you have two). Some folks use a Circuit breaker anyway as a handy on/off switch for servicing/debugging solar charging system or install a breaker/combiner box for future expansion.

    For lightning "interception", there is the "300 foot diameter/150 foot radius" rolling ball model... anything that touches the rolling ball, is higher risk for string strike... Here is a quick diagram of what a single tall lighting rod can protect (not as much as you think):

    https://www.lbagroup.com/products/lightning-protection-dissipaters-portable-masts-rods

    -Bill


    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • softdownsoftdown Solar Expert Posts: 3,496 ✭✭✭✭
    mike95490 said:
    alexj said
    but it seems like with with a metallic pole mount sunk into the ground that it should not necessarily be a significant lightning attraction, but nor would it or a proper grounding system likely protect the other system equipment in the case of a direct strike. Therefore a ground rod and grounding system for the pole mount likely doesn't have any a advantages. Thoughts?

    A golfers umbrella has enough metal to ruin many a day on the greens.   Don't fool yourself.
    Florida Man needs to make synthetic ribs. Floridians get whacked on the golf course with regularity. 
    First Bank:16 180 watt Grape Solar with  FM80 controller and 3648 Inverter....Fullriver 8D AGM solar batteries. Second Bank/MacGyver Special: 10 165(?) watt BP Solar with Renogy MPPT 40A controller/ and Xantrex C-35 PWM controller/ and Morningstar PWM controller...Cotek 24V PSW inverter....forklift and diesel locomotive batteries
  • Dave AngeliniDave Angelini Solar Expert Posts: 5,807 ✭✭✭✭✭
    edited August 2019 #11
    You really have to balance if the work is worth the loss. A couple solar panels ungrounded on a grounded pole are pretty useless in protecting anything electrical. But in this case the cost of a loss from a nearby strike may be acceptable. It sounds like this is what you want to hear and it has merit.  Keep doing what you are doing and maybe use the ground rod to keep a bolt away from people as much as you can.

    After some time reevaluate. This plan is not good for a large system.
     It is interesting how a small system gets big very quickly, sometimes. 
    "we go where power lines don't" Sierra Mountains near Mariposa/Yosemite CA
     http://members.sti.net/offgridsolar/
    E-mail [email protected]

  • alexjalexj Registered Users Posts: 25 ✭✭
    edited August 2019 #12
    Thanks Bill and Dave for your input. I definitely understand that this this design is not suitable for a system that is very much larger - I spent more time working with larger offgrid systems, but always in the US.This forum is great because people have experience and are capable of thinking outside of codes and really exploring technically why something should be done or why not. This is vital in places like Peru as cost is a very real constraint and there are many companies, NGOs, and government institutions working in the field installing systems in many different ways with limited to no oversight. It is the wild west, but the systems don't function as poorly as one might expect.
    I will explore equipment here for grounding in the extremely rocky soil. I think that burying a ground rod 2.5 feet or so is going to be more feasible than driving one 8 feet down. And see what I can do with regard to a GFI breaker.
    Cheers!
  • EstragonEstragon Registered Users Posts: 4,494 ✭✭✭✭✭
    I used plates something like this here in the rocky Canadian shield:
    https://www.homedepot.com/p/ERITECH-7-5-in-x-38-5-in-Copper-Grounding-Plate-UGP738P5/202194160

    Tied a pair of plates together, roughtly 12x24" each, about 2' apart and 2' deep in a somewhat depressed area (most likely to stay moist).

    Another possibility:
    https://en.m.wikipedia.org/wiki/Ufer_ground

    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • Wheelman55Wheelman55 Registered Users Posts: 135 ✭✭✭
    edited August 2019 #14
    Interesting read from Erico. I used a chemical ground rod from Harger, similar to what’s on page 26 in the document below.  My install site was in rocky, dry West TX. This solution is expensive. 

    https://hickslp.com/wp-content/uploads/2015/02/Grounding_Products_and_Systems.pdf


    Building Off-Grid in Terlingua, TX
    14 CS 370 watt modules. HZLA horizontal tracker. Schneider: XW6048, Mini PDP, MPPT 80-600, SCP. 1 Discover AES 48 volt LiFePO4 battery 130 ah
  • jonrjonr Solar Expert Posts: 1,384 ✭✭✭✭
    edited August 2019 #15
    I agree with alexj that a small,  ungrounded high voltage system is less likely to electrocute someone.  And that an inverter is an OCPD.  Regarding lightning, I expect that grounded systems cause more problems than they solve.    Use a separate system (up high rod direct down to ground) for lighting protection - and SPDs for non-direct lightning.


    I am available for custom hardware/firmware development

  • NikiNiki Registered Users Posts: 21 ✭✭
    I agree grounding a “floating system” I would think actually is worse as it creates the potential for a shock or overcurrent fault.  In UPS Systems the battery string many times is ungrounded even up to 120 VDC meaning no potential current path to ground exists.  Once the negative battery buss is grounded then a high current fault can occur if any battery in the string is accidentally grounded causing a very big spark to happen.  However all metal work and cabinets should be grounded due to lighting potential.  
  • Dave AngeliniDave Angelini Solar Expert Posts: 5,807 ✭✭✭✭✭
    Go back and read them all again. One of the big mistakes on the internet is responding to the last post as if it is more refined.
    "we go where power lines don't" Sierra Mountains near Mariposa/Yosemite CA
     http://members.sti.net/offgridsolar/
    E-mail [email protected]

  • alexjalexj Registered Users Posts: 25 ✭✭
    edited August 2019 #18
    Hi Dave,
    I am confused by your last post. I read all of the posts here and had already previously read the links that Bill put forth - they were part of the reason for this post as there was quite a bit of dissension in those threads. Can you explain further what you think that I am missing here?
    Thanks!
  • Dave AngeliniDave Angelini Solar Expert Posts: 5,807 ✭✭✭✭✭
    Not meant for you Alexj   Cheers!  I just finished "Lost in the Andes" or "Alive" the old story of the plane crash between Chile and Argentina, a bit south from you in 1972.  Still an amazing story told by Nanado Parado.  
    "we go where power lines don't" Sierra Mountains near Mariposa/Yosemite CA
     http://members.sti.net/offgridsolar/
    E-mail [email protected]

Sign In or Register to comment.