Advanced Level Grounding Discussion for All-in-one Solar Power Systems
ฝัง
- เผยแพร่เมื่อ 19 ก.พ. 2022
- All in ones that I recommend: www.mobile-solarpower.com/all...
Does off-grid solar confuse you?
Check out my DIY friendly website for solar system packages and product recommendations, and so much more! www.mobile-solarpower.com
Join our DIY solar community!
#1 largest solar forum on the internet for beginners and professionals alike:
www.diysolarforum.com
Check out my best-selling, beginner-friendly 12V off-grid solar book (affiliate link):
amzn.to/2Aj4dX4
If DIY is not for you, but you love solar and need an offgrid system, check out EnergyPal! They will price out a system and get your house offgrid: energypal.com/diysolar/
My Solar Equipment Recommendations (Constantly updated! Check here first):
12V/48V Lithium Batteries: www.mobile-solarpower.com/sol...
Solar System Component Directory: www.mobile-solarpower.com/sol...
Complete 48V System Blueprint: www.mobile-solarpower.com/48v...
Plug-N-Play Systems: www.mobile-solarpower.com/ful...
Complete 48V System Kits: www.mobile-solarpower.com/com...
DIY Friendly Air Conditioner/ Heat Pumps: www.mobile-solarpower.com/sol...
My Favorite Online Stores for DIY Solar Products:
Signature Solar
Offgrid One-Stop-Shop. Best Value 48V LiFePO4, Victron and Offgrid Specific Heat Pumps:
www.signaturesolar.com/?ref=h...
Current Connected
SOK, Victron, Mr.Cool Heatpumps and High Quality Components:
currentconnected.com/?ref=wp
Ecoflow Delta
My favorite plug-n-play solar generator:
us.ecoflow.com/?aff=7
AmpereTime
Cheapest 12V batteries around:
amperetime.com/products/amper...
Renogy
A classic 12V solar store that has been around for ages!
renogy.sjv.io/n1VjXx
Rich Solar
Renogy's biggest competitor! Similar products, but at a better price:
richsolar.com/?ref=h-cvbzfahsek
Battery Hookup
Cheap cell deals
bit.ly/2mIxSqt
5% off code: diysolar
Contact Information:
I am NOT available for personal solar system consult! If you wish to contact me, this is my direct email: williamprowsediysolar@gmail.com
Join the forum at diysolarforum.com/ if you wish to hang out with myself and others and talk about solar
FTC Disclosure Statement and Disclaimers:
Every video includes some form of paid promotion or sponsorship. Some links on this youtube channel may be affiliate links. We may get paid if you buy something or take an action after clicking one of these. My videos are for educational purposes only. Information is subject to change/update at any time. Electricity is DANGEROUS and can kill. Be smart and use common sense :)
DIY Solar Power with Will Prowse is a participant in the Amazon Services LLC Associates Program,
An affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to Amazon.com
Just wanted to congratulate you. Even licensed electricians get this subject wrong. I’m a master electrician with 23 years experience in the field and I see this confusion with the earth ground way too often. Thanks for spreading the right information on your channel. Keep up the good work.
Great video!
HI william, i have a 5kw hybrid inverter. its charges my 48v battery and can put out 5000watt AC.
when i switch on the AC output, the frames of my solar panels start tot have around 50VAC power on them. i even can get a shock from this.
when the AC output is switch off. nothing is wrong and the pannels stay safe.
do you have any idea what i have to do to fix this? i relly hope you can help me. thanks
Same setup/scenario as above. I'm feeling very confused about things, going to disconnect my ground rod and also the ground/neutral bond wire at Service Panel (which is main panel, off grid no AC Input with a Chinese Hybrid Grid tie inverter) except for a sole earth ground connection at battery negative. I have more to read about positive grounding before dabbling.
One, That's how many ground/neutral bonds we need. With grid power input on an inverter or multiple inverters in this case. And that is already done at your main panel or meter panel. One ground location per service structure, now requiring two rods a minimum of 6 feet apart at the first means of disconnect. And in a separate structure (out building) sub panel or even a sub panel in your house never interrupt or divert your neutral or insulated ground from going back to the power company transformer. Meaning never give it an alternative path (ground bond to secondary ground rod) from getting back to the power company transformer. Because your sub panel in a separate structure does now require a ground rod also. And that's death waiting to happen if bonded. For example, your "incoming" ground, 4th wire, in any sub panel is a "ground fault current path" back to the transformer if the neutral gets broken or is shorted to either hot leg and, it must be insulated, not bare. Only your main service entrance wires from the meter to first means of disconnect can use an uninsulated ground. All this is another reason why separate auto transformers should never be used and should be banned if grid power is used on the inverter. In low frequency inverters with their own transformer they are protected if a neutral is lost, the inverter will shut down output AC power. But not separate auto transformers. Ever wonder why there is no breaker or disconnecting means on a neutral, death is why. Neutral IS A CURRENT CARRIER all the way back to the transformer and from there, all the way back to the generator at the power station. And ground rods just dissipate some of that current draw on its way back and keeps the voltage at zero between neutral and ground. It takes a completed loop to use electricity. If you interrupt that neutral path back to the transformer and send it to ground you can be shocked and everything attached to your ground bars in your sub panel will be energized. Be safe. Everyone has this false sense of safety when it comes to grounds, ground this, ground that, ground here, ground there when in fact a ground can be just as much deadly as it is safe. Mike Holts videos are great to watch he shows you exactly how people are being killed by not understanding neutral/ground bonds and ground fault current path. Great video and topic Will.
Dude! Thank you for this information AND especially directing us to Mike Holtz!!!
Mike Holz is a great resource. It's funny though how even Very knowledgeable people's perspective evolves, if you look at his videos over time. In earlier videos he is against multiple ground rods (not tied) and he is right. Later, there is a video where he and a Power company employee installed 2. 50 foot ground rods that were tied with 6 gauge copper and they tested it. They point out how low the ground resistance is and how much SAFER this setup makes the electrical system! Great resource.
Ya'll Take Care and be safe, John
This is a very complicated subject that even has many professional electricians scratching their heads. I would encourage more in depth coverage of the subject (that even I get lost on).
I totally agree I know nothing about grounds or grounding
It is not that difficult, this just common mode, engineer tends to use complicated words that makes a mess
Yes. More coverage needed. What to do with the ground because the wrong thing can be done in an attempt to do it right. What in the heck is the purpose of the ground on these inverters. The fact there isn’t even a slight suggestion in the manual makes it worse.
This is why right to repair is important, actual board schematics need to be public even just for safety of the end user.
I'm glad you're working through this. I'm totally comfortable with DC systems but have to work my brain pretty hard to figure out grounding issues during this brainstorming phase of an offgrid system i hope to build in the coming months.
Introduction: Each inverter has its own current flow path, each one needs a reference ground point and a short current path. Two of them are combined to make a split phase system which shared the neutral wire. In this split phase system, current flows from each hot wire of it's own inverter and return back itself through the neutral wire. The ground is for reference point and a short current flow path. If the voltages of two hot legs are different, there will be current flow from one inverter to the other, does not matter you have one or two ground-neutral bond(s).
1. For transfer mode (Inverters using power from grid to supply power to the main panel by built-in relays), ground-neutral bond should be at the first panel (after the meter). This was already done when you have electric meter installed. The inverters are using ground-neutral bond from grid, this is correct. Do not create any ground-neutral bond to any panel in your house.
2. For off-grid, the ground-neutral bond should be at the inverters (grid neutral wire should be disconnected). If in case there is a short, the current flow will be terminated at the inverters. Remember, ground-neutral bond is used for reference point and short current flow path. One ground-neutral bond will work, but two are better. Current loop occurs only if your neutral wires are so small and two inverters are mounted far apart (long distance). For conclusion, keep the ground-neutral bonds on all inverters. You cannot count all inverters have bonding as multiple points, it is still single point of bonding because they are together as one power source.
The neutral and ground are tied together and bonded to the transform on the pole. The neutral and ground are tied together and bonded to metal and connected to a ground rod at the main service disconnect feeding your house. The inverter would normally float the neutral ground connection on pass-through of utility power and bond them on an outage and switch internally to disconnect from utility power to inverter power (produced from the batteries) and also in the same switch tie the neutral to ground. This is why it's important to ground all metal cases on an inverter. It's great to see you diving so deep on this issue.
I made a video showing this ground /neutral issue last year on my channel. The neutral should only be bonded at the main panel but there is still an issue with these inverters and the only way this could be fixed in my thoughts is upon transfer the ground and neutral bond should change it's state at that point.
I was in contact with the MPP solar engineers and they even watched my video, but I couldn't get them to understand the issue.
The reason these are wired like this (I was told) is that these are shipped to many places with varying code requirements.
I am no electrician but I was in the dairy industry and when wiring a milking parlor all sub panels ground and neutral wires were to be separated. If they were not it caused a condition called stray voltage which harmed the animals. If I remember right the only place they are connected is at the main panel because all power eventually the power goes back to the power stations. We did have some cases where there was power coming back to the farm by the ground wire, then the power company would have to add more pole grounds. This probably don’t shed any light on the subject but that was my experience in grounds and neutrals.
Great explanation and demonstration, this is much better than a 'code-parrot' just repeating NEC ad-nauseum... You showing and talking through the thought process is extremely helpful to all viewers. Ground loops that close with that big of wire are not nearly as bad as ones with a long run between them, and would probably never have any issue in your lifetime, but for continuity(pun intended), following a standard is best. Thank you for your work.
After a lighting strike that went up my ground rod and fried my inverter I talked with a home power engineer and explained how the ground here has coleagy "water proof ash" 12 inches down and the rain sits on top of it and the lighting hit the ground and went up my ground rod and straight to the inverter. He said to disconnect it and only use the battery ground. It's been 28 years now and lots of lighting but no more problems.
Are you saying to ground the inverter chassis to the Negative of the battery?
@@coreydean9277 I wonder sounds kind of like what he's saying
I plan to use my home's slab as an ufer ground. I'll have multiple access points to bond any/all panels, etc to this single bond. I also plan to install several GEM trenches (ground enhancement material) to further diversify my grounding. So, in short, my grounding would not be a loop, but all bonded to the same grounding source. Ground-Neutral bonding should be as close to the "utility" as possible. So either at the outside main panel. I would think if the inverters feed directly to this panel, their own neutral-ground bonding would not be an issue. Oh, I also get to add in a 75 ft antenna tower into my fun....
Great video. I've been thinking on this for installation of two units to provide 50AMP power for my RV. The grounding has been my boggle, but I think in an RV if I were to bring the AC Input from the automatic transfer switch that the T-switch should handle the ground-neutral bonding for me as it does between the generator and shore power now. Still working on this (on and off as I have time), the total lack of documentation from the All-In-One manufacturers really makes it a pain to figure out how to execute safely.
This is an excellent article on grounding, with particular attention to lightning protection, a much neglected aspect of all types of solar pv installations.
It amazes me how often lightning isn’t addressed in grounding considerations. Most folks just float the whole thing, but I have metal vintage camper. It’ll light the whole thing up like a Christmas tree if I didn’t create my own grounding system (which I have).
Grounding in split phase systems can rapidly become confusing, particularly when there are subpanels in detached structures involved; in such cases NEC Article 250 *requires* a distinct grounding electrode in the detached structure, although a single *load* in a detached structure can be fed from another panel without need for a local grounding electrode. All electrodes must be bonded together (so they're unipotential at low frequencies -- all bets are off in the low megahertz range, as you'd see with a lightning strike). If you have RF systems like me, a perimeter ground and single point ground panel get added to the mix, so I have electrodes (two each) for all of my outbuildings (including the one with all of the PV gear) tied together with buried #6 that surrounds the compound and ties into the SPGP for the antenna systems.
This relatively straightforward issue gets massively confused when using UL-approved inverter-chargers. UL 458 *requires* that inverter-chargers with integrated transfer switches bond neutral to ground when running off of battery or the PV array, but not bond them (for obvious reasons) when powered from AC-in.
A lot of this complexity is a consequence of the fact that the US uses ground both to dissipate charge build-up (earthing), and to clear faults. The latter is why we bond neutral and ground at the service entrance; fault current on ground will trip the protection device and clear the fault. The reason we don't bond them in subpanels isn't to avoid ground loops, it's to avoid neutral current being carried on the ground conductor. Note that none of this has anything meaningful to do with lightning protection; minimum conductor side for lightning protection grounding is 1.5 to 2 inch 16 ga. strapping.
I suspect the LV6548 ships (sometimes) with the bonding screw installed is because not every country does things quite as whacky as the US. It just would have been nice if it had been in a more...convenient...location.
You should join our forum. Diysolarforum.com
Made a new video with ground neutral bonding discussion, now have the bond screw removed
Us Building Biologists call this net current : "The reason we don't bond them in subpanels isn't to avoid ground loops, it's to avoid neutral current being carried on the ground conductor," parallel paths cause imbalanced return currents to occur because electricity takes all paths back to source (transformer of power company). This is serious code violation because a AWG 14 lighting circuit wiring isn't designed to carry more than 15A, but it can when commingled neutrals from another other branch circuit that is rated at 20A and uses AWG 12.
Three way stairwell lights often have this issue, and although the switches work, the wrong gauge wires might be smoldering inside the wall, and that's why it must never be done. Yet some 80% of homes have N-N issues or N-G miss wiring issues.
We trace these complicated to isolate wiring errors in junction boxes, by means of magnetic field meters, as such looping (where they shouldn't) AC currents generate large and deleterious magnetic fields
This is exactly the headache I've been having with my split phase system. My solar shelter is located about 75ft from my shop. I have a panel inside the shelter powering the ac inside the shelter & a Native DC split providing climate control for my battery bank, that's connected directly to my bus bars. I then have 75' of 2awg running to my "main panel" for the shop. My PV array is on its own grounding rod & I had my shelter on its own grounding rod & also one at the shop panel. After watching this video, I think I don't need my shelter on its own grounding rod & instead have a floating ground/neutral bond in the panel inside the shelter. Everything has been working fine. My shop lights are now powered by the shop panel, but when I started flipping on outlet breakers, I got an output fault at my inverter. I had the feeling it had something to do with the grounding possibly making a ground loop, even though I'm completely off grid. My shelter panel isn't ground neutral bonded, yet my shop panel is. I only have power wires running to my shop panel without a ground wire connecting the two panels. I'm not an electrician & basically youtube'd my way through this massive project. I live in the mountains & it's impossible to get a solar professional out here to look over my system for safety. If you have any advice for my set ups grounding, I would be in your debt. I'm running a 28kwh 13000w split phase off grid solar setup, but my particular set up with my solar shelter being 75' from my shops panel is confusing. My shelter is a refurbished Korean era military radio shelter. It's a metal box with aluminum mesh lining for faraday protection I assume.
I have an all in one inverter using the reliance 6 circuit transfer switch. The transfer switch is using my utility ground so therefore no need for a ground for the inverter. My panels have their own ground to limit a static charge and prevent a lightning strike but it stays separate.
Hey Will, can’t wait to see your review of the new Bluetti AC500 and B301 battery pack.
Thanks Will as I have been of grid for thirty years and to earth at inverter and distribution panel made no sense and equipment suppliers were vague about this also. All information I read or watched stated to ground but I decided for the reasons you outlined to use the "floating earth" at the equipment end. Using the Inverters internal GN bond. And to quote part of Chris's comment " The reason we don't bond them in subpanels isn't to avoid ground loops, it's to avoid neutral current being carried on the ground conductor". Finally getting some sense on this important issue.
I'm no electrician, but I was always taught that, so long as you only had one ground for a common rail, it didn't matter which ground point it was. Ground loops are only an issue when multiple points are grounded differently, and there's potential between them due to the varying resistance. On a small system like this one, I'd probably stick with an earth ground in the panel so all my disconnection points are in one place. If all the (edit: ground points) are connected together, and only one point is actually grounded, I don't know that it matters. 🤷♂️👍️
100% spot on. Would you like to see what Sigineer says on how to hook it up like you have your setup done? Guess I could email/message you the link. Only ground once. Only bond the neutral and ground ONCE.
I know from the Victron Inverters, that there is a ground relay, that automatically bonds ground and neutral in inverter mode.
The relay opens, once a AC input is connected. But the ground relay can also be disabled in software.
Yes, everyone should copy victron. I am working with one of their inverters right now and the ground relays are fantastic. Fixes all of these issues.
Hi Will.
Yep. Definitely bond / ground earth and neutral only at the source so that your RCD'S (UK) / GFCI's will trip if a fault condition occurs. An earth loop impedance test procedure (ZE) proves this.
One pole of a suitable N.O contactor relay can be used to achieve this bonding function ...when power becomes "lost" from the grid making the inverters " off grid".
This way the original grid bond will be established once grid power is resumed and the contactor re-opens breaking the "off grid" neutral to earth connection.
Its losing the neutral when the supply is disconnected that makes this so important !
My quest is to fully understand the grounding arrangements and benefits gained from earthing my 48 volt Pylontech battery ?
I have my solar (DC array, framing and metalic enclosures) grounded via a thick cable going into an earth electrode rod in the back garden ( not linked at all to the grid / AC earthing system.
Should i bond the battery and metallic enclosures to the output side of the negative fuse terminal ?
Assuming zero volts exists there and it just provides a low-resistance path for the DC battery current to pop a fuse under a short-circuit fault scenario ?
It really becomes risk management in worst case scenario, and thank you for specifying this is not for beginners
Have your checked the code book. Grounding is for lightning protection also.
Ground wire is a backup track from any device to the source in case a live wire is shorted to the panel of the device. And there should be only one return path to the source and at the source ground wire is bonded to the neutral( either it is a battery based source or a transformer). In case of battery, it is a neutral bond inside it. In case of utility it is at the main panel. In your case when inverter is transferring utility power to house, it should connect ground of utility ( coming from main distribution panel of house) side to its loads. When in battery mode, it should disconnect that ground and instead source a ground from neutral bond. But I don’t know if your inverter has this switching happening inside. My two cents…
I knew my bro Will was a Bitcoiner. Just knew it.
Thanks for the awesome information, man. Building a solar array because of you.
I agree and this was a concern I had recently when setting up a system, because there are many systems that don't specify this but when you test you find it to be true. This is exactly the reason why the NEC says not to have a neutral ground bond in sub panels
@@realnealsmith my limited understanding is the differential between the ground rods causing a voltage difference that is more dangerous than the previously accepted lack of bond
@@ShaleAudio Think of it under an extreme condition like lightening, there is a tremendous voltage difference in just a few feet from the strike point across the ground, if you have two separate ground rods then those points have a potential voltage difference from just the static electricity in the air.
@@dean5263 doesn't nec require two grounding methods (ie: two rods 6'+ apart, one rod + metal water pipe, one rod + ufer, etc)? (practically speaking, since it's not reasonable to prove the ground resistance of
People, remember true earth ground,” grounding rods” is ONLY for lighting protection. Period. That’s it. Multiple grounding rods is a fire hazard if lighting did strike and travel through that ground circuit or energy from around the ground from a strike traveling through the ground circuit. Plus you could get cross arching that cause’s the fire and death or burns to anyone in between. In addition it would most likely fry any electronics connected to it. This has been a documented fact when they use to earth ground huge cnc machines years ago.
I had in interesting time getting it 'right' on a UK barge. Although our 240V single-phase is simpler, the occasional use of shoreline utility power needed extra thought. We ended up not bonding N to E within the boat as that often caused the onshore utility outlet, or the on-board RCD before the inverter input to trip. All Earths are connected onboard & with the DC -ve of both 12V & 48V systems. But while diagnosing the tripping we ended up with a 3-way switch to govern how the on-board earth connects with the shore-line earth. In one position it is a direct connection, in the other position it goes via a galvanic isolator. The mid-position provides isolation (just for testing, with the all-in-one configured to not use bypass). It's all stable & reliable now.
Here is a fun advanced idea to make a video about: with the days getting long, how can you make a "dump load" that will turn on a device to use excess solar when the batteries are fully charged.
Yeah I was wanting to see a video on that too
There are inverters doing this. Try Sol-ark or Deye, the Gen connectors act as a smart load and you can configure the Gen output to be a dump load (water heater)
I would think Will Prowse would design a way for excess energy to mine bitcoin, not just heat water.
I wonder if you could use a voltage-sensitive relay and a contactor for that function, when the voltage of the battery gets high enough itd close the contactor and plug in that device. ON multi inverter arrays i also wonder if one of the dry contacts could be programmed separately from the rest to have the same function? Food for thought ^,.,^
I don’t know but it seems that the High freq split phase inverters really don’t have a neutral ground like a low freq split phase inverter transformer .The same for a pole pig transformer connected to your house . Could there be a ground/ neutral difference among the two since the high freq split phase inverter is all electronic and the low freq split phase inverter has just a transformer ? Might can do a test differential video among the two outputs with neutral /ground to find the best way to connect them to alleviate a common ground loop problem . May the solar be with you Will Prowse or any prepper interested in this matter .
Nothing should flow through ground during normal operations. I think the electric code specifies there should be only a single Ground Neutral Bond in a system. As far as I understand you can have multiple grounds in a system but only one Bond between ground and neutral. I mean when you mount a breaker box against a metal building part - it's already grounded. I think what you are concerned about is a "neutral loop" Which can happen when you got two grounded and bonded panels.
Personally, only the generator (floating ground) will connect to AC input,
but usually genny only drives a floating charger connected directly to the battery, multiple chargers if the load requires.
Ground is a local reference and 2 different grounds may not be the same reference aka ground loop.
The power company may have 100 to 150 KV on the pole and taps on a transformer for a specific circuit.
A transformer may supply multiple circuits.
A transformer with 220 secondary and dual (180 phase shift) 110 secondary circuits with a center tap, the new (CT) ground reference is different from the input ground. One would not want to tie the input and output grounds together.
The power company knows you are going to ground your breaker box and designs their system accordingly.
When using solar/ grid tie configuration, the equipment should handle any grounding issues.
If one was completely off grid, from a purely safety perspective it would be prudent to ground the box to a copper rod,
however, once done, the solar panel may increase the likelihood of becoming a lightning rod if not otherwise present.
Letting ground float might offer some protection.
The Schneider Conext MPPT 100/600 has two ground, one for positive and one for the negative but only one can be active. my system is negative grounded so the fuse can go only in the negative fuse holder.
The instructions are clear for mobile application ground and neutral have to be bonded and gound strap is required for vehicles but on grid applications grounds from house distribution box and for auxiliary inverters have to be bonded.
I’ve had issues with all-in-one inverters like this in the past. Mostly sol-ark, and I found that if i don’t bond the neutral to the ground in the AC output load center, I get fault codes, and I even got shocked from the ground out of the inverter. But once I tie the neutral to ground in the panel, problem solved. Very strange as I was under the assumption that they were bonded within the inverters themselves.
Seems a lot easier to just keep it as a off grid isolated system. Even for 240V loads like charging the Tesla I would use dedicated 240V equipment. And lastly, I would use several independent systems. One for ground floor, one for upstairs, one for basement and another for the garage/workshop. Might be more expensive but if one system fails, the others still work. Instant resilience to power loss.
You have such great skills (I am in awe of your verbal skills) and insight in explaining what you are talking about but as an old retired master electrician, I just cannot understand all of this without schematics and symbols. And words seem to often fall short on tecnical problems. When I went into programing industrial automation, honestly words just could not much help. However, fortunately my limitations are not universal, and keep up the great work. And ground faults seem to be such a misunderstood problem and for good reasons. Let us all try to help each other stay safe! Says the old arc flash safety trainer.
I had bonded with a jumper my neutral and ground when running a single LV2424 pip since my ground would float after the sun went down and the charge controller relay would kick off. I was grounded to the AC input, but it didn't seem to matter. -The inverter output would float the ground after sunset. I have since added a 2nd LV2424 to make a split phase system and put in a panel where the ground and the neutral are not bonded. This was in theory that the both the AC inputs had neutral/ground bonding in the main panel. The floating issue after dark has since gone away. I have a Reliance transfer switch for critical loads that also ties into the main panel ground/neutral bar. So at this point, my ground for the whole system is bonding to the main panel at the AC input. I have both PIPS connected to main panel on their own leg from the utility. It seems good, but I'll check to see if I am getting leakage on the grounds during the day and at night to see if something is screwed up.
You should be OK with an additional ground rod if you bond it to your utility ground, preventing a ground loop and a voltage potential. Use at least #6 solid wire.
My only questions is with the ground either looped or in one panel how would this effect GFCI circuits? The other issue is on a totally off grid system, if these is a generator backup is that considered grid power? Or how would you connect the generator? Do you first ground the generator or just run ground to the first inverter system? You bring up great issues that get you thinking.
A GFCI (ground fault circuit interrupt) looks for the difference of current on the circuit conductors. For GFCI to work properly you cannot bond ground and neutral. On a Generator they have ground lug so you can connect to a ground point and provide the unit chassis/earth ground, and without it would be a floating ground, all chassis would have the same potential, but no earth connection.
If you have your Inverter grounded with an earth ground via a sub-panel, the ground wire in your generator cable would provide a chassis ground to the generator.
As you said .. Ground should have only on one place never in two places because of ground loop issue. That is why usually on AC input is ground. You can connect chasis of each inverter into ground of your input system. ..For those who use that..ofc on some system where there is no ground it is the best thing to make ground..but ofc only one point..i had inverters that were from plastics so they didn't have 3rd wire so yeah that is also option as they had floating voltage..to be honest i'm using that kind of inverter for outside stuff ..just in case there is contact with water..it is abit safer
I found this to be true with my Growatt. I contacted them and found that removing a couple of grounding screws on the main PCB it would break the bond - all good now.
This is an awesome topic Will, and one that I think is significantly overlooked. Would you mind clarifying: If you have parallel MPP Solar units, should every AC input ground be tied to the main breaker panel, or just the master unit? Since there is a neutral-ground bond at the main panel, connecting the ground to every input would effectively create a neutral-ground bond to every MPP Solar unit. Also, if you have BOTH AC Input (connected to main panel) and AC Output connected (to wherever), the AC Output ground should be connected to a non-earth ground (such as the sub-panel). But what if you have a grid-tied MPP Solar unit (AC Input connected to grid-connected main panel), and the AC Output ALSO connected to the grid-connected main panel (separate breakers), for when you are energy-shifting to maximize TOU? Assuming that we aren't breaking any laws, of course. Do we connect the AC Output ground to the main panel in that case (which would have a neutral-ground bond)?
Thanks for doing these videos Will. Please can you do a video on using super capacitors or capacitor banks to increase the surge capability of an inverter system? For example, running loads with high inrush/startup current. I believe there are people doing this currently and it "works". Or if you think it's a terrible idea, let us know. Also, my 2.3kW Goodwe EM inverter says not to run an air conditioner on it, is that because of the surges, or it being an inductive load? If it's just the surge issue then I'd be tempted to use capacitors on the circuit. If it's going to kill my inverter and/or AC even with capacitors, then obviously I would avoid that.
You must keep the ground and Neutrals separate in this case. The only connection made between the ground and the neutral conductor is made on the service side at the meter base. Everything beyond that is a bonding conductor bonding the system the the grounded system conductor. All electrical systems only have One grounding conductor. That is generally the #6 CU conductor that is terminated to your grounding electrode. The green, or bare conductor connecting your equipment to “ground” is the bonding conductor. I believe this is just confusion on the proper use of wording. The conductor referred to as a “Neutral” in this case is, by definition, the conductor that caries the unbalanced load in a three wire electrical circuit supplying two different voltage loads (ie: your range in your home is supplied by a 3 wire circuit, supplying 120 volts to your digital timers and clocks while 240 volts is supplied to your elements, the unbalanced load than travels the neutral conductor) - The white conductor often referred to as the “neutral” conductor in a two wire circuit is the “Identified” conductor, which is the grounded circuit conductor, and it is bonded to the grounding conductor at the meter base. You can not create a ground loop in this case, as current will always take the shortest path to ground, never going back on itself or traveling multiple paths. Though the additional bonding would be overkill (creating a “super bonding” type situation) I’m unsure of what hazard you would be creating. This is not armoured single conductor cable, so sheath currents would not apply.
PS: Love your videos, the tear downs, and explanations on batteries and renewable energy networks is outstanding.
Yes I agree
“You got to keep ‘em separated “ blasts in the background.
So in the case of having inverters as the source in an off grid system the ground rod to the main panel should be the only ground? And the inverters create the neutral to ground bond? All other grounds and neutrals should remain separate?
while I don't have one of these units in front of me (yet) I would assume they are using some form of the contactor to switch between the Grid-tied and islanded states. This contactor would be wired to have the Grid side simply pass straight through the inverter with the charging circuit tapped onto its output. If this is the case, you DONT need any special wiring as the inverter does all the thinking for you I believe. You would simply use the neutral and ground bus in the subpanel like you would normally a standard 240v outlet (it shouldn't need a ground connection to the case as the ground bus is already electrically connected but check for yourself always). This lets the inverters Operate like a simple off-grid DEVICE (NEC CODE TERM), using the panel neutral when it is grid-tied and creating its own bonded neutral when islanded. All inverter chargers I've seen inside of including growatt and victron operate this way as do some RV generator transfer switches its effectively the same as plugging in a cord to an outlet and in fact you can do just that. All the same connections are needed in the same format for each inverter. The inverters in Islanded mode will create a Unified neutral at the point of generation (they act like one big inverter) that effectively BECOMES the main panel with the bond.
One last thing. ALL of the inverters should be fully grounded to the panel or sub-panel, but there should only ever be one ground rod from the main panel as you've pointed out or a ground loop WILL occur. even when islanded the inverters will "borrow" that ground for operation although it is not necessary for function(think RV or mobile applications) same reason many generators have an extra ground lug on them so you can Borrow the house ground for extra safety but will run just fine without them. In the islanded state however the neutral and hot are electrically disconnected from the inverter and not performing any functions outside of isolated sensors and thus no strange loops or secondary bonds are occurring.
When in doubt grab your meter and test it out. should operate just as if you pulled a plug from the wall when islanded except a ground connection was maintained but the inverters have created their own bond. If feeding a sub panel with the inverters you would not need that panel to be bonded in islanded OR grid-tie use after the inverters
Just two cents from a nerdy solar lover and Electrician apprentice ^,.,^
excerpt
If power for IT equipment is served from a separately derived source independent of the building ground system, noise might be generated on the system. This is a form of common mode noise, in which the power-source ground is referenced to a different point from the equipment ground. These stray currents-in other words, noise-can find their way onto grounded equipment, thus energizing the equipment.
Never clicked so fast. Thank you will for all you hard work!
greta video again even if I don't untherstand the problem
The technicality is way over my head, but i think I got the main point from this vid...For my off grid 120V set up with just one AIO inverter, DONOT bond N-G in the breaker panel because the AIO inverter bond N-G internally when using power from battery, and if i need to use gas generator to power the AIO to run bypass mode and charge battery, the generator need to be bonded N-G. Basically all the bonding need to be done at the AC input of the AIO inverter, not the OUTPUT. Thanks Will and Filterguy
Yep, this is causing issues for me. I'm using a single LV2424 to power some critical circuits as a backup/free solar.
The AC output of the LV2424 is going to a sub panel which I have combo AFCI/GFI breakers in, however these are occasionally tripping, along with the 30A GFI breaker in my Main panel that serves the LV2424.
Like you, I have measured 1 volt or less between various points of neutral and ground in my system. I am thinking the bonded neutral in the LV2424 is causing a very small voltage leak to ground that is tripping both input and output GFI breakers.
But for safety concerns I want the LV2424 input to be GFI protected in case something catastrophically fails in the inverter to where the casing could be energized.
Not sure where to go from here. For now I replaced the GFI breakers with standard thermo-magnetic ones. Which is not ideal and I want to find a solution ASAP but it keeps my circuits active for now...
I should note that I have the LV2424 output circuits going into transfer relays so if the inverter fails, the transfer relays fail closed and the critical circuits still have power from the Mains panel since the LV2424 AC relay is a fail-open type.
You need a bonding wire and all metal objects that you can touch with your hand needs to go through this bonding wire to a ground. So I would string a bonding wire from each unit to your panel to ground it all assuming your panels go through your AC supplier. As for rounding loops you're supposed to have two grounds on the house that's a loop as well they tell you I believe by code 8 to 18 ft apart depending on what locations you at work parts of the country. That's why a lot of them use your copper pipes coming into your house as the second Bond.
Awesome happy you are bringing this up because I was looking into getting one of these.
Thank very interesting
This would a great subject for a longer in depth video, @WillProwse.
Great discussion, Will. Just to clarify when you are using the AC input (and grounding through the AC input), are you still grounding the inverter chassis? Or are you saying we don’t need to ground the inverter chassis? If we should ground inverter chassis, would tying the ground to the ground from AC input suffice? Thanks for all you do!
All a little bit over my head but it occurs to me that what Will is doing here is a pretty standard installation, pairing two of these units in an off-grid configuration and that any manual should offer a "reference design" for the most common deployments. There really shouldn't be too much debate about the standard way to do this and discussions should be limited to when you decide to get a little funky with your design. Manufacturer's need to do a better job with documenting reference designs and maybe pay some people who know what they are doing to develop these designs and document them. I think an extra $50 on the price of a unit would be well worth the reduction in research and headaches while trying to figure this stuff out.
I dont really know what is your elecrical system looks like, but our in Europe looks like this : 230V (50Hz) supply consists of 3 wires. 1-hot, 1-neutral, and 1-safety ground. 230V (50Hz) is obtained between the hot and neutral lead. Double-pole main circuit breaker protects both the hot and the neutral leads. Branch circuit breakers are single-pole, switching the hot lead. Has hot bus connected to the line side of the branch breakers. Has separate buses for neutral and safety ground. Standard European practice and ABYC requirements include ground-fault protection for the entire AC system as well as critical branch circuit GFCI protection. I think, no matter what we do, and how we do our configurations : There must be a LINE IN from the GRID connected to the Charge controllers (and never use the grid power but the Neutral and Grounding) or a dedicated grounding wire connected to the inverter(s) and other devices that requires grounding wire. And never ground twice! ONE SINGLE COPPER ROD IS MORE THAN ENOUGH!
You have to bond the panel. If one of the LV6548 isn't connected, there is more neutral current than what the system can handle. You would have to wire each invertor with the maximum current that the system can handle. Each invertor isn't designed to handle all of that possible neutral current. It has to be done at the load distribution panel.
@Will, I think I've been saying the same thing maybe in an awkward fashion. In the example you provide here the panel should have a floating neutral and the ground should be bonded to the case. It is effectively a "subpanel". Equipment bonding to ground and neutral ground are 2 different animals. Supplemental equipment grounding is okay and even encouraged but you only ground the neutral (current carrying conductor) at the point of origination. Subpanels ALWAYS float the neutral. That describes U.S. grid systems. I would expect the same set of requirements essentially to apply to an off-grid or separately derived system. Is that correct? Are we saying the same thing?
Thank you. I will remove the bond wire in my sub panel and float the neutral and allow my off grid ecoworthy 5k to manage the bond internally. I was alarmed and reacted when I saw a 33vAC potential between N and G at the panel.
I think the best solution for bonding multiple inverters is actually simple -- a dedicated non-current carrying wire from the neutral bar within the inverted panel back to each inverter's bonding relay. Without current there are no voltage differences between each unit, and thus no ground loops. It's a controlled parallel path situation, where there is only ever current when you want it to be there -- namely a ground fault. I don't like only having one unit bonding as the inverter may fail, but the rest of the system still operating without issue and without noticing (think parallel 120v -- not split phase).
The neat part here is that it's basically what's already being produced and it would ship from factory with a N > Nb jumper installed. Then if you are connecting many in parallel, you can remove with 5th wire. Additionally, if you are connecting to an existing bonded panel say for backup generator purposes, you can remove entirely (similar to software disable).
3:10 You can indeed connect a ground rod to the AC output of the inverters. But it must be done to the ground terminal, not to the neutral terminal (to avoid ground loops and objectionable current).
Similarly, in telecommunication systems, the NEC requires you to add a new ground rod to the grounding/bonding system (but not the neutral) of the telecom system if the distance between the telecom grounding system and the power grounding system is greater than 6 meters.
I think people forget that these Hybrid solar inverters were originally designed to be added to an existing AC mains consumer panel which has its own Neutral grounding scheme from the loads back to the panel.. its supposed to have utility power line installed directly into the AC input, then the AC output fitted to the existing distribution panel. The Inverter acts like a whole house UPS, switching to to to the mains when the solar power or the battery power becomes insufficient and vice versa.
Hi Will, at about 2:30 you stretch both arms to make contacts. Please consider a spring loaded "alligator clamp" for the test wire so your arms don't provide a possible circuit.
I actually have a lot of those but they were on the other side of the shop
Grounding is a multifaceted beast, and has many different implication as whether it is AC or DC and/or even RF. An electrical diagram will show many items tied to ground rather than draw out the wiring for a ground wiring path. In AC circuits it typically refers as a chassis ground to bleed of static and provide a path of least resistance when/if the device has a short.
When we get to Inverters and especially the All-in-one units, it is recommended we treat them as a sub-panel. This Video shows the wiring and use of the sub-panel giving detail for ground and neutral connections. The ground wire is to provide a chassis ground to the cabinet of the invertor. When the AC loads are connected to the Inverter with a sub-panel the ground is still for Static and Chassis connections and not a current wire, the sub-panel passes the ground back to the main service panel and connects to earth with its bonding. The bonding of the main panel to earth is primarily to prevent lightning entry.
When you connect your batteries to your invertor they are completely isolated from all AC connections and bonding. On your Battery you can have a Negative Buss Bar to connect all the DC negative connections together. Never connect the batteries to earth ground, should the invertor generate a short you could send AC through the batteries and have an explosion.
Off-grid systems are wired with a sub-panel also, but as it does not have a main panel with a ground bonded to create the static/chassis connection we need to connect this panel to a ground rod to give that static/chassis connection. We DO NOT bond this panel! Again the Ground and Neutral in this sub-panel are NOT connected. Your Neutral wire is a current carrying wire to complete the circuit, the ground is for Static/Chassis and helps the circuit trip when there is a fault.
When paralleling units in a 120v configuration, connect only the Line1 together and then only connect the Neutrals together. Your Neutral is your return current wire in powering the circuit.
Thank you that really does sound logical I have to read it 10 times try to get it through my brain
@@VinceBadovinatz88 glad it helped
Just a quick question I have a small inverter 12v little bit off topic but on topic of small inverters 1200watt for camper van (Giandel)it has a ground wire on the back can I just run that to the 135ah LiFePo4 battery negative? or actual ground do I even need to use it on small inverters?
Actually can you make a ground loop with a small battery’s and small inverters and give us a demo on what not to do that may help some people understand loops better
Great videos by the way
Will, I hope you find out definitively and post an exhaustive video for both off-grid and grid tied.
I contacted MPP and they agree with my solution. Grounds and neutrals separate.
Any chance of diagramming what you're talking about? It's over my head but I'm very interested in what you're saying.
Nice shirt!
I have two LV6548 units connected to the grid for bypass mode, so the neutral and ground are bonded at the grid master breaker. It is important to not bond the neutrals and grounds anywhere else, including any subpanels, thereby avoiding current sharing between the neutral path and ground path. Current sharing is bad - high currents may result in voltages appearing on the ground, which leads to the possibility of electric shock from equipment cases. No current should flow through ground during normal operation. The LV6548 units disable ground-neutral bonding when in bypass mode - so far so good. However, in battery mode, each unit will connect its output neutral to ground using a relay. This works if there is only one LV6548 in your system, but for multiple units, the extra connections need to be disabled. If you remove the case of an LV6548, you will see a small inverted circuit board just left and below the center. If you peer vertically down along the top edge of that board, you will see a screw head on the circuit board below. Remove that screw to disable any bonding.
diagrams, even hand drawn, would help tremendously in explaining. :)
Good T-shirt 😁 not Markting at all 🤣
When my growatt 3000 LVM-48P is running just off the battery, there also is a GN bond created... However the bond is CRAP. I see continuity, but there is a TON of resistance there. so much that my furnace won't run unless i re-create my own GN bond.
That's interesting. I was always under the impression that a system should never have current flowing on the ground path. If the inverter has a neutral bonded to earth the question is will some of the current of the output find its way back on the earth ground. I'm not completely convinced that it will because it is artificially generated by the inverter itself. If this just a single phase input splitphase output transformer than that center tap would be bonded to earth ground or locally grounded to earth directly if the distance was far enough. 🤔
Grounding the solar panels and supporting alluminum structures (DC side) to the same ground rod for panel and inverters AC side)
Or 2 separate ground rods for each side. (NEC unifilar ongrid description.)
Could you please do a simple test for me? I assume that you have your panels all connected to ground? Could you check voltage from the panels to the ground rod?
During the daytime I have about 17vac running to ground, at night when the sun has long set I have about 5 volts running to ground. There is no voltage in the house ground wire. I have the combiner box also hooked to the same ground rod as the panels because of the lightning protection circuits need a ground and I don't want to tie that into the house wiring.
I need to check each panel from the output positive to the frame, and see if there's anything bleeding from any of the panels. I found the voltage when I was connecting each panel to a common ground wire running to a ground rod. I felt voltage run through my hands when I touched the next panel and the ground wire I was connecting. I don't like having voltage running on my ground wire so I am a little curious where it's coming from and if it's normal.
I want to use a separate grounding at more than 10m distance, under 4ohms, with a permanent link between N and PE on AC out. For home circuits I will use the home grounding, completely separated from it.
Thanks for posting this, as it will get a lot of attention. I've been trying to get a definitive answer on this topic for a while.
Looked for someone to mention this...couldn't find.
The reason for bonding neutral to ground is for a low impedance path for fault current to return to "source" This completes a circuit to enable over-current protection devices to work promptly when a "ground fault" (a path from hot to case, for example, occurs in some load-side device.) In inverting mode, I'm convinced the inverter is a separately derived system & requires the AC neutral output to be bonded to equipment ground for the purpose I described. The inverter is the "source" where ground-fault current must return. In bypass mode, I assume the AC inputs and outputs are continuous & the AC output should not be bondedin addition to the prescribed bonding at the electrical service main shutoff, as Will explains. Only if the AC input and output neutral are always internally continuous, could we elect to bond once, but on either side. I use a random manufacturer 240VAC split phase inverter/charger with no connection to AC input neutral. I have to believe that bypass mode is *not* simply pass-through mode, & so I must bond neutral output and ground to get a ground-fault-return-to-source path for either 120V leg.
The statement about AC charging was unclear: load devices are always connected to AC mains in parallel, and connecting in parallel to a DC bus is not a challenge, up to the charging current limit of battery.
Would the variables make more sense if the 240v breaker panel was considered as a sub panel for code compliant earth/neutral practices?
Ok so I know that this was advanced users but as someone who wants to set up a dual All-in-one system for off grid only, how do I set up the neutral and ground at the load panel. I got confused by the different scenarios discussed.
I'm new to this topic but very interested in it. Would it be possible to ground your secondary panel with a rod and connect the input AC to three inverters without connecting the ground from the mains panel?
You are using a sub panel, but what about using your 2 inverters configuration on the main panel and off grid system where the main panel has a ground and neutral bounding per code in the US? Thank you.
Hi Will, Here in the UK, we just have an earth ground (Green), live (RED or Brown) and a neutral (Black or Blue) wires keeping it simple. Why does one need to connect a third to the neutral wire?!! That's putting like a live-sh part (Neutral side of a Supply power) to what you are calling a ground wire, which is supposed to be Just that an earth grounded wire into the earth. Thus any metal surrounding power supplies casings or electrical goods are then save to touch, as they are all connected to ground earth. Which by the sound of things is being turned into the negative side of all power supplies!! Am confused with this!! Makes no sense to me this, a ground wire is either a common negative separate to earth ground is it not? Regards George.
One reason to make a neutral ground bond is for a GFCI interrupting device if you do not have that bond your interrupting device will not work properly.
I have watched the Advanced level grounding video several times. I am installing two Rich Solar LV6548 to get split phase.100% off grid. I want to understand the grounding.
1. The LV6548 are ground/neutral bonding from factory.
2. I am going to install a main ac disconnect in the solar shed. That main disconnect will then feed a house panel.
So, from the video I will not do a bond at the main disconnect or the house panel. Correct?
Earth ground to 1 LV6548 inverter only on the ac input. Correct?
Do I earth ground the house panel?
Thank you in advance
I wonder if the the 240v single phase versions are different. That earthing arrangement could cause alot of issues on modern installs in the UK and Europe. And all but unusable on a split phase supply in the UK (we do have split phase its just very rare, it normally manifests itself as two completely separate 250v supplies due to our phases running at 400-415v)
Does all your ground faults work?
Not having a neutral wire seems like they wouldn't work.
Grounding is 1 of most important things in electricity. In 120v you have 3 wires. 1 is live (black) 1 is neutral (white) 1 is ground bare wire. Neutral and ground is same . Difference is that the neutral is isolated and the ground is used for protection and not to carry electricity.
Say you have a washing machine with default live touching case and not grounded then the machine become live. If someone Touch the machine and a nearby copper water pipe you get a free hairdo at the least. If the machine would of been grounded the breaker would open the circuit and all danger gone.
If there are two inverters next to each other then having the same ground for both isn't an issue because there wouldn't be any resistance between the two inverters bonding to create that Earth loop.
When the MPP inverters are grid-tied does the inverter separate the ground neutral bond?
I believe there is no risk in combining grounds from multiple grid tied inverters onto one bus bar as long as there isn’t also another neutral/ground bond on that bus bar. The point is to prevent neutral current from traveling on a ground…unless something bonded becomes accidentally energized. Then you want a path to ground to dissipate the charge/trip the OCP. The neutral is a current carrying conductor under normal conditions. The ground is only there for the safety of us and our equipment.
Problem is you create a ground loop which circumvents circuit breakers if there’s a short inside the loop.
@@cybertrk How will parallel paths to ground cause OCP device not to trip? Current will flow breaker will trip.
I am just getting into solar and to me this grounding stuff is the most confusing thing there is to learn. Do you have any recommendations on videos that do a good job of explaining grounding requirements? There seems to be a lot of confusion even among the seasoned vets on grounding. Last thing I want to do is fry my brand new equipment when setting it up or even worse electrocute myself.
So if you have multiple inverters, would you tie their grounds together on a bus bar and then earth ground the bus bar?
Nice information. But what about if the grid fails. Maybe something happened and there is no power from the grid. How will the system be bonded to ground. Because now we just loose our grid. Where our ac input. Ground and neutral is bonded.
Part of the advertising for the LV6548 says "custom parameters can be set according to any battery between 40 and 62V". Does that mean I could use a higher voltage battery bank, say adding a couple cells as long as I keep the high voltage cutoff at 62V or less? I am replacing a 120VDC 1640AH telephone battery bank and am bummed that the highest voltage solar battery banks have got to is 48VDC.
Hi.... I would like to ask you please
between Must, Teckfine and Sako which inverter do you think it's better for home use
Thank you
Will, would you still suggest the new LV6548"V" as the overall best unit to start out with (assuming the 6000w is what one needs) ?
Hi Will, are those inverters pure sign wave? Are any other manufacturers not pure sign wave, so I can avoid them. Thanks
Hmmm. I am not an electrician. Having said that it has also been 20+ years since messing with electrical circuits. I begin thinking 1:1 isolation transformers would be the ticket to interrupt any DC bias. However if the inverters are outputting pulsed DC and not AC, I have no clue.where to go. just 2 cents that is probably inapplicable.
Would the rules for a normal (gas powered) generator apply in this case? Since they have been around for years as secondary power sources they must of figured it out?
I agree having an earth ground anywhere on this sub-system is a bad idea if there is another earth ground on the AC input. Even connecting just one of the inverters to an earth ground is a bad idea because in your AC output panel all the grounds and neutrals from the multiple inverters are tied together, meaning they would all be tied to earth ground close to this system, and far from the main earth ground. As you correctly stated this is dangerous. The only way I would connect an earth ground to even one of these inverters is if there is no grid-tied AC input with its own earth ground.