Grounding/Setup Questions for Off Grid

[wpg205]

Hi All, I am 90% confident in what I am about to do, but wanted to get final thoughts from the forum. I'll be using the inverter in a totally off grid setup in my shop with one main panel that will in turn have one connection to earth - thus neutral is also earth ground. I was planning on grounding the inverter chassis to neutral, grounding battery negative to neutral, as well as the body of my solar charge controller (victron) to neutral.  I'll be putting a fuse on battery positive, and non-polarized DC breakers between SCC and the battery and another between inverter and the battery. 

Does anyone anticipate any issues? I'd appreciate any suggestions. I'm also still looking for a good 100 amp DC fuse, as well as large busbars, if anyone has any suggestions. Thanks for your thoughts. 

[InPhase]

There are any number of functional ways to electrically connect all of your equipment. But I am an electrician by blood and knuckles, so I will speak in terms of an installation that meets the NEC. 

The code would allow you to bond the neutral in either the inverter or the first disconnect, but not both. So bonding the transformer case to the neutral AND bonding neutral to ground in the breaker panel would be a no no. And if you used 4 wires to connect between the inverter and panel, you will have created a parallel path for neutral current, which is another no no.

 

I think your best route would be to use a main breaker panel as your main disconnect. Bond the neutral and ground with the nice little green screw that comes with the panel. Bring your earth ground to the neutral bar of the panel, as well as your metal water pipe and/or structural metal bond wire. Leave the neutral unbounded in the inverter. Use a 4 wire cable or 4 wires in conduit to connect between the inverter and panel. Any separate inverter case ground will go to the panel, not the inverter neutral.

As far as bonding battery negative to the AC neutral, I guess that depends on how the inverter works. I wouldn't guess there would be a problem on a LF inverter since the transformer should isolated the output, but I don't know if there may be a connection internally that would cause a problem. Sid would know for sure.

[TheButcher]

At least over in Australia, and as far as I know, no need to bond the - terminal on the battery where there is a transformer (low or high frequency) providing isolation between the AC and DC side.  If it's a transformerless design, so grid tied, the battery needs to be bonded but it'll also be a high voltage battery and should be completely enclosed to keep sticky fingers away anyway.

[Sid Genetry Solar]

8 hours ago, InPhase said:

I wouldn't guess there would be a problem on a LF inverter since the transformer should isolated the output, but I don't know if there may be a connection internally that would cause a problem. Sid would know for sure.

The GS inverters have true and complete isolation from the DC battery side and the AC side.  There are decoupling caps to the chassis on both battery and AC sides, but no electrical connection or non-isolated parts.

@InPhaseI really like your above statements/description, they are very clear and concise.  Seems that in a whole-house setup, internal inverter ground-bonding is more or less useless and/or actually problematic.

[wpg205]

Thanks for the comments all - really helpful.

@InPhase Thank you for your comments. You're saying not to connect/bond the inverter neutral at all. I haven't seen that suggested elsewhere but if you and @Sid Genetry Solar think it makes sense then that is good enough for me. For connecting the inverter to the panel, you recommend using four wires? Two for each phase? I was just going to use welding cable (one for each phase) in conduit. Is this a problem?

As for grounding the battery, what is the consensus? @Sid Genetry Solar suggests it shouldn't be an issue as the AC-DC sides are isolated, but @TheButcher thinks it isn't necessary. I've seen a lot of discussion on both sides - is the point of grounding battery negative that if somehow battery positive energizes something in the AC system, it can flow back to negative (and hopefully blow the fuse I have on positive)? Whereas without the battery negative grounding, if positive somehow the positive energizes the AC system, someone or something could close the circuit accidentally?

Again, thanks all for your thoughts. 

[Sid Genetry Solar]

31 minutes ago, wpg205 said:

You're saying not to connect/bond the inverter neutral at all.

Not quite what @InPhase was saying...instead, he was saying that the ground-neutral bonding takes place at the electrical panel. 

The confusion seems to be stemming from where to bond ground-neutral, not whether to bond ground-neutral.  @InPhaserecommends bonding at the electrical panel, NOT the inverter terminals.

33 minutes ago, wpg205 said:

For connecting the inverter to the panel, you recommend using four wires? Two for each phase?

4 wires from the inverter to the panel, as follows:

• L1
• N
• L2
• chassis ground

Unless you're doing a single-phase 240v setup and don't need the Neutral?

 

34 minutes ago, wpg205 said:

As for grounding the battery, what is the consensus? @Sid Genetry Solar suggests it shouldn't be an issue as the AC-DC sides are isolated, but @TheButcher thinks it isn't necessary. I've seen a lot of discussion on both sides - is the point of grounding battery negative that if somehow battery positive energizes something in the AC system, it can flow back to negative (and hopefully blow the fuse I have on positive)? Whereas without the battery negative grounding, if positive somehow the positive energizes the AC system, someone or something could close the circuit accidentally?

Grounding and I are admittedly not on speaking terms, so I'll let those who know the codes do most of the talking here 😉.

Personally, well I don't have anything grounded (at my own risk), but it's really nice to have the AC and DC sides of the system truly isolated.  Particularly for me, as I have a lot of equipment running on direct battery DC...if the AC sides and DC sides are tied together with a common ground, it becomes extremely easy to blow up equipment if you aren't thinking...because of unexpected voltage potentials.  Most AC equipment is isolated to some extent--but most DC equipment is NOT isolated.

At your own risk.

[wpg205]

Ahh ok, this is totally due to my lack of clarity in my original post. When I said "neutral" I meant main electrical panel neutral. So I think @InPhase and I were originally on the same page - connect inverter neutral directly to main panel neutral/ground and connect inverter body to directly to main panel neutral/ground. So the four wires are as @Sid Genetry Solar describes above. Thank you for clarifying Sid.  

As for grounding the battery - my first instinct was to leave it ungrounded. The actual battery connections should be fairly limited and in an encased space. I think grounding the battery opens up a can of worms and it is just much easier for me to understand what is going on with the battery floating. 

[Sid Genetry Solar]

Battery grounding would apply if you were grounding more than just the solar panel frames...i.e. trying to ground the negative solar input wires would also ground the negative battery lines as well, as most MPPTs have a common negative bus.

Not saying that I've ever heard of grounding the solar lines...just the solar panel frames.

[TheButcher]

4 hours ago, wpg205 said:

Whereas without the battery negative grounding, if positive somehow the positive energizes the AC system, someone or something could close the circuit accidentally?

I haven't put a lot of thought into this answer but I suspect that the inverter would not still be operating, ie producing AC output, if this ever happened.  Getting DC + to the AC side would mean a loss of isolation in the transformer and if something that bad has happened odds on all the transistors have long since taken their hats off, sat down, and had a good smoke to relax.

The concern with battery bonding is as I understand it really with grid tied transformerless inverters where all components, inverter, battery and panels can be live at mains voltage and have full mains current available on them.  Poking fingers where they shouldn't be will see the pokee with mains voltage across them.  This also explains why many countries require fuses on the + and - leads of a grid tied solar array - to handle the case where either side of the array goes to ground for some reason.

[Sid Genetry Solar]

1 hour ago, TheButcher said:

I haven't put a lot of thought into this answer but I suspect that the inverter would not still be operating, ie producing AC output, if this ever happened.  Getting DC + to the AC side would mean a loss of isolation in the transformer and if something that bad has happened odds on all the transistors have long since taken their hats off, sat down, and had a good smoke to relax.

You can safely short either battery positive or battery negative to any of the AC output or input leads on a Genetry Solar inverter without any problems or issues.  It's truly and totally isolated.

 

1 hour ago, TheButcher said:

The concern with battery bonding is as I understand it really with grid tied transformerless inverters where all components, inverter, battery and panels can be live at mains voltage and have full mains current available on them.

HF inverters are especially problematic here, as they usually have a "flying neutral" (explaining why you can get shocked by touching just the neutral).  For the high wave, the neutral is shorted to the negative of the high voltage capacitor...and for the low wave, the neutral is shorted to the positive of the high voltage capacitor.  Still, the boost circuit is usually isolated.

Wouldn't surprise me if some grid-tie inverters aren't isolated.  Whatever, I digress 😉.

[TheButcher]

🙂 I was more thinking along  the lines of how exactly would the + side of the battery even get to the AC output terminals.  The only reasonable way (ie, not opening it up and wiring it across) would be for a bridge between the LV and HV winding on the transformer to happen and if that happens things have gone spectacularly wrong and odds on the transistors driving the LV side didn't like working into the, probably, large number of shorted turns that are now present on the LV side.

[The Blind Wolf]

hmmm.  Well my solar system is natually grounded lol.  the whole frame is made out of wood, well one of the array is.  Hmmm, will have to go take my talking voltage meter and see if the frame on the solar panels are conductive, for some reason I think they are not, since well if they where, you wouold get a shock during the rain/sunny day, and well I've never got shocked even when the panels are soaking wet and the sun is fully out.

[The Blind Wolf]

46 minutes ago, Sid Genetry Solar said:

You can safely short either battery positive or battery negative to any of the AC output or input leads on a Genetry Solar inverter without any problems or issues.  It's truly and totally isolated.

 

HF inverters are especially problematic here, as they usually have a "flying neutral" (explaining why you can get shocked by touching just the neutral).  For the high wave, the neutral is shorted to the negative of the high voltage capacitor...and for the low wave, the neutral is shorted to the positive of the high voltage capacitor.  Still, the boost circuit is usually isolated.

Wouldn't surprise me if some grid-tie inverters aren't isolated.  Whatever, I digress 😉.

Funny, I got a 3d printer here, and I'm getting sap while its off or on, I think the power supply on it is a HF lol. and I got it on a battery back up, 

[Sid Genetry Solar]

11 minutes ago, The Blind Wolf said:

hmmm.  Well my solar system is natually grounded lol.  the whole frame is made out of wood, well one of the array is.  Hmmm, will have to go take my talking voltage meter and see if the frame on the solar panels are conductive, for some reason I think they are not, since well if they where, you wouold get a shock during the rain/sunny day, and well I've never got shocked even when the panels are soaking wet and the sun is fully out.

The frames may be grounded (and often are on commercial installs), but due to the series-parallel wiring methods of solar panels, it's not possible to tie either positive or negative to the grounded frame.  Kinda a weird thing if you ask me...a ring of grounded lightning rods around each solar panel it seems 😉

[TheButcher]

It can be a safety precaution.  If it's a high voltage array and there is some leakage due to moisture ingress there can be quite a few milliamps available and that in turn can liven up the whole roof via the racking.  All it then takes is for some hapless person to climb up on the roof and ...  It's one of the risks taken when purchasing 2nd hand panels.  Trina had a whole heap of panels sold in Australia go leaky a few years ago due to moisture ingress from the edge of the cell sheet.

[The Blind Wolf]

2 hours ago, Sid Genetry Solar said:

The frames may be grounded (and often are on commercial installs), but due to the series-parallel wiring methods of solar panels, it's not possible to tie either positive or negative to the grounded frame.  Kinda a weird thing if you ask me...a ring of grounded lightning rods around each solar panel it seems 😉

That was my thoughts.  I've heard of folks useing some kind of shark washer with the ground wire coming off of it, forgot what they call them, but really, I think if your solar going to get hit by lighting, then I don't care how much you got a ground wire on it or the solar charger, you going to have one bad day.  I mean, I've never ever heard of solar panels being a lighting magnet.  *laughs*  You can't even put a magnet to the frame it won't stick.

[InPhase]

The code requires systems operating above 50 V nominal to be grounded. A 48 V battery system doesn't HAVE to be grounded. But a 90 VDC PV system would. Grounding a system like that would automatically tie the AC and DC sides together. In a contracted system, that is, one not likely to be tinkered with by the customer, blow ups only happen from failed components of lightning. But I have a feeling systems represented on this forum probably blow up weekly from wandering screw drivers🤣.

The pic is just one way it could be done. I hope it is clear enough.

[The Blind Wolf]

12 minutes ago, InPhase said:

The code requires systems operating above 50 V nominal to be grounded. A 48 V battery system doesn't HAVE to be grounded. But a 90 VDC PV system would. Grounding a system like that would automatically tie the AC and DC sides together. In a contracted system, that is, one not likely to be tinkered with by the customer, blow ups only happen from failed components of lightning. But I have a feeling systems represented on this forum probably blow up weekly from wandering screw drivers🤣.

The pic is just one way it could be done. I hope it is clear enough.

Well, 😛 I can't see that pic, anyhow, well then I'm good, cause my system is sat at 60pcv.  and I'm only dealing with 24v.  I just bought a 250ah 24v for $1k 😛  wish these batts came on sale last year or showed up I would have switched to 48v $2k for 13kw isn't bad and considering these batts never been cycle woohoo.  Just wow, they are big!!! 15: long 2.2: wide and I think like 6.5: tall.  yeah boy. 

[kuhrd]

On 4/14/2021 at 6:55 PM, The Blind Wolf said:

That was my thoughts.  I've heard of folks useing some kind of shark washer with the ground wire coming off of it, forgot what they call them, but really, I think if your solar going to get hit by lighting, then I don't care how much you got a ground wire on it or the solar charger, you going to have one bad day.  I mean, I've never ever heard of solar panels being a lighting magnet.  *laughs*  You can't even put a magnet to the frame it won't stick.

The point of grounding the entire PV array is not directly for lightning and the required grounding wire gauge will likely never handle the instantaneous current of a direct lightning strike though it may take a good portion of it to ground.  The reason for bonding or grounding a PV array is to ensure that any charge accumulation on the array makes its way safely to ground while providing a way to ensure the potential between the racking, modules, and charge controllers never exceeds the max rated system voltage. If it ever does due to hardware failure, tracking, corrosion, or short and becomes energized, that charge has a way to go safely to ground to prevent the chance that the system may be floating at several hundreds or even thousands of volts and create a lethal shock hazard.  

Lightning rods are installed on high structures to prevent a lightning strike by allowing the accumulated charge between the earth and the atmosphere (charged clouds and ionized air) a low resistance path to equalize before a lightning strike occurs lowering the potential and reducing the chances of a strike from taking place.  So the intent is that they will not be struck by lightning as they are preventing it by allowing the charge differential to equalize.  If the charge buildup is happening faster than the rate at which it can be dissipated through the lightning protection system then you still have the potential for lightning to strike but at least it might cause less heat damage on it's way to the ground since you are providing it a low resistance path to ground.  If your array does take a direct lightning strike, you will likely be replacing/repairing some things to bring everything back to working order but proper grounding helps reduce the risk and helps keep the entire system at or near the same potential.

[kuhrd]

If you are interested in reading a bit more about this.  Here is a link to the basic fundamentals involved sourced from the Feynman lectures.