DC generator and inverter ground

[PacificGreen]

If you have that proper ground through the 3-wire AC power you shouldn't need to have a separate connection to a grounding rod to earth. If in doubt then your grounding rod connection would give you one for sure. But it should be buried several feet to meet code here in the US.

I have 6000w of inverter power that I will use on board the boat. The only time I will ever use shore power is for the battery chargers that I will seldom if ever need to use and yes I will use them with a proper extension cord.

I will be working on the boat and using the inverters on the boat to power my electrical needs during the time on the dry not running extension cords up to my boat from shore power.

So it seems I will have to have a grounding rod attached to the boat.

[exMaggieDrum]

Correct.

[PacificGreen]

So having resolved that what are your thoughts on grounding my 12vdc runs with my 48vdc runs to the same DC negative bus bar that by way of connection also supplies the non-current carrying green wires a path of least resistance to the ground or sea?

My 12v loads will be supplied with a 120vac to 12v converter.

[roland stockham]

There is some argument about this! European regs ask for all grounds to be interconnected. N America is the opposite. I prefer to go for independent DC, AC and radio grounds. I keep radio separate to reduce the chance of interference. I keep AC and DC separate so that in the event of the ground bolt failing or other machanical failures it is not possible for fault current to flow into the DC neg bus. My DC system is floating i.e. no connection to either shore ground or seawater. AC & radio grounds are to small external bronze plates (made from a foot of scrap flat bar). If using keel bolts check it is not an encapsulated keel! and also be wary if using the keel as the ground plane for radio aerials. Interference again but also a fault current could fry the radio.

[PacificGreen]

Originally Posted by roland stockham

There is some argument about this! European regs ask for all grounds to be interconnected. N America is the opposite. I prefer to go for independent DC, AC and radio grounds. I keep radio separate to reduce the chance of interference. I keep AC and DC separate so that in the event of the ground bolt failing or other machanical failures it is not possible for fault current to flow into the DC neg bus. My DC system is floating i.e. no connection to either shore ground or seawater. AC & radio grounds are to small external bronze plates (made from a foot of scrap flat bar). If using keel bolts check it is not an encapsulated keel! and also be wary if using the keel as the ground plane for radio aerials. Interference again but also a fault current could fry the radio.

While I have four or more easily accessible keel boats. The picture will give you a better idea. What voltage is you DC system? My testing of my system was done for the most part floating but only with 48v as you say but in real life while floating I'm inclined to ground via water with brass bolts and brass bar stock.

Not sure how these keel bolts will like being removed and what is involved in making sure they don't leak after but sure to find out now. So far as I've read over at Grounding Your Circuits - Electric Seas
there was good discussion around 12v being grounded with other systems but that only relates to a 12v system using a 12v battery for current and not a 120vac to 12v converter.

https://www.tc.gc.ca/eng/marinesafet...ection8-67.htm I just found this so it nicely condenses ABYC E-11 for me.

[EllisElectric]

Originally Posted by ramblinrod

I repeat, your AC grounds must be tied to earth to prevent electrocution.

If your AC equipment "grounds" are floating at some potential other than "earth", should your body come in contact with any floating component and earth, you could get zapped.

Should a fault occur where a line conductor comes in contact with anything metal that is ungrounded, it will be live and at AC voltage potential to earth.

If you touch this casing (or anything electrically connected to it) with one hand, and touch seawater (connected to earth) with the other. Good night Irene, forever!

The only exception would be if the vessel was equipped with an ELCI (very good idea by the way), and I still recommend a proper AC ground, as ELCIs can fail.

I can't think of a "good" reason not to ground any marine AC electrical system.

Ramblin Rod
About Sheen Marine

First, what's a ELCI?
Secondly, Ungrounded systems are supplied either by a transformer
or by an independent supply of power (e.g. battery, generator,
etc.). The special aspect of these systems is that no
active conductor is connected directly to ground. In the
case of a fault to frame or ground fault, no short circuit
current can flow as it would do in the grounded systems.
Instead, as a consequence of the lack of a complete circuit,
there will only be a small fault current with a magnitude
defined by the isolation resistances RF and the capacitance
Ce of the conductors to ground.

[senormechanico]

ELCI / GFCI Electrical Shock Protection | West Marine

[sailorchic34]

If one studies the circuit diagram of the typical DC to AC inverter, you will note that the final transformer looks like an isolation transformer. This is for US 120V, but would be true for 220V too.

While shore power is hot, neutral and ground. What you have at the inverter is not hot and neutral, but hot leg to leg. The only current path is between the two legs of the transformer inside the inverter. One leg to ground is not in circuit, so there is no current path and no shock danger.

I should note that with switching power supplies such as a laptop brick, it might be possible for the ground pin on the brick to become hot with a transient should something fail inside the brick. But under normal conditions with resistive and inductive loads there is very little chance of shock, really none, from an inverter on the AC side.

To my poor eyes the typical inverter will behave exactly like an isolated power supply. You can add a ground, but it's never in circuit.

[smac999]

Originally Posted by sailorchic34

If one studies the circuit diagram of the typical DC to AC inverter, you will note that the final transformer looks like an isolation transformer. This is for US 120V, but would be true for 220V too.

While shore power is hot, neutral and ground. What you have at the inverter is not hot and neutral, but hot leg to leg. The only current path is between the two legs of the transformer inside the inverter. One leg to ground is not in circuit, so there is no current path and no shock danger.

I should note that with switching power supplies such as a laptop brick, it might be possible for the ground pin on the brick to become hot with a transient should something fail inside the brick. But under normal conditions with resistive and inductive loads there is very little chance of shock, really none, from an inverter on the AC side.

To my poor eyes the typical inverter will behave exactly like an isolated power supply. You can add a ground, but it's never in circuit.

a marine rated inverter like a magnum has an internal transfer relay that joins the output N to G while inverting. which is an ABYC requirement to have a N-G bond at a source of power (for gen, shore, and inverter), and opens it up while passing though so the bond goes back to the gen or shore instead. the cheap inverters that float the H and N (and sometimes blow up if you connect the N to G...) do not meat this requirement. those I would not use without GFCI. or at all

[ramblinrod]

Originally Posted by EllisElectric

First, what's a ELCI?
Secondly, Ungrounded systems are supplied either by a transformer
or by an independent supply of power (e.g. battery, generator,
etc.). The special aspect of these systems is that no
active conductor is connected directly to ground. In the
case of a fault to frame or ground fault, no short circuit
current can flow as it would do in the grounded systems.
Instead, as a consequence of the lack of a complete circuit,
there will only be a small fault current with a magnitude
defined by the isolation resistances RF and the capacitance
Ce of the conductors to ground.

This is not something to fool with. If you don't know why a marine AC electrical system should be grounded, set down the tools and step away, until you learn it, and never suggest others should have an ungrounded marine AC electrical system.

1. An ELCI is similar to a GFCI receptacle, except that it protects the entire on-board AC system, rather than just the devices plugged into a GFCI protected receptacle. Probably the single most important safety advancement in marine AC electrical systems in the last 5 years.

Anyone who operates an ungrounded marine AC electrical system, without ELCI protection, is risking electrocuting the vessel occupants.

2. Again, there is no good reason to have an ungrounded AC electrical system (120 or 240V) on a boat.

To not ground the AC system is a great way to kill somebody.

If the vessel AC system is not grounded, an accidental or fault connection from the AC line to anything conductive that should be at earth ground potential, will not cause the respective AC breaker to trip (as it does when the AC system neutral is connected to earth ground).

The surface that should be at earth ground potential, is actually live with the AC system voltage. If you touch that surface and anything else that is connected to earth ground (in contact with seawater) with your body, you will most certainly receive an electrical shock.

Ramblin Rod
About Sheen Marine

[Frankly]

A little loose with the truth going on here.

A couple of definitions to begin:

Grounding Electrode “ a device to establish an electrical connection with the earth” (dirt) typically a rod driven at the point of entrance to the building electrical service.

Grounding Electrode Conductor “the conductor used to connect the grounding electrode to the equipment ground” the wire that runs from the rod to the equipment ground connection (typically the equipment ground connection is a ground bus)

Grounding Conductor “a conductor used to connect equipment or the grounded circuit of a wiring system to a grounding electrode or equipment ground connection” typically green or bare conductor

Grounded Conductor (notice the difference grounding vs grounded) “a system or circuit conductor that is intentionally grounded” typically the neutral conductor or white wire

If we drive a ground rod and connect up a grounding electrode conductor to the equipment ground have we created a grounded electrical system? Hardly, we have just grounded the electrical system but it is not a grounded system.

What is the advantage of installing this grounding rod system. (1) provide an electrical path to the earth for lightning. (2) to minimize the step and touch potential between the earth and the electrical service entrance (don’t want to shock the electrical meter reader. (3) provide an electrical path to the earth for static charge build up. (4) There are some additional esoteric reasons that have to do with limiting the voltage when faults occur on the ELCO end of the system. While all these could be associated with preventing electrical shock probably not what you were thinking. One could argue that by establishing this ground connection the system is more hazardous. In reality there is enough leakage resistance and capacitive coupling for any real system that the small current necessary to cause heart problems (doesn’t take much) would exist even without the ground electrode connection. Advantages to having it, so we have it.

How do we create a grounded electrical system and why do we want a grounded system. A system becomes a grounded (notice the ed vice the ing) electrical system when we take one of the current carrying conductors and intentionally connect it to the ground electrode system. In most of the systems we are exposed to this is required by the NEC, and the conductor we connect to the ground is called the grounded conductor (fancy that) and is indicated with white insulation (some call this the neutral conductor). This neutral to ground connection is typically made at the electrical service entrance (at only one point). The reason the NEC requires this arrangement is to establish a low impedance (resistance) path for ground fault currents (typically measured in thousands of amps) that can be detected and cleared in fractions of a second by circuit breakers or fuses. If the high side (usually black wire) contacts the skill saw case (think metal saw case) a high level of current will flow back through the green ground circuit through this connection to the neutral at the service entrance and the breaker will open removing the electrical supply from the skill saw. When you pick it up it won’t run, but you will live to drive down to Home Depot to buy a replacement. The powers that be have decided that this electrical arrangement provides a reasonable level of safety for a reasonable level of complexity.

Are all electrical systems grounded (ie have this neutral to ground connection). Not at all. For instance think about say a big scrap steel operation where a big electromagnet is used to move/ lift tons of steel from location A to location B. A ground fault develops in the circuit feeding the magnet, the breaker trips, current is lost and scrap steel is falling like hail. Probably not an ideal situation. #2 gun turret on the Missouri and a Russian Mig is heading your way. Just as he is in your sights a ground fault in the electrical system feeding the machine gun, breaker trips and all you can do is shoot him the bird as he rakes you over. Also not a good situation. An ungrounded (ie no neutral to ground connection) system will allow operations to continue with a single ground fault. To monitor the condition of this ungrounded system a ground detector is used. A light comes on indicating a ground fault, the system is still operational, and the operator has been alerted that trouble is afoot and the system should be fixed ASAP but hopefully at a more opportune time. Yea I know no Russians were mixing it up with the Missouri. Ungrounded systems (AC and DC) abound just not in your typical everyday encounter, and a ground rod does not make a grounded electrical system.

I write this in the hopes to filter out some of the BS and that by understanding the thinking behind modern electrical systems the OP and others can better understand the problem/lingo to help make their own informed decisions.

Using the house bank or start bank to power a windlass is a matter of opinion, some of the rest of this is serious stuff. Although the context of this is more aimed at shore side electrical systems most concepts still apply to boat/ yacht systems.

As a related but side note, these Honda generators that abound in the cruising fleet have an ungrounded electrical output. That is the reason many question why their reverse polarity lights on shore power inputs glow weakly. They work fine in this manner or if the light bothers you just connect a wire between the neutral and ground pins on a plug and insert it in the extra receptacle slot. Works that way also (now grounded electrical system). The real secret to safe 120 VAC power is GFCI protection.

[sailorchic34]

Originally Posted by smac999

a marine rated inverter like a magnum has an internal transfer relay that joins the output N to G while inverting. which is an ABYC requirement to have a N-G bond at a source of power (for gen, shore, and inverter), and opens it up while passing though so the bond goes back to the gen or shore instead. the cheap inverters that float the H and N (and sometimes blow up if you connect the N to G...) do not meat this requirement. those I would not use without GFCI. or at all

There isn't a neutral on an inverter output. You have two hot legs. The relay would connect G to G and N to N when on shore power. On inverter, you have hot and hot. Actually need to connect the two legs to have current flow. There is no current flow with either leg to ground. So ground would serve no purpose there.

I am aware what ABYC says, I disagree with it in this case as I do with a few other thingys.

[ramblinrod]

Originally Posted by sailorchic34

If one studies the circuit diagram of the typical DC to AC inverter, you will note that the final transformer looks like an isolation transformer. This is for US 120V, but would be true for 220V too.

Correct.

While shore power is hot, neutral and ground. What you have at the inverter is not hot and neutral, but hot leg to leg.

Correct, if and only if the inverter is not wired properly and the inverter AC output ground is not connected to earth. Wired correctly, with the inverter AC output ground connected to earth, you still have line, neutral, and ground, it is just the ground is independent of shore power ground.

The marine inverter isolates the AC output ground from the shore power AC ground, as you cannot have two (distant) earth ground connections to neutral as they could be at different ground potential.

So IF the inverter has a built-in shore power transfer relay, when the inverter is operating, it disconnects the shorepower AC ground from the inverter output AC ground. It is imperative that the inverter AC output ground, be connected to earth (seawater).

The only current path is between the two legs of the transformer inside the inverter. One leg to ground is not in circuit, so there is no current path and no shock danger.

If the inverter AC output ground is not connected to earth ground, and the inverter output is not GFCI protected, any metal AC appliance connected to the inverter AC output that has a line or neutral fault connection will be floating at the AC system voltage. If you touch this and earth ground (anything metallic touching seawater) there most certainly is an electrical shock hazard. This is exactly why marine AC electrical systems have been required to be grounded, well, forever.

I should note that with switching power supplies such as a laptop brick, it might be possible for the ground pin on the brick to become hot with a transient should something fail inside the brick. But under normal conditions with resistive and inductive loads there is very little chance of shock, really none, from an inverter on the AC side.

To my poor eyes the typical inverter will behave exactly like an isolated power supply. You can add a ground, but it's never in circuit.

You must connect an inverter (or shore power AC isolation transformer) AC output ground to an independent earth, for the reason above. It is not safe to leave an AC neutral floating with respect to earth ground.

If the inverter output is GFCI protected, that should cause the fault described above to be avoided. However, GFCIs are notoriously unreliable, and the reason every one of them must have a TEST and RESET button, and the reason the manufacturers are mandated to instruct routine tests.

There is no good reason to have a marine AC electrical system floating with respect to earth ground, it negates the AC panel breaker from tripping on a line fault to ground and risks occupant electrocution.

Ramblin Rod
About Sheen Marine

[sailorchic34]

Originally Posted by ramblinrod

If the inverter AC output ground is not connected to earth ground, and the inverter output is not GFCI protected, any metal AC appliance connected to the inverter that has a line or neutral fault connection will be floating at the AC system voltage. If you touch this and earth ground (anything metallic touching seawater) there most certainly is an electrical shock hazard. This is exactly why marine AC electrical systems have been required to be grounded, well, forever.
Ramblin Rod
About Sheen Marine

Might want to study up on isolated power supplies. Then study the schematic of the typical inverters. The typical inverter output is not hot and neutral, but two hot legs. The ONLY circuit path is from leg to leg. Exactly like an isolated power supply.

There isn't a current path from leg to ground, as ground is not a current path in a inverter output. It's isolated. When on shore power you are correct, a ground is required. Agree 100% there.

On inverter away from dock, the only current path is leg to leg. Touch either leg and ground and zero current flows as there is no circuit path. Once on shore power there is a ground reference and a ground wire, properly installed will be required.

[PacificGreen]

So SC and EE are pretty much saying the same thing in a dissimilar fashion. EE begins defining that an inverter can be an ungrounded system. Then said not to ground beyond the DC bus and let it float. He then denotes that if there were a fault in the 48vdc system it would not short circuit and I'm reading into it create an arc. He then relates a small fault current of unknown amount denoted by RF and Ce which is derived from probably at one point experiencing this which would possibly come from a failed inverter or generator head.

SC gives some theory on transformers and inverters further defining their function Goes on to say if there were a fault present in an inverter conductor the shock hazard would be negligible

SC goes on to relate possible suspect electronics susceptible to ground faults that may precipitate transient stray currents. I have to agree as I went through probably 9 laptops with one reseller for unkowon reasons the machines would simply refuse to boot.

I'm sure everyone is rubbing their poor eyes at this point over how expensive an inverter installation can be. Specifically me.

I only found out today that I need 1/0# from my inverter chassis ground instead of #8awg like the manufacture called for. Plus remove reinstall attach grounding plate replace bolt attach 1/0# to ring termainal and torque nut and affix B to main DC Negative Bus Bar.

Or and just to be clear can I float my simple electric system at 48volts and forego the grounding to water?

Also I'm specifically looking for an answer as to how I would ground my 12v system as that will be supplied by a 120vac to 12v converter without a battery inline supplied from inverter that is in turn supplied from the main 48v bank?

Can it be A B or C?

A) Ground the 12v to the 48v negative dc bus bar?
B) Make a separate 12v ground?
C) Leave the 12v from the converter float?

Also has anyone had to attach a 1/0# wire from an inverter that only has a lug that utilizes a screw down abutment on the aft of the case that only accepts up to 6mm diameter wire. So I'm sure if I tried I could squeeze in a 2awg wire but loose a few strands.

Also I make some assumptions at times in this post and need to be clear on things concerning my grounding and bonding project. I did study quickly on ELCI and find that it concerns shore power dead faults and breaks circuits.

I feel I'm adequately protected by the inverters own GFCI receptacle and that a dead short from there would trip the circuit. If a leg of the 48v were to contact another leg so positive to negative short it would spark and perhaps melt something usually tripping circuit protection.

Regardless the inverter will get a chassis ground and so will the dc generators and the battery case. The question remains for me to bond to water or not given my system?