grounding of ac dc systems question (few different scenarios)

pltek

New member
I have a battery disconnect on the negative of the starter battery. It disconnects the chassis from the negative terminal of the starter battery. The negative cable from the Sterling is connected to the chassis side of the battery disconnect. The negative side of my house batteries and inverter are also connected to the chassis in the back, so there are really two paths for the Sterling to ground. If the Sterling negative wire was connected to the starter battery minus side of the disconnect, the disconnect would not be effective as there still would be a fairly poor connection from the battery to chassis via the long wire to the Sterling.

Regards,

Mark
hi Mark,

so i am in final phase of the install and i was going to run a long wire from sterling (back of the van) to the quick disconnect (near the pedals) i was trying to unscrew the post of the disconnect planning/hoping that i can stick the lung under that and screw ther post back in. That thing is in there permanently.

So is there another way to connect to the battery negative, say under the driver seat where i am going to tap the positive wire to the distribution block?
 

marklg

Well-known member
hi Mark,

so i am in final phase of the install and i was going to run a long wire from sterling (back of the van) to the quick disconnect (near the pedals) i was trying to unscrew the post of the disconnect planning/hoping that i can stick the lung under that and screw ther post back in. That thing is in there permanently.

So is there another way to connect to the battery negative, say under the driver seat where i am going to tap the positive wire to the distribution block?
I've got a T1N, so the battery and the disconnect are under the hood, completely different from yours.

You really do want the negative of the Sterling on the chassis side of the disconnect or the disconnect won't work and the engine ground current will go through the Sterling negative wire, which can cause all manner of problems. You can connect to any chassis point near the battery or engine too.

I don't know about the disconnect in your vehicle, bu if it is a corroded bolt, have you tried Kano Kroil penetrating oil and an impact driver? That's never failed for me.

Clean it with a solvent and add some Noalox or similar when you put it back together.

Regards,

Mark
 

Kevin.Hutch

2011 Mercedes 313 906
There is a common misconception as to the terminology of powered wires.

All wires that have power are "live" wires and they come in pairs only sometimes active/hot/neutral.

As soon as one live wire is connected to earth it becomes a neutral and the other related to it is active or hot.

The earth or ground connection is a "protective" ground and should only carry current in case of a fault.

DC systems commonly found in vehicles are chassis return as most vehicles are steel or conductive. A protective ground is irrelevant as vehicles sit on rubber tyres insulating them from the real ground, so by definition are isolated systems unless connected to an outside power source that has a neutral connection (most generators or inverters do not). Then a protective ground connection is required, usually by law. The neutral/earth connection is never made within the vehicle or standard GFCI/RCD devices will false trip on connection.

When you introduce nonconductive components in the vehicle construction the power system generally becomes a two-wire system and if less than 100v no protective earth requirement.

In the case of a boat, isolated systems for all power connections are the most commonly used to reduce the problem of electrolysis caused mainly by ground loops, yet metallic parts are generally bonded with all components individually connected to a common grounding point in a tree like fashion.

The safest system is an isolated system as it takes two faults to create a dangerous situation, but only if a method of detection of a faulty live connection to ground can be detected. Under 40v DC and 100v AC systems are generally not considered dangerous so the earthing decisions only become an issue when an external source of 100v+ power is involved.
 

avanti

2014 GWV Legend 3500 I4
A protective ground is irrelevant as vehicles sit on rubber tyres insulating them from the real ground, so by definition are isolated systems unless connected to an outside power source that has a neutral connection (most generators or inverters do not).
This is incorrect. There is nothing magical about the "real" earth. You can think of it as a big chassis that is convenient for tying lots and lots of returns together. It will work the same way on Mars. And, it works the same way (on a smaller scale, obviously) on your van. In both cases, the protective ground is protective precisely because there are many paths by which accidental faults will find their way to the "chassis", thus blowing a fuse or tripping a breaker. The fact that your chassis is not bonded to "real" earth does not eliminate this value..
The safest system is an isolated system as it takes two faults to create a dangerous situation, but only if a method of detection of a faulty live connection to ground can be detected.
Also incorrect. As has been thoroughly explored elsewhere on this list, isolated systems are more reliable (because faults are harder to create), but LESS safe, for exactly the same reason. All that exposed steel all over the place provides many opportunities for faults to find their way to "ground", blowing a fuse in the process. Blown fuses on single faults are a GOOD thing. They increase safety, rather than decrease it. Requiring two faults before the fuse blows is a BAD thing (for safety, but not for reliability). Moreover, a un-bonded chassis can easily become "hot" in many fault scenarios, which can itself be a danger (sometimes an extreme one). I do realize that you stipulated that an isolated system requires "a method of detection of a faulty live connection to ground can be detected." But, such a method is inherent in a properly fused bonded chassis setup.

There is an immense amount of passion around here on this topic by folks who think they understand the logic of grounds but don't. I'm sure we will hear from them. I am not going to participate further, since as I said, this has been exhaustively explored and explained elsewhere.
 
Last edited:

Kevin.Hutch

2011 Mercedes 313 906
This is incorrect. There is nothing magical about the "real" earth. You can think of it as a big chassis that is convenient for tying lots and lots of returns together. It will work the same way on Mars. And, it works the same way (on a smaller scale, obviously) on your van. In both cases, the protective ground is protective precisely because there are many paths by which accidental faults will find their way to the "chassis", thus blowing a fuse or tripping a breaker. The fact that your chassis is not bonded to "real" earth does not eliminate this value..
Can not help to respond here an isolated system is one that is not connected to ground/earth and ground/earth is as different to the chassis, when not connected by a protective ground/earth conductor, as is the other live power connection. It requires a circuit to be completed for current to flow and if not electrically connected it is just confusing calling a common ground or earth. In a van the most dangerous situation is standing on the ground and touching a live metal part of the van, not possible with a properly setup isolated system.

Also incorrect. As has been thoroughly explored elsewhere on this list, isolated systems are more reliable (because faults are harder to create), but LESS safe, for exactly the same reason. All that exposed steel all over the place provides many opportunities for faults to find their way to "ground", blowing a fuse in the process. Blown fuses on single faults are a GOOD thing. They increase safety, rather than decrease it. Requiring two faults before the fuse blows is a BAD thing (for safety, but not for reliability). Moreover, a un-bonded chassis can easily become "hot" in many fault scenarios, which can itself be a danger (sometimes an extreme one).
Isolated systems can easily be set up to verify when the first fault occurs by the use of a Residual Voltage Detector that simply detects if more than 40v exists between any live conductor and the chassis that is usually identified incorrectly as the ground. With this extra protection an isolated system is intrinsically safer and is regularly used where safety is paramount. In marine where salt water provides a good conductor to earth and earth return systems are rare the regulations mandate the detection of earth faults that GFCI or RCD units do not, as they only identify a current differential between the two live wires.
 
Last edited:

avanti

2014 GWV Legend 3500 I4
In a van the most dangerous situation is standing on the ground and touching a live metal part of the van, not possible with a properly setup isolated system.
You are confused. The so-called "hot skin" danger can only occur if the source of the current is itself bonded to ground. If the current is coming from and on-board source, such as a genset, you cannot be harmed by standing on the ground and touching live metal, whether or not the genset is bonded (as it certainly should be). Moreover, as you yourself explained, code REQUIRES that the chassis be bonded to earth when power is coming from an external source, for this very reason. "Hot skin" faults are only possible when this requirement is not met. In no way are they an argument against chassis grounds.

Professionally-designed vehicle power systems virtually always use chassis grounds. DIYers would be well-advised to follow suit.
 

Kevin.Hutch

2011 Mercedes 313 906
You are confused. The so-called "hot skin" danger can only occur if the source of the current is itself bonded to ground. If the current is coming from and on-board source, such as a genset, you cannot be harmed by standing on the ground and touching live metal, whether or not the genset is bonded (as it certainly should be). Moreover, as you yourself explained, code REQUIRES that the chassis be bonded to earth when power is coming from an external source, for this very reason. "Hot skin" faults are only possible when this requirement is not met. In no way are they an argument against chassis grounds.
No that is correct, it is not possible to have a "hot skin" situation if the onboard power system is an isolated one, that is isolated from ground.
Further AC systems and DC should not be electrically connected together.
 
Last edited:

avanti

2014 GWV Legend 3500 I4
Further AC systems and DC should not be electrically connected together.
Why is that?

Every time you use a "real" ground, you are doing exactly that--connecting your circuit (quite literally) to billions of other circuits in the world, AC, DC, random voltages, etc. Works fine. Once again, there is no logical difference between "earth" and your chassis. The earth is equivalent to a huge wire.
 

Kevin.Hutch

2011 Mercedes 313 906
Spot on when there is no logical/electrical difference between the chassis and ground then all references are equal.
When you change to an earth isolated system then any faults can take paths via any (ac or dc) circuits and safety is compromised, not to mention ground loops and electrical interference.
 

avanti

2014 GWV Legend 3500 I4
Spot on when there is no logical/electrical difference between the chassis and ground then all references are equal.
When you change to an earth isolated system then any faults can take paths via any (ac or dc) circuits and safety is compromised, not to mention ground loops and electrical interference.
I give up. I recommend that anyone considering following this advice to dig deeper.
Common AC/DC references cause ground loops? :crazy:

EDIT: Obviously, we are talking about connecting the DC return to the AC safety ground, not the AC neutral. Per code, one and only one such bonding must be present.

I'm out.
 
Last edited:
No. (but, then: what's "proper" in your eyes?)

The inverter's metal box is *not* connected to the DC negative wiring.
If part of the inverter's 120 volts AC section had a spectacular (or simply Murphy's Law) failure, it could make the inverter's metal box "hot".

If the box was not "grounded" (by either the "bolt the feet down" or "dedicated wire" method) then anyone (or anything) touching the box would also now be "hot". (which, inside the typical RV, would probably still be fairly safe).

--dick
Maybe I'm not understanding your comment, but Victron shows many wiring diagram examples with the inverter casing grounded to the negative bus bar and the bus bar grounded to the chassis (or earth). I've followed their design in my system....

A ground that I never see illustrated is on Victron Smart Solar Controller 100/50. The casing has a grounding point, so I grounded it...

1611602458752.png
 

pltek

New member
I've got a T1N, so the battery and the disconnect are under the hood, completely different from yours.

You really do want the negative of the Sterling on the chassis side of the disconnect or the disconnect won't work and the engine ground current will go through the Sterling negative wire, which can cause all manner of problems. You can connect to any chassis point near the battery or engine too.

I don't know about the disconnect in your vehicle, bu if it is a corroded bolt, have you tried Kano Kroil penetrating oil and an impact driver? That's never failed for me.

Clean it with a solvent and add some Noalox or similar when you put it back together.

Regards,

Mark
i got the post removed with wrench/socket placed lug on chassis bolt, screw the post back on and now have this ground running back to my sterling negative. The only other thing that is confusing as I finalize this and overall grounding is that the sterling diagram shows that aside for the cable i just ran, there should be another negative cable that connects starter battery and auxillary battery bank. is that really necessary??
 
Last edited:

autostaretx

Erratic Member
Maybe I'm not understanding your comment, but Victron shows many wiring diagram examples with the inverter casing grounded to the negative bus bar and the bus bar grounded to the chassis (or earth). I've followed their design in my system....
Which is a fine way to go... this means that, in the event of an internal inverter failure making its box "hot", the casing being grounded to the negative bus which is ALSO bonded to the vehicle chassis means that YOU won't end up acting as that wire if you touch the box and the frame at the same time.

Scenario when (without that bonding) you could get hurt: inverter fails. You open access panel and reach in to turn it off (or switch master disconnect or "just to look"). You use your left arm to support your leaning weight by the edge of the access hatch (cassis metal). You reach in with your right arm to flip the switch ... it accidentally (it's dark in there) touches the inverter's box (or just a screw head on the box). You now (without that case-ground bonding to the frame) potentially have an across-the-heart electrical situation. Not good.

--dick
 

Top Bottom