[kenryan]

I would look at just doing double stacked cables from the batteries to the Class T Holder. BEP marine's version has radiused ends allowing the lugs to be not just straight on or 90 degrees. BEP Marine Class T Holder. If you need a clean 3 terminal bus bar BEP makes one of those BEP heavy duty bus bar.

Thanks @Scalf77 Don't want to use cables, but thanks for the tip on the 3 terminal heavy duty bus bar. Appears to be exactly what I need.

 

[cycle61]

AIC=amps interrupting current. It's a common concept in engineering industrial and commercial electrical protection systems but not so much in smaller applications. It means the maximum fault that a device can safely interrupt. The device, a fuse or breaker, needs to have a higher rating than the maximum amperage that the source is capable of delivering under worst case scenario. For instance, say a dead short across your battery terminals can produce 7000 amps for a few fractions of a second...if your protective device is rated at 5ka it might theoretically fail to interrupt the fault and leave your system to burn.

It's not directly related to the size or voltage of your battery or energy source, and it's not about the current rating of the fuse or breaker. For large high voltage systems we do all sorts of math, starting with data from the utility and then calculating based on cable size, conductor length, what devices are installed, additional sources, etc. For a van build, since I couldn't find good data on Victron Lithium battery fault current, I installed T-class fuses which were the highest AIC available.

 

[gregoryx]

I keep reading the thread... and I keep thinking how I thought that the built-in switch and fuse in the batteries I got was cool... now I'm leaning more and more toward, "every storage battery should have a fuse and switch built in." 🤔

 

[njvagabond]

I keep reading the thread... and I keep thinking how I thought that the built-in switch and fuse in the batteries I got was cool... now I'm leaning more and more toward, "every storage battery should have a fuse and switch built in." 🤔

Send a letter to your congressperson. 😛

 

[natecostello]

Thanks. Very helpful for the AC side, which I have yet come to a good understanding of yet.

Asking about concern was asking about the charger output from an inverter charger and the t-class fuse was at the "end" of that circuit several feet away. It would seem to me that it might be good to have have a circuit protection device close to the charger output (/inverter input). Say an ANL fuse (or 187 breaker) of equal amperage to the t-class at the battery end. I'd appreciated you educated input on this question.

@kenryan, I'm really sorry for polluting your thread with so much text, but some of this is relevant to your design.
@njvagabond, since you asked:

I found a full version of E-11, but its the 2003 version. If anyone has access to a later version that isn't $195 I'd be much obliged. Here's the relevant DC requirements:

11.12. OVERCURRENT PROTECTION
11.12.1. FOR DC SYSTEMS
11.12.1.1. Battery Charging Sources
11.12.1.1.1. Each ungrounded conductor connected to a battery charger, alternator, or other charging source, shall be provided with overcurrent protection within a distance of seven inches (175mm) of the point of connection to the DC electrical system or to the battery.
EXCEPTIONS:
1. If the conductor is connected directly to the battery terminal and is contained throughout its entire distance in a sheath or enclosure such as a conduit, junction box, control box or enclosed panel, the overcurrent protection shall be placed as close as practicable to the battery, but not to exceed 72 inches (1.83m).
2. If the conductor is connected to a source of power other than a battery terminal and is contained throughout its entire distance in a sheath or enclosure such as a conduit, junction box, control box or enclosed panel, the overcurrent protection shall be placed as close as practicable to the point of connection to the source of power, but not to exceed 40 inches (1.02m). Overcurrent protection is not required in conductors from self-limiting alternators with integral regulators if the conductor is less than 40 inches (1.02m), is connected to a source of power other than the battery, and is contained throughout its entire distance in a sheath or enclosure.
11.12.1.1.2. In addition to the provisions of E11.12.1.1.1, the ungrounded conductor shall be provided with overcurrent protection within the charging source, or within seven inches (175mm) of the charging source, based on the maximum output of the device.
EXCEPTION: Self-limiting devices.

11.4 DEFINITIONS
Self-limiting - A device whose maximum output is restricted to a specified value by its magnetic and electrical characteristics.

This is pretty consistent with the ABYC requirements for AC (which I referenced originally) and clarifies things a little more than Nigel Caldor's rephrasing in one case.

Per the exception to 11.12.1.1.2 for "Self-limiting devices", defined as a device whose maximum output is restricted to a specified value by its magnetic and electrical characteristics, overcurrent protection is not required at or within 7" of the charger or alternator.

My (and others) interpretation of that means something like the victron multiplus only needs one fuse or breaker at the point of connection to the system, and not at the multiplus. I would consider it self limiting because it isn't capable of sourcing much, if any, more than its rated DC output current. A similar, but perhaps less defensible, argument can be made for an alternator.

 

[kenryan]

Not at all @natecostello appreciate the inputs. My system will be better thanks to this discussion.

 

[Michael Ophus]

I used to have a lot to say about these kinds of things but copd and the lack of oxygen make it too hard to think.
My Magnum MS2012 has been working great for going on seven years, I see no need to ever replace it.

 

[njvagabond]

@njvagabond, since you asked:
I found a full version of E-11, but its the 2003 version.

THANK YOU VERY HELPFUL!
Not that I am probably not the top polluter of threads, but I figured it might be relevant to the discussion.

Still need to wrap (unintentional pun) my head around the wire protection requirements in the cited exception.

While looking for the ABYC definition of sheath I did come across the 2003 version of the ABYC electrical standards that you mentioned. No help to you but maybe for others. Also came upon this document that correlates some of the relevant CFRs with the ABCY standards that are relevant to this discussion.

 

[kenryan]

I would be interested in answers to the following questions:

1. Is there a standard for RVs (or motor vehicles in general) that serves the same function as ABYC does for boats?
2. If there is, what does it have to say with regards to ELCI protection?

 

[natecostello]

I would be interested in answers to the following questions:

1. Is there a standard for RVs (or motor vehicles in general) that serves the same function as ABYC does for boats?
2. If there is, what does it have to say with regards to ELCI protection?

I believe its the NFP something or other that makes the residential NEC that is applicable. However, the last time I looked I couldn't find anything in their code that would preclude using solid core copper Romex or wire nuts in an RV. That was enough to make me look past it towards a more conservative and applicable code. But, I haven't done an exhaustive search of their code and I could be wrong. I had asked a similar question elsewhere on this forum but never got an answer that challenged my initial review.

 

[njvagabond]

I would be interested in answers to the following questions:

1. Is there a standard for RVs (or motor vehicles in general) that serves the same function as ABYC does for boats?
2. If there is, what does it have to say with regards to ELCI protection?

RIVIA has standards, but that is a organization funded and run by the industry for their benefit. Most comments I have seen do not have much regard for their standards. I believe there parts of the FMVSS that apply, there may be other federal regulations on vehicles that apply. There are probably also ASME and ISO standards for vehicles.

ELCI - I have only found info on that as it relates to boats, usually in discussions of ABYC standards. What confuses me is why that over GFCI. GFCI trips at 1/10 of ELCI, so I would think that would be better (and typically less costly) protection. All I could glean was that ELCI is intended to cover the whole vessel were there could be multiple circuits each with small leakage. So maybe to avoid nuisance trips which might lead to users overriding or removing the protection. Along the same line to provide some allowance for crappy shore power. TOTAL GUESS on my part, thinking maybe analogous logic to code of minimum distance that smoke detector can be to a stove?

I see @natecostello is in the house at the moment. Maybe he can add some insight into the GFCI vs ELCI since I am sure he did his usual informed due diligence to decide on his configuration that he described earlier in this thread. Also, on the AC distribution side I was looking at a the dual function arc fault/ground fault device that @jkmann brought to my attention. Seems like that might make a lot of sense for 120V wiring in a metal box.

 

[kenryan]

My immediate issue is that I have already purchased BS 8077 and the much more expensive BS 8100 appears to be unavailable for the foreseeable future. I am also concerned about nuisance tripping. When I built my house about 25 years ago, arc fault breakers had recently been required for bedroom circuits. My electrician warned me that they are subject to nuisance faults, but I have not had any problems with them, so who knows? Admittedly the bedroom circuits get very little use.

I want to be safe, but I have this feeling in my gut that my current plan which involves using good installation practices and gfci outlets on each 120v circuit is probably good and sufficient. I am always open to having my mind changed. If I was starting from scratch, and the BS 8100 was actually available, I would go that route. But reality intrudes.

 

[njvagabond]

But reality intrudes.

Maybe something from here?
GFCI Store

 

[kenryan]

Maybe something from here?
GFCI Store

Interestingly, Magnum provides specific guidance in their manual:

2.5.3 Recommended GFCI (Ground Fault Circuit Interruption) Outlets
In compliance with UL standards, Sensata tested the following series of GFCIs and found that they
function properly when connected to the inverter’s AC output:

• Hubbel #GF520EMBKA
• Cooper VGF15W
• Leviton 8899-A, N7899, GFNT1, or GNNT2

 

[natecostello]

I see @natecostello is in the house at the moment. Maybe he can add some insight into the GFCI vs ELCI since I am sure he did his usual informed due diligence to decide on his configuration that he described earlier in this thread. Also, on the AC distribution side I was looking at a the dual function arc fault/ground fault device that @jkmann brought to my attention. Seems like that might make a lot of sense for 120V wiring in a metal box.

So I learned a lot about ELCI and GFCI back on Van 1.0. Back in '17 I was weighing whether to leave my DC system ungrounded, as others have done (including Orton). In researching this, I first came across some of Caldor's writing on the subject which intersects with GFCI/ELCI and the philosophies behind each. Unfortunately I can't find the specific article where he discusses it at length, but I think what I could find will answer your questions:

RCD = "Residual Current Device" which encompasses both ELCI and GFCI

The Europeans and Americans have different philosophies when it comes to RCD devices (which measure leakage of current from the ‘hot’ and neutral wires to earth, and trip when the current exceeds a certain level). In Europe, the trip limit is generally set at 30 mA (30 milliamps) whereas in the USA it is set at 5 mA as a result of tests conducted many years ago that established 5 mA as the level at which leakage current in the water began to effect muscle control in humans (this was determined by suspending student volunteers in a swimming pool and ramping up the current!). In the damp marine atmosphere 5 mA is low enough to be susceptible to nuisance trips, so the Americans protect individual outlets on the boat (this puts only one circuit, and the leakage current from that circuit, on each RCD, as opposed to putting the whole boat, with its accumulated leakage currents, on a single device) whereas the Europeans put the 30 mA RCD on the incoming shorepower line and use the single device to protect all the circuits on the boat. European standards then assume that this will protect against leaks to earth and as such when a whole boat RCD is fitted permit the boat to be wired without the AC earth circuits connected to DC negative (something that is otherwise important from a safety perspective). On boats that do not have an isolation transformer, not making this connection helps to prevent galvanic corrosion (especially important with metal boats). However, in the USA a considerable amount of data has been collected from domestic RCD use to show that failure rates are high (principally due to lightning-induced voltage surges) so even with a whole boat RCD the American standards advocate the AC earth to DC negative connection, and then recommend either a galvanic isolator, or an isolation transformer, to break the galvanic circuit back to shore. (I know some readers will be scratching their heads at this point! Galvanic isolation is a another complex topic I do not have space to address here.)

Here is a link to the source of the above quote. Also, it was based on his writing that I dropped the ungrounded DC concept.

EDIT: If you want to read a crazy article that deals with RCD on boats and demonstrates how fundamental physics can bear on pedestrian subjects like ELCI (specifically freshwater vs saltwater risk). Check this one out. You'll be shocked to find that what we often think are drowning deaths are actually related to electric fault induced loss of muscle control. I know...terribly dark pun...I couldn't help myself.

 

[gregoryx]

Great article there, @natecostello. TFPU.

Still leaves me with the same question I always come back to on land-yachts: isn't the whole thing isolated from earth by the tires? And doesn't that make it rather different than a boat? Especially a boat in salt-water; but I'd think even "fresh water" would allow more conduction than through our tires. 🤔

It always takes me back to the ground-straps that high-end stereo folks used to run - typically from differential or suspension point to the road - and how those /seem/ like the answer... but wouldn't survive my off-road adventures.

I haven't found much that addresses this. Thoughts? Or at least some additional insight of why the boat examples for ground seem so directly relevant? Maybe I'm wrong and there is more "earthing" going to earth through the tires than I think?

 

[natecostello]

It is iff

Great article there, @natecostello. TFPU.

Still leaves me with the same question I always come back to on land-yachts: isn't the whole thing isolated from earth by the tires? And doesn't that make it rather different than a boat? Especially a boat in salt-water; but I'd think even "fresh water" would allow more conduction than through our tires. 🤔

It always takes me back to the ground-straps that high-end stereo folks used to run - typically from differential or suspension point to the road - and how those /seem/ like the answer... but wouldn't survive my off-road adventures.

I haven't found much that addresses this. Thoughts? Or at least some additional insight of why the boat examples for ground seem so directly relevant? Maybe I'm wrong and there is more "earthing" going to earth through the tires than I think?

There are certainly differences between the boat and van situation. Here’s the conclusion I came to back when I was designing and immersed in this stuff, from my notes:

“
An ELCI will be used for the Shore Power Disconnect Circuit Breaker
Use of an ELCI as the shore power disconnect circuit breaker will provide a trip:

• if a hot-to-ground fault occurs upstream of the Main Panelboard ELCI while connected to shore power with a functioning ground.
• if a latent hot-to-ground fault exists upstream of the Main Panelboard ELCI with an open shore power ground, when a person completes the path to earth (e.g. touches van metal while standing on earth).
  
  In the two cases above, and ELCI or GFCI at the shore power source should also provide the same protection, but will not be relied on. Additionally, if the fault occurs downstream of the Main Panelboard ELCI, it would also provide the above trips.”

 

[gregoryx]

It is iff

There are certainly differences between the boat and van situation. Here’s the conclusion I came to back when I was designing and immersed in this stuff, from my notes:

“
An ELCI will be used for the Shore Power Disconnect Circuit Breaker
Use of an ELCI as the shore power disconnect circuit breaker will provide a trip:

• if a hot-to-ground fault occurs upstream of the Main Panelboard ELCI while connected to shore power with a functioning ground.
• if a latent hot-to-ground fault exists upstream of the Main Panelboard ELCI with an open shore power ground, when a person completes the path to earth (e.g. touches van metal while standing on earth).
  
  In the two cases above, and ELCI or GFCI at the shore power source should also provide the same protection, but will not be relied on. Additionally, if the fault occurs downstream of the Main Panelboard ELCI, it would also provide the above trips.”

Makes total sense. Thank you. Leaves me curious about the way I've been doing "shore power" at least from home: the outlet I use is a GFCI - but at the house, not inside the rig. Inside the rig... even worse and weirder, since I'm coming from an isolated (apparently) inverter into the Victron Multiplus.

I'm still curious about the ground-to-body premise for DC and AC circuits while NOT connected to shore power. When on shore power, the premise of routing all the grounds to the shore "earth" make a lot of sense. IIRC, this is part of the reason for the "Ground relay" function in the Victron Multiplus.

And that reflects back up to the above question as well: obviously, ground relay is ideal when plugged into shore-power; but should probably be /disabled/ when running off my isolated DC-AC-DC inverter? Ugh... so many variables!

Ground relay
Used to enable/disable the internal ground relay functionality. The ground relay is useful when an earth-leakage circuit-breaker is part of the installation.

When the internal transfer switch is open (inverter mode) the Neutral of the inverter is connected to PE. When the transfer switch closes (AC input is transferred to the output) the Neutral is first disconnected from PE.

Warning: Disabling the ground relay on "120/240V" models (split phase models) will disconnect the L2 output from the inverter.

 

[natecostello]

Makes total sense. Thank you. Leaves me curious about the way I've been doing "shore power" at least from home: the outlet I use is a GFCI - but at the house, not inside the rig. Inside the rig... even worse and weirder, since I'm coming from an isolated (apparently) inverter into the Victron Multiplus.

I'm still curious about the ground-to-body premise for DC and AC circuits while NOT connected to shore power. When on shore power, the premise of routing all the grounds to the shore "earth" make a lot of sense. IIRC, this is part of the reason for the "Ground relay" function in the Victron Multiplus.

And that reflects back up to the above question as well: obviously, ground relay is ideal when plugged into shore-power; but should probably be /disabled/ when running off my isolated DC-AC-DC inverter? Ugh... so many variables!

At this point the topic of grounding, GFCI, and ELCI probably deserves a dedicated thread. Maybe there is one. Alternatively, if folks want to have a video conference to go through some of that stuff it might be worthwhile and I'd be game. It would be fun to put faces and voices to names.

@gregoryx can you elaborate on your shorepower path? Is it house GFCI -> some inverter with a built in transfer switch -> Multiplus mains AC input?

Regarding ground-to-body while NOT connected to shore power, I don' t THINK this is a safety concern. If you are standing on the ground and reach in and touch a hot wire, the flow path to the AC system ground would be hot->you->your shoes->earth->tires->van chassis->inverter. If the van was wet, and salty, maybe. I still wouldn't try it. In any case, if there was and ELCI or GFCI in between the inverter and what you touched, if that resistance is lower than I think it is, and you draw 30mA or 5mA respectively, the trip would protect you.

Regarding your last question, many inverters have the neutral connected to the chassis which is in turn grounded to the vehicle structure. If this is the case the system behaves the same way. Can you provide details on which model inverter you have for your orton method implementation? Also, is your house DC system grounded to the vehicle chassis? You give me that and I can try and work out whats best/safest.

 

[gregoryx]

I'm game for a chat-session if others are. I've enjoyed meeting a few of the folks here - even got some skiing and biking time with a couple.


I feel like I'm answering these questions in an odd manner... but see if the overall input makes sense...

We have a Giandel 2200W inverter on the DC-AC-DC side. Nothing fancy about that, as far as I know. No transfer switch on the Giandel, obviously. And we didn't install a transfer-switch for the above reasons. But now that I think about it, though... I think there's a chassis-ground for the Giandel that I never connected - assuming it was already covered by the ground side of the Giandel unit. But I bet it's not since the docs say that the AC neutral is not bonded to ground and they recommend attaching the chassis ground to earth. I have not connected it. Hm... interesting one.

The Giandel inverter is on the far right in the photo.

That unit feeds a single "AC out" that is non-GFCI. I think that was an early-testing thing that I completely forgot about until just now - probably should be a GFCI there, right? That outlet is typically plugged into the "shore power in" plug just to the right of it with a 1' cord.

That "shore power in" (obviously inside the van) feeds the Multiplus 24/3000/70 unit and is seen as "shore power" when in DC-AC-DC mode. When actually connected to shore-power, we run an extension cord through the window and plug it into that plug. We do it very rarely, so it works fine for us. But when we do it's almost always from our house with a dedicated GFCI straight off the main panel - seemed the right thing to do since I'll push that thing as far as 20A when I'm using it (as above, it's a 20A GFCI). Makes me wonder, though... should I have a GFCI or ELCI between that shore-power input and the Multiplus? 🤔

The Multiplus has a built-in transfer switch and the ability to pass through the shore-power ground or not (as well as combining shore plus inverter for extra power and/or UPS functions). I typically leave the ground pass through feature turned on as it seems like the right thing to do when on shore-power... but I'm unsure if that's a bad thing when in DC-AC-DC mode, since the Giandel is likely isolated - though I'm not sure about that. This is one thing I'm unsure of: might be better to NOT pass it through and rely on the chassis-ground path since that is more commonly in use? 🤔

The output side of the Multiplus goes into two GFCIs - one on each side of the van that feeds outlets and loads. Each of the two GFCIs has additional outlets strung off the back of them.

Interesting thing: at one point, someone dumped a bunch of dishsoap down the back of the galley and swamped an outlet back there - when I initially only had the one GFCI shown above with everything off the back of it. It popped the GFCI. And, of course, it was not just going to "reset" until that outlet was removed from service. That's when I broke it into two sets of GFCIs - so the whole system wouldn't be down for one such mistake. Initially, I installed a GFCI in place of the ruined outlets and that did NOT work. Duh. Can't string GFCIs in series. Things I don't think about... 😬

Let's see... yes, the house DC system shares a ground to the chassis; but it's a fairly small wire (#4, IIRC) relative to the possibility of the Multiplus trying to find it's way back there. I based it on the likely load paths, not on the worst-case failure path. And I still don't really understand all the implications thereof.


On the transfer switch thing... my brother recently set up almost the same config and hadn't installed his ATS when he plugged into shore power and forgot to unplug the Giandel he's using... fried that thing.


Okay... educate me! And thank you! 🙏