Can I use a Victron Isolation Transformer on the input side to prevent GFI's tripping while charging

@SamuelSullivan As Kevin has suggested, an earth to neutral bond downstream of the shore connection GFCI will cause it to trip. A picture of the inside of the distribution panel would be helpful. Upon ruling out an incorrect bond, I stand by my original assessment.

@kevind1 North American GFCI are rated in the neighborhood of 5ma, our European counterparts are somewhere in the neighborhood of 30ma. There are MANY systems here that have been verified to be properly wired and installed that have presented this issue, and Victron has acknowledged the issue is a result of internal component leakage. While I agree that anyone experiencing this issue needs to do their due diligence in ensuring proper system wiring and configuration, we can’t automatically assume faulty a installation.

This is a common enough issue that we warn our customers about using a 15A or 20A outlet for charging from an outlet inside the garage or on the outside of the house because it likely has a GFCI device in the circuit. I always suggest a 30A or 50A RV outlet if they can do it, wire a new circuit with a 20A circuit w/o GFCI as a dedicated RV charging outlet (though this likely won’t pass any house inspection), or remove the line side GFCI device on an existing garage circuit (same issue with safety and inspection).

It’s annoying and I don’t expect Victron to address the root cause of the issue unless one the OEM RV manufacturers force them.

I have been doing the same. Unfortunately I agree with you, we will probably not see a fix for this from Victron. I imagine the North American mobile market is probably just a drop in the bucket compared to their global market.

Yes, but I would bet the mobile market is Victron’s largest market in North America, followed by off-grid stationary. I might be wrong, but I don’t think so. In other areas of the world, Victron has grid interactivity, but not here, so Victron is trying to make inroads with RV manufacturers here in the USA to grow their business (I know this on good authority) and I have asked my Victron rep what those OEMs are saying about this (if they have figured it out yet). We’ll see. Happy Independence Day!

I wonder why the USA has a far lower GFCI current setting (5mA) than the EU and the rest of the world (30mA)?

Lots of equipment worldwide – including Victron – is designed to not cause any problem with the 30mA standard (e.g. leakage through internal noise filtering capacitors), but may have issues with a 6x lower limit in the USA…

@kevind1 @thomasinaz Sorry for the wait. Here are the photos and a wire diagram I put together real quick. Let me know if you have any questions that you would like clarified! I may have left something out of the diagram, so if anything looks weird I can double check for you. Kevin, you mentioned a long extension cord could be the issue. What length would be the maximum you would recommend for a 15amp cord to run trickle?

Wire Diagram.pdf (26.0 KB)



I’m not an international code expert, but I have always been told that in the USA RCD’s are designed to protect people first and not equipment. In the USA, our electrical code is governed by fire codes, and they have chosen 5mA, which causes lots of nuisance trips on the equipment side, but protects the people side. Regardless of whether or not we think it should be such a low trip current, the fact is the code requires and equipment sold in the USA shouldn’t have a leakage current above 5mA. I’m not stating that as a legal matter, just a design matter. But in practical terms, it’s not a safety issue, just a minor nuisance that can be designed around.

It’s funny that the rest of the world – including the EU, which is if anything keener on safety in general than the US – seems to think that 30mA is a safe GFCI leakage threshold to protect people, it’s certainly nothing to do with protecting equipment.

Or are you seriously suggesting that the US alone is right and all the other electrical certification authorities are wrong?

Of course it could be designed around, but two points to consider – why should the ROW (much bigger market!) do this just to suit the USA, and what is the downside of doing this?

(for example, does the EU have more stringent EMC requirements than the US which mean more input capacitance/filtering is needed, which causes more earth leakage current?)

Samuel

Wiring in panel looks good except for lack of ferrules.
and it appears no ferrules on the NSI connectors in the junction box.

If you have a multi meter, set it to continuity (when you touch the two probes together you get a beep or indicator) and only proceed if you feel comfortable doing the steps below safely
wear safety glasses and have a buddy.

Turn off both inverters (using switch on front of each) and disconnect from shore power

Touch one probe to top left lug where white wires are and top right where green wires are in the panel, you should not have continuity, no beep.
This will test if there are any wiring issues inside the trailer, called a neutral isolation test.
if you get a beep / continuity, disconnect the two big green wires at the top right in panel and test again touching probe to ground bar where the wires were in the panel and the white wires on the left.
if you still get a beep you have a wiring issue in the trailer, if you don’t you have an issue with the inverter ground relay.

If no beeps above then connect your 120v shore power cord to a GFCI outlet and it should not trip.
then turn on to charge only the inverter with AC2 connected only and see what happens.

You should have a GFCI on the load side of the inverters, either GFCI outlets, but that won’t work for your 30amp and 240v circuits, so a GFCI between the inverters and the panel would be easiest for your setup

https://www.homedepot.com/p/Square-D-QO-60-Amp-2-Space-4-Circuit-Spa-Panel-Main-Lug-Load-Center-QOE260GFINM-QOE260GFINM/312035158

or equal GFCI breaker wired between your inverter and panel
(you cannot replace the existing 50amp main breaker in your panel with a GFCI breaker, because you cannot back feed it, it will need to go in another enclosure like the attached link)

or switch the breakers in your panel to GFCI breaker you have room but some will need to move to the other side as GE does not have a slim line GFCI breaker, also GE is now ABB in case you have trouble finding breakers

https://www.homedepot.com/p/GE-20-Amp-1-Pole-Ground-Fault-Breaker-with-Self-Test-THQL1120GFTP/206602317

https://www.homedepot.com/p/GE-30-Amp-Double-Pole-Ground-Fault-Breaker-with-Self-Test-THQL2130GFTP/206602329

could add GFCI outlet to the 20amp 120v circuits and breakers to the 30 and 240v circuits, to save cost.

An electrical contractor in your area could help you with this if beyond your comfort level.

Ian and Ed

5ma is for life safety , because 10ma is the let go threshold, meaning at 10ma of current you can no longer let go, at 30ma you have respiratory paralysis and can no longer breath. Above that you get ventricular fibrillation, so defeating or “designing around” this “nuisance” is not a good idea.

In re to EU being safer, they still have shelter in place as the standard in some areas during a building fire. The issue is two different electrical system and codes. RCDs are considered supplementary circuit protection and in most cases protect all of the circuits in a panel not just a single outlet so 30ma is a compromise to avoid nuisance trips at the cost of life safety. There are also varying earthing requirements across Europe, some areas have floating neutrals being the building earth is not bonded to the utility neutral so an RCD is critical to detecting and protecting from a line to earth short but they are set to 100ma, RCD are primarily for supplementary protection with a bonus of some life safety.
But I’m getting ready to celebrate treason day and my independence from the above electrical system so my thoughts maybe biased.

Also AC gives you a chance 50 to 60 times a second to let go, DC does not, DC voltages above 50volts can kill well before the 100s of amps fuse trips just FYI.

No, of course not. I think you have misinterpreted what I said. My point was the standard in North America is 5mA. I don’t know if 30mA is less safe or not. That’s irrelevant. My point was, if the standard here is 5mA, shouldn’t the equipment be designed to work within that if it’s sold here? No where did I say, or mean to imply, that our standards are better. Please don’t put words in my mouth.

Thank you. I didn’t have the facts so I didn’t want to make bold claims. And you are correct, designing around the problem is not a great idea. But, if a customer uses a 30A RV connection in their garage, that circuit almost certainly won’t be GFCI protected. So, is the answer to stop selling Victron systems that don’t have dedicated, directly attached AC inputs? I don’t think that’s the answer either.

FWIW, I appreciate talking things out like this. I don’t have all the answers and every day I try to be a little less stupid than the day before. haha.

You seem knowledgable about this. What is a proper and safe solution? Thanks!