AC charging on one circuit is tripping GFCI breakers on other circuits

AC charging on one circuit is tripping GFCI breakers on other circuits in my electric panel.

I have tested charging the EP500 on three different 20 amp circuits and each time one or two GFCI breakers on different circuits trip immediately after turning on the AC charging. The GFCI breakers cannot be reset until the Bluetti unit is no longer charging (I disconnect it).

The 20 amp circuits I am using for testing have for years run a variety of kitchen appliances (microwave, induction cooktop, mixers etc.) and vacuum cleaners and have never caused GFCIs on different circuits to trip.

There seems to be a wiring fault in my unit. FYI, the AC cord being used, pin 5 is the ground, pin 4 is the hot, and pin 2 is the neutral. Pins 1 and 3 are not connected to the 3 prong plug.

I am posting this to see if anyone else is seeing a similar issue.


Adding this [edited] from the Kickstarter comment section where I responded to a question:

[The circuits I was using for charging] just happen to be 20 amp circuits since I am in the kitchen, testing the refrigerator for run time.

And yes, circuits the EP500 is plugged into [for charging] do not trip. It is one or two completely different circuits (happen to be GFCI) that are tripping.

Months ago I looked into getting a second EP500 to tie into the electric system. Ran across lots of info on how important a floating neutral would be for this. The EP500s use a bonded neutral. An online search turning up a discussion about one circuit tripping another circuit suggested the possibility of a white wire being used as a part of a switch leg that is somehow feeding back into a neutral.

Maybe the ground is involved too.

Are you sure that’s true? I’m still learning about this too, but I don’t think the neutral and ground are tied together. I didn’t

In order to charge my Tesla M3 I had to jumper the neutral and ground together (the Tesla charger needs to see a ground). Only doing this with and A/C cord NOT plugged in

The EP500 ground seems to tie to the house ground through the A/C plug and I didn’t want the EP500 neutral strapped to my house ground.

Yes, Bluetti said the units use bonded neutrals. This is for output. I don’t know how this relates to the AC input issue I am seeing.

6 months ago on the comment thread I asked about floating and bonded neutrals and the subpanel:

"…Regarding the UPS box (sub panel), will the EP500 be considered a separately derived system requiring a switched neutral?..

"You replied earlier that the EP500s are bonded neutral units. In a home’s electric supply, the neutral can be bonded to the ground at only one point.

"If the neutral in the EP500/EP500 Pro is changed from a bonded neutral to a floating neutral, the neutral passes to the main panel and uses the main panel bonded neutral. If it is not changed to be a floating neutral, the house neutral must be disconnected from the EP500 UPS.

"Here is one method of achieving this:
X-Series Neutral Switching Panels - Products - Reliance Controls Corporation

“So will the EP500/EP500 Pro exist only as a bonded neutral unit (meaning it would need to be installed as a separately derived system) or will it be changed to a floating neutral when using the UPS panel box and connecting to the house?”

Bluetti replied:

"Hi there,
"Thanks for your explaination.

"The EP500/EP500 Pro exists as a bonded neutral unit when using it separately. When using the UPS panel box and connecting to the house,the EP500/EP500 pro could be changed to a floating natural."

So the standalone unit you tried on your Tesla should have charged it and not required your bonding the neutral. I take it you simply used the included portable Tesla charger at 120V, plugging it into one of the outlets in the front and plugging the standard Tesla connector into your car? Can you be more specific about how you bonded it? I think the online videos I saw using the EP500 to charge a Tesla showed normal charging and did not mention anything about having to bond the neutral and ground together.

If indeed they are sending out unbonded units (my refrigerator did charge fine), that could possibly affect how the AC charger was functioning and the issue I am seeing with the GFCI breakers tripping.

If I have time tomorrow, I’ll test the EP500 output on my MY.

Yea, so when I just plugged in the tesla mobile connector (with the 120v, 20amp plug), the tesla connector threw "no ground error.

Then I took a test plug… an extension cord with the female side viciously chopped off and twisted the ground and neutrals together and plugged it into a different plug on the front of the EP500. The Tesla then charged fine.

Just now… with the EP500 unplugged from the wall & the battery switch off I put an ohm meter between the Neutral and Ground on a front plug and read an open. So, assuming I understand what bonded neutral is, my unit does not appear to have a bonded neutral (this is a US model).

If yours is different, then one of our units is “broken”.

I just checked my cable and it’s wired the same as yours. Also, with the batt off, I attached the A/C cord and still read an open between neutral and ground at the mail wall plug.

You are right. When a 120 outlet is energized, using a simple testing light, the neutral does not seem to be bonded. There’s a Bluetti EB240 in the house too and it does not seem to be bonded either. I don’t seem to have continuity across the ground and neutral using the tests you used either. On a standard house outlet, using a simple testing light, you see it light up between the hot and ground. Not on the EP500.

The refrigerator I tested for operation only has a two prong plug, not a grounded plug and was probably unaffected. [edit - this was a 3 prong plug, not two and WAS supposedly grounded]

This is not how they said it should be. I wouldn’t be surprised if this was causing the GFCI issues I am seeing.

I seem to remember the advertising for this unit showing that it could charge a Tesla. It would have to be bonded to do so. Yes, there is a picture of the unit charging a Tesla about a quarter of the way down on this page:
https://www.kickstarter.com/projects/bluetti/bluetti-ep500-and-ep500pro-the-new-era-of-home-backup-power/description

It’s easy enough to bond the outputs if you think it would solve the problem. But, I’m not sure that would solve the problem. I’m also not sure you’d want it bonded at the same time you’re say powering something with a ground & have it plugged into an A/C outlet. Too many bond points (with all the grounds tied together).

With your GFCIs tripping did you try with and without anything plugged into the front EP500 outlets? To trip a GFCI you’d need a leakage (or noise) path to the ground line. Seems like you’d have less chance of trouble in an unbounded configuration (with the 500 not messing with the ground).

I did some charging off the AC today and don’t have any issues with GFCI tripping (so far).

I’m still trying to think how the blu-etti on one circuit is tripping the GFCI on another. The GFCI is looking for a difference in current flow between the hot and neutral lines of what is plugged into it. I wonder if it would happen if the GFCI is completely unloaded. You could try unplugging all the devices plugged into the GFCI(s) and see it it makes a difference. That would include downstream outlets tied to the GFCI (in my case two 2nd floor bathrooms and two garages plugs are all off the same GFCI in the main floor bathroom).

I wonder if the EP500 could have an internal leakage path to ground. You could also try charging it on one of the GFCI Plugs. I just plugged mine into a GFCI plug and it’s charging fine

PS: My tesla mobile connector is a different story. The electrician put in a 50 amp GFCI breaker and it was occasionally tripping. I decided I didn’t need that GFCI that bad and replaced it with a regular 50 amp breaker (could have saved myself $80). The mobile connector likes to “test” the ground path by leaking the current into it which can trip the GFCI.

If you’re dying to bond the EP500 outlets and don’t want to twist two wires together:

Some interesting suggestions.

All tripped GFCI incidences happened with no load on the EP500. One GFCI circuit has an led night light that doesn’t even register a current draw. The other has a trickle charger on it for an old car battery.

The 240V 20A GFCI car charging circuit has worked flawlessly and only tripped once randomly in the +year I have been using it. I figured with the amount of driving I do (not a lot) the easy to wire 20A 240V circuit would be adequate and it is.

I plugged the EP500 directly into each of two GFCI outlets on the two GFCI circuits and it tripped immediately.

So the GFCI circuits detect a fault if used directly and also if the EP500 is plugged into a non-GFCI outlet. Not looking good.

EDIT: more testing on two different exterior GFCI receptacles not on GFCI circuits. Charging was possible with no tripping of any GFCI receptacles in use or either of the two GFCI circuits not in use. However, charging using one of the interior circuits that the GFCI receptacle was on tripped one of the two other GFCI circuit breakers that originally tripped when testing on the three kitchen circuits.

What I worry about is a situation that could be a fire hazard. The uncharacteristic tripping of the original GFCI circuits should not be happening, IMO.

EDIT#2
Tesla gen 2 mobile connector T flashes red twice. Error code if not charging is that the ground is lost. So mine is definitely not grounded.

*2 flashes *
Ground loss. The Mobile Connector detects a loss of ground.
Make sure the power outlet is properly grounded. Consider connecting to a different outlet. If uncertain, ask your electrician.

I wonder if your EP500 has a leak path to ground (on the A/C charger side).

If you had a well stocked test bench I’d suggest a oscilloscope current probe on the ground line.

When you did the continuity test of your charge cable did you also check for shorts between all the conductors ?

Have a friendly neighbor? Roll it over to the next door garage and try in on one of their GFCI outlets. It did not trip my GFCI when plugged directly in and something is probably wrong if it’s tripping yours.

When you were talking about tripping different GFCI’s I was wondering if your ground rod might be marginal. I’m still not sure how one circuit is effecting another.

Only other idea I have (not sure I’d recommended it… but I’ve also shocked myself changing wall switches w/o flipping off the breaker :sunglasses:) would be to try charging w/o a ground (get one of those old stupid 2 prong adapters). Be careful. If you have a hard ground fault the case could become energized. I’d want to then put a volt meter between the ground screw on the back and the house ground to see where it’s at (assuming it then doesn’t trip the GFCI).

I guess it could just be a high resistance path to ground leaking a small amount of current (also a defect).

You may end up having to try to contact Blu-Etti. Dunno if you could change out the A/C charge board yourself.

Yes. Made sure the only connections were the three to the prongs and the remaining two were not connected to anything.

Unfortunately, no oscilloscope (have never used one). I did double check several outlets to make sure they were “good” (i.e. grounded, no significant leakage, ground rod likely fine - car charger is happy on the 240V circuit), comparing voltage hot to neutral and hot to ground (pretty much the same) and insignificant voltage neutral to ground (supposedly less than 5% is acceptable).

I edited the post above this and mentioned successfully using a couple GFCI receptacles (not on the GFCI circuits).

Online found a suggestion for testing an appliance online.

I don’t currently have a clamp meter and if I could find one, might need to rig a mini extension cord with the three wires separated so I could clamp around each individually.

Thanks for your ideas. I’m also thinking about replacing one or both GFCI breakers to see if a newer one might not be as sensitive. GFCI design may have changed in 30 or 40 years. These do appear fully functional.

I am still suspicious.

That’s a little confusing. What is:

GFCI Receptacle
vs
GFCI Circuit

Presumably both are GFCI protected.

I tried to distinguish the circuit protected by a GFCI breaker from an individual GFCI receptacle that only protects that particular receptacle or those after it, depending on how it is wired. The breaker for the individual GFCI receptacles is a standard breaker. So either a GFCI circuit where every outlet is protected or an individual GFCI receptacle that can be reset right there. Hope this is clear enough.

The GFCI breakers are old.
The GFCI receptacles were installed recently on standard circuits with no GFCI protection.

I recommend getting a new GFCI breaker. If it doesn’t solve the problem, you can return it for a full refund at all mayor electrical suppliers.

got it.
you have a lot of GFCIs there.
I didn’t realize you’re dealing with some breaker and some plug types.

You’re problem is a bear. It’s really strange that you can’t reset the circuit breaker until the blu-etti is disconnected (so not an inrush thing).

Are the GFCI’s on 220 circuits? I guess you can try the new breaker. I assume the GFCI breakers have the neutral wired correctly through the breaker.

Let us know now it turns out.

I’m putting this on the back burner for now but will likely replace one or both GFCIs.

Thanks all, for your suggestions.

Wow! I try to read all your comments in communications between the 2 of you. I received my EP500 about 2 weeks now. I have been testing it to charge a hybrid plug-in EV. I have a similar problem. First, it does not charge the EV if it is only plugged in 1 of the A/C Out ports unless a Bonding Plug (BP) is inserted in a second port or used and plug a BP on a multi-outlet strip. However, if this setup along with the A/C Input port plugged in the 110V wall outlet at the same time. It trips the GFCI on the circuit. I tested it on a different circuit without GFCI. It trips the associated breaker.
Note: The only time it works as expected without a need for the BP is when it runs in UPS Offline mode. If I change it to UPS Online mode, the EV Charger would report a no-ground error.
David_L312 Were/are you able to charge your Model 3 without a need for the Bonding Plug?

What you write is what I would expect. The bonding plug should only be used when not connected to the grid. I needed a bonding plug for my EV and it will only charge when not plugged to the grid and the bonding plug must be used.

Keep in mind charging an EV from an EP500 is inefficient and slow. If you have solar panels, still inefficient and slow but it’s something.

Yep, that exactly correct.

Need the bonding plug to charge EV AND only use the bonding plug when NOT tied to the grid.

(and, yep… charging the EV at 120v 20 amps is pointlessly slow)

By the way, if you have a generator that you are using to charge the EP500 and it has a floating neutral instead of a bonded neutral, I think it may help to plug the bonding plug into one of the gas generator’s outlets and use another outlet to properly charge your EP500.

I am revisiting this topic.

Out of laziness, have not replaced the ground fault breakers. I charge the EP500 infrequently (only two power outages since purchase) and just manually throw the breakers if I need to charge.

HOWEVER, what I just received in the mail (and what probably all of you will receive), is a new charging cable.

The new cable, pin 2 is hot, pin 4 is neutral, pin 5 is ground.

The old cable, pin 2 is neutral, pin 4 is hot, and pin 5 is ground.

The new cable has a new sticker that says “enhanced AC charging” and another one that says, “Warning: In a split phase system, ONLY the Fusion AC Input Cable is able to charge EP500.”

I just tried to charge with the new cable and unfortunately that one tripped the breakers too. I was hoping the new cable was a cure but apparently not!