Bonded Generator, Hybrid inverter, Fully off grid, one ground rod. How to handle the neutral ground bonding in the panel?

[gimiq]

Hello, noob here. Finally plucked up the energy and courage to ask on here to help figure out this issue.

Been trying to figure out the neutral bond conundrum for ages and cannot figure out the best way forwad.
Living in france off grid.
Bonded hyundai 6600 selr 3000rpm generator (with rcd, gfci fitted) ()
connected to the hybrid epever upower hi5000 inverter. (floating neutral in bypass mode and when its supplying power from panel and or batteries)
7kw solar panels split n two series strings and connecting in parallel at the combiner box.
Fully off grid no utility.
4 pylontech us2000 (2 of which are us2000c model) in one stack connected to the inverter.
240v 50hz single phase. 1 hot, 1 neutral,
Output of the inverter goes into the main breaker panel for the house. which has a rcd the neutral is bonded before the rcd and the rcd feeds a few mcbs which then feed the circuits for the house.
panel bus bar connected to safety ground which is a ground rod outside the house.

Now as you know there is an issue when the generator is running as we have two NG bonds, one in the generator one in the panel. Seems like if the generator was not bonded I would be good to go.

Only way I can can get the genertor to not trip with this setup is to not connect the generator ground to the main panel. Luckily have not had to use the generator too much. But i need to solve this situation asap as I cannot rest at night When solar and batteries are the power source everything is honky dory. But when i need to fire up the generator I am paranoid about what and who I could fry.

is there a way to solve this? relaiably disconnect the neutral bond in a safe way when the generator is on by some sort of automatic transfer or relay/ contactor?

I dont want to remove the bond from the generator as I took a look and there is too much wiring going on in there for my liking. Plus if someone used it standalone etc could be issues

Or am i stuck with having to sell either the inverter or generator and get one that switches the bond internally (inverter) or an unbonded generator? both options I would like to avoid if possible. If i had to though I would probably opt for changing out the inverter

Please help. I have read most of the other threads and they seem to talk about automatic transfer switch but as far as i can tell that wont work in my case because the generator is needed to both charge the batteries and feed the house when the batteries are low.

thanks in advance

 

[gimiq]

another point is that when running dishwasher etc on solar it all works fine. When generator is going through inverter in bypass mode the inverter relay switches on an off and the batteries end up supplying the power when in high demand.

the dishwasher runs fine from the generator directly.

i cant tell if this is due to either the generator finding hard to lock on to the dirty electricity produced from a 300rpm diesel generator or beaqcuse of the double NG bond that exists when the generator is running.

 

[sunshine_eggo]

I say both. The EPEVER is almost certainly like every cheap Chinese Voltronics inverter that is completely intolerant of even slightly dirty generator power.

Can you adjust the charge current to the absolute minimum allowed and change the behavior?

I personally don't see a huge issue with not connecting generator ground.

Generator is N-G bonded... this also implies it's connected to its own ground rod, so the generator is fully earthed, grounded and bonded.

Panel is N-G bonded, and also earthed, so anything coming into it is N-G bonded and earthed.

The only possibility of a ground loop is with the earth itself, which is presumably a high resistance connection.

Is this code compliant or best practice? No.

Is it high risk or dangerous... not at the generator and not at the panel, so almost certainly not in between.

 

[gimiq]

yes generator has its own ground rod. Which brings up the question if i need to bond the two ground rods outside the house?

yes mistake getting the epever (fml). I will look at victron next probably if it has internal NG bond switching and doesnt cost and arm and a leg. But until I can cough the money up for that would like to figure this one out in the mean time.

So while not compliant the risk is minimal then?

And a proper long term solution for this situation and potentially being able to handle the generators dirty electricity would be a good inverter with internal NG switching? any recommendations and sizing would be appreciated. If i could get a lower powered one that could be daisy chained with another one later on that could help getting hit with the upfront cost in one go?

the earth safety ground wire to ground rod is the thickest cable excepting the dc battery output cables into the inverter

 

[sunshine_eggo]

yes generator has its own ground rod. Which brings up the question if i need to bond the two ground rods outside the house?

yes mistake getting the epever (fml). I will look at victron next probably if it has internal NG bond switching and doesnt cost and arm and a leg.

You technically should. Here in the U.S., manufacturers have STOPPED recommending a ground rod at the generator because there is no hard requirement to bond the two ground rods, and a nearby lightning strike produces a large voltage differential between the two unbonded ground rods and creates quite a large current between them - typically using the system wiring between them to disperse the difference (see video in link #7).

Of course, if you bond the two ground rods... you're back to 2X connected N-G bonds.

 

[gimiq]

@sunshine_eggo "Can you adjust the charge current to the absolute minimum allowed and change the behavior?"

when i reduce the charge current the epever locks on better. the problem is worse at higher current. The main draw being the battery charging (max utility charge) is set to 15amps. if i go above that it starts to switch the relay quite often. lights flicker, water pump will show signs of flickering. This will happen every couple of minutes.

Below 15amps it will do it once every 5-10 minutes for a short period then lock back onto the generator.

I cannot run the dishwasher while generator running without near constant flickering and relay switching. even if i adjust to max utility charge current down to 2a. Often, even just charging the batteries needs the MUC to be manually stepped up to connect without flickering on and off. i.e set muc down to 2-5a. then 2 mins later upto 5-10, wait two minutes then to 15amps

this brings me to the other concern i am only getting like 700watts from generator before it start sounding like its getting bogged down with the current setup when its going through the inverter. Does not sound like that when running the dishwasher from the generator directly. Generator plus inverter is not great unless not running high energy appliances at all and MUC is set to less than 15 amps. but then I feel I am not running the generator hard enough as apparently they should be run at between 30-80% of capacity

 

[sunshine_eggo]

The change in ability to charge with changing charge current limit really points to dirty generator power.

The N-G bond issue MAY be at play, but for my money, it's the charger on the inverter hating what's coming out of the generator. They REALLY want to the see stable power of an inverter generator.

This problem is pervasive on this type of unit. There's a reason the Chargeverter exists.

 

[gimiq]

nice to know that does exist. With this I could just bypass the inverter altogether when the generator is fired up. I would need an automatic transfer switch and then connect this directly to the batteries leaving solar only for charging via the crappy inverter and providing power when there is enough sun etc

this would solve dirty power from generator
any NG bond issue (provided the neutral ground link was before the transfer switch) ( my assumption here may not be correct)
would need approx 400 spent on a chargeverter plus a suitable automatic transfer switch.

Or use a quality inverter with internal ng switching that may handle dirty power better (but may not depending on what this hyundai is providing).

If better quality inverters are better at coping with diesel generator dirty power and has internal switching I think i would prefer that route rather than messing about with transfer switches. Thought hyundai 6600 selr would of been halfway decent so it really leave the epever being the weakest link afaik. (I know 1500rpm diesel are supposed to be much better but the ones I have seen are just out of the budget right now

 

[toms]

Unfortunately the Hyundai generator has nothing at all to do with the company that makes Hyundai motor vehicles.

It’s a shame they are able to use the same name to convince people to pay twice as much for a very low quality generator as they should. I’ve seen this many times, your generator isn’t suitable for AC input of a hybrid inverter.

 

[gimiq]

yeah i didn't know that it wasnt part of Hyundai. that is a bit of a con. So what would be suitable @toms ? What to look for regarding ac input of a hybrid generator? minimum requirements and characteristics?

Thinking about it I might just throw more panels at the situation so that there is minimal need for a generator aside from really bad weather periods.
We have plenty of power most of the year with just the winter period being a bit tricky. I might look at oversizing the panels so that they produce enough in the winter. And in the process slowly transition to a victron multiplus and charger incrementally. I can use these epever ones either as back ups, sell them or external cabins

a new diesel generator 1500rpm is looking like 8k min. Then there is the need to fuel them and also require a lot of maintenance and replacement parts etc. doesn't seem worth the hassle i live in france charente region the winters get cold but nothing too crazy

 

[toms]

yeah i didn't know that it wasnt part of Hyundai. that is a bit of a con. So what would be suitable @toms ? What to look for regarding ac input of a hybrid generator? minimum requirements and characteristics?

Thinking about it I might just throw more panels at the situation so that there is minimal need for a generator aside from really bad weather periods.
We have plenty of power most of the year with just the winter period being a bit tricky. I might look at oversizing the panels so that they produce enough in the winter. And in the process slowly transition to a victron multiplus and charger incrementally. I can use these epever ones either as back ups, sell them or external cabins

a new diesel generator 1500rpm is looking like 8k min. Then there is the need to fuel them and also require a lot of maintenance and replacement parts etc. doesn't seem worth the hassle i live in france charente region the winters get cold but nothing too crazy

I have used a Kubota J108 with a SMA Sunny Island. At the moment my system is configured to use a 240V - 48V charger from the generator though.

The most common inverter (in Australia) I have seen in use for quality off grid systems is the Cummins Onan.

 

[sunshine_eggo]

I'm assuming that's a Voltronics-type inverter, and it's not going to respond well to anything but an inverter-generator that outputs very clean power - or the Chargeverter.

 

[gimiq]

Thanks for all your replies. As far as the generator goes it will probably have to be the Chargeverter route. Although I cannot see it for sale in Europe so far. Will keep looking though. `

Mid to long term I will look to get a better charger and inverter, probably victron. Will look on other threads for what brands are recommended.

I think if I doubled the amount of panels and added one or two more batteries I would rarely need the generator at all bar some the worst stretches of weather. The first year I had the system I used the generator for about 3 hours a day for approx 7 days throughout the whole year.

I will do some more research on how much I can over panel. I have over panelled already I believe. Am I right in thinking that:

Given:

Inverter Specifications (EPEVER UP5000-HM8042):​

• Max PV Open Circuit Voltage: 500V
• MPPT Voltage Range: 120~400V
• Max PV Input Power: 4000W
• Max PV Charging Current: 80A
  
  inverter set at 25a max charge current
  15a max utility charge current

Panel Specifications:​

DMEGC Solar Panels (340W, 24V)
Vmpp: 24V
Impp: 10.01A
SR-310-60M Panels (310W, 32.3V)
Vmpp: 32.3V
Impp: 9.6A

One String with 7 SR-310-60M Panels: This is a series connection of 7 panels.
Another String with DMEGC Solar Panels String: Panels: This is a series connection of 9 panels.

9 DMEGC Solar Panels (340W, 24V):​

• Total Wattage Output: 3060W (3.06 kW)
• Current (Impp): 10.01A (remains the same in series)
• Voltage (Vmpp): 216V (24V per panel x 9 panels in series)

7 SR-310-60M Panels (310W, 32.3V):​

• Total Wattage Output: 2170W (2.17 kW)
• Current (Impp): 9.6A (remains the same in series)
• Voltage (Vmpp): 226.1V (32.3V per panel x 7 panels in series)

Combined Characteristics of the Solar Panel Strings: (2s2p?)​

Combined Current (from both strings in parallel): 19.61A (10.01A from DMEGC string + 9.6A from SR-310 string)
Voltage of DMEGC String: 216V
Voltage of SR-310 String: 226.1V

(I know they should ideally be same spec panels. But they didn't have stock when I looked at the second lot)

By my count, which could be off. I have not violated the below conditions:

1. Respect Open Circuit Voltage (Voc) and PV Input Current Limits: Ensure that the total Voc of the solar panels in a series string does not exceed the maximum PV open circuit voltage limit of the inverter.
2. Similarly, the total current from the panels (particularly in parallel configurations) should not exceed the inverter's maximum PV input current limit.
3. Configure the solar panels such that their combined voltage at maximum power (Vmp) falls within the inverter's MPPT (Maximum Power Point Tracking) voltage range.

I have gone over the total wattage, please let me know if I

a) Have already gone too far with the over panelling and need another inverter.
b) If not how many more panels I could add.
c) could configure them in a more optimal way
d) the two strings are within approx 10v and 1a of each other is this discrepancy too much or acceptable?

They are all pointing south right now. I could go half south east, half south west perhaps.

The goal is to increase production just enough in winter to reduce reliance on the generator. Most of the year we have enough or more than enough than we need.

Any advice appreciated. Hopefully I havent already made a big mistake with the current over panelling.

Thanks

​

 

[sunshine_eggo]

That voltage mismatch is acceptable.

You don't appear to be violating Voc limits.

I don't see that you supplied PV input current limit.

inverter set at 25a max charge current
15a max utility charge current

25A output? vs 80A?

It seems you're deliberately crippling your PV charging.

a) Have already gone too far with the over panelling and need another inverter.

Not sure yet.

Alternatively, you could just simply get another MPPT rather than a fully all-in-one.

b) If not how many more panels I could add.
c) could configure them in a more optimal way
d) the two strings are within approx 10v and 1a of each other is this discrepancy too much or acceptable?

They are all pointing south right now. I could go half south east, half south west perhaps.

SE/SW might probably net a larger total yield.

The goal is to increase production just enough in winter to reduce reliance on the generator. Most of the year we have enough or more than enough than we need.

If you really have limited your PV charging to 25A, then setting that to 80A should do it.

 

[gimiq]

Yes max charging current was set to 25A when trying to figure out what is going on with the generator.

Well I obviously don't fully understand because if the max current the panels can produce in the current configuration is about 20A afaik.
So I don't know where the extra current would be able to be supplied from. So I left it at a bit over. But by your comment I obviously don't understand how it's working.

Then the cables are 6mm2 some of them about 30m long (the ones furthest from the barn where the inverter is) So have been erring on the side of caution. 6mm can carry 70A at 60'c.

the MCB between the solar panels and inverter is 40a.

The data sheet reads MAX PV charging current 80A for the epever upower hi 5000 inverter.

I have some panels left over that I haven't connected and can order more of each if needed.

 

[sunshine_eggo]

Yes max charging current was set to 25A when trying to figure out what is going on with the generator.

Well I obviously don't fully understand because if the max current the panels can produce in the current configuration is about 20A afaik.
So I don't know where the extra current would be able to be supplied from. So I left it at a bit over. But by your comment I obviously don't understand how it's working.

Then the cables are 6mm2 some of them about 30m long (the ones furthest from the barn where the inverter is) So have been erring on the side of caution. 6mm can carry 70A at 60'c.

the MCB between the solar panels and inverter is 40a.

The data sheet reads MAX PV charging current 80A for the epever upower hi 5000 inverter.

I have some panels left over that I haven't connected and can order more of each if needed.

Don't confuse panel voltage/current with battery voltage current. Panels operate a high voltage/low current. The charger works at battery voltage/high current.

Recommend you adjust charge current to 80A. That value is the PV charge current + utility charge current TO the battery, not from either source.

Example:

PV: 300V * 10A = 3000W

At a 56V battery, the 3000W array yields: 3000W / 56V = 54A

 

[gimiq]

oh man ?‍. Thanks for this! Obvious now you mention it and I have seen this explained before. I have 4 pytlontechs stacked. they recommend charging at 25a (per battery) so 80a should improve this a lot. Thanks again.

Will see how this goes before slinging up any panels. But for future reference and allowing a margin of safety

It seems I could feasibly double the amount of panels if they were arranged in the same manner as described above resulting in 4S4P? which would take me to about 40a from the panels with the voltage remaining the same as present? please correct me if I am wrong here.

dont think i will ever run utility and solar at the same time. Its usually in the evening we put it on if the sun has not done its job

 

[sunshine_eggo]

oh man ?‍. Thanks for this! Obvious now you mention it and I have seen this explained before. I have 4 pytlontechs stacked. they recommend charging at 25a (per battery) so 80a should improve this a lot. Thanks again.

Will see how this goes before slinging up any panels. But for future reference and allowing a margin of safety

It seems I could feasibly double the amount of panels if they were arranged in the same manner resulting in 4S4P? which would take me to about 40a from the panels? please correct me if I am wrong here.

dont think i will ever run utility and solar at the same time. Its usually in the evening we put it on if the sun has not done its job

Please link the manual for your all-in-one.

 

[gimiq]

https://www.epever.com/upload/file/2108/UPower-Hi-Manual-EN-V2.1.pdf its the UP5000-HM8042

 

[sunshine_eggo]

It does appear there's an implicit 40A maximum Isc for that model. It's not a specification per se, but it's referenced in the wire sizing table.

On that basis, Isc 40A max.
Array Voc 440V max (they've built in a 13.5% cold temperature margin) @ 25C (should be good down to about -9°C).

If your total array does not exceed either of the above, you should be fine... However...

It does state that max INPUT power is 4000W, so that implies that over-paneling isn't permitted.

Over-paneling is very common, and I would certainly consider it. I'm accustomed to Victron where they publish BOTH Voc and Isc limits on the input side and explicitly state that you can over-panel up to both limits. My current MPPT 250/100 can only deliver 5800W, but it can handle just over 10kW.

EPEVER standalone MPPT have a "50%" rule where the array power is not more than 50% higher than the rated output of the MPPT in watts. In your case, this would be a 6000W array.

If you've had your max charging, setting #21, set for 25A, you're limited to about 25A * 56V = 1400W. Set to 80A and see how it goes.