EG4 6000 XP - Splitting Out the 240 Split Phase Into 2-120V Circuits

[Marobel]

So the EG4 server rack battery is up and running on SA, the 8 - 545W panels are up and pumping power into the now fully charged batteries. Now it’s time to design and install the AC output into my shop. I’m looking for a sanity check on this next phase.

I would like to separate the 240V split phase AC into 2 - 120V isolated circuits to run the shop lights, dust collection. I would also like to have at least one plug for a 220V to run a new tablesaw. (I can wire this for 120V but will lose power doing so.

From the L1, L2, N and Ground terminals on the 6000XP, how could I do this?

(Asking for a friend)

@timselectric

 

[timselectric]

Install a sub panel and feed whatever circuits you need.

 

[marionw]

Install a subpanel with two interlocked breakers, one from your 6000XP and the other from your "Main" panel. The interlocked breakers will allow you to power the subpanel from either the "Main" panel or your 6000XP (not both at the same time, either/or)
@timselectric and others can provide you guidance on managing the neutral/ground bonding between the inverter and the subpanel which is extremely important.

If you do not want the option to power from the "Main" panel then neutral/ground bonding becomes a bit easier

 

[Marobel]

Thanks for the quick reply and for confirming my plans. I thought this very much like the 220V feed into the main house. I’ll run L1 into one leg and L2 into the other. Will the neutral (N) then be tied into the ground in this sub panel or will I keep these apart? There’s been so much discussion over this so sorry for re-asking this.

 

[Marobel]

Install a subpanel with two interlocked breakers, one from your 6000XP and the other from your "Main" panel. The interlocked breakers will allow you to power the subpanel from either the "Main" panel or your 6000XP (not both at the same time, either/or)
@timselectric and others can provide you guidance on managing the neutral/ground bonding between the inverter and the subpanel which is extremely important.

If you do not want the option to power from the "Main" panel then neutral/ground bonding becomes a bit easier

This will not have access to the house main power. The new shop was built some 150 ft away from the house.

 

[marionw]

Great, one "Master" breaker from the Inverter powering the subpanel (really a "Main" panel)

I will leave the neutral/ground bonding to @timselectric and others as it is above my pay grade.

I expect that since it is separate building and not connected to the "Main" panel that you will need a Grounding Electrode Conductor (GEC, I get lost in all the abbreviations) where you have several grounding electrodes (or meet the NEC requirements in some other manner, depending on the building construction) and that this GEC and the ground and neutral from your inverter are bonded in the subpanel (which by the way may be better described as the main panel)

 

[timselectric]

Thanks for the quick reply and for confirming my plans. I thought this very much like the 220V feed into the main house. I’ll run L1 into one leg and L2 into the other. Will the neutral (N) then be tied into the ground in this sub panel or will I keep these apart? There’s been so much discussion over this so sorry for re-asking this.

Run the L1, L2, N, and G from the 6kxp to the service panel.
You will have to check if the 6kxp takes care of the N/G bonding. This should be the default, but you never know. If not, then you will need to do it in the panel.
The service panel can be a main lug type, as there's already a breaker in the 6kxp.

 

[dfw_dude]

The default setting on 6000xp for N-G bonding is "disabled" - you can enable it in the software. I'm not an expert by any stretch but when I went down the rabbit hole on that often discussed topic, for OP's setup: it seems that since the 6000XP would be doing the bonding, there's no need to do it in his panel as well.

 

[timselectric]

If that's the case, then just do it at the panel.
And yes, you can only have one N/G bond.

 

[Marobel]

Getting back to this project just today. I’ve wired the L1 and L2 to a small sub-panel that allows me to keep the two feeds separate electrically. Meaning the twin 15 amp breaker’s of L1 is on a separate circuit that the twin 15 amp breaker’s of L2. The inverter N is tied to the neutral bar in the sub-panel and the ground to the box housing itself. (10awg, All pretty straightforward ). I’ll get into the 6000XP software to configure the N/G bond option soon. In testing what I’ve done so far, something comes up I can’t explain. The frequency (Hz) across L1 and L2 is 59.97-59.99. What should the frequency be from L1 or L2 to the neutral or ground bar? I was expecting ≈60 as this is what I get from a typical 120v plug in the house. But the readings I get are all over the place. Some as high as 160 and then down to low 100’s and then back up. No pattern.

 

[TM48]

Do you have an earth ground?

 

[Marobel]

I do not, but I could. This is a stand alone system powered by 8-545 panels with 3 EG4 48-100Ah batteries. There will likely never be an AC feed to this. I can add some background if needed. Why do you ask?

 

[timselectric]

The frequency (Hz) across L1 and L2 is 59.97-59.99.

This is fine. (Close enough) 60hz is not always exact.

But the readings I get are all over the place. Some as high as 160 and then down to low 100’s and then back up. No pattern

This doesn't make sense.
I would double check that and the testing process.

I do not, but I could. This is a stand alone system powered by 8-545 panels with 3 EG4 48-100Ah batteries. There will likely never be an AC feed to this. I can add some background if needed. Why do you ask?

You should install a ground rod, and run a #6 AWG GEC (Grounding Electrode Conductor) to the panel ground bar.

 

[Marobel]

This is fine. (Close enough) 60hz is not always exact.

This doesn't make sense.
I would double check that and the testing process.

You should install a ground rod, and run a #6 AWG GEC (Grounding Electrode Conductor) to the panel ground bar.

So I rechecked the frequency and it’s still all over the place when I test L1 to N/G and L2 to N/G. When tested across these two (L1-L2) I get 59.99 - 60.00.

Going to get a message out to EG4 later this week but was hoping for some other expert considerations here. Maybe @EG4_Jared can add his thoughts. Is this an inverter setting situation or am I’m asking too much from the inverter to give me 2-120volt legs?

 

[timselectric]

There are two 120v inverters inside a split-phase unit.
They are connected in series to provide 120v/240v split-phase.
So, the only way to get 240v is if both 120v sections are operating correctly.
Frequency is identical and in sync, for any output.
I think that the testing process is flawed, somehow.

 

[EG4TechSolutionsTeam]

Would you be able to post a few pictures of your testing as well as export the data history on the inverter?

 

[Marobel]

Jared, I’m out of town now and the system is off. I’ll get back to you asap though. Thanks for your reply.

 

[Marobel]

So to update all on this issue. I went back and forth with SS, from whom I bought the 6000XP from in March. After conducting a series of tests while making short videos of the results, the conclusion was that the inverter is faulty. They’ve arranged a replacement to be shipped asap and I’ll return this unit for inspection and repair. My testing abilities were not flawed at this point.