Safe Grid Use of the 5000ES and transformer

[timselectric]

If not, youtube does.

 

[Hedges]

Hard wired.

Big enough, then.
Most of the auto-transformers offered for inverter users are sized such that they can't handle full wattage.
Would your system support 120V loads with wattage totaling inverter (240V) wattage? Or did 120V breakers limit it, or just the user (you)?

I plan to add a switched auto-transformer to help balance across stacked 120V inverters. Disconnected on-grid, connected off-grid. It will have breakers and thermal protection (also fan), but OK if it disconnects because inverters maintain neutral, will shut off if overloaded.

 

[msllarson2]

I take it to mean you can't use center-tap of neutral to supply neutral of loads, such that the only path to grid neutral/ground is through windings of auto-transformer.
You can tie the center tap to grid neutral (will now try to rebalance grid)
You can tie load's neutral to grid neutral and not use auto-transformer center-tap while fed from grid. If you disconnect from grid, then connect load neutral to center-tap.

So this sounds like you could make this work in a few different configurations:

1. Completely isolated from the grid with a ground-bonded generator. Connect generator to inverter AC inputs and pass the generator neutral thru the inverter input and then from the inverter output terminal to the load panel.
2. Connected to grid/main panel through a mechanical interlock, optional generator with a floating neutral connected to the inverter input. The grid/main ground-bonded neutral is connected to the load panel neutral. The grid neutral would then be electrically connected to the load and generator neutral. This way the grid transformer and the auto-transformer would never have the hot legs interconnected.
3. Or do what timselectric has done with a relay switching the auto-transformer input off when grid is being passed through (requires soldering to inverter mother board)

I haven't heard of any option to use the auto-transformer without a generator or the grid that would comply with 215.11

 

[timselectric]

Would your system support 120V loads with wattage totaling inverter (240V) wattage?

Yes, at the current setup.
But will add a second one, on expansion of the system.
Thought about switching to an isolation transformer.
But more losses and space required.

 

[timselectric]

Connected to grid/main panel through a mechanical interlock, optional generator with a floating neutral connected to the inverter input.

You can't float the grid neutral. It's already bonded.

In any scenario there should only be one N/G bond.
And for safety sake, that should remain a permanent connection.

 

[msllarson2]

If your generator has a N/G bond. It needs to be removed.

This is only if you have the N/G bond somewhere else because you need 1 (and only 1) N/G bond somewhere correct? Can't this be either the generator, or the main (once it is installed). Correct.

You can't float the grid neutral. It's already bonded

Sorry, generator neutral would be floating.

 

[msllarson2]

And for safety sake, that should remain a permanent connection.

Hmm, this sounds like it would be better having rhe N/G bonding in the load panel only until the main panel is connected would be better than having the N/G bond at the generator.
Then removing the load N/G bond once connected to the main neutral.
(Trying to develop raw land, with hope of getting grid power within a few months)

 

[timselectric]

Hmm, this sounds like it would be better having rhe N/G bonding in the load panel only until the main panel is connected would be better than having the N/G bond at the generator.
Then removing the load N/G bond once connected to the main neutral.
(Trying to develop raw land, with hope of getting grid power within a few months)

This would be my recommendation.

This is only if you have the N/G bond somewhere else because you need 1 (and only 1) N/G bond somewhere correct? Can't this be either the generator, or the main (once it is installed). Correct.

Maybe, if the generator is hardwired.
But if it's cord and plug connected. It's not a permanent connection.

 

[msllarson2]

Would one of these configurations be able to meet NEC code requirements, or will this Growatt system always need to be under the table and a potential insurance liability if there is ever a fire?

My wife is  concerned about my purchase/competence (even more than when I first bought solar and she didn't know the code existed :D)

 

[Hedges]

Thought about switching to an isolation transformer.
But more losses and space required.

Think twice, and find a transformer that is nice to your inverter.
Most are optimized for cost, draw a high but acceptable current from the grid as they go partially into saturation.

What I'm doing is operating windings below their rated voltage. I can use 240V primary windings as 120V windings in auto-transformer configuration. But I don't have one with enough windings at the right voltage for 120/240V to 120/240V (or 240V to 120/240V) isolation transformer while operating below rated voltage. I could do it with two such transformers, each taking in 240V and creating one of the 120V outputs.

transformer excitation current - isolation transformer used as auto-transformer

So you think you can backfeed a transformer secondary, either as step-up or as auto-transformer? Turns out these things aren't so ideal and reversible as one might expect. It isn't just the greater inrush (primary is often wound outside secondary to reduce inrush current), but also idle current...

diysolarforum.com

Here is a 3-phase Delta-Y connected isolation transformer. Draws higher current driven rated voltage. I rewired the Delta primary as Wye, applied 120/208Y to the 200V legs and got 277/480Y from the 460V legs. Then I drove the 230V taps to get 240V from 460V legs. I'm still having issues with how I make use of neutral.

3-Phase SMA Sunny Island (15kW) + Sunny TriPower (30kW)

Showing a work-in-progress assembling 3-phase SMA system. 3x SI-5048-US are configured for 3-phase and produce 120/208Y, rated 15 kW continuous. Sunny TriPower 30000TL-US-10 is 480V delta (or 277/480Y), rated 30 kW. TriPower combiner box will be used, but for now just one string being connected...

diysolarforum.com

Somewhere in SMA's documents, I read a note related to this issue, and they recommended high efficiency transformers available from a Spanish company. Can't fit it now, though.

(Trying to develop raw land, with hope of getting grid power within a few months)

You get to be your own AHJ, many ways to make it safe.

 

[Hedges]

Would one of these configurations be able to meet NEC code requirements, or will this Growatt system always need to be under the table and a potential insurance liability if there is ever a fire?

My wife is  concerned about my purchase/competence (even more than when I first bought solar and she didn't know the code existed :D)

Last I saw, GroWatt's transformer was UL listed. But none of their inverters was. That may have changed.

Situating questionable electronics and batteries in a bunker away from the house would help avoid its fire getting to the house.
But loss of neutral or loss of one phase could do bad things to circuits in your house.
I like Victron's current & temperature measurement of 14/28A auto-transformer disconnecting 100A breaker feeding load. That allows largest 120V load and largest 240V load. Looks good for off-grid. Doesn't appear to comply with NEC rules on connecting neutral.

 

[timselectric]

Would one of these configurations be able to meet NEC code requirements, or will this Growatt system always need to be under the table and a potential insurance liability if there is ever a fire?

My wife is  concerned about my purchase/competence (even more than when I first bought solar and she didn't know the code existed :D)

Not UL listed. So, almost impossible to pass an inspection.

 

[timselectric]

The only Growatt inverters that are UL listed, that I know of. Are the grid-tied ones.

 

[timselectric]

Think twice, and find a transformer that is nice to your inverter.
Most are optimized for cost, draw a high but acceptable current from the grid as they go partially into saturation.

Not too much of a concern for my 40kw system.
At least not for me.

 

[Hedges]

If you do feed a transformer, try determining no-load VA and W. Assuming it is accurate with funny waveforms, I think that can be done with clamp AC ammeter on transformer wire, and clamp DC ammeter (or inverter display) for battery current (difference with/without transformer.)

I think I've measured that most of the no-load current draw is reactive power, recovered by Sunny Island. The added power consumption was about half of apparent power. That matters to me with small battery to ride through the night. Also, considering having 3 out of 4 (25W no-load consumption) inverters sleeping (4W consumption), just one inverter operating, and auto-transformer to create missing phase.

Measurement of auto-transformer balancing low-frequency inverter

I tested the effect of a 9kVA toroid auto-transformer connected to 120/240V split-phase low-frequency inverters. Inverters are SMA Sunny Island 6048-US, four wired series parallel. I also turned off output breakers to test with two wires in series. Imbalanced load was applied with four...

diysolarforum.com

But I don't know how well high-frequency inverters perform with reactive power, so other people's measurements would be interesting.

 

[zanydroid]

You can tie the center tap to grid neutral (will now try to rebalance grid)

Would adding an isolation transformer on the input to the inverter address this? With common neutral between the AT and grid neutral. Now the inverter-L1 and inverter-L2 voltage will shift up and down (relative to grounded conductors) without causing issue with balancing grid.

Of course this would limit the inverter output to the capacity of this isolation transformer.

Another way of getting this is to put 240V circuits on a 240V only subpanel fed directly by inverter from 240V. And then put 120/240V circuits on a 120/240V subpanel fed by an isolation transformer with center tap secondary, with the center tap bonded to utility neutral. This way the bottleneck from the isolation transformer is only incurred on some of the loads and not all of them.

 

[Hedges]

Yes, to all points.

Since grid is a bottomless source, cheap used isolation transformer between grid and inverter is quite viable.
About $250 to $500 (plus a few $hundred shipping if you can't pick it up) will get you 25kVA, 100A at 120/240V.

Isolation transformer for 120V loads in theory sounds good, problem is most (i.e. those available used) are designed such that idle current is maybe 4% of full-load current. That's reactive power, not real, but your inverter has to drive it and may have additional losses.

If you set up just one 120V phase rather than split phase, a single 240/480 to 120/240V transformer could be used at half its rated voltage (and power). I got about 1/6the the idle current that way.

 

[timselectric]

Transformer secondary neutral does not get connected to primary source neutral. It gets bonded to the grounding system. And yes, both systems (primary and secondary) are bonded to ground. So, technically they are connected. But there's no electrical connection between the two systems. Since a circuit needs a circular path to flow current.

 

[Hedges]

I think it is fine to connect transformer secondary neutral to primary source neutral.
If you want to backfeed the main breaker panel from inverter (e.g. through transfer switch or interlocked breaker), you probably have to do that anyway. If separately bonded to ground, additional switching will be necessary.

 

[timselectric]

I think it is fine to connect transformer secondary neutral to primary source neutral.

You shouldn't do it. It creates a parallel path with the grounding conductor. The only thing that the two sides of an isolation transformer should have in common is the grounding system.

If you want to backfeed the main breaker panel from inverter (e.g. through transfer switch or interlocked breaker), you probably have to do that anyway. If separately bonded to ground, additional switching will be necessary.

The secondary neutral is isolated from the primary conductors. Maybe you are thinking of an autotransformer. Which is required to be connected to the primary source neutral. (If one exists)