Design Review - Growatt SPF 5000 ES Grid Backup & Neutral Bonding

[automatikdonn]

*must* only If a circuit is derived from autotransformer.

I think it would be better to switch center-tap of autotransformer, so only derived when not fed from grid.

While in pass-through mode, autotransformer is not needed to establish neutral. When disconnected from grid, it is needed.

If neutral of branch circuit is hardwired to neutral of grid, that provides neutral/ground bonding (for fixed, not mobile applications.)
When inverter disconnects output L1 & L2 from grid and starts producing floating 240V output, connect center tap of transformer to neutral.
This could be accomplished inside inverter with 3PDT relay. Two poles connect L1 & L2 to grid in one position. Third pole connects N to autotransformer in other position.

(Either inverter is always connected to L1 & L2, or gets connected in other position, perhaps depending on whether it is grid interactive or an off-grid inverter.)

This avoids auto-transformer trying to rebalance grid and carrying current even when you have no loads (as Ian or someone showed), it avoids violating NEC because it doesn't derive a circuit while on grid, and it establishes a neutral (still ground bonded) when off grid.

Is there a way to control an external relay to accomplish this with GroWatt? Signaling relay programmable to show "on grid/off grid"? Tap into control signal of ground bonding relay? Tap into (isolated) grounding terminal of ground bond relay?

Ground bond relay connects neutral to ground when off grid. If screw has been removed to defeat dynamic bonding, that relay terminal (and PCB ring around screw hole) is open circuit when on-grid, connected to neutral off-grid. So just get a SPST relay, 120V coil, and wire coil from L1 to screw hole (but isolated from ground)?

For fail-safe (avoid lost neutral), use 3PDT relay. One position connects loads L1 & L2 to grid. Other position connects loads L1 & L2 to load panel on output of GroWatt and connects auto-transformer center-tap to neutral. Autotransformer hard-wired to load panel, no breaker. Load panel fed by GroWatt through breaker (sized for auto-transformer).

I think that is safe, but significantly limits 240V power (including two 120V loads on opposite legs) which can be powered on that loads panel with autotransformer. You can have a separate 240V-only loads panel.

The Victron autotransformer gets around current limit of autotransformer with a 3-pole breaker and monitor circuit. Breaker passes up to 100A through to loads. Autotransformer centertap is monitored by current transformer, and temperature is monitored. If either limit is violated, 3rd pole is remote tripped, tripping the other two poles. The Victron has dynamic bonding of neutral to ground. It does not dynamically switch neutral.

I think it does not comply with the NEC rules quoted above (derives neutral for loads and does not connect to neutral of grid.) Perhaps I could rewire it so it does.

https://www.victronenergy.com/upload/documents/Datasheet-Autotransformer-32A-and-100A-EN.pdf

Those are all extremely valuable points. I have been testing this theory of grid tied autotransformer to the grid transformer balancing. I see my A/T doing some work when a motor turns that is connected to the house panels, but the A/T is connected to the shop panel. The work it's doing is very very small (less than 1 amp), but none the less it does happen. I think for my situation this validates Ian's points in the video.

The setup for my system could not be any more complex. I have to navigate how to do this all and not cause some grid side issue. After evaluating the potential to use isolation transformer's VS auto transformers - I am again leaning back towards auto transformer. I also need to have a method for easy operations my unskilled users.. so the system needs to be simple enough it doesn't take a 100 page instruction manual to operate.

I am going to start a different thread - because this one is so full of focused information on the OP post - I want to steer clear of hijacking.

 

[Larry Oswald]

Those are all extremely valuable points. I have been testing this theory of grid tied autotransformer to the grid transformer balancing. I see my A/T doing some work when a motor turns that is connected to the house panels, but the A/T is connected to the shop panel. The work it's doing is very very small (less than 1 amp), but none the less it does happen. I think for my situation this validates Ian's points in the video.

The setup for my system could not be any more complex. I have to navigate how to do this all and not cause some grid side issue. After evaluating the potential to use isolation transformer's VS auto transformers - I am again leaning back towards auto transformer. I also need to have a method for easy operations my unskilled users.. so the system needs to be simple enough it doesn't take a 100 page instruction manual to operate.

I am going to start a different thread - because this one is so full of focused information on the OP post - I want to steer clear of hijacking.

Just to chime in. I have a 3KVA isolation transformer with split primary and secondary so it is versatile. I hooked it up using the primary as an autotransformer to create the neutral from my EASun 240VAC output for the critical loads panel. Works OK but not challenged or stress tested. (Running my computer and internet now.) So there is the unused secondary. What to do???? I wired it as a 2:1 stepdown so the secondary produces 120VAC. Connected it to a plug in type receptical with one side as neutral and the other as "hot". I also connected THIS neutral to ground. Remember it is just the secondary of the isolation transformer. I plugged my "emergency well" into this outlet. (I do not want to ever pull that pump and I hope it is somewhat more protected.) Works 'well'. Does laundry.

 

[Hedges]

Is your primary 240/480V? or 120/240V?

Using ammeter, how much current does primary as auto-transformer draw from your inverter?
How much more current drawn from battery when feeding it?

 

[Larry Oswald]

Is your primary 240/480V? or 120/240V?

Using ammeter, how much current does primary as auto-transformer draw from your inverter?
How much more current drawn from battery when feeding it?

Mine is like this GE

No reliable current measurements. Maybe soon.

 

[Hedges]

Mine is like this GE

That should work well. Exactly what I tried (link below.)

120V applied across 240V winding does not get net saturation, so current approximately sine wave. And low.
second 240V primary winding is auto-transformer, makes another 120V.
120V + 120V secondary windings produce 120V.

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

No reliable current measurements. Maybe soon.

Harbor Freight 1000A clamp meter is what I have. Down to 0.01A DC
On AC scale, will register 1.23A, but for 0.23A it shows 0.00

I use current transformer into scope to see waveform. (for a fancier probe, needed for higher frequencies but not just distorted sine wave.

 

[Larry Oswald]

That should work well. Exactly what I tried (link below.)

120V applied across 240V winding does not get net saturation, so current approximately sine wave. And low.
second 240V primary winding is auto-transformer, makes another 120V.
120V + 120V secondary windings produce 120V.

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

Harbor Freight 1000A clamp meter is what I have. Down to 0.01A DC
On AC scale, will register 1.23A, but for 0.23A it shows 0.00

I use current transformer into scope to see waveform. (for a fancier probe, needed for higher frequencies but not just distorted sine wave.

Some confusion. I connect 240VAC to H1 and H4. H2 and H3 tied together become the neutral for two 120VACs which hopefully have approximately balanced loads meaning almost no energy passes TROUGH that part of the transformer. The secondary surprised me with 55VAC on each half. (implying it is a 2:1 step down) So I tied X2 and X3 yielding 120VAC between X1 and X4. X1 became that part's neutral and also got grounded. X4 is 'hot' Anything plugged in to the secondary all goes THROUGH the transformer. But we are talking 3KW, a lot for my mini critical loads. The primary if balanced will do more but my EASun is rated 3.6KW max.

 

[Hedges]

OK, same difference. You apply 240Vrms to a pair of windings 240+240=480V, so well below saturation.
Transformer current limit at half voltage means half the VA, able to transfer 1.5 kVA to secondary.
Each primary winding meant to handle 1.5kVA, half voltage 0.75kVA each, which is what can transfer from from one phase to the other.
I think this can handle 1.5kVA of imbalance, so 1.5kVA (12A) on one 120V circuit with 0A on the other.

You can probably get away with 3kVA for a while, but windings may overheat. Your only place to measure is outside.
Here's a funny idea - wire secondary in series such that it outputs zero volts, not 120 volts. Then measure resistance. Temperature coefficient of copper is 0.4% per degree C. Or if aluminum, look that up.

Be sure to protect against over-current, and protect connected loads from loss of neutral if this disconnects, but disconnecting them.