diy solar

diy solar

Should the neutral wire from an autotransformer be bonded to ground in the breaker box?

Where is your primary bond from neutral to ground?

How long was the neutral wire from wherever the primary bond is to this panel?
There is only one panel according to the OP's post on page 1.
This system has no AC input.
The distance from the autotransformer which is creating the neutral from the L1 and L2 inputs to the neutral bus appears to be approximately 2 feet.
Certainly much shorter than from the service transformer to the primary panel in a grid derived system.
 
This system has no AC input.
The distance from the autotransformer which is creating the neutral from the L1 and L2 inputs to the neutral bus appears to be approximately 2 feet.
Certainly much shorter than from the service transformer to the primary panel in a grid derived system.
Thanks, but my question was not from service transformer to mains panel (since neutral from service transformer does not connect-through) but from primary bond after meter to mains panel).

I have L1/N/L2 coming from grid/pole/service transformer over into house and into utility meter, but only L1 and L2 come out of utility meter into my main breaker box containing only a single mains breaker. Ground from a nearby ground tod as well as bonded to water main enters that main breaker box and is bonded to house neutral there, from which L1/N/L2/G pass through ~4” of conduit to my mains panel on the interior side of that exterior wall, where the mains panel is.

Ground and Neutral are on distinct electrically isolated bus bars in my mains breaker panel (meaning the panel with all my home’s breakers but no main service breaker, since that is in the small breaker box on the other side of the wall and connected to the service meter).

So if I understand that earlier post correctly, he saw better-balanced L1 vs L2 voltage by bonding neutral to ground again within the breaker box where the breaker connected to the Autotransformer was located.

I believe there is only supposed to be a single bond between ground and neutral in the system, so unless he is connecting L1/L2 from a 240V inverter to an Autotransformer that goes on to power 110/120V loads, a second bond directly in a mains panel does not make sense.

If we are taking about an isolated ‘Critical Loads’ panel serving 110/120V circuits (in addition to 240V circuits), then of course a new primary bond between ground and neutral within that isolated critical loads panel would be necessary.

But in the case that energy is being served both back to the mains panel (where the primary bond should already be) as well as on to a critical loads panel, I believe neutral can be ‘passed through’ from mains panel to critical loads panel.

We’re talking about a 240V inverter, correct? So no neutral present.

So if you just run L2/N/L2 from the mains panel to the inverter, only connect L1 and L2 to the inverter AC input and pass N on to the Autotransformer along with the inverters L1 and L2 outputs, you should be able to provide full split-phase service to a new critical loads panel in a manner that does not require a new ground-to-neutral bond (and continues to work just fine when the grid goes down).

A new ground-to-neutral bond in a new critical loads panel should work fine as well, but may cause issues with how well load balance of the non-critical-loads portion of the house load performs.

Passing-through neutral from the primary bond around/through the inverter and on to the Autotransformer and Critical Loads panel seems like the safer was to assure there is only one neutral for the entire system (with or without grid active).

And if I’ve misunderstood and this is a true off-grid system, then of course a single proper primary ground-to-neutral bond needs to be created, ideally in the main panel.
 
@fafrd
The system discussed in this thread has no grid connection.

That discussion is taking place here.


My comparison between the subject system in this thread and an on grid system was to point out the distance from the transformer to the bonding point can be quite long, and still be perfectly safe. As you point out there should only be a single bonding point. Or at least only a single bonding point "active" at any given time. The discussion in the thread I linked goes into that in depth.
 
@fafrd
The system discussed in this thread has no grid connection.

That discussion is taking place here.


My comparison between the subject system in this thread and an on grid system was to point out the distance from the transformer to the bonding point can be quite long, and still be perfectly safe. As you point out there should only be a single bonding point. Or at least only a single bonding point "active" at any given time. The discussion in the thread I linked goes into that in depth.
My apologies - I guess I believed the need of a primary neutral to ground bond in any AC system should be so obvious that I just assumed that earlier post was referring to adding a second[/b] ground to neutral bond in the panel connected to the Autotransformer…
 
So let me recap to assure I’ve understood correctly:

If you locate an Autotransformer on the AC input side of a hybrid inverter, the balancer can balance whole-house load (up to it’s maximum capacity) but only while grid is active and once grid goes down and transfer switches open, critical loads will only be powered by 240V and not 120V (so basically useless unless backup is a non-priority).

If you locate an Autotransformer on the AC output side of a hybrid inverter (Critical Loads panel), you’ll have 120V power for the critical loads when the grid goes down, but balance for the whole-house load while the grid is up will be limited by the maximum current rating of the transfer switches within the hybrid inverter as well as the max current rating of the Autotransformer itself (whichever is smaller).

So to avoid any issue, you need to properly fuse the main breaker in the critical loads panel appropriately for the maximum output power of the inverter and you need to use an Autotransformer which is rated to handle at least 50% of that same power rating and protected by it’s own breaker with appropriate rating for it’s maximum current.

So, for example, an Autotransformer with a 25A maximum current rating can be used with an inverter of up to 6kW without issue, and the input breaker of the critical loads panel in that case would be sized at 25A while the Autotransformer in that case would also be protected by another 25A 2-pole breaker within the critical loads panel.

If the inverter is more powerful, main breaker would increase in size but breaker protecting Autotransformer would remain at 25A (and this would mean that the Autotransformer is unable to fully-balance a 50% imbalance and 100% of inverter output cannot be delivered to either single leg).

While a less-powerful inverter would use a smaller critical loads main breaker while a 25A breaker could still be used to protect the Autotransformer (which should never trip). And this would mean full inverter output could be directed to either leg when in backup mode but whole-house balance when in grid-tied-mode would be limited to 25A or 3kW (in this example).

And if there is no need of backup power (offsetting of self-consumption only), an Autotransformer can be connected directly to the AC output of the inverter (critical loads) as long as the inverter includes an integrated OCPD which is sized no higher than the Autotransformer’s max current rating or otherwise some fuse or breaker/switch sized to Autotransformer max current needs to be added to the wire between inverter AC output and Autotransformer input.l. In this case, whole-house balance can reach max rating of Autotransformer or Hybrid Inverter transfer rating, whichever is lower.

Or alternatively, for that last no-backup-power scenario, the Autotransformer can be connected through an appropriately-sized breaker in the mains panel (which, again, means only 240V backup power when the grid is down). This would allow for whole-house balance up tio the full rating of the Autotransformer regardless of how small the rating of the transfer switches within the hybrid inverter are (because no current will ever pass through those transfer switches).

Did I get all that right?
 
You would not want to install the auto transformer on the ac input side at all.
Would appreciate to understand why (unless it’s the reasons I already mentioned I my recap post).

The negative is you will not have any split-phase backup power, I understand.

The positive is that you can balance the whole-house load with an Autotransformer rated for higher current than the transfer switches.

If this causes another issue I’m missing, I’d appreciate an explanation.

I’m actually mulling this exact use case because my inverters offer no backup-power (they are GTILs and only output power through their AC input).

Because I have one 1kW inverter offsetting 120V (or that leg’s share of 240V) consumption on each leg, load imbalance results in one inverter being underutilized while the other will be driven harder at a minimum and can be maxed-out / saturated in the worst-case.

So I’ve got no choice but to add an Autotransformer on the ‘AC input side’ of the inverter(s), since their is no AC output side (ie critical loads AC output) and the benefit of balancing the house load right out of the main panel is that a balanced load will result in my dual-120V inverters being balanced and more effective (average power of the more highly-driven inverter will be lower and total power output between 1kW and 2kW total will often be higher).

Is there a downside / risk I am overlooking?
 
Have you read the thread I linked? Lots of info there.

But the AC side is the grid, it already has split phase 120/240.
You'd be installing a transformer in parallel with the grid transformer.
A 5kW transformer isn't going to do anything, you're already hooked in to a 37.5 or 50 KVA transformer from the grid.
The neutrals linked together might be an issue (I don't know if they are additive and if they compete).

The circuits downstream of the Growatt inverter need to be in a separate panel from the grid side, even if they are powered primarily through AC pass through from the grid. The Growatt is not a grid tied inverter, it only uses the AC input for battery charging or to pass through utility power to the loads. You risk backfeeding the grid if the circuits it powers are on the grid side.
 
I'd recommend starting a new GTIL thread to avoid confusion.

What you are describing as load imbalance is due to the phases you have your common (always on) 120VAC loads hooked to.
It's a consumption issue. You're trying to maximize the offset. So you should move some of those loads to the opposite phase in the panel.
The utility transformer is perfectly happy with the imbalance. Just as the auto transformer would be in an off grid set up as discussed in the OP.
 
Have you read the thread I linked? Lots of info there.
Just scanned that whole thread (carefully reading all of your posts).
But the AC side is the grid, it already has split phase 120/240.
You'd be installing a transformer in parallel with the grid transformer.
A 5kW transformer isn't going to do anything, you're already hooked in to a 37.5 or 50 KVA transformer from the grid.
The neutrals linked together might be an issue (I don't know if they are additive and if they compete).
I agree that a separate thread on GTILs makes sense to not hijack this thread which seems more focused on the backup-power / critical loads panel use case, but on both the Growatt thread and here, I see what I believe is a basic misunderstanding.

Yes, the ‘grid transformer’ also balances at the pole and forms it’s own neutral through a bond to a local ground rod at the pole, but this is not really ‘in parallel’ and has little to do with the neutral formed at the primary neutral-to-ground bond of the house.

The grids neutral does not pass though to the house. Grid/pole neutral and house neutral are only connected through common parallel connections to earth ground which will typically by separated by over 100 feet if not 100 yards.

A 5 kVA Autotransformer at the home is going to do something. It is going to balance load at the home so that less balancing needs to be performed at the pole (you will pay the 0.1% efficiency loss rather than have is absorbed / paid for by the utility).
The circuits downstream of the Growatt inverter need to be in a separate panel from the grid side, even if they are powered primarily through AC pass through from the grid. The Growatt is not a grid tied inverter, it only uses the AC input for battery charging or to pass through utility power to the loads. You risk backfeeding the grid if the circuits it powers are on the grid side.
In title at least, this thread is about a general question about neutral bonding for Autotransformers.

For backup inverters that only use the grid to charge a battery and form a new island to a critical loads panel, I agree with what you state above but repeat my assertion that the neutral formed by the primary bond can be extended to neutral of the Autotransformer and the Critical Loads subpanel.

At worst, the Autotransformer may balance some of the house load imbalance but as long as it is properly protected by OCPD, that should not be a safety issue.

If house imbalance is so large that the OCPD protecting the Autotransformer trips repeatedly, then sure, letting the transformer at the pole do the heavy rebalancing and forming an entirely new ground for an entirely now primary bond to an entirely new neutral for the critical loads panel may make more sense.

But if house imbalance is not large enough to ever trip the OCPD, just extending house neutral to both Autotransformer and critical loads panel is a LOT easier…
 
Very Interested: Is anyone in this DIYSolarForum using one of those? Look like a step down in quality as compared to favored MPP brand or GroWatts. ... I am Interested in any type of "auto Power Limiter" options for grid tie experimenting. like on an old mechanical meter type connections (outside usa) ... like in Costa Rica.
Re the $300 option on Amazon, and a private chat with a guy in California using two of them (one one each 120vac leg of breaker box/ only need one ) ... I the line of such lower cost equipment at lower prices on Aliexpress for anyone interested n lower cost experimenting - type learning projects ... I like the grid tie idea with power limiter options: https://www.aliexpress.com/wholesal...21&SearchText=Grid+Tie+Inverter+Power+Limiter ... options include for with or without batteries depending on which unit one selects .The Amazon link w reviews I studies to know about such options (posted above/ was) https://www.amazon.com/gp/product/B07GC53QBD/ref=ask_ql_qh_dp_hza
 
Last edited:
Re the $300 option on Amazon, and a private chat with a guy in California using two of them (one one each 120vac leg of breaker box/ only need one ) ... I the line of such lower cost equipment at lower prices on Aliexpress for anyone interested n lower cost experimenting - type learning projects ... I like the grid tie idea with power limiter options: https://www.aliexpress.com/wholesal...21&SearchText=Grid+Tie+Inverter+Power+Limiter ... options include for with or without batteries depending on which unit one selects .The Amazon link w reviews I studies to know about such options (posted above/ was) https://www.amazon.com/gp/product/B07GC53QBD/ref=ask_ql_qh_dp_hza
Probably more appropriate for another thread. You could start a new one or use this already-established thread: https://diysolarforum.com/threads/recommended-vendors-for-gtil-inverters.15928/
 
Back
Top