I assume the wiring from JB to 60A will either be full sized for 400A or feeder tap rule calculations / wiring methods confirmed if you want to reduce the size. And these calculations will need to include the solar backfeed adjustments (could end up being 0 adjustment but calculation should be done to confirm).
I checked the feeder tap rules and since I am within 10 feet, my conductor size only has to be 10 percent of the supply so 40 amps. Therefore if my understanding is correct, I should be good with 60 amps. The minimum wire size for the feeder taps in my JB is #4 so that's what I'll use. However, in learning about the other rules, it says you cannot have one feeder tap feed another. Bummer. The L1 and L2 feeder taps have an available port, but N doesn't since the 4th port is being taken up by the N-G bond wire. And I assume it is illegal to put two wires into one port. I wonder why one feeder tap can't supply another if they are sized appropriately. Even though though there is a N-G bond at the meter 300 feet away, I don't see how I can remove the N-G bond at the house since they didn't carry the ground wire from the meter the house. So it seems my only options are to get a 5 port feeder tap, or figure out how to connect the inverter's grid connection to a backfed breaker in the main panel. The problem there is that even though the bus bar is the correct size, I have no available space.
Your diagram uses isolation transformer symbol.
Oops, I didn't even pay attention. I will fix it.
The output overcurrent protection may be sufficient to satisfy code of protecting the wiring before the autotransformer. However a breaker might be good to make it easier to service, otherwise you will have to shut down the inverter and confirm that the critical load output is isolated from grid input.
I was mistaken. It turns out the breaker on the Victron is on the input side. So I'm guessing that means I don't need a breaker on the output side since the input breaker protects the transformer and anything downstream?
3. If output is on and going through an unloaded autotransformer, the parasitic losses are probably pretty small (just the internal resistance of the transformer). If there is a circuit from AT output to grid through neutral then the loses will go up because the AT will do some balancing work in parallel with grid (assuming output is on). Potentially you can interlock the AT neutral against the grid neutral so they can't touch, however this may not be allowed. And if you do this you will probably need to have an additional N-G bond for the AT side specifically since there will be no path to the N-G bond at the service.
TO SAVE TIME, PLEASE SEE EDIT 2 BELOW WHICH MAY MAKE THIS QUESTION MOOT. Just to clarify, the auto transformer will only try to balance if the backup output of the inverter is on when connected to the grid (not in battery backup mode)? Let's say it is on, which I'm not sure why it would be, how does that affect the autotransformer downstream? I don't see how there is a completed circuit since the only possible connection is through Neutral with no return path. This all may be moot since I planned to use a 3 pole manual transfer switch. So unless I forget to switch it to 'off', is that the only way it can try to balance? And if I did forget, is it dangerous or could it damage the transformer? If the answer to that last question is yes, how about adding a relay so that it can't connect unless the grid is off?
EDIT: Timselectric did something similar as described in post #1 of this thread:
https://diysolarforum.com/threads/growatt-spf-5000-es-at-and-n-g-solutions.36370/
To solve all of these issues, this is what I have done.
1) I ran a neutral conductor from the main panel to the loads panel. (Not bonded at the loads panel)
This provides a neutral for the loads in bypass mode. And provides the N/G bond from the main panel, in all modes.
2) install a relay on the AT feed conductors. Which is controlled by the inverter grounding relay. To only turn on the AT in battery mode.
EDIT 2 Actually, is this even an issue since my main breaker is interlocked with the generator/inverter circuit breaker and therefore the autotransormer can never be connected to the grid.
4. Switching neutral. Probably not. Switching neutral is used to avoid double N-G bond, if you don't have that then I think you should avoid switching.
I saw on one video from U of Maryland (
) time stamp around 3:35 and maybe elsewhere that if the inverter is transformrerless, you may want to switch the neutral so you can fuse the neutral. Does that make sense?
One additional question. The 400 amp disconnect at the meter currently supplies 3 panels. One near the meter (100 amp) one at the house (200 amp) where the taps are, and one at the barn (200 amp) 75 feet away from the house. This totals 500 amps, but is it like the breakers in a panel, where the total of the individual breakers can exceed the main rating, since all loads are not on at the same time?