I'm failing to see why the transformer or line cares from which direction that alternating current originates.
But I'm still teachable.
That's good
I have this habit of walking into a room full of experts who've been on a project for a decade, and telling them what they've done wrong.
Oh, did I say I am a rocket scientist?
(Although it is more common for me to do that to an academic hired by industry, than to people with actual hands-on experience in the field.)
We all know the neutral of a transformer carries difference between current in L1 and L2. Except, I've come to realize that is only true if source of power is the primary winding. For an auto-transformer, it can be the sum of L1 and L2.
Imagine a 10kW PV system backfeeding 40A, 240V. And a neighbor drawing 5kW 40A 120V on L1. Also a second neighbor. We're now using transformer secondary as an auto-transformer, sending our 10kW 40A 240V to loads drawing 10kW 80A 120V. Probably so far so good.
Now let another customer draw additional power, say 5kW 40A 120V also on L1 (which means 5kW comes from grid). Neutral wire from transformer and along the poles is now carrying 120A, which utility never would have though possible for the 10kVA transformer if primary (or L1 and L2 of secondary) fused appropriately.
It takes a big PV system or a small transformer to get much of an overload. I've grumbled about the 120% rule for PV breakers in a panel, said 200% would be fine. But I now realize neutral busbar and wire from grid can carry that 120% or 200%.
It's just Kirchhoff's current law, but it does take a bit of thinking to understand how the auto-transformer does this.