Wye not delta?

[fsck]

First, I appreciate my fellow's engineer's patience on what, to people who build MW power systems, must seem like a gnat's worth of watts.

Here's another view of the proposed system. The issue is the backup inverters do not supply 277 P-N, just 120V. So to feed the long cable, we need a step-up transformer of [208/120:480/277].

The 480 will be wye-fed by the poco, but when we move to backup, that's out of the picture.

The xfmr LV will be wye, given the source. To avoid the 480 floating, the HV side would be wye as well. (Y:Y)

True? Or should we just use delta on the HV, and create a reference above ground [three resistors?]?

Here's the other shoe. When in PGE mode, I'd like to switch [not shown] the transformer around to step 480 down to 120 for several functions. (One use is feeding the inverters to recharge their battery plant.) Is that a feasible idea, and if so, does it alter the wye-delta picture?

 

[1000V]

Don't forget the & you maniac!

 

[fsck]

So I'm still looking at this {simplified} scheme:


[tt]

480/277Y
PG&E X2 b.u. invrt.
---------o o--[480:208]----[3x120V P-N.]-[gen]
/ |
ATS1 / [Battery plant]
o
|
| 3x277 P-N
---[X4]--X----- grid-tie
| invtr.
|
|
.... [~800 ft 350MCM+N]
|
|
[480Delta:208/120Y] X3
|
|
Loads


[/tt]

with an eye on T2. First, it runs both ways; it steps up power from the 3phase inverters to feed the load, but when the solar arrays have sufficient output, they backcharge the battery plant. [The two inverter systems conspire over RS-485 to accomplish this...]

My concern is X2. Should it be delta:wye, or wye:delta? If the HV side is delta, that means the long leg has no neutral reference outside of the grid-tie array. If it is the other way, then I'm feeding the inverter wye output into a delta winding. I think that's acceptable but I'm less then 101% sure. Education welcome.

(If it changes anything, X4 is a far smaller, say 5kva, transformer to run misc 120V loads in the inverter building.)