Isn't there one approach that answers more questions than raising new ones?
In your latest speculations in the Romulan BoP thread, you made an interesting proposal suggesting that each nacelle (in the dual-concept only) had a different charge (+ versus -) to enable the creation of a warp field.
The one thing that left me wondering was how to create different charges as the plasma source seems to be essentially the same.
Assuming you had two M/AM reactors (one for the port, one for the starboard nacelles) wouldn't it be easier to have each M/AM reactor output to be "charged" at the source for either port or starboard nacelle applications?
Thus, with a four-nacelled vessel, both of the port nacelles could have a negative and both of the starboard nacelles a positive charge.
Couldn't that be an approach with some practical advantage to help solving the riddle?