Different initial saturation guesses in 2p2c and 2pnc
This emerged from #371 (closed).
The initial saturation values when switching from one phase to two phases are different in the 2p2c and 2pnc model:
|nPhase -> both Phases||wPhase -> both Phases|
|2p2c||Sw = 0, Sn = 1||Sw = 0.999, Sn = 0.001|
|2pnc||Sw = 0.0001, Sn = 0.9999||Sw = 0.9999, Sn = 0.0001|
In principle, the Newton should take care of the different initial values. But since the Newton also has a threshold, small differences might add up and lead to intolerable (?) differences. What should we do to synchronize both models?