Had a long think about Balsa's scenario:
If you maintain zero sideslip with rudder, the relative airflow will always come from straight on. However, the laeral mass imbalance will cause a wing to drop because he weight vector and lift vector will have a lateral distace between them, causing a couple which will roll the aircraft.
So you are now rolling but keeping sideslip zero, which will prevent the dihedral effect from happening. As you roll, the lift vector will point away from vertical, which should cause sideslip except you are using rudder to point the nose back into the relative airflow. Therefore as the roll continues, the nose points more and more into the turn.
This would lead to a spiral dive with increasing roll angle, eventually ending up with a near vertical dive. Not a nice thing to happen at night if you have lost your AI!
Following this thought experiment, my thinking is that if you lose gyros at night, it is essential to let sideslip do what it wants to do. that way a lateral mass imbalance would lead simply to a steady heading sideslip, which is way more manageable than a spiral dive!
Last edited by WeekendFlyer; 5th Jun 2013 at 10:42.