Tulipa kolpakowskiana?

Late last night (well not that late in fact) I finished implemented a draft of the thermal blankets opening sequence, which I was talking about in my previous post, applying the 3rd theory considered:

As it may be visible on the screenshots, I throw in a bit of randomization to some of the animations parameters, in an (futile?) attempt to make it more realistic looking. Then, I sent a few screenies to Castorp to see what he thought of it and he made the very valuable comment that the motion of the vessel at the moment of the separation of the BO shouldn't have such a large outcome, nor should there be that much difference between each blankets (opening angle & speed). And he's right, as there is no actuator forcing the blankets to open, it is only the motions of the vessel that should make them open (or close), starting with a small backward motion due to the jettison of the orbital module, then with some more of the same motion upon jettison of the service module. Now, here's what I think would be more accurate:

1. Upon separation of the BO, the impulse given to the vessel make the blankets (which are loose at the forward part of the SA) to start opening slowly
2. The separation of the PAO jolt the blankets to open more and the SA "pushs" its way out of the blankets as the PAO is going in the opposite direction than the SA.

Now, I'm having an hard time warping my mind around the behavior of the blankets (attached to the PAO) when the motions occurs ... should a force applied to push the BO from the SA make them open or close? The more I think about it, the more it seems that they shouldn't open on their own as the force applied to the vessel during the separation will be transmitted to the blankets throught their attachment point .... which mean that they should rotate towards the SA and not away from it ... hmmm .. how puzzling ... :-