Observable and interaction

[lalbatros]

Let us consider the http://en.wikipedia.org/wiki/Stern-Gerlach_experiment" experiment.

The inhomogenuous magnetic field produces a drift force that can separate particles according to their spin and offer a possibility for a "measurement". This (very) small magnetic interaction can be embedded in the Hamiltonian of the system as an interaction. If this interaction is active during enough time the separation and also the "measurement" of the spin "observable" are made possible.

This famous experiment illustrates a link between:

an interaction energy
an observable (a "measurement")
a interaction time​

I would be interrested to know more about this link.

Thanks.

Michel

 

[lalbatros]

In other words: how can an observable be "observed".
Are there general rules, or is it enough that this observable in some way appears in the Hamiltonian ? (With some sensibility?)

Michel