'Degree of liberty' definition

1. Feb 10, 2005

quasar987

Could someone write a formal definition for "number degrees of liberty of a system". Thank you very much.

2. Feb 10, 2005

dextercioby

The minimum number of INDEPENDENT coordinates to describe the physical state of the system.

For example,the mathematical pendulum:a priori 3 coordinates for the bob,but once u impose the 2 customary constraints,u'll end up with one degree of freedom.

Daniel.

3. Feb 10, 2005

quasar987

So if we have 2 identical simple pendulums, separated from each other by a fixed distance L and oscillating in a plane at the same angular frquency and in phase, then the angle that one makes with the vertical is sufficient to give the exact position of each bob and so the degree of liberty of this system is 1. Correct?

4. Feb 10, 2005

dextercioby

Ther are 2 degrees of freedom (one for each pendulum).The fact that they oscillate in phase only says that the two functions (angle of time) are identical as functional dependence,but the # of degrees of freedom is still 2.

Daniel.

5. Feb 10, 2005

quasar987

But isn't there something missing in the definition then? I mean, isn't it true that one angle sufice to describe the entire system ???

6. Feb 10, 2005

dextercioby

Nope,there are 2 independent systems.If the oscillators are not coupled (meaning that between the 2 bobs there's no rigid rod to ensure they will always oscillate in phase),then there are 2 systems,each of them described by an angle which is the degree of freedom.

Daniel.

7. Feb 10, 2005

quasar987

Oh ok... right NOW (i.e. for these particular initial conditions) they are described accurately by one angle, but it's POSSIBLE that we might need 2.. and that's what's meant by "independant coordinates".

8. Feb 10, 2005

dextercioby

Yes,you could say that they can be described by one angle vs.time,which would mean solving only one ODE,but the # of degrees of freedom is still 2.

End of story.

Daniel.