0
$\begingroup$

Consider the action $S = \int dt \sqrt{G_{ab}(q)\dot{q}^a\dot{q}^b}.$

Now for computing the Euler-Lagrange equations, we need the time derivative of $\frac{\partial L}{\partial \dot{q}^c} = \frac{1}{\sqrt{G_{ab}(q)\dot{q}^a\dot{q}^b}}G_{dc}(q)\dot{q}^d$.

Do we also need to take the time derivative of the denominator? If you do then the equations of motions become ugly, and my intuition says that they should be pretty normal.

The only explanation I can think of, is that the root is a scalar since it is fully contracted, so there is no need to consider the time derivative.

$\endgroup$
2
  • $\begingroup$ A scalar does not necessary mean that it is a constant of motion. $\endgroup$
    – Keyflux
    Commented Jan 11 at 13:59
  • $\begingroup$ Possible duplicate: physics.stackexchange.com/q/149082/2451 and links therein. $\endgroup$
    – Qmechanic
    Commented Jan 11 at 16:48

1 Answer 1

2
$\begingroup$

In general you do have to take the derivative of the inverse square-root factor. However if you choose $$dt^2 = G_{ab}dq^adq^b$$ then the factor becomes $1$.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.