0
$\begingroup$

When considering whether or not the total energy of a system varies over time, we consider the time dependence of the Lagrangian, namely by looking at $$\frac{\partial L}{\partial t} \, .$$

Why is it that we only consider explicit time dependence in this case, and not the dependence of each variable within the Lagrangian? My notes on this topic are fairly vague and I just wondered if there was a simple reason.

$\endgroup$
1
$\begingroup$

Short answer: it can be shown that the Hamiltonian satisfies $$\frac{dH}{dt}=\frac{\partial H}{\partial t}=-\frac{\partial L}{\partial t}.$$Long answer: since $H=\dot{\mathbf{q}}\cdot\mathbf{p}-L$ we have $\partial_t H=-\partial_t L$, and by Hamilton's equations $$\frac{dH}{dt}-\frac{\partial H}{\partial t}=\frac{dH}{d\mathbf{q}}\cdot\dot{\mathbf{q}}+\frac{dH}{d\mathbf{p}}\cdot\dot{\mathbf{p}}=\frac{dH}{d\mathbf{q}}\cdot\frac{dH}{d\mathbf{p}}-\frac{dH}{d\mathbf{p}}\cdot\frac{dH}{d\mathbf{q}}=0.$$

$\endgroup$
0
$\begingroup$

\begin{align} \frac{\mathrm dE}{\mathrm dt} &\equiv \frac{\mathrm d}{\mathrm dt} \left(\frac{\partial L}{\partial \dot q}\dot q - L\right) \\&= \left(\frac{\mathrm d}{\mathrm dt} \frac{\partial L}{\partial \dot q} \right) \dot q + \frac{\partial L}{\partial \dot q} \ddot q - \frac{\partial L}{\partial q} \dot q - \frac{\partial L}{\partial \dot q} \ddot q - \frac{\partial L}{\partial t} \\&= \left( \frac{\mathrm d}{\mathrm dt} \frac{\partial L}{\partial \dot q} - \frac{\partial L}{\partial q} \right) \dot q - \frac{\partial L}{\partial t} \\&\overset{\text{eom}}= - \frac{\partial L}{\partial t} \end{align}

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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