but about the vertical axis, L is constant, and τ is zero
we solve this sort of problem with the ordinary linear F = ma equation
There are two problems in my book. The first one is an example with its solution. As I told you before the solution is [tex]\vec{\tau}=mglsin\alpha \vec{e_{\theta}}[/tex] for the problem [tex]\vec{\tau}= \vec{r} \times \vec{F}[/tex]. The second one is a problem with no solution which asks "Is [tex]\vec{\tau}= \frac{d\vec{L}}{dt}[/tex] correct for [tex]\vec{\tau}= \vec{r} \times \vec{F}[/tex] in the first example". I have been trying to show that it must be correct. I mean I have to show that [tex]\vec{\tau}= \vec{r} \times \vec{F} = \frac{d\vec{L}}{dt}[/tex].
I think my book solved the first example in a bad way!
I was very naive about [tex]\vec{L}[/tex], because I wrote [tex]\vec{L} = mlr \omega \vec{e_{r}}[/tex] which is wrong! Because [tex]\vec{r} = lsin \alpha \vec{e_{r}} - lcos \alpha \vec{k}[/tex] and [tex]\vec{v} = r \omega \vec{e_{\theta}}[/tex] so [tex]\vec{L} = \vec{r} \times mv = mlr \omega cos \alpha \vec{e_{r}} + mlr \omega sin \alpha \vec{k}[/tex].
Now we have [tex]\vec{\tau}= \frac{d \vec{L}}{dt} = mlr \omega ^{2} cos \alpha \vec{e_{\theta}} = lFcos \alpha \vec{e_{\theta}}[/tex].
[tex]\vec{\tau} = lmgtan \alpha cos \alpha \vec{e_{\theta}}[/tex]
[tex]\vec{\tau} = mglsin \alpha \vec{e_{\theta}}[/tex] which is the right answer!