Post History

No, if you're integrating with respect to time, you can't treat functions of time as constants.

Remember that, by the fundamental theorem of calculus, the result of your integration must be a function that, when differentiated, yields your original integrand. So if you incorrectly concluded that \\(\int \frac{l}{\dot\theta r \dot r}dt = \frac{l}{\dot\theta r \dot r}t\\;(+\\;C)\\), you would need to be able to show that \\(\frac{d}{dt}\frac{l}{\dot\theta r \dot r}t = \frac{l}{\dot\theta r \dot r}\\). By the product rule, \\(\frac{d}{dt}\frac{l}{\dot\theta r \dot r}t = \frac{l}{\dot\theta r \dot r} + t\frac{d}{dt}\frac{l}{\dot\theta r \dot r}\\), so that's the right answer only if \\(\frac{d}{dt}\frac{l}{\dot\theta r \dot r} = 0\\)—in other words, only if those functions combined in that way really is a constant.

To correctly solve the above integral, you'd need more information about those functions—if you know the definitions of the functions, you could substitute those in and then (maybe!) the integral will be solvable.