I don't think there's a set limit either way. As long as you make sure to discuss all the relevant key ideas and theorems, then you have formed a sufficient proof. Try to be as direct as possible and it doesn't hurt to take shortcuts now and then. Remember to always clearly state what the goal of the proof is, and set up the boundaries needed (eg: x is a nonzero real number). Also, make sure you connect the "given" to the thing to be proved. I've noticed a lot of students forget to do this.
If possible, try to explain the concept to someone outside the class so you can try to get a better handle on the proof. At the same time, you don't want to spend too much time going over the very fine details. Again it's all about balance in my mind.