This may sound very basic, but I've just learned about the potential well with infinite barriers at +a and -a and I had a doubt. If we measure the momentum of a particle inside the well, it collapses to an eigenstate of the momentum operator, so, the uncertainty will be zero. Accordingly, the uncertainty of the position shouldn't be infinite? But how can it be possible if the well is finite (from -a to +a)? This obviously is just a misinterpretation, but could someone tell me what am I doing wrong? Thanks, Humble.