Hi Bagus, There are a couple of things going on here. First, you are plotting out only one half of the frequency array, so you only see one peak, the positive frequency one. Amplitudes are determined by dividing the fft result by N. You are dividing by (N/2) instead which doubles the size of the positive frequency peak. You end up with one peak instead of two, but of twice the height. That's all right when you fft a real function since in that case the pos and neg frequency peaks are the same size**. It would definitely not be all right if you were to fft a complex function.
As for the height of the pulse, it's less than 14 because the duty cycle of the wave is less than the full 15 sec. It looks to be about 90%, so a peak of 12.5 or so is about right.
**they are complex conjugates so same absolute value, not the same phase angle.