We should review the code thoroughly and ensure the normalization of complex numbers is consistent throughout (i.e. peak vs RMS, leading to 1/2 factor differences). One potential inconsistency is Floquet ports, requiring a $\sqrt{2}$ factor to ensure consistent S-parameters when mixed with lumped or wave ports, as noted in #802 and fixed in #804, but there might be others. We should also make sure our choice is clearly stated in the documentation. Closes #507.
We should review the code thoroughly and ensure the normalization of complex numbers is consistent throughout (i.e. peak vs RMS, leading to 1/2 factor differences). One potential inconsistency is Floquet ports, requiring a$\sqrt{2}$ factor to ensure consistent S-parameters when mixed with lumped or wave ports, as noted in #802 and fixed in #804, but there might be others. We should also make sure our choice is clearly stated in the documentation. Closes #507.