I used model Philip´s for Si3N4 like the core of the waveguide, for VarFDTD, is it correct the position of the mode source? , and I set up a time of 11000 fs for the simulation, but the situation is the same, I can't see the propagation of the light. 

On another hand, I´m confused about the loss. In my case, I want to calculate the total transmission of the waveguide and after calculating the loss, but if the model that I used in MODE is the Philp for the core waveguide of Si3N4, and the result of modal losses is 1e-12 dB/cm for mode 1, is correct only multiplying these losses with the length of the waveguide for this case?.
I understood that model Philip´s has virtually zero absorption and the modal losses that I have are from PML, so, for this reason, I ignored this result of the modal losses for my waveguide in MODE, so, I set up a simulation in 2D-FDTD of the waveguide of 1 mm of length with 470 nm of thick of Si3N4 core with the model Philip´s, I put a Frequency-domain field monitor in the input (T1) and out (T2) of the waveguide, to obtain the T (T: Transmission as a function of frequency/wavelength). The result of T2= 0.82 at the end of the waveguide for a length of 1mm, so I calculated the Loss like -10log(T2/T1), is this process correct to use the monitors and to calculate the total loss? 
Thank you so much