ADS EM co-simulation differs between simulation with S parameters (using S2P_Pad3 and pad dimensions) and the real component

[druiz]

Hello,

I am doing 2 EM co-simulations in ADS for a simple OMN. One of the simulations is with the extracted S2 parameters and the other is placing the component of the library from ADS: "sc_atc_100_CDR13BG_G_19960828". The S2 parameters has been extracted trought "SPOutput" in magnitude and phase from "sc_atc_100_CDR13BG_G_19960828" component. The ports to extract de S2 parameters are 50 Ohms in 1 and 2. In the other hand, to validate the EM co-simulation, I am doing a schematic simulation of the same OMN and comparing the 3 simulations. The obtanied results and the layouts used are:

Simulation obtained with S parameters and S2P_Pad3

Simulation obtained with component
sc_atc_100_CDR13BG_G_19960828 placing in the layout

Layout for EM co- simulation with S2P_Pad3

Layout for EM co- simulation with
sc_atc_100_CDR13BG_G_19960828 component placed

Blue lines in S parameters graph are the EM co-simulation and red lines are the schematic simulation. As how can see, the simulation with the component "sc_atc_100_CDR13BG_G_19960828" differs a lot in relation to the schematic simulation and the S2P_Pad3 component, and the simulation with S parameters and S2P_Pad3 is closer to the schematic simulation than the simulation with the component.

Why this happens, if the S parameters in the SP3-Pad has been extracted from the component "sc_atc_100_CDR13BG_G_19960828"? I think in the case of the component instead of the S parameters with S2P_Pad, the EM co-simulation has considering the encapsulation of the component but not in the case of the S2P_Pad component, where the only reflected are the pad's size.

Whic one should be the correct simulation for an EM co-simulation and is this the right way to do it?

 

[BigBoss]

Co-Simulation places an extra Port in EM simulation and this Port is coupled to Real PADs Ports.
Therefore, this error occurs.
Don't forget, the Ports are coupled between each other, and EM simulator engine calibrates these ports to extract s-parameters. The difference between schematic and layout EM simulation is normal. The models are not %100 accurate for planar structures.

 

[volker@muehlhaus]

Also take a look what port configuration and port calibration you have for the SMD device. If your co-simulation treats the device as series 1-port (port cal type SMD), that means no parasitic shunt effect of the device is included. To include the device shunt effect, the device must be simulated as 2-port with both ports referenced to PCB ground.