Difference in S11 paramter between hfss and momentum

hfss s11

i have been trying to compare momentum simulation to hfss simulations. i am using a directional coupler, and when i simulate using both programs i am getting all the S-params to match very closely, but S11 is very different. i'll attach the plot, dashed lines are the hfss params, and solid lines are momentum(ignore the colors, they got screwed up in matlab). i really can't figure out why the s11 param is so much different. i have tried as many different mesh's and different port sizes in hfss and i can't get them to change to much.

layout is just 2 parralel microstrips, with spacing of 1.5mm on substrate of 1mm hieght, dieletric const=4.5, strips are 1.81mm each. i am wondering if i should got out 90degrees from each ustrip so when i set up the ports they will be far apart from each other, but i didn't want to do this unless i had to becuase i didn't want to add anything more into the circuit that i didn't have to, i was just trying to keep it simple

just getting frustrated, wonder if anyone has anyideas why this is this way,. thanks for the help

In HFSS simulation, you set one wave port to each trace? If like that, the size of wave port is too small, that may result the bad result for S11 parameter. I recommend you set one wave port to 2 traces, and assign 2 terminals, assign them integration line accordingly. Let me know the results.

Best Regards,

i do have 1 have port with 2 terminal lines, and the thought did come to mind that it was too small so i increased the size until it was huge and no difference. but i do notice that when i look at the impedance it starts off at 48 ohms and goes up to about 55, maybe momentum and hfss may be getting different impedances and possibly that is causing the difference,

How can one S-parameter be much different while the others are close?
How about the Conservation Law?

try adjusting your wave port definitions..and reducing the length of your guard trace

Calculate the power balance from both S-parameter sets to see who's lying.