Microstrip Line Simulation

[nikosnte]

Hello dear friends, i have two microstrip lines. One is straight with 18[mm] length, and the other one is constructed of many curved parts to be "bent" (second and fourth image attached). So, in order to achieve very small phase difference between these two lines i tuned the length of TLIN37 and TLIN40 to be 0.0965 [mm] and 0.6133 [mm] respectively. I simulated both lines in the schematic and i came with a very small phase difference for the whole frequency band as it can be seen (third image attached). Later i simulated the same lines in Momentum but the results were quite different (fifth image attached). I also got this warning

Now, i know that the results in Momentum may differ from the results in the schematic, but i don't know if i can trust these results because of this warning. My line width is 0.4557mm and i use dielecetric of constant 6.15. The frequencies are above 27 GHz. So the port width is bigger than λg/8 at some point and this is the reason i get this warning. Should i completely not trust my momentum results, or my schematic results? Thank you

 

[vfone]

In the schematic, even the models for curved and straight transmission lines are accurate, the circuit doesn't take in consideration any coupling when meandering the lines, and at 27GHz may give this difference in results.
So I would trust more Momentum, than the schematic.
I don't know exactly about the error you get, but you can right click on the pin, edit, and change the shape and its dimensions.

 

[volker@muehlhaus]

If the rest of the Momentum model is fine, you can trust results. Momentum warns quite early about the port size. If the warning is 29GHz and you simulate up to 33GHz, that is fine.