I recently ran into an issue with a multi-layer board simulation where my Momentum simulation (and accompanying TDR view) disagreed with the results from the Controlled Impedance Line Designer (CILD). In order to isolate the problem, I created a new layout with a single layer board to see if the same mismatch occurred. It did - CILD as well as Si9000 and LineCalc all predicted a 50 Ω line width of about 5.5 mils. However, my Momentum simulations show this line as inductive. This is very confusing, since both the CILD and Momentum simulations are referencing the exact same substrate. I have to increase the line width to about 10 mils for Momentum to see it as 50 Ω. For reference, the basic parameters of my substrate are as follows:
Copper - 2 mils
Dielectric 2.62 mils (Dk = 3.06)
I have had a few other engineers take a look at this, and we can't figure out where the simulation is going wrong. Has anyone else run into this problem before? I have attached an archive of my ADS workspace for reference.