Spiral inductor synthesization using ADS Momentum EDA tools

Hi all,

I am using Momentum to synthesize the spiral inductor. I have created the substrate description file for Momentum usage. I have checked that the input parameter is correct. Then I use this to synthesize my simple spiral inductor. This is using the Top metal (M6) to make a spiral inductor path and M5 for bridge connection. The Via5 for M6 and M5 connection. After synthesiztion, I use the synthesized model and compared with the PDK inductor model. I want to see whether it close to the PDK inductor.
A special result is given out. During the Momentum synthesization, I have done 3 different Via5 simplification.
(Case 1): In the original layout, there are 7x7 Via5 for M5-M6 connection. I have done one of this model. (one Via5 is 4um x 4um, and Via5-Via5 seperation is 2um)
(Case 2): I have simplified the 7x7 Via5 and merge them in a one big square (the size of the big square is 40um x 40um).
(Case 3): I have simplified the 7x7 Via5 and merge them in a one little bit smaller big square (the size is 16um x 16um).
Then, I compared this synthesized inductors with the PDK inductor. A strange reuslt is given. The original PDK inductor give a Q=10 @10GHz.
Case 1 Ind: Q=12
Case 2 Ind: Q= 9
Case 3 Ind: Q= 8
How could I explain this result??

Thanks,

Hi,

This apparent discrepancy between simulation and measurement is common. Many unaccounted affects could result in this difference. Among them could be wrong port setup, unaccounted de-embedding, low density meshing, and especially your definition of Q might be different than what was used in measurement.
Also, the measured value of Q is not something written on stone and could be slightly different than the real, after all engineers who measure them are human and subject to mistakes.
In my opinion the Case 1 in your via settings is the closest to reality. Simplifying vias should be done with especial care.