Question about Momentum: How to simulate patterns in GND plane
时间:04-07
整理:3721RD
点击:
Dear experts,
Let's say that I am trying to simulate an on-chip CMOS inductor which its GND plane has pattern on it (Pattern Ground Shield).
How I should define this GND plane in Momentum?
I think the way for doing this is defining a finite GND layer with pattern on it.
First, to make sure that I am doing it right, I designed a branch-line coupler on layer cond and then I tried to remove GND by making it open boundary in substrate definition and then added a layer say cond2 and attached ground reference ports on it to make it GND. (Every single port on cond layer has its associated ground reference port on cond2)
But unfortunately simulation results of two cases (case1: infinite implicit GND plane, case2: finite explicit GND defined by cond2 and ground reference ports) don't match. I mean there is a great difference.
How should I simulate GND with pattern on it using Momentum?
Since in on-chip CMOS inductors using pattern ground shield is a common practice, I am wondering how people deal with this problem.
Let's say that I am trying to simulate an on-chip CMOS inductor which its GND plane has pattern on it (Pattern Ground Shield).
How I should define this GND plane in Momentum?
I think the way for doing this is defining a finite GND layer with pattern on it.
First, to make sure that I am doing it right, I designed a branch-line coupler on layer cond and then I tried to remove GND by making it open boundary in substrate definition and then added a layer say cond2 and attached ground reference ports on it to make it GND. (Every single port on cond layer has its associated ground reference port on cond2)
But unfortunately simulation results of two cases (case1: infinite implicit GND plane, case2: finite explicit GND defined by cond2 and ground reference ports) don't match. I mean there is a great difference.
How should I simulate GND with pattern on it using Momentum?
Since in on-chip CMOS inductors using pattern ground shield is a common practice, I am wondering how people deal with this problem.
Solved:
I was doing it right:
First, to make sure that I am doing it right, I designed a branch-line coupler on layer cond and then I tried to remove GND by making it open boundary in substrate definition and then added a layer say cond2 and attached ground reference ports on it to make it GND. (Every single port on cond layer has its associated ground reference port on cond2)
But for some unknown reason simulation results of two cases (case1: infinite implicit GND plane, case2: finite explicit GND defined by cond2 and ground reference ports) didn't match.
Now they match though there is very small discrepancy between them.
