RF design - How are you clearing plane metal under component pads?

Discussion created by bburski on Apr 20, 2012
Latest reply on Apr 29, 2012 by dale.rebgetz

Reaching out here to Expedition RF users to see how they handle clearing plane metal under component pads. I’ve submitted idea D8394 which has not seen a lot of activity so I’m wondering if my explanation wasn’t clear or if others don’t need to do this. So if you use controlled impedance RF lines connecting soldered components where the reference plane is a few layers away, I would like to hear from you.


In many cases we want to replicate a part vendor’s reference design. These reference designs typically are simple construction, two layers with adjacent layer reference planes. Our boards are multilayer with the reference planes 2 or 3 layers away with ground filled in every open space on every layer.


All RF nets are assigned to a Netclass to ensure consistent clearance to the same layer ground plane. The Group Clearance Rules work great for projecting this clearance through adjacent layers between the meanders and the reference plane. The problem is the plane metal directly between SMT component pads and the reference plane does not get cleared like meanders because component pads can’t be added to an RF group, therefore do not apply to the clearance rule.  This will result in different clearances between the top and intermediate layers…potentially causing performance problems.



I would like clearances identical from 1 to 2 to 3… With these results we cannot replicate the vendor’s reference design structure and depending on clearances, pad-to-plane coupling can increase capacitive and impedance discontinuities. So, what are others doing about this?