polyam
|
Hi,
I am using one of the top metals (for the sake of discussion let's call it m5) in the given technology to ship a high frequency complementary (differential) clock from one point to the destination (by high frequency I meant +50GHz). Let's say I have 100-200um differential traces (m5) and side-shielded by the same metal (m5). Do you think putting lower metals (let's say m2) as a ground plane underneath of the clock path and connect it to a substrate via a substrate contact would reduce the substrate coupling? how does it reduce the coupling? how to model that ? which tools can model the substrate coupling/noise ?
Thanks
|