Weird resonance for backplane channels
时间：03-30
整理：3721RD
点击：
I've been trying to generate a very rough model for a typical copper backplane channel. I have been using SONNET Lite to generate the model and then porting it over to Virtuoso for circuit simulation purposes. For some reason when I try to use the model for simulation, the response always displays resonance around 10GHz. Here I've plotted S31:
The schematic in SONNET looks like this:
I tried playing around with the unit length of the model and also the unit cell size but the aforementioned behavior still persists. The metal layer is just copper and the dielectric constant of the sub is 4 (I just used generic values from wikipedia).
Anyone got any pointers on how to work around this? I'm expecting a copper trace to just have a monotonic insertion loss.
The schematic in SONNET looks like this:
I tried playing around with the unit length of the model and also the unit cell size but the aforementioned behavior still persists. The metal layer is just copper and the dielectric constant of the sub is 4 (I just used generic values from wikipedia).
Anyone got any pointers on how to work around this? I'm expecting a copper trace to just have a monotonic insertion loss.