Strange group delay at lower frequencies in Momentum
时间:04-04
整理:3721RD
点击:
Hi,
Say I have a simple CPW line and I want to see its group delay up to 40 GHz. I use Keysight's Momentum for my simulations.
My understating is that group delay will be almost flat and eventually will increase past a certain high frequency (say 30 GHz).
But to my surprise, I get an increased group delay at lower end of my simulation bandwidth. it doesn't matter whether I set the lower simulation frequency to 1GHz or 100MHz.
I get nonsense results when I try to go to very low frequencies (compared to the higher end of 40 GHz) like 1MHz or 10 KHz.
Since I don't see any reason for the increased group delay at lower end, I suspect that this might be a shortcoming of the simulator or my setup.
Is it safe to assume that group delay in CPW (which has a good group delay flatness) shouldn't increase at lower end of the desired bandwidth?
Please advise
Say I have a simple CPW line and I want to see its group delay up to 40 GHz. I use Keysight's Momentum for my simulations.
My understating is that group delay will be almost flat and eventually will increase past a certain high frequency (say 30 GHz).
But to my surprise, I get an increased group delay at lower end of my simulation bandwidth. it doesn't matter whether I set the lower simulation frequency to 1GHz or 100MHz.
I get nonsense results when I try to go to very low frequencies (compared to the higher end of 40 GHz) like 1MHz or 10 KHz.
Since I don't see any reason for the increased group delay at lower end, I suspect that this might be a shortcoming of the simulator or my setup.
Is it safe to assume that group delay in CPW (which has a good group delay flatness) shouldn't increase at lower end of the desired bandwidth?
Please advise
Is the load on the CPW equal to its characteristic impedance?
It's a normal CPW line with a characteristic impedance of 50-ohm terminated on 50-ohm ports at both sides.
it should be relatively flat, unless there is dispersion at the higher frequencies.
I am not up on momentum, but if it is a meshing software, maybe your mesh size is whacky and it is predicting poorly.
Momentum is 2.5 D electromagnetic solver, that uses meshing to solver structure. You can re-run simulation with dense mesh setting and take more than 50 cells/wavelength.
Strange frequencies Momentum 相关文章: