### Wideband loss factor / kappa

Posted:

**Fri 13 Mar 2020, 14:49**I created a simple microstrip example after looking at some of the tutorial files (then rewrite the code from scratch to check if I got the things right).

Good news first: Results are within the expected range.

But (as always ): I found some odds due to loss factor of the substrate. I am using Er=4.6 and a kappa value calculated from loss factor. The standard material is somehow narrowband (as the kappa from loss factor formula is based on a fixed frequency term). The frequency area of interest for my example is pretty wideband (0-35GHz).

I got very good results for ~5GHz segments if I run a simulation with a kappa calculated for the center frequency of this 5 GHz segment.

Is there a simple example/howto showing a frequency dependent modeling of a (FR-4) substrate with a given loss factor?

For the current use-case it would be sufficient to assume the loss factor being constant.

I currently use:

kappa = 2*pi*fcenter * EPS0*EPS_FR4 * loss_factor; % substrate losses

Best regards

Gerald

Good news first: Results are within the expected range.

But (as always ): I found some odds due to loss factor of the substrate. I am using Er=4.6 and a kappa value calculated from loss factor. The standard material is somehow narrowband (as the kappa from loss factor formula is based on a fixed frequency term). The frequency area of interest for my example is pretty wideband (0-35GHz).

I got very good results for ~5GHz segments if I run a simulation with a kappa calculated for the center frequency of this 5 GHz segment.

Is there a simple example/howto showing a frequency dependent modeling of a (FR-4) substrate with a given loss factor?

For the current use-case it would be sufficient to assume the loss factor being constant.

I currently use:

kappa = 2*pi*fcenter * EPS0*EPS_FR4 * loss_factor; % substrate losses

Best regards

Gerald