Simulating a waveguide-coax transition in HFSS

[LioTa]

Hello, I am currently working as an intern. My mission is to find the length of a tapered section that would make a transition from a WR430 waveguide to a 10x2.5cm waveguide filled wirth FR4-Epoxy, the goal being getting low VSWRs so that the Epoxy heats homogenously.
I have watched a lot of tutorials on HFFS regarding the simulation of rectangular waevguides and already done a lot of simulations using wave ports directly on the faces of the waveguide, but my boss recently asked that I simulate the entire structure and showed me the waveguide that was going to be used, and I realised that it was a WR430 to coax adapter. So now I've got this question : How do you simulate the wave coming from a coax, do I have to model the probe or do I just have to do as I did so far, like in the picture attached.
Also, if anyone knows if there is a way to calculate the guided wavelength in a tapered section of a rectangular waveguide, I'd really appreciate it because I can't find anything about it.
Hope I was clear enough, if I wasn't or said anything that looks stupid to you, don't hesitate to tell me.

 

[PlanarMetamaterials]

Yes, if you have been asked to model the whole structure, you should probably include the adapter to a high a degree of fidelity as you can.

That being said, I'm not sure why you'd need it in this case, unless the tapered sections are filled with epoxy as well. You should be able to infer from the data sheet that the transitions work reasonably well -- for additional confidence, you could perform a co-simulation with the model you've shown and the S2p file of the adapter (if available) included on both ends.

 

[LioTa]

I see, thanks a lot for the response!
My boss said the tapered section could be filled with epoxy if I found that it would lead to a better matching.
Also, when the probe will be included, should I keep my wave ports as they are how is there a better way to simulate the wave entering the adapter?

 

[PlanarMetamaterials]

If the adapters are present, you should simulate the signal entering from the coaxial ends, yes -- as long as you have a good model for the adapters. This will ensure you're modelling all possible effects. That being said, it might not be necessary if the epoxy is far enough away from the adapters.

 

[LioTa]

Thanks a lot for your input, I'll try to simuate it entering from the coax transition then.