While my old PNA has limited support TRL, it doesn't support waveguides. They may have offered it with later firmware but I have the most recent version my hardware supports installed. I have attached a paper that discusses TRL, using WR90 as an example.
****
I thought I should double check that statement and it does support it. See attached photos. Which had I paid attention, I actually showed it previously:
https://www.eevblog.com/forum/rf-microwave/experimenting-with-waveguides-using-the-litevna/msg4801919/#msg4801919
It is not so old
I have done some waveguide TRL in 1995 on a HP8510 so before that become Agilent. (On the HP8510 itself, it is not outside using wincal or other software. I have not yet look in detail on METAS VNA Tools but I have seen you use it and that they say: “ Supported calibration types: One Port, SOLT, GSOLT, QSOLT, Unknown Thru, TRL, LRRM, Juroshek, Reflection Normalization, Transmission Normalization, One Path Two Ports, LHKM, LHKM_TRL_LRL, LHKM_TRM_LRM and Optimization.”. I have not yet have the time to look in detail on what do this software)
I was on groups.io (https://groups.io/g/liteVNA/topics) where someone had asked about TRL and WG. I did post a link to this thread. With you having a similar interest, you may want to join that group and see if the two of you could work together on it.
Thank you for the link. To be clear on my plan: My plan is first to see if I am able to do nice full two ports measurement. Then if it is ok, I want to play with calibration. The NIST has share a calibration method called multiline also in the 90s or beginning 2000. This is an extension of the TRL method. One of the limit of the TRL method is that the length of the line must represent a phase shift between
20 and 70 degres. (edit mistake 20 and 160 degrees). This doesn’t work well with small or big phase. (For waveguide it is not a problem because the waveguide is mostly used when you have only one mode (so between 2 cut-off frequencies and you can find a line that will meet the phase criteria). In the case of the NIST, it was for planar lines so with a bigger range of frequencies and they do multiple lengths to cover a wide range of frequencies. So back to the beginning, my plan is to play with that and with planar lines. That I can do line like the ring resonator I have shown before but I will not play soon with waveguide, I have not the skills to fabricate and there are nothing at low cost on eBay from France. I am still in the process of thinking my plan and assembling the bricks. For example, at this point, I do believe that the ham radios community has not seen scikit-rf. (Or I have miss why it is not used). I am interested in what you show and I learn a lot like how to do a transition between coax and waveguide or where is cut the half waveguide (you speak about that when comparing your fabrication and a Narda waveguide (perhaps the sliding load but I have forget, the important information for me was where should be an half guide cut
).
Edit: looking a little bit more on Metas, in release notes
- Agilent PNA Driver improved (support for point average added
- LHKM TRL LRL improved (choosing the eigenvalue for the unknown reflection).
So in your case that is perhaps also an option if the trl is not already in the pna.
And for me I have to work more but perhaps I could use also if I have full two port measurement.