What is port direction for S11 and S22 at 6GHz?

[BrucceJason]

The specification only said port direction for S11 and S22 is better than 10 dB. but what is the value for it?

[jankae]

I am not entirely sure what you are referring to. Could you please point me to where that value is specified? Maybe I have forgotten something but I can not seem to find it anywhere. I am also not sure what "port direction" is supposed to mean (maybe you mean port directivity?)

[BrucceJason]

The data comes from a lot of online retailers. For example, Seesii LibreVNA 2.0
and AURSINC LibreVNA NanoVNA Vector Network Analyzer

Yes, I mean port directivity. Sorry for my typo.

BTW, I saw a recent update on VNA from Henrik who have designed VNA which could work up to 15 GHz Designing a Low-Cost, High-Performance 10 MHz–15 GHz Vector Network Analyzer
However, He only provided the schematics but not PCB layout. Hopefully, someone can facilitate the layout process.

[jankae]

The data comes from a lot of online retailers.

In that case I am curious how they got that number because I did not specify that anywhere. My guess is that someone came up with a method to measure it in some way and then everyone else just copied the data.

In both your links they also claim a directivity > 10 dB at 9 GHz, for which the LibreVNA is not even specified. Yes, you can measure something there using the harmonic mixing mode but in my experience the data at 9 GHz is not that useful anymore and certainly not guaranteed.

They also do not specify whether that number is raw directivity or calibrated directivity. All in all, I would not trust that specification very much.

BTW, I saw a recent update on VNA from Henrik who have designed VNA which could work up to 15 GHz

Yes, I have also found that blog post recently. Very impressive design! It shouldn't be too hard to replicate the layout, only a few areas on that PCB handle RF frequencies. Personally, I am not interested in simply copying the design. First of all it wouldn't be fair to Henrik and second of all the block diagram has a few downsides:

• 10 MHz low frequency cutoff: sure, not an issue if you are only interested in high frequencies. But for a general purpose VNA, I would prefer something that can go lower
• Very little amplitude control for the stimulus power
• No source filtering

That is not really criticism for his design, I am sure he is aware of the limitations and chose them intentionally. It is just not the target I have set myself for a new LibreVNA version.