On Wed, May 6, 2020 at 08:39 PM, Radu Bogdan Dicher wrote:
I'm invariably getting about 5dB lower readings on the left/lower side of the response plot. I'm definitely willing to persist quite a lot, but I'd really love to figure out where the deficiency is...Give it a break, and come back to it after a couple days. If your unit might be defective, maybe look at the limit for returns pronto tonto.
It would be useful if you filled in the blanks in the middle data area. Use the mouse to place the cursor in the plot area. Put the cursor in the middle of the filter response and click to move the cursor there. This will add data to the cursor 1 field. And that data will show numbers for ESR, L, C, impedance, phase, gain, etc derived from both S11 and S21 that can be used to help determine what is different in your fixture, cal, or hardware / firmware nanoVNA, as indicated in your plots.
OTOH, this current plot looks better than all previous ones.
There appears to be an odd noisy effect just below 10.7 MHz that hardly shows on the upper response curve, but is jarring on the group delay plot. Oddly, it also corresponds to a deep notch in the left side filter response at the bottom, that may be related to the broad lower left side pull down. I cannot explain it.
You will be really glad you bought that HP54522A Scope when I show you more tricks you can do with it to measure the IF Path in tuners, and raw filters. It will give you an alternate path to verify all this stuff, including filter response. and tuner IF filter response.
In the mean time, feed the nanoVNA CH0 into the 54522A. Below 300 MHz you will see a nice square wave with a 600 ps risetime. This guy can cause resonance problems with all that upper frequency energy being pumped out.