Hopefully those who read this blog survived the assault on the Jellybeans, colored eggs and chocolate bunnies. Needless to say, those who practice dentistry are a bliss with the thoughts of new cavities being formed arising from the Easter holiday.
But we are here to talk about amateur radio and so we must begin. I was asked a question the other day on why I was such a Debbie Downer on Nano VNA's.
A recent post had this photo above showing the plot of a four-pole crystal filter I built and tested using my Nano VNA. This plot told me a lot. But behind the plot itself was a knowledge base of what the plot should look like after following the design process.
My issue with the Nano VNA is that it is nothing more than a tester (assuming its calibrated, working properly and the test setup has a load on the DUI) that depicts the response of what actually exists. In the plot above I can see that there is a good match to the filter by the flatness of the pass band, that it favors LSB by the slope of the curves and shape factor that with the wide skirts is as broad as Mary Jo. Will it work in a circuit? Yes! Will it be good on CW? Likely No!
The Nano VNA's use as a design tool is limited to physically changing hardware and then making a run and reporting that result. That takes a lot of valuable bench time.
If we move to Band Pass Filters, then an approach I prefer is to simulate the filter in LT Spice where its use as a design tool enables me to look at the pass band frequencies, the shape of the filter and the all-important section coupling capacitor. Changes are quickly done with a click of the mouse, and you don't need a soldering station.
Simulation with 1pF Section Coupling Cap.
Simulation with 5pF Section Coupling Cap.
Simulation with 0.5pF Section Coupling Cap.
Having done this simulation over and over you soon recognize the pattern of a double hump response means the section coupling cap is too large and making it smaller results in a flat top response curve. If that cap is too small you get a narrow, peaked response. The key is that the simulation approach gets you there faster.
Once built you can then run your Nano VNA to affirm what was actually built is the response predicted. Here is the key point! If the two don't match, then most likely it is not the design but what was built that is the issue. Things like a bad solder or shorted solder joint, a wrong value capacitor or wrong number of turns on the inductor are the issues not the design!
Bare basics: a screwdriver is a poor substitute for a hammer.
Memories of Easter's past. I grew up in western Pennsylvania, where we had large enclaves of immigrants such as Italian, Irish, Polish and Ukrainians. Many came to work in the aluminum or steel factories and coal mines. Of note about the Ukrainians, they hand painted their Easter eggs and these were works of art. I also remember the Ukrainian YL's seemed to physically mature more quickly and were quite hot! Mary Jo may have been Ukrainian!
73's
Pete N6QW