3

I am new to antenna design and teaching myself how to optimize antennas using a VNA.

I have created a monopole antenna on a PCB (i.e. a single length of track.) The antenna can be fed via a PI network and is connected to my NanoVNA via an SMA connector. My mission is to understand what values of L and C I would use in my PI network to match this antenna.

I have attached the Smith chart and readings, along with the explanation of how I think I do this, (also captured below.)

Smith chart and readings

To make this an ideal 50Ω antenna, is the following approach correct?

  1. Calculate value of a shunt inductor to move value to point X following the red arrow on the chart to, for example, on the 50Ω impedance line
  2. Calculate the value of a series capacitor to move value along the green line to point Y. Is 50Ω the ideal point?

Using the readings from my NanoVNA, how do I calculate the shunt(s) (L or C ) and the series (L or C) values?

JRE
  • 67,678
  • 8
  • 104
  • 179
Daveo
  • 49
  • 5

2 Answers2

2

You didn't mention the operating frequency. – Andy aka yesterday

Doh - thanks Andy.. the centre frequency is 868Mhz – Daveo

enter image description here

Now it's solvable. I choose initially to look at the equivalent parallel impedance of your antenna and note, that to tune-out the inductance (443.69 nH) at at 868 MHz you need a parallel capacitance of \$\dfrac{1}{4\pi^2 f^2 L}\$ = 0.076 pF. Now 0.076 pF is not really doable with any certainty because the parasitic capacitance that surrounds the joining of this added capacitor is going to be about that amount.

So, you series tune the equivalent series inductance of 5.18 nH. That requires 6.49 pF. This means that at 868 MHz, the impedance of your antenna looks purely resistive at 258.39 Ω.

To match that to 50 Ω requires an L-pad like this: -

enter image description here

Above image taken from my randomly focussed and somewhat crappy and sporadic website.

Simulation: -

enter image description here

Andy aka
  • 434,556
  • 28
  • 351
  • 777
  • 1
    a massive thanks Andy - thats a fantastic explanation! I followed it and completely understood it. You are clearly a very talented engineer! thanks again - Dave – Daveo Jan 05 '22 at 23:06
0

Same situation here, just wanted to add an alternate approach. The NanoVNA can export Touchstone (s1p) files for further analysis. Export it either through USB drive mode or with nanovna-saver.

There is an awesome "free" tool called Atyune that can read Touchstone files, show frequency bands and generate various matching networks. You can even simulate and look at the expected Smith chart.

Even it you want to do this by hand for learning, I suppose this will certainly provide a good sanity check on whether your calculations make sense.

Insa
  • 103
  • 2