I am attempting to compute an impedance matching network by following the guidelines outlined in the ST Application Note - AN5457. However, on page 24, the formula and calculations appear perplexing to me due to the presence of a negative value under the square root.

enter image description here

Even though the provided example calculation values for L1 (4.45 nH) and C1 (5.53 pF) are intended for a 14 dB power setup at 868 MHz, they don't correspond with the component values found in the reference design, such as the MB1720. For instance, besides L1, the reference design includes a 0 Ohm resistor, and for C1, it features 0.7 pF. enter image description here

What further confuses me is that these values are allocated to a high-power reference design - 22 dB components too. Kindly advise how a matching network is calculated? Does the reference design show wrong component values or are the Application Note calculations misleading?

  • \$\begingroup\$ You have little red markers on the schematic implying that every node is 50 ohm and, that definitely will not be the case so, please explain what that is all about. To explain each component in detail proper figures are needed for the IC output impedance and antenna impedance. \$\endgroup\$
    – Andy aka
    Feb 14 at 9:52
  • \$\begingroup\$ I have revised the post and included a link to the Schematics along with the more detailed screenshot. \$\endgroup\$
    – Atrahasis
    Feb 14 at 11:13
  • \$\begingroup\$ You still haven't explained what the 50 ohm red markers are all about. These cannot be correct at all and, although you are distracted by the process, I am distracted by those markers. This doesn't help me explain to you what might be going on here. I mean, if you just wanted the theory on how a pi-filter can match different impedance that's easy but, it's all the unknowns such as antenna impedance and source impedance that makes this a mammoth process to explain. Start with humbler beginnings is my advice then learn from that level. \$\endgroup\$
    – Andy aka
    Feb 14 at 11:30
  • \$\begingroup\$ And please provide a direct link to the source document. I won't download anything from unknown sources for my own protection. \$\endgroup\$
    – Andy aka
    Feb 14 at 11:32
  • 1
    \$\begingroup\$ If that's what they indicate then it's a mess because using a 50 ohm trace of any significant length with an impedance source or load that isn't 50 ohm is going to produce problems. If on the other hand those traces are short, then there's no need to state that they are 50 ohm. That's my confusion here. \$\endgroup\$
    – Andy aka
    Feb 14 at 11:35


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.