Does the matching network really matched?

[anydream]

The RF impedance of ESP32-D0WDQ6 is 30+j10Ω, other RF components such as QPF4219 & RFSW8009 & DEA162450BT is 50Ω.

I noticed that there is some 12pF/56pF/3.3nH/7.5nH between them, I think some of these component values are wrong or unnecessary.

The correct matching value between 30+j10Ω and 50Ω should be something like shunt 2.8pF + series 2.6nH + shunt 2.0pF, and there is no need for any matching between components with impedance is 50Ω.

All these RF IC except RFSW8009 have internal DC block, so some DC blocker is unnecessary too.

Can you explain the purpose of these matching component values?

Thanks!

[bisonscience]

All the device are matched according to the S parameters measurement at each port. The completed measurement for each section is shown on the Crowd Supply website. Each section shows correct power delivery which is possible with proper/correct matching.

https://www.crowdsupply.com/bison-science/esp32-m1-reach-out

Please find here the datasheet for RFSW8009. The datasheet illustrated the Evaluation Board contain coupling capacitors similar to ESP32-M1. This is not new in RF to place coupling capacitor according to the resonance frequency.
https://www.qorvo.com/products/p/RFSW8009#documents

[anydream]

Thanks for your reply. I noticed that due to some parasitic capacitance and inductance effects, the real world is very different from the theoretical simulation.
Can you give me the final Bill of materials? (after S parameters matching)

[bisonscience]

Currently, I am doing parts replacement because of the part shortage. I will update the schematic if necessary with new parts.