Hi Guys, I'm designing a Wilkinson power splitter to work around 160 MHz using lumped element components. I've designed the splitter as per the standard design procedure for a series L / shunt C topology (values are 13pF, 27pF, 68nH) and it works well. However I need to add some high pass filtering to my circuit and I'm wondering can I design a high pass version of the wilkinson splitter (instead of the standard low pass version) to combine the two functions? I've added some high pass filtering in before the splitter but my in-band insertion loss becomes too high.

Does anyone know where would I find how to do the calcs for this, if it's possible?

Many thanks,
Dave

 

You can simulate a quarter wavelength line either as a low pass or as a high pass. For a 2-way Wilkinson splitter the quarter wave lines need to be √2*Zo. So for a 50 ohm system, that is 70.71 ohms. A 70.71 ohm quarter wave transmission line will consist of 3 reactive components, each with a reactance of 70.71 ohms.

So a 70.71 ohm quarter wave line at 160 MHz, configured as a high pass would be a shunt 72 nH, a series 14 pF, and a shunt 72 nH.

A Wilkinson divider would look like the attached.

 

Great, thanks so much, that's a big help. Could I go one step further and ask how I could apply this to an unbalanced topology, say where i wanted the 3 dB split to be weighted more towards one arm?

Many thanks,
Dave

 

Try this: https://www.everythingrf.com/rf-calculators/wilkinson-power-divider-calculator