DIY RF splitter?

[awallin]

Hi all, is it possible to DIY RF-splitters with decent isolation?

I would guess a splitter with just three resistors in Y or Delta configuration provides very little isolation between input-ports (e.g. when used as a combiner)!?
https://www.radio-electronics.com/info/rf-technology-design/coupler-combiner-splitter/rf-resistive-splitter-combiner-divider.php

mini-circuits has lots of splitters for 30-60eur/piece - so I guess the trade-off is how much tinkering one has to do to DIY a splitter with similar performance (I would be looking at 10-100 MHz, not GHz!)

[TheUnnamedNewbie]

A resistive type is going to provide little isolation by design - even with ideal components it just wont pull it off. What you need for high isolation is something like a Wilkinson power combiner/splitter. This requires one resistor (100 ohm in the case of 50 ohm transmission lines) and a number of transmission line sections of various impedance. If space is an issue, you can also do it with inductors and capacitors (done on chip sometimes).

[awallin]

This requires one resistor (100 ohm in the case of 50 ohm transmission lines) and a number of transmission line sections of various impedance. If space is an issue, you can also do it with inductors and capacitors (done on chip sometimes).

hm, ok, although the mini-circ connectorized ones are a bit pricey there are SMD splitters with good specs for about 10eur/pcs. maybe that's the way to go, e.g.  https://ww2.minicircuits.com/pdfs/LRPS-2-1.pdf
for $9.35 I guess DIY tinkering to get 30dB isolation is not worth it..

[Wolfgang]

Hi all, is it possible to DIY RF-splitters with decent isolation?

I would guess a splitter with just three resistors in Y or Delta configuration provides very little isolation between input-ports (e.g. when used as a combiner)!?
https://www.radio-electronics.com/info/rf-technology-design/coupler-combiner-splitter/rf-resistive-splitter-combiner-divider.php

mini-circuits has lots of splitters for 30-60eur/piece - so I guess the trade-off is how much tinkering one has to do to DIY a splitter with similar performance (I would be looking at 10-100 MHz, not GHz!)

I would say that a frequency range from DC to 100MHz is uncritical. You could easily homebrew this.

https://electronicprojectsforfun.wordpress.com/rf-module-gallery/the-rf-splitter-and-combiner-gallery/a-resistive-combiner-for-imd-measurements/

What you could try is a 19.5dB splitter if you need more isolation. I have made a lot of BNC/leaded resistor stuff and measured them on a VNA.
Up to 100MHz results are quite good (30) if you keep leads really short:

https://electronicprojectsforfun.wordpress.com/measuring-a-picotest-j2111a-current-injector/verification-of-some-homebrew-bnc-calibration-standards/

When you try SMA connectors and chip resistors on transmission line PCBs you could easily go up to 500MHz.

Above that I would buy some Mini Circuits stuff because homebrewing gets too tricky at several GHz.