#805

[@stevemahrer7097]

You should terminate any unused ports, that will greatly reduce ripple and improve isolation. Steve

[@robertdixon8238]

This is a Wilkinson splitter / combiner. It has less insertion loss than a resistive splitter, but operates over a reduced bandwidth. The trace lengths are related to the frequency of operation.
More information here:
www.microwaves101.com/encyclopedias/wilkinson-power-splitters

[@k7iq]

I love the way that analyzer traces the peaks !

[@bobblacka918]

I needed a 2-Way RF Splitter for ham radio use so I could use one antenna on two radios (receive only). On a whim, I measured the Insertion Loss and Frequency Response of several splitters I had in my parts bin designed for TV use. I discovered the GE Coaxial Cable Splitter which is spec'd over a frequency range of 5 to 2,500 MHz actually works well down to 1 MHz and the loss was only slightly higher than the theoretical 3.0 dB. This was clearly the best of the bunch and beats out many other well-known splitters in all essential parameters. Some engineer did a remarkable job designing this device to be able to sell it for such a low price. It can be purchased from many different sources for less than $10.
The down side is it uses F-Type RF connectors, but adapters are easily obtained to convert F to BNC. Keep in mind a 4-Way Splitter will have double the loss of a 2-Way so use the minimum number of ports that is required for the application. Note also I measured a very wide variation in performance among TV splitters from various brands even though they were all spec'd essentially the same so do not assume any TV splitter will work just as well. The GE was without doubt the best performing one.

[@IMSAIGuy]

If it is only receive, did you try a simple BNC T? I don't know why you need isolation between receivers. If you do, just use two resistors in a Y.

[@bobblacka918]

@@IMSAIGuy : Yes, that would work, but sometimes the Local Oscillator of one radio could feed back into the other radio. Depends on the frequencies and quality of the radio, but generally I like to keep everything at 50 to 75 Ohms all around.

[@hanaverike5409]

Hi, what was the f and attenuation of the pronounced dips at 5:27 on your video , do you recall? And, did you ever discover the origin of this device? Thanks

[@williamdenbeste9703]

It would be interesting to see this tested using the nanoVNA which can go up to 3GHz. Also would like to see all outputs terminated during all testing.

[@deBug67]

I think I have seen an explanation of these kind of splitters and the inductors are actually high impedance transmissions lines that gets lower impedance at each stage after the resistor bridges (hence the wider traces)

[@IMSAIGuy]

That sounds reasonable. This also keeps the phase information intact.

[@viocaia]

Nice! Please let me know , if I have 3 device on output , the left port should be terminated with 50ohms resistor?

[@IMSAIGuy]

yes

[@Mr.Leeroy]

7:56 spice
pure, unrefined spice..

[@NimbleJack3]

First time I've ever heard of a splitter being called a coupler, usually a coupler is more like a bridge.

[@bobblacka918]

A coupler is a more general term and splitter is one type of coupler (assuming non-resistive).

[@danc8278]

On that 4 port splitter would you have got more isolation from... Ah you just said while I'm typing on my phone lol

[@RB9522]

For heaven's sake, terminate the unused ports.

[@IMSAIGuy]

while you are correct, in these systems the first is split then those two are split. so pairs effect each other much more, the second pair not so much

[@RB9522]

@@IMSAIGuy I'd like to see you do the same tests again and compare the terminated results with the unterminated ones.