Cheap Coaxial Splitter...how does it work?

[Peeps]

So I decided to search for images of the insides of those cheap cable-tv coax splitters and came across this:



And now I'm pretty confused about how it actually works. I mean the input loops around the ferrite bead ONCE and is...shorted to itself . Then the output wire just passes through the middle...a tiny single strand. How is that enough to couple anything? It all just looks wrong to me.

[AG6QR]

I think you've missed the way the wire is routed through the rightmost ferrite.  I believe the wire starts at the rightmost jack, goes throught he middle of the ferrite, loops around it, where a tap goes out to the input jack, and then loops back in through the middle of the ferrite, continuing on to the left where it's connected to the tap on the other ferrite.

That loop around the ferrite just splits the signal in two, while maintaining an impedance match.  There's a similar loop around the leftmost ferrite, and in the same way, it splits its signal in two while maintaining the impedance match.  Thus, the left two output jacks get a lower signal level than the right output jack.

There's a DC connection all the way through, between all the inputs and outputs.

[Peeps]

Ooh well that does make a lot more sense. Thank you. For some reason I couldn't get it out of my head that they weren't  directly connected.

[Rigby]

the right-most output is 50% of the input signal, then the other 50% is split between the leftmost and middle outputs, yielding 25% signal strength on those.

those splitters used to have embossed db loss characteristics on them.  One would say -3.5db and the other two would say -7db.

[strangelovemd12]

The newer one I have laying around still has the loss markings, but it is a 2 way.  Comcast sent it to me a couple months ago, so presumably cheap but not the cheapest.

[hagster]

@AG6QR Is this how you think it's wired? Not sure I understand how the ferrite maintains the impedance match.

What's the frequency response of something like this?

[AG6QR]

@AG6QR Is this how you think it's wired? Not sure I understand how the ferrite maintains the impedance match.

What's the frequency response of something like this?

Yes, I believe you've drawn the wiring correct.

Here's a discussion of splitters in general

http://www.minicircuits.com/app/AN10-006.pdf

The pictured splitter appears to be roughly similar to the splitter described and analyzed in Figure 3 of the linked PDF, except missing the resistor between the output ports.  It's just the center-tapped transformer, lacking all the other parts.  It probably does a poor job of presenting an impedance match on the input, and its function may be more to provide a little isolation between the two output ports.  It really is made to a (cheap) price point.


Interstingly, you got me reading about splitters.  I found this interesting comparison between a good one and a bad one, with photos:

http://www.elso.sk/media/download/datasheet/Fnet-Splitter-CIQ.pdf

The cheap one shown looks identical to the one that sparked this thread, except that it's just a two-way split, rather than three way.

What I wonder is how I can find a splitter designed like that "good" one in that example.  Consumer level splitters for TV are normally just marketed as black boxes (or shiny gold colored boxes) with no information about the nature of the impedance matching network inside.  But it seems like the really cheap ones are common.

Here's another analysis, showing the same type of splitter with just a single ferrite bead.  Scroll down, and/or search for "VH476" on the following page.

http://www.cliftonlaboratories.com/6_db_hybrid_combiner.htm

[KJDS]

What frequency range do you want to cover?