homebrew connector impedance - how to figure?

[WBear2GCR]

Ok, so...

I have a few assorted coaxial type cables of different configurations, none of which are particularly standard compared to a SO-239/PL-259. Since we hams tend to be stuck in that world for connectors, and N type are a royal PIA, expensive and still won't fit these cables...

Anyhow I have three different cables that I want to use and interface back to SO-239/PL-259 world. One is a 50 ohm (yellow jacket) Ethernet coax that is slightly thinner than standard RG-8 (I think, gotta take a really really good look at that) and has braid + foil or else it is two layers of foil or something along those lines... I and 100% certain it is 50 ohm, fwiw. I also have some rather large coax of about 1"+ OD, a very heavy thick inner conductor, and it has a standard nomenclature that I looked up a year or two back, it too is 50 ohm. Then I have some CATV hardline, 1/2" nominal OD. I have the splices for it and I think an end termination that comes out to a center pin on the rear...

Anyhow, for the two first ones I was thinking about using a section of copper tubing, sliced along the lengthwise, two or 4 ways to permit a clamp effect, and then an endcap soldered on one end with a "standard" PL or SO connector mounted there.

Then there is the idea of a small metal mini box or PCB soldered up box, with the cable clamped on the input side and a connector on the other side and the wires run through up to the connector...

All this sealed with appropriate goop...

BUT, I read that the difference between the  50 ohm BNC and the 75 ohm BNC is only the difference in distance between the inner and outer and nothing more.

SO, here's the question, will one of these sorts of termination ideas alter the impedance at the connector in a meaningful way - and regardless what needs to be done if one wanted to maintain the "50 ohmness" of the homebrewed termination??

                 _-_-bear

[W1RKW]

I'll take a general stab at this.

I would think if you created the connector system you are talking about and ran a HF digital signal through the cable and connectors you should see whether there is a blip or bump in the square wave signal at the receiving end (50ohm load) if there is a mismatch in the circuit.   Sort of like using a TDR.  Ideally you should see an exact replica of the source signal for a smooth impedance characteristic along the length of the circuit.  I don't remember if it would be the leading or trailing edge of the squarewave but whatever the edge, it will indicate an upward swing or downward swing depending on the impedance.  Ideally a very short pulse like that used in a TDR would be the best approach for checking impedance matching.  That's a very general approach.

[Gito]

Hi,Bear

Do you have the ARRL antenna hand book?
In this book ,there"s an article of the Characteristic air insulated of air-insulated co-ax line

The characteristic impedance formula is

Zo =136 log b/a

Zo= the characteristic impedance
b = inside diameter of outer conductors
a = outside diameter of inner conductor (in same unit as b)

So 50 ohm = 136 log b/a

using my calculator ,I found that b : a = 2.3
50 Ohm = 136 log 2.3

using coper tubing to connect the co-ax to the BNC  connector
(for instance), if you used 1 mm for the diameter of the inner conductor,than you must used a 2.3 mm for the inside diameter of the outer conductor.(coper tubing)

Regards

Gito

[WA1GFZ]

Bear,
There is no difference in space in a 50 ohm vs 75 ohm BNC. The 75 ohm has air dielectric between center pin and grounding flanges. 50 ohm has a ring of teflon or plastic. to raise the dielectric constant between conductors.

[W2DU]

Whether any difference in impedance from one end of the connector to the other is significant depends on what frequency it will be used on. In the MF and HF regions it won't even be noticed. At VHF it might be noticeable, but will not cause any trouble. However, at high UHF frequencies it could be troublesome if the difference in impedance is large.

Walt,W2DU