[TowerTalk] Fwd: RG-149: 50 ohm/70 ohm - does it matter?

[Michael Tope]

On 12/15/2013 9:05 PM, Jim Lux wrote:
<SCENARIO 1>
>> I think the real question is whether a such a hybrid/coupler/line
>> sampler would have a characteristic impedance? That is, if I have a
>> 75 ohm source and a 75 ohm load and I put a line sampler (designed
>> for 50 ohm systems) in between, will the 50 ohm source see a
>> mismatch.  I think not. Wrong! If you put a "line sample, made for 50
>> ohm, in a matched 75 ohms line, the sampler will show a 1.5;1 SWR. It
>> doesn't care what source you have, only what impedance the load
>> have.
>>
<SCENARIO 2>
> I'm not so sure.  If I make a short length of 75 ohm transmission line
> (or, for that matter, use a PL-259/SO-239 barrel, which is NOT 50 ohms)
> in a typical 50 ohm system.  So it's, say, 10 feet of 50 ohm line, 2" of
> 75 ohms, and 20 feet of 50 ohms, terminated in a perfect 50 ohm load.
>
> I doubt that if you hooked up an SWR meter it's going to show anything
> remotely like 1.5:1.

This scenario you just described is not the same as the one above it. In 
the scenario above (scenario 1), you are putting a 50 ohm calibrated 
meter in a matched 75 ohm system. If that meter uses a Bruene circuit 
(classic VSWR bridge) it will read 1.5:1 VSWR. In the scenario you just 
described (scenario 2), you have what amounts to a matched 50 ohm system 
with a 2" long 75 ohm impedance bump. At low frequencies where 2" is a 
small fraction of a wavelength that 2" x 75 ohm bump won't have much 
impact on VSWR (for the exactly the reasons you describe below). In any 
case, in scenario #2 you don't indicate whether the VSWR meter is 
calibrated for 50 or 75 ohms. That is what matters. If it is calibrated 
for 50 ohms, the meter will read ~1:1 VSWR. If it is calibrated for 75 
ohms it will read 1.5:1 VSWR.

I think the confusion here has to do with the characteristic impedance 
of the short line that passes thru the current sampler (typically a 
current transformer composed of a straight wire passing through center 
of a toroid with the secondary wound on the toroid). That impedance 
doesn't really matter for exactly the reasons you describe below. What 
matters is the ratio of the voltage sample (capacitively coupled sample) 
to the current sample (inductively coupled sample).

> Sure, if I hook up 10 feet of 50 ohm line, then 20 feet of 75 ohm line,
> terminated in a 75 ohm resistor, THEN the SWR meter is going to read 
> 1.5:1.

Again it depends on whether the bridge is calibrated for a 50 or 75 ohm 
load impedance. If it is setup for 50 ohms, it will read 1.5:1 VSWR. If 
it has been calibrated for a 75 ohms system it will read ~1:1 VSWR.

> (neglecting the loss in the line in all cases)
>
> Here's why..
>
> Call the starting point A, the transition from 50 to 75 point B, the
> transition from 75 to 50 (2" later) point C, and the ending load point D.
>
> My power starts at A propagating down the line.  It hits point B, and
> some power reflects back.. it continues on and hits point C, also
> reflecting some power back.  The amount of power is almost exactly the
> same, and because the distance from B to C is so short, the phase
> difference is negligible, except that the sign is flipped.  So both
> those reflected waves essentially cancel, and the *net* reflected power
> is very low.
>
> This is exactly what happens when you have non-constant-impedance
> connectors in the line: something widely acknowledged to not cause SWR
> bumps, loss, etc. (yes, you can detect it with a time domain
> reflectometer or a swept frequency measurement, but it's still pretty
> darn small)
>
> I contend that it is EXACTLY the same if the BIRD or whatever has a 50
> ohm line inside, and you're operating it in a 75 ohm or 92 ohm or
> whatever system.  The meter will show forward and reflected readings,
> but they won't be numerically correct (because the meter is calibrated
> for a 50 ohm system), but the ratio is probably right..
>
>
>
> I know you can buy SWR meters made for 50 or 75 ohms but I
>> haven't seen anything for ladder lines (300 - 600 ohms). Maybe
>> because nobody is looking it. Who cares what SWR you have on a ladder
>> line, you always connect that to a tuner anyhow. Hans - N2JFS
>
> And I think the difference in 50 and 75 ohm meters is more in the dial 
> calibration than in the actual wiring inside.

If by "dial calibration" you mean meter scale factor, the I disagree. 
It's the bridge termination resistors and the nulling capacitors which 
change when going from a 50 ohm calibrated bridge to a 75 ohms 
calibrated bridge. Those components determine what load impedance will 
null the bridge. That is more than a scale factor. The meter scale 
calibration might change too (I'd have to think about that a little 
more), but the bridge will only null (read zero reflected power) when it 
see a load impedance at its output port equal to the load impedance for 
which it is calibrated.

73, Mike W4EF.....................