Towertalk
[Top] [All Lists]

[TowerTalk] 50:75 ohm transformation

To: <towertalk@contesting.com>
Subject: [TowerTalk] 50:75 ohm transformation
From: w8ji@contesting.com (Tom Rauch)
Date: Thu, 11 Feb 1999 10:43:24 +0000
To: <towertalk@contesting.com>
> Date:          Wed, 10 Feb 1999 21:28:51 -0500

At great risk we tread on the edge of the thin conjugate match ice! 

> > I use 3/4 inch CATV and a run of about 180 feet to my
> > tri-bander.  The worst case I have is 2.2 to 1 on 15 cw.  I 
> > look at it as 2-2.5 to 1 is negligible on HF, whats it
> > going to do, reduce your delivered to the antenna power a couple
> > of watts?  The other end will never know..
> 
> Um, well...a 2:1 SWR will eat up about 10% of your power, so 2.2:1
> will be marginally worse.

14% is the "% reflected power" at 2.2 : 1 SWR. But those numbers are 
meaningless in our applications. Let me explain why....
 
My FT-1000 runs 200 watts out, into a 2:1 SWR it folds back to 50 
watts of actual forward power (power forward minus power 
reflected) delivered to the line. That means I have at least 6 dB 
power reduction without a tuner from a paltry 2:1 SWR.

I also have a solid state rig that puts out MORE true power into a 
1.4 : 1 SWR of the correct impedance (it does fold back with even a 
very slight mismatch in the WRONG impedance direction). That isn't 
unusual because rigs are seldom exactly 50 ohms source Zo unless they 
are adjusted for maximum power transfer or efficiency with a 50 ohm 
load. Most solid state SSB rigs are optimized into loads higher than 
50 ohms. That's to prevent flat-topping when the load is mismatched, 
since a high impedance load always causes flat topping to appear 
sooner than a low impedance direction mismatch.

If I run the tuner in the FT-1000, the power reduction is 
eliminated and the only loss, other than the loss in the tuner, is a 
very small additional loss in the transmission line (assuming the 
line is a fairly low-loss line).  

An amplifier with an adjustable network that can match the line's 
impedance will not suffer the mismatch power reduction of a untuned 
PA stage, so in that case the 14% figure is totally meaningless. The 
only SWR caused power loss would be the small additional loss in the 
transmission line primarily caused by increased current in sections 
of the line, but that loss would have absolutely nothing to do with 
the SWR measured with a 50 ohm instrument connected to a 75 ohm line.

A high SWR does not always increase system loss. In a typical 
pi-network PA loss in the tank decreases with higher impedance 
loads, and increases with lower impedance loads! It's a very small 
change, but it's there.

Also, loss on a fractional wavelength transmission line can be 
reduced by operation with a high SWR, if the impedance movement is in 
a direction that reduces line current integrated over the entire 
length of the line. That's because line loss is primarily 
due to conductor (I^2R) loss at the frequencies where we operate and 
with the lines we use. Once the line is 1/4 wl long, we can't reduce 
current because any increase in SWR always results in an increase in 
current somewhere on the line, so for longer lines an increase in 
SWR always causes an increase in line loss. 

The true SWR on the 75 ohm line, measured with a 50 ohm bridge as 2.2 
to one, could be as high as 3.3 to one  in a lossless line and higher 
when line loss is counted!

We can not accurately measure the 75 ohm line SWR with a 50 ohm 
bridge inserted at one place on the line. The 50 ohm bridge measures 
the SWR normalized to 50 ohms, not normalized to 75 ohms as it 
should be measured on a 75 ohm line!  A 2.2 : 1 reading only 
tells us that the line presents an impedance somewhere between 22.7 
(50/2.2) and 110 ohms (50*2.2) at point where the 50 ohm-design-
meter is inserted. That means the worse case impedance presented to 
the 75 ohm line could be more than 75/22.7 ohms, or 3.3 to one!!!

When  things are measured with the wrong equipment, and the 
incorrect data  is used to estimate the results of the incorrect 
measurements, and we then misunderstand what some fancy words 
really describe, the final conclusions get pretty far from the truth.

What does this all mean??? It means we often have no idea what 
the heck is actually going on in our systems when we mix in odd cable 
impedances. We measure the 75 ohm cable's SWR with the wrong 
impedance instrument at the wrong point in the line, and have little 
or no idea how our equipment ore the cable behaves with the higher 
than normal SWR. 

Fortunately if we are happy, that's all that really matters. As with 
the best antenna, the best SWR is the SWR that makes you 
happy.

73, Tom W8JI
w8ji@contesting.com

--
FAQ on WWW:               http://www.contesting.com/towertalkfaq.html
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com
Search:                   http://www.contesting.com/km9p/search.htm


<Prev in Thread] Current Thread [Next in Thread>