Towertalk
[Top] [All Lists]

[TowerTalk] Measuring 1/2 Wavelength coaxial lines with an MFJ-259

To: <towertalk@contesting.com>
Subject: [TowerTalk] Measuring 1/2 Wavelength coaxial lines with an MFJ-259
From: W8JI@contesting.com (Tom Rauch)
Date: Mon, 24 May 1999 10:37:23 -0400
Date sent:              Sun, 23 May 1999 20:29:09 -0700
From:                   "Robert L. Shohet" <kq2m@mags.net>
Send reply to:          kq2m@mags.net
To:                     towertalk@contesting.com
Subject:                [TowerTalk] Measuring 1/2 Wavelength coaxial lines with 
an MFJ-259

> Here's what happens...
> 
> I measure an odd multiple of 1/2 wavelength for let's say 21.250.
> 
> At the 3rd harmonic, instead of seeing an swr "dip" at 63.75, I might see
> the dip at 64.25 instead (3rd harmonic of 21.417).  At the 5th harmonic
> instead of seeing the "dip" at 106.25 Mhz, it might dip at 107.30. (5th
> harmonic of 21.460)

What are you using for a dummy load and connectors? Why are 
you using the SWR dip?

> Why does this happen?  Shouldn't the 3rd and 5th harmonics be exactly 3x
> and 5x the original frequency?

Not if there are impedance bumps, or reactances in the system. If 
there are, then the feedline has to add compensating reactance to 
null SWR. 

What test setup are you using, and why are you trying to use high 
order harmonics like the 3rd and fifth in the process???  Measuring 
in the VHF range will actually decrease accuracy if you have 
anything that mixes Zo's in the test setup. One source of mixed 
impedances is at the connector of the 259. The Zo of the SO-239 is 
about 38 ohms. That single connector won't have any major effect 
until you get up around 150 MHz, but how many other connectors 
and lines that are not 50 ohms are in your test setup? What about 
the load you are using?
      
> More importantly, which freq. do I use as the correct measurement for the
> 1/2 wave coaxial line, the 1st, 3rd or 5th harmonic or some combination?

I'd never use SWR. Using a measurement method with a broad-
nosed peak that depends many other variables to be accurate to 
center something is bad form.

I'd tune for the lowest Z with the line shorted at the far end at the 
desired frequency or the lowest Z with the line open at HALF the 
desired frequency if I wanted a half wave line at some frequency.

Where did you get that measurement protocol?


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>