[TowerTalk] measuring coax loss with an MFJ-259B

[Pete Smith]

I have just been trying to get a baseline loss measurement on my two feedlines from shack to tower, using the MFJ-259B in Coax loss mode.

According to the manual, the unit can be set to any frequency within its range, and will read out a direct return loss number at that frequency, if the coax is open at the other end.

In practice, I find that around 28 MHz, for instance, the displayed loss value for one feedline can be anywhere from .9 to 2.1 dB, depending on the frequency.  The displayed SWR on the unit is at a minimum when the loss figure is maximum, and vice versa.

I think I may know why, if not how, this is happening.  My feedlines both utilize long runs of 75-ohm CATV hardline, cut to a half-wave multiple at 1750 KHz, and measured at 28 MHz.  On either end of the 75-ohm runs are runs of 50-ohm RG-213 to the antennas and into the shack.  My theory in doing things this way is that the 75-ohm will reflect the input impedance at the output, and as a practical matter it seems to work rather well.  But my hunch is that particularly at 28 MHz, relatively small frequency excursions can result in the 75-ohm section translating to quite another impedance than the 50 ohms at the input end, and that this in turn results in the changing "loss" numbers.

Do I have this right?  If so, is any of the numbers I'm seeing an accurate reflection of the loss in the total composite feedline? 

73, Pete N4ZR
The World HF Contest Station Database
was updated on 23 October 2005
Over 3000 contest stations at 
www.pvrc.org/WCSD/WCSDsearch.htm