[TowerTalk] USING MFJ-259B TO DETERMINE LENGTHS OF COAX

[jimlux]

On 10/28/17 5:36 PM, Greg Best via TowerTalk wrote:
> I have several lengths of RG8 (and other types) coax that are more than 25 years old and i am trying to sort good stuff from stuff to throw away.  Most of them are over 100 feet. I have used my MFJ-259B using the distance-to-fault feature to try to determine the length.  I tested one coax that was 30.5 feet exactly to see if I can get reliable figures for the length. The MFJ-259 indicates it is only 29.4 feet after multiplying the displayed length times the appropriate velocity factor.

that's a 3% error - the velocity factor could be that far off (0.68 
instead of 0.66, for instance).  It's only "sort" of well controlled - 
the thing the manufacturer controls is the impedance.





  In this case, it was RG-142 with teflon dielectric.  The only 
potential I see for any error is how the coax is terminated.  The manual 
says to just leave the coax unterminated on the other end when the 
measurement is made.  But it also says, that more accurate measurements 
might be obtained if it is terminated in a resistive bad match as 
opposed to a reactive bad match.

That's more important at higher frequencies where the fields from an 
open connector "add" to the apparent length of the coax. But it's easy 
to solder up a "short" and try it.




> Does anyone have experience with this and found a reliable method?
> On the good side I have checked some of the coax lengths and found that the loss measurement feature of the MFJ-259B tracks within 0.1 dB of the measurement made with a professional quality tracking generator and spectrum analyzer.

That's basically a "reflected power" vs "forward power" measurement and 
depends on the quality of the bridge.  IN a "unterminated" case, most of 
the power is reflected, so you're comparing the ratio of two signals at 
the "high accuracy" end of the measurement.

It's measuring the quality of a matched load that's hard (because you're 
trying to distinguish between -20 and -21 dB reflection, for instance).





> GregN9GB
> _______________________________________________
> 
> 
> 
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>