[TowerTalk] Coax Loss -- RG-Numbers Don't Tell us Much

[jimlux]

Richard (Rick) Karlquist wrote:
> Jim Brown wrote:
> 
>> Not as ideal as you might think. The problem is that most coax is 
>> shipped on reels with access to only one end. To make a transmission 
>> loss measurement, you must unreel it to get a connector on both 
>> ends. 
> 
> No, see below
> 
>> Now, if you happen to own instrumentation that allows you to measure 
>> that loss on a 1,000 ft spool from only one end, that's another 
>> story. :) 
> 
> See below
> 
>> 73,
>>
>> Jim K9YC
> 
> I happen to own instrumentation that allows me to measure loss from
> one end; it's called an MFJ-259B.  Cable loss is a built in function.
> As long as the far end is open, cable loss is half of return loss.
> You can also calculate loss from VSWR.  For example, 3:1 means your
> loss is 3 dB.
> 

And a fancier widget that does some computation and actually does vector 
detection can do even better (e.g. measure the length of the cable).

The question I would have is how much precision is needed?  If you're 
measuring 1000 foot spools, then a 1 dB measurement precision will give 
you 0.1 dB/100 ft kinds of precision.  Measuring 1dB out of, say, 10-20 
dB seems doable.