[TowerTalk] ground and conductivity

[Jim Lux]

At 02:14 PM 2/28/2005, Tom Rauch wrote:
> > A Canadian gov't website claims that the conductivity of
>the Great Lakes
> > water feeding into the St. Lawrence is about 300 uS/cm
>(=30 mS/m).  I grant
> > that this is at the downstream end of all the cities on
>all the lakes.
> > Milwaukee's water department reports that their intake
>water (from Lake
> > Michigan?) is around 200 uS/cm (20 mS/m)
>
>I only know what we measured and extrapolated to FCC
>conductivity Jim using stations at the high end of the AM
>BCB.
>
>We did that by measuring the slope of attenuation with
>distance.
>
>I care less what someone measures at DC.


ACtually, water conductivity measurements are made with AC (to avoid 
polarization effects).  60 Hz or 1kHz would be common frequencies.

I doubt that the conductivity is that frequency dependent, although I don't 
know enough about the ionic mobility in solution to say for sure.

I do note that the standard conductivity measurements for Seawater (i.e. 
4500 mS/m) are pretty close to those accepted values for RF (5000 mS/m).

As far as measurement technique goes.. It seems that a direct measurement 
in a test cell with calibrated equipment might be a better measure of 
conductivity than an extrapolation from RF field strength measurements, 
which can be affected by other factors (like the quality of the ground on 
the "land" side of the antennas and probes).

On the other hand, if the goal were to measure and predict received field 
strengths, then what it happens to extrapolate to is less important, as 
long as the measured data fits the model.

Much like adjusting the ground parameters in a NEC model to match observed 
impedance and/or pattern data.  The adjusted parameters may or may not 
reflect what the actual soil properties are, but with the adjusted 
properties set, the radiated field predicted by the model should be "right".


>  I would like to
>measure the attenuation at HF, but a 1.5MHz large area
>sample is certainly closer than a 60Hz sample.
>
>73 Tom