Bob,
I agree that consumer GPS readings may not be accurate enough to compensate
for terrain with lots of variation. Elevations on my steep hillside location
can vary considerably in just a few feet, and I've spent many an hour trying
to verify the tower locations against Tiger landmarks by interpolation.
Although I use a GPS unit with WAAS and averaging (I usually take at least
1000 samples), the accuracy is probably no better than 10 feet, which can
make for interesting differences in HFTA.
However, there's another possible source of error. Did you use the
"seamless" NED data or the straight USGS DEM data? A while back several of
us were looking at proposed tower locations at NT1Y and found that when we
overlaid the Tiger data the tower locations were considerably off from
visual observations. For example, NT1Y took a GPS reading on one side of a
road, but the location showed up on the other side of the road in MicroDEM.
The error was at least 30 feet, more in some cases. At first we questioned
the accuracy of the GPS readings, but further analysis by N6BV led to
speculation that the seamless NED data was not lining up with reality. Dean
thought that the smoothing algorithms used to produce the NED data might
lead to inaccuracies in certain areas, particularly with very irregular
terrain, which is the case at NT1Y.
I've not done an exhaustive study, but my recollection is that the NED data
didn't match my tower locations very well, either. For this reason, I use
the straight USGS data. The disadvantage is that you might have to combine
several files if the location is near a map segment boundary. This is not
hard to do in MicroDEM.
My feeling is that the inherent inaccuracy of consumer GPS and USGS data
mean that HFTA results should not be taken literally -- e.g., "If I move the
tower five feet I'm going to pick up 3dB!" I think the results are really
best thought of as relative. For example, I have two main tower locations on
my property, separated by 200 feet or so. There are fairly dramatic
differences between the two locations, attributable both to a significant
difference in elevation and the angle to hills between my QTH and Europe.
Moving the two towers around in HFTA can result in some different gain
readings, sometimes big ones, but the average relative performance is pretty
consistent and correlates very well with on-the-air observations of gain in
all directions.
73, Dick WC1M
> -----Original Message-----
> From: W3YY [mailto:w3yy@cox.net]
> Sent: Thursday, January 13, 2005 7:20 PM
> To: TowerTalk List
> Subject: [TowerTalk] RE: Downloading terrain data for use with HFTA
>
>
> To All -
>
> Thanks to Pete for posting this, as the process has, indeed, changed.
>
> I want to note a related observation. I have very detailed
> local topographic maps of my property and the surrounding
> area. These were obtained from my County.
>
> Using a GPS receiver I "located" my two towers, then
> downloaded the USGS data, and ran elevation profiles.
> Something seemed wrong. When I checked the USGS elevation at
> the base of each tower, the elevation figures didn't seem
> right. This conclusion was based on where I know the towers
> are located relative to my property lines and my physical
> topo maps. I suspect the errors in the GPS are causing the
> problem, i.e., the location as specified by the GPS receiver
> is not sufficiently accurate and might be off by 25ft or more.
>
> The topography on my property is quite irregular, so "moving"
> the towers even 25 or 50 ft can make a noticeable difference
> in elevation and elevation profiles.
>
> The bottom line - if you are on highly irregular property,
> I'm not sure I would trust results based on the USGS data.
> This is not because the the USGS data is bad, it's because
> you can't really determine the correct position of your
> towers, unless you have a military-grade GPS unit.
>
> Maybe there is another explanation. Other thoughts are
> welcome. I'd like to be able to confidently use the USGS
> data, if possible.
>
> 73, Bob - W3YY
> >
> >
> > ----- Original Message -----
> > From: "Pete Smith" <n4zr@contesting.com>
> > To: <towertalk@contesting.com>; <antennaware@contesting.com>
> > Sent: Wednesday, January 12, 2005 8:29 PM
> > Subject: [TowerTalk] Downloading terrain data for use with HFTA
> >
> >
> > > This is intended for anyone who has tried to download
> terrain data
> > > from USGS to use with HFTA, only to discover that several
> aspects of
> > > the downloading process have changed. N6BV has just posted an
> > > article that
> I
> > > wrote on the ARRL web site,
> > >
> > > http://www.arrl.org/notes/9043/HFTA-Changes.pdf
> > >
> > > which gives the new procedure with screen shots and
> active links.
> > > Both Dean and Peter Guth, the author of MicroDEM, were a
> huge help
> > > in pulling this together and making things work again.
> > >
> > > 73, Pete N4ZR
> > > The World HF Contest Station Database
> > > was updated 13 December 2004
> > > 2843 contest stations at
> > > www.pvrc.org/WCSD/WCSDsearch.htm
> > >
> > > _______________________________________________
> > >
> > > See: http://www.mscomputer.com for "Self Supporting Towers",
> > > "Wireless
> > Weather Stations", and lot's more. Call Toll Free, 1-800-333-9041
> > with
> any
> > questions and ask for Sherman, W2FLA.
> > >
> > > _______________________________________________
> > > TowerTalk mailing list
> > > TowerTalk@contesting.com
> > > http://lists.contesting.com/mailman/listinfo/towertalk
> > >
> >
>
>
>
>
_______________________________________________
See: http://www.mscomputer.com for "Self Supporting Towers", "Wireless Weather
Stations", and lot's more. Call Toll Free, 1-800-333-9041 with any questions
and ask for Sherman, W2FLA.
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|