[Top] [All Lists]

[TowerTalk] Modeling

To: <>
Subject: [TowerTalk] Modeling
From: (Robert Shohet)
Date: Sun, 3 Feb 2002 11:20:30 -0500
Hi Guys,

It seems we have two "sides" to this issue.  Those who believe in the value
of modeling and those who are skeptical.

Among those who are "believers", some believe that they should always use
the model, others believe that they should rely on it but treat the results
with some skepticism.

Among those who are skeptical - some seem to argue that we should not use
models at all, and others would agree that they are of some value but are
not very useful because they can never be 100% accurate.

Ok, so we have four main categories with a lot of additional shades of gray.

Let me add this to the debate:

The value of using modeling software is directly proportional to how complex
the terrain is and how many antennas will be used.

This may seem counterintuitive at first, but carefully consider that there
is no way that any human being can predict "by eye" what the optimum heights
will be over complex terrain for a given arrival angle.  This alone makes
good modeling software invaluable.

Then also realize that the more antennas one contemplates putting up on one
band, the more proof there will be that modeling works or doesn't work.  The
whole point of using models to predict the best wave angles/optimum heights
is so that as many wave angles as possible can be covered with the antennas
by themselves and/or in stacks to the various parts of the world.

Again, it is NOT possible to do this by guessing.  Some people might achieve
good guesses, others not.  Some guessed heights will work better over some
types of terrain, others not.

I can not give you mathematical or physical proofs one way or the other, but
I can tell you this from my experiences of building a station with stacks on
several bands over VERY complex and uneven terrain here in Southwestern CT:


Modeling as to the optimum spacing and antenna elements/spacing is VERY
accurate for predicting patterns and front-to-back, especially when you are
a perfectionist as far as building antennas EXACTLY to the proper specs with
the proper materials.

Modeling as to the best heights for antennas for a particular wave angle for
a phased array is INCREDIBLY valuable!

How do I know this?

Simple - when I was rebuilding my station last Summer/Fall from devastating
multiple lightning hits, I spent several WEEKS modeling my antennas over my
terrain with all possible wave angle combinations on 10, 15, 20 & 40 meters
for EU, JA and Oceania.

Prior to doing this I had operated some 30 contests and made over 100,000+
qso's from this location with 2 towers and stacks over the past 3 years.  I
had a darned good idea of what bands and directions I was loud in and when,
which phasing combos worked well and which did not and under what condx.

Based on the results on of the modeling, I changed my antenna heights on 10
from 25/45/70/101 to 23/38/66/101, and on 20 from 42/84/131 to 58/82/131.

I kept 15 the same as it agreed almost perfectly with what the models

The differential on 10 was AMAZING!  Where before the phased antenna combos
showed little difference, the new heights now provided dramatic differences
in signal strengths to EU as well as to JA.  In fact, as the model
predicted, the 38' antenna to JA was often superior to the one at 101'!

On 20 I went from being weak in the afternoon and evening, with the lower
stacks not being very good, to being LOUD in the afternoon and evening with
dramatically better and more consistent EU runs.

Perhaps the most interesting difference was on 40.  My lower 40-2cd was at
55' before, which seemed like a good height to phase it with my 40-2cd @

Unfortunately it performed more like a dummy load.  By raising it to 66', it
is now an effective antenna and there is some stacking gain with the top
antenna at 123', although you would not think that 57' foot separation would
be very good on 40.

The point is this:  MODELING WORKS!  and regardless of how well someone can
guess at the "best heights" for their location, no one can guess as well as
a good model can predict.  The more complex the terrain, the more
counterintuitive the "right" heights will be.  The more antennas you use,
the more valuable the model will be in helping you sort out the optimum and
counter intuitive stacking distances.

Astute and experienced contesters are likely to notice small differences the
most because they need to be loud in the high volume (fairly high angle)
areas as well as the low angle areas. DXers will benefit too, but will
likely see less differences because they are not focusing on volume but
probably more low-angle paths, where lower antennas are generally not as

Casual guys will benefit enormously too, since it will help them plan their
setup and antenna types and heights BEFORE they actually spend the money.

This may be the greatest value of all!

As a practical businessperson, I can tell you the value of time not-wasted
and frustration not experienced makes any of the software programs an
exceptionally good value.

One of these programs for $100+ bucks learned and used well, may ultimately
provide more value and signal strength than spending an extra $1000 for a
super-duper tribander or $2000 for an amplifier.  Not to mention that a
well-modeled setup will likely use SMALLER antennas with less elements and
thereby have much less wind-loading and ice-loading and therefore better
storm survivability.  I used a good modeling program to shorten my 20-4cd's
from 31.5' boomlength to 25.5' boomlength because of the extreme winds and
frequent ice-storms at my qth.  Something to think about!

I'm a believer in modeling.  With all its warts, the programs are enormously
helpful, efficient, fairly easy to learn to use and very gratifying when you
discover things about your system BEFORE you go to the time, effort and
expense of putting them up, or spending days in the tower experimenting.

73 and good modeling!

Bob KQ2M

Where do you get ICE bandpass filters & beverage matching boxes?  The
same place that pays for the hosting of this list:  The eHam Store.
Order online at

FAQ on WWW:     
Administrative requests:

<Prev in Thread] Current Thread [Next in Thread>