[TowerTalk] Modelling SteppIR Elements

Thu Apr 9 13:09:13 EDT 2015

I went through a rigorous attempt at modeling the four element Steppir 
on all its bands several years ago.  My results did not differ 
substantially from those who used a less rigorous approach.  I have all 
of our results tabulated, if you're interested in the data.  Here's how 
I described my approach to the issue of wire diameter and resistivity.  
(BTW, W9CF teaches physics or EE at ASU. and you can get to the 
equivalent diameter calculator here: 
http://fermi.la.asu.edu/w9cf/equiv/index.html).

Wire diameter

The Steppir elements are made from a rectangular perforated Cu-Be alloy 
tape that is 0.56" wide and 0.018" thick.  The perforations are 0.125" 
in diameter and there are five perforations/inch.  If one calculates the 
surface area of the tape, accounting for the perforations, and then 
calculates the equivalent cylindrical wire diameter, a value of 0.33" is 
obtained.  (Based on postings on the Steppir reflector, this is the 
diameter that Steppir recommends for modeling purposes.)  However, if 
one obtains an equivalent diameter based on self-impedance using W9CF's 
on-line calculator, a value of 0.30" is obtained.  Although the 
difference between the two calculation methods is not major, I used the 
0.30" diameter in my models.  (Since the calculator didn't account for 
the perforations in the Steppir tape, I asked W9CF about their effect.  
He said that the impact on equivalent wire diameter would be small and 
would tend to reduce it.)

Wire resistivity

It appears that the standard Cu-Be alloy used in electronic applications 
is hardened 2% beryllium, and I am assuming this is the material used in 
the Steppir tape.  The conductivity for this alloy is reported as 22% 
ICAS or a resistivity of 4.55 times that of copper (1.74E-8).

The W9CF calculator also provided an equivalent cylindrical diameter for 
a rectangular conductor based on resistivity and the result was 0.20".  
I asked W9CF why this value differed so much from the equivalent surface 
area calculation (0.33") and he explained that for a rectangular 
conductor most of the current is confined to the edges.  Since 
resistance per unit length is inversely proportional to cross-sectional 
area, I accounted for the effect of the resistivity based wire diameter 
by multiplying the Cu-Be resistivity by the ratio of the square of the 
diameters.  Thus, I used a resistivity of 4.55 X 1.74E-8 X (0.32/0.22) = 
1.78E-7.  As you will see, the difference between this calculated value 
and the resistivity of copper only reduced gain by about 0.2 dB.

73, Joe
K2XX

> Jim Lux <mailto:jimlux at earthlink.net>
> Wednesday, April 08, 2015 11:54 PM
> On 4/8/15 1:42 PM, Richard (Rick) Karlquist wrote:
>
>
> In general, you can use a round element with circumference equal to 
> the width of the tape (or half the width of the tape.. opinions vary).
>
> I just did a bunch of models of a "tape" dipole at work where we were 
> comparing the results from HFSS (modeling a 1cm wide tape) and NEC2 
> (modeling it as a tube with radius 1mm.. 0.628 cm circumference) and 
> they were "pretty close".
>
> The variations from the dielectric properties of the tube will have a 
> bigger effect.
>
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>
>
> Richard (Rick) Karlquist <mailto:richard at karlquist.com>
> Wednesday, April 08, 2015 4:42 PM
> The fiberglass tubes lower the resonant frequency by an
> uncontrolled amount, since they are just fishing poles
> that do not have any exact permittivity value.  It is
> analogous to FR-4 PC boards where the impedance is
> to some extent an accident of the characteristics of
> the epoxy glass.  SteppIR has sourced the tubes from
> various vendors over the years which further complicates
> the issue.  Any correction factor you would use would
> be specific to a particular run of tubes.  SteppIR
> seems to gloss over this issue.  There is a parameter
> you can set on the controller that scales all elements
> by a fudge factor that can be determined experimentally
> for the particular copy of the antenna that is being used.
>
> Another issue is: what size round element is equivalent
> to the copper tape?
>
> Rick N6Rk
>
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>
>
> William Osborne <mailto:wosborne44 at gmail.com>
> Wednesday, April 08, 2015 4:05 PM
> Does anyone have experience modelling an antenna in EZNEC and then loading
> those elements into a SteppIR antenna (I am using 4E so that is of most
> interest). If you have, how does your antenna perform with your new 
> element
> dimensions versus the factory default ones? Mine does not seem to resonate
> at the same frequency as the model, so I think a correction factor is
> required to the EZNEC model elements?
>
> Bill--K5ZQ
>
> ***********
>
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>
>