[TowerTalk] Measuring resonance of a yagi element

[jimlux]

On 4/22/20 9:34 AM, Jim Brown wrote:
> On 4/22/2020 7:51 AM, Robert Harmon wrote:
>> Good luck on getting the files.   They won't give you theiir Eznec files.
> 
> Who is the "they" in this post? I looked all through this post and 
> quoted stuff to find "them." If you're referring to mfrs, what makes you 
> think they use NEC for design? There are multiple modeling programs.

True enough.. I wouldn't think that HF antenna designers are doing much 
with HFSS or CST studio and that type of modeler, though. It's not clear 
that FDTD or modal analsyis buys you much for a Yagi.  But MMANA and 
FEKO are popular.

MMANA is also a method of moments (MoM) code, using the MiniNec engine 
so should produce results almost exactly the same as NEC for 'easy' 
models (no dielectrics, no funky segementation).

FEKO does MoM for this kind of antenna - it has nifty stuff for more 
complex antennas that makes the model more computationally tractable.

ADS Momentum has a MoM modeler too.

Underneath it all, they use the same modeling techniques- they solve the 
Electric Field Integral Equation numerically - the big difference is in 
handling "special cases" and in things like auto-segmentation and 
meshing.  And, fancier packages handle the transition from "wires and 
grids" to "3D objects" better.


> 
> It's easy enough to produce an EZNEC file for an antenna that doesn't 
> include inductors or capacitors -- all you need a a measuring tape and a 
> caliper to determine element diameters. In his very useful 
> self-published "Array of Light," N6BT includes detailed descriptions of 
> many  Force 12 antennas he designed for the company he owned.

For the kind of "does it make a difference" sensitivity analysis, a 
pretty simple model with lengths and diameters would work well.

However, if you're trying to match measurement to model, it might become 
a grad student thesis project.

It kind of depends on how closely you want to model. Do you taper the 
elements in the modeling code? or do it manually? Or apply a "taper 
adjustment"? What about modeling the element/boom attachment? I don't 
think I've seen anyone model U-bolts with a meshed grid, but you could.