Home: https://ky8d.net/spiral
Series:
[1]
[2]
[3]
[4]
[5]
Iteration Set #5: Having decided upon an OD of lambda/2 for 16.1MHz (CLI arg --f_lo 16.1) during Iteration Set #4, that having been performed upon spiral of high resolution (CLI arg --div 32), now I'm needing to simplify that geometry to something reasonable for physical construction. That is to say, settle upon a spiral aliased more coarsly. One with fewer radii, thus vewer and longer flat sides per full turn. So here I iterate once again, keeping all the same CLI args args as before except '--div', which I'll allow a full range of values: 8, 10, 12, etc..
Skipping over 4 and 6, as never do they seem to perform well at all. Particularly not when the spiral is planar. And going past 16, all the way up to 32, so as to be well informed on the concept.
Color Key for Paragraph Text: White (greater contrast) presents information crucial to interpretation. Red presents a delving into various fiddly details about the surounding main text in white. Stuff which I feel some readers may find self-evident, uninteresting, or even beside-the-point. Tidbits I couldn't convince myself to simply delete.
Obviously, I wouldn't be building a spiral support web having 32 radii. But since previously I'd chosen div = 32 as a baseline for iterations on f_lo, and from that chose f_lo = 16.1 as providing best dBI and Elev angles on the two WARC bands, here I include div = 8 to 32 as my final narrowing down.
Overshoot: You will note that, even though the spiral approximations progress through their devisions by linear steps, the wire lengths and wire-end positions seem a bit random. This is owing to the antenna-generating program enforcing a rule that spiral ends must terminate at a whole-division radius. Combining this rule with the target OD occurring between said radii, spiral ends nearly always require extension.
Well, it looks as if I didn't get much by going smoother. May as well just stay with 8 as the number of spokes. I am tempted at this point to try something not supported by log-spiral theory. You will have noted that none of my spirals start at the center. Instead of that, all of my log-spiral antennas tested so far have begun their cuves at a certain inside diameter. And that always it has been the same.
The particular inside diameter which I've been using is lamdba/2 for 50MHz. This as opposed to log-spiral theory, which suggests that said inside diameter need not proceed any closer to center than lambda/2 for the highest frequency used: in this case 30MHz. But I had tried a couple like that, and they handn't performed all that well. Thus did I quickly tire of trying out random designs, and began this series of iterations. Starting out, I had to pick something, and so went with a dimension I had found trustworthy on designs for log-periodic arrays.
Having got to this point, I think I shall now revisit that question.
And so, that's what I'll try next: Re-Iterating on ID.