[TowerTalk] 2 element 40m steppir yagi

[Jim Lux]

At 10:39 AM 10/7/2004 -0700, Michael Tope wrote:
>What I was thinking about for my small suburban lot was a
>big IR driven element attached to a mast pipe in concert
>with a single regular IR parasitic element cantilevered on a
>short boom off the mast. This would give 2 element yagi
>performance on 13.8 to 54 MHz and rotary dipole performance
>on 6.9 to 13.8 MHz, and still be reasonably "neighbor friendly".
>The big IR driven element could be grounded to the mast with
>a relay and used as tuneable top loading for a 160 meter
>shunt-fed tower. For eighty meters I would probably just use
>an inverted-vee.
>
>As far as 80 meter yagi's go, I think the hot ticket would be
>a center loaded element with small_IR element halves installed
>beyond the center loading to serve as a continuously adjustable
>elements tip. The IR element tips could then be adjusted from
>the ground for optimized yagi performance from 3.5 to 4.0 MHz.
>Getting the control wiring past the inductor might be a
>challenge (you might be able to put it inside the coil tubing).
>
>73 de Mike, W4EF......................................

I'm not sure that some sort of adjustable loading network in the middle of 
a fixed element might not work just as well.  When all is said and done, 
what changing the tip lengths would do is just change the reactance of the 
element, and a single motor driving a vacuum cap or low loss inductor at 
the center of the element might work just as well.  Folks have done this 
sort of thing with 2 states and relays as the actuator (say, to reverse a 
beam, or to switch between CW and phone subbands on 80/75).

I've done a bit of modeling of fixed length elements with variable 
reactances in a 3 element yagi scheme, and it looks like it wouldn't have 
performance that is markedly different from the SteppIR, assuming that you 
can manage the losses.  That is, all my modeling has dealt with lossless 
networks.  AND, this is most important, it's really only viable for single 
band designs. (like your proposed 80 meter mini-behemoth)

One minor complexity (although one that is not hard to manage) is that most 
of the Yagi design programs don't address the concept of a variable loading 
reactance in the elemnet.  For my purposes, I just used a general purpose 
optimizer in 4nec2 and let it grind to find the loading X for each 
element.  One would also need to do some sort of sensitivity analysis to 
see how critical the tuning of those reactances is, but, to a first order, 
I can't imagine its any better or worse than the sensitivity to the lengths 
of the elements in a SteppIR. It's not going to change the gain much, but 
the F/B will be radically affected.

Where the SteppIR concept has real value is when you need to accomodate BIG 
frequency changes (like band to band octave steps).  Trying to do this with 
a variable network and a fixed length element makes for many compromises. 
Either the element is much too short, and losses in the network eat you 
alive; or, the element is too long, and likewise.

Sure, one could conceive of a scheme where you have elements with switches 
along the element to add and remove segments at the end, and then a 
variable tuning network at the feedpoint to do the fine control. Such 
schemes have even been described in the professional literature at least as 
far back as the 60s.  The problem is the complexity. You'd have many 
actuators to worry about, instead of just one, as in the SteppIR.