[TowerTalk] Method of calculating phase delay variation when stacking two dissimilar yagi antennas?

Jim Lux jimlux at earthlink.net
Sun Sep 1 13:22:23 EDT 2013


On 9/1/13 10:06 AM, Mark Beckwith wrote:
>> So my question at this moment is - which is the
>> proper method for me to calculate the proper
>> delay line length to properly phase the two
>> dissimilar SteppIR yagis?


The calculations are decidedly non trivial.  The two antennas will 
interact, small changes in the lengths of the elements (SteppIR, after 
all) or their relative orientation (wind) will change the feed point 
impedances, particularly the reactive component.  That won't change the 
"match" (in terms of VSWR) but will change the relative phasing.

>
>
> This is not meant to be a snarky or otherwise "cute" response.  Just
> more like a reality check.  The answer is:
>
> Feed them with equal length feedlines and see how they play.  If you
> like the way they work, then you're done.  I suspect you will like the
> way that they work and I also suspect changing one feedline length by as
> much as 8 feet would not affect your overall conclusion.
<snip>
> Get on the air and work guys. I bet you work more guys with two antennas
> at you disposal than just one antenna, and I bet when you do
> top-bottom-bip-bop* tests you will find some times when some stations
> are lounder under some of those four conditions.  I suspect this would
> be the same conclusion whether your coaxes are the same length or
> different lengths by 4 feet or 8 feet.
>

Exactly this.  You can have a few jumpers of some shortish length (1/4 
wavelength ish) and try substituting them in and see what happens.

Here's the deal..
For phased arrays  you can have fairly large phasing errors and the 
*gain* doesn't change much.  What changes is the depth and position of 
the nulls.  So the whole deal on the BIPBOPTOPBOT thing is that you're 
moving the nulls around to suppress the stations you don't want to hear, 
while slightly changing the gain in the directions of the stations you 
DO want to hear. (for that matter, most likely, what you're doing is 
moving the null OFF of the desired station).


So, you could model and measure and carefully calibrate your system (you 
*are* allowing for the thermal expansion effects, aren't you <grin>), 
and maybe not have any better effect than you would by essentially doing 
the "try it and see".






For what it's worth, having two SteppIRs makes for a very interesting 
scheme.  You could feed with fixed lines, and just bump the driven 
element a bit shorter or longer to shift the phasing.   I don't recall 
the exact numbers, but I seem to recall that a 1% change in length off 
resonance is about 5 degrees in phase. 10% is about 30 degrees.

There will be all sorts of interactions among the elements, so you 
really need to model it, but it's an interesting approach.


More information about the TowerTalk mailing list