[Top] [All Lists]

[TowerTalk] Reflectors and directors

To: <>
Subject: [TowerTalk] Reflectors and directors
From: (Tom Rauch)
Date: Fri, 2 Feb 2001 05:32:39 -0500
> At 08:46 AM 2/1/01 -0800, you wrote:
> snip.
> >
> >
> >I have long wondered why the reflector on a yagi is longer than the
> >driven element and why the directors are shorter.  Also why do
> >multiple directors increase the gain but multiple reflectors do not?
> This is incorrect when speaking of multi (high density) elements
> or optimum elements per array wavelength.
>  When you have multiple directors then multiple reflectors
>    DO  assist the gain as well as INCREASE the front to rear/back
> ..... especially when TOTAL BAND SPREAD is considered.

Multiple reflectors ONLY noticeably increase gain or F/B when they 
are offset from the main axis of the array, or when the original 
element is NOT optimized. 

As a side note  reflector does not reflect, and a director does not 
direct. Both re-radiate signals, just as if they are fed with a 
feedline. They ALL increase gain, or change F/B, by their re-
radiation cancelling signals in unwanted directions. 

If they do not re-radiate an appreciable amount of energy compared 
to the energy in the direction you are trying to null out, they do not 
help remove unwanted energy from needless directions.

If they do not reradiate a substantial amount of energy compared to 
the overall energy radiated by the array, they can not greatly 
change gain.

They are all like dipoles, excited by the fields impinging on them.

A "reflector" causes a null away from the exciting source because 
it is self-resonant or nearly self-resonant at the operating 
frequency. When a system is totally non-resonant (like a very large 
antenna above a small one would appear, or the ground below the 
antenna) or when it is exactly self-resonant or has lagging phase in 
the re-radiation, it nulls the signal away from the exciting source.

It does this because reradiation is 180 degrees out-of-phase with 
the source when the structure or element being excited is exactly 
resonant or virtually non-resonant. Think of what happens when an 
additional element excites the reflector. The signal "arrives" and 
causes a current flow in the element that is 180 degrees (if it is self-
resonant or very large and non-resonant) out-of-phase with the 
exciting signal. 

Looking at the case of a signal from a driven element .25 wl away, 
the signal is delayed 90 degrees in space. It is inverted 180 
degrees in the element. Energy is reradiated with a 270 degree 
phase delay from the driven element.  270 degree lag is the same 
as a 90 degree lead, so you have the equal of a two element driven 
array with 90 degree lag in the driven element. Since phased arrays 
with close spacing null in the  direction of leading phase, the 
energy leaving the array to the rear is cancelled.

Since the energy has nowhere to go but the other directions, it 
goes that way. The array fires, like any two element close spaced 
array, towards the lagging phase element...the driven element.

Once you have a reflector that is properly tuned to null in the plane 
of the reflector, or if you had a series of directors tuned to provide a 
perfect null in the direction of the reflector, so much energy is 
removed there is little current in the reflector. It does nothing, or 
next to nothing to overall pattern.

Directors, because they are significantly shorter than a 1/2 wl, have 
a positive phase shift. They look capacitive, and so have leading 
phase in the currents. The phase lead internally added by the 
element is up to 90 degrees for a very short element, but 
unfortunately current diminishes rapidly as the element is moved 
from resonance by being made shorter.

In this case the spatial phase delay is offset by the leading phase 
in the element's internal currents. The effective phase- shift with a 
short director spaced 90 degrees from the exciting source 
normalized to that source is somewhat less than -90+ 90 + -180, 
or less than -180 degrees. The radiation leaving that element 
follows the rule-of-thumb given in the reflector example above, and 
cancels the most signal in a direction towards element(s) having a 
leading phase. The radiation that would have gone that direction 
has nowhere to go except in directions where it is not canceled, 
and so now it fires in the direction of lagging element phase.

This is towards the direction of the director.

So you see, it is easy as pie and quite simple. Reflectors do not 
reflect, and directors do not direct. They both re-radiate and that re-
radiation removes energy from some directions. Since nothing 
much is dissipated as heat, the energy has to go in the direction 
where it is LEAST cancelled.

This is also why sharp pattern antennas of a given physical 
construction, especially small ones with close spacing, have higher 
element currents than broad pattern antennas. The array "fights 
itself" in every direction, and so current has to increase 
substantially in order to radiate the same power!

Remember behind all of this, it is the charge acceleration in the 
elements that makes things happen. If we impede radiation in 
many directions we have to see an element current increase (more 
charges accelerating) to get rid of all that energy from the 
transmitter in the narrow direction where radiation can still occur. 
The antenna, in effect, fights itself.

Obviously if we add an element in a direction where there is little or 
no energy, it can have little or no effect on the system. Nothing 
amazing here at all. 


 If you make a mistake with a reflector it is best to err long 
because the inductive reactance causes     


The reason a reflector  

> Reflectors in such cases are NOT always longer than the driven element
> ( must not exceed 3 in number ,total 13 elements max per wavelength of
> array) ......because the self impedance of an element is of more 
> importance than 
>  physical length 
> Seems like this should be a  subject other than 'shadowing'
> Art Unwin
>  I suspect it has something
> >to do with the geometry and transit time but its only my WAG. 
> >Hopefully "several technically will-versed people" will enlighten us!
> >
> >k7puc
> >
> >
> >
> >--
> >FAQ on WWW:     
> >Submissions:     Administrative
> >requests: Problems:                
> >
> >
> >
> >
> --
> FAQ on WWW:     
> Submissions:     Administrative
> requests: Problems:                

73, Tom W8JI

FAQ on WWW:     
Administrative requests:

<Prev in Thread] Current Thread [Next in Thread>