[NLRS] 10GHz feed horn design & question

[Barry VE4MA]

I would not bother making a new feedhorn up for such a small difference as 
Donn says concentrate on getting the best out of what you can with your 
exising DSS horn

Barry VE4MA
----- Original Message ----- 
From: "Baker, Donn B" <Donn.Baker at UNISYS.com>
To: <W0ZQ at aol.com>; <nlrs at mailman.qth.net>
Sent: Monday, November 29, 2004 12:56 PM
Subject: RE: [NLRS] 10GHz feed horn design & question


Hi Jon,
The feed horn affects the "edge taper," or the "illumination" of the dish. 
Ideally, the feed horn will uniformally place all the energy on the surface 
of the dish, and none beyond the the surface.  In practice, this is 
impossible:  the rule of thumb is for -10dB edge taper... i.e., the edge of 
the dish is illuminated 10 dB down from the center of the dish.  If the 
pattern of the feed horn does not provide the correct beamwidth, it cannot 
provide the correct illumionation.  The higher the gain of the feed horn, 
the narrower the pattern.  Since your dish is slightly larger than the 
(usual) DSS dish, you would want a slightly lower gain feedhorn, just like 
the patterns produce.  (Funny how that works out, isn't it ?)

That said, it may not be worth the effort to change.  A few tenths of a dB 
at best.  On the other hand, making upo a new feedhorn isn't hard... maybe a 
hour's work.

My TRW "Antenna Communications Calculator" shows slightly smaller numbers 
than your calculations:  33 dB gain @ 50 % eff.; 33.8 at 60% eff.  Have you 
measured the dish as it is ?  If its within a dB of the either of these, I'd 
be tempted to leave it alone.

The place of the phase center of the feedhorn at the focal point of the dish 
is more important that having exactly -10dB, etc. illumination.  At 10 GHz, 
we have maybe 1/8" to play with.  Thats in 3-space: Left-right, up-down, and 
in-out.  I'd bet (at least a quarter !) that you'll get more gain out of the 
dish playing with this than you will with a new feed horn.

73 Donn
WA2VOI/0



> -----Original Message-----
> From: nlrs-bounces at mailman.qth.net
> [mailto:nlrs-bounces at mailman.qth.net]On Behalf Of W0ZQ at aol.com
> Sent: Wednesday, November 24, 2004 1:30 PM
> To: nlrs at mailman.qth.net
> Subject: [NLRS] 10GHz feed horn design & question
>
>
> I finally measured the dimensions of my "larger" 10gigs
> offset dish that I
> have been using this summer.   Here are the stats:
>
> Largest axis (top to bottom) = 654.05 mm
> Shortest axis (left to right) = 596.90 mm
> Deepest spot along largest axis = 58.42 mm
> Deepest spot from bottom along axis = 292.10 mm
>
> Using hdl_3b3, this dish, at 10368.1 has the following attributes:
> Focal length = 357.9 mm
> Full parabola f/D = 0.3
> Focal point is located 357.92 mm from bottom edge and 611.73
> mm from top  edge
> Large axis tilt (to get beam on the horizon) = 67.2 degrees
> Use a feedhorn design of f/D = 0.68
> With 50% efficiency, gain is 33.2 dBi
> With 60% efficiency, gain is 34.0 dBi
>
> So then I punched the f/D = 0.68 data into the horn design
> and  generated a
> template for an 11.29 dBi horn (at 10368).   The horn that
> is currently on
> this dish is a RCA DSS horn with a gain of 11.49  dBi.   When
> I hold the two
> templates up against each other over a  strong light source
> to compare them, they
> are indeed just a little bit  different.
>
> So, for the dish guru's out there, will changing the feedhorn
> from the
> current RCA DSS feed horn (11.49 dBi) to a more appropriate
> feedhorn (11.29  dBi)
> provide better illumination for the slightly larger dish and
> hence a bit  more
> gain ?   My thinking is yes, but the improvement may be on
> the  order of one
> dB or so (its not the difference in the gain of the two
> feedhorns ..... is it
> ?), certainly not a BIG deal, but a deal none the  less.
>
> 73,  Jon
> W0ZQ
>
> _______________________________________________
> NLRS mailing list
> NLRS at mailman.qth.net
> http://mailman.qth.net/mailman/listinfo/nlrs
>
_______________________________________________
NLRS mailing list
NLRS at mailman.qth.net
http://mailman.qth.net/mailman/listinfo/nlrs