[Moon-net] 2 D model of Parabola that incorporates an RMS error factor.

Christoph Fischer chris.the.fish at gmx.net
Sat Sep 27 18:21:01 CEST 2008


Hello Pat, Rasto and the rest of the list,

as you have the complete 2D data of the dish you could try some raytracing: You start a ray from the phasecenter of the feed, reflect it on the surface of the dish (using the available data) and trace it to a reference plane in front of the reflector. At that referenceplane you calculate the current phase of the ray modulo 360deg and repeat that for each point of the reflector you have in your datafile. With these phaseerrors and some assumptions on the illumination of the dish you can synthesize the antennadiagramm using some formulas from THE antennabook "Antenna Theory: Analysis and Design" by Constantine A. Balanis.
If this idea sounds applicable i can give you some more detail and some formulas when i am back at the university on monday. At our institute we use this technique to calculate the reflectorplates of planar reflectarrays.
In a dissertation about these reflectarrays i read that a random phase error does not affect the gain of a reflectorantenna much but results in a (slightliy?) higher sidelobelevel.

73
Chris
DC6GF


-------- Original-Nachricht --------
> Datum: Sat, 27 Sep 2008 08:51:20 -0700
> Von: Pat Barthelow <aa6eg at hotmail.com>
> An: om6aa at yahoo.com, aa6eg at k6bj.org
> CC: moon-net at list-serv.davidv.net
> Betreff: [Moon-net] 2 D model of Parabola that incorporates an RMS error	factor.

> 
> Thanks, Rasto, 
> 
> For the reply and the reference Book.  I will have to look for that one.
> 
> Perhaps it could be done in two dimensions, just a slice/profile of the 
> surfaces through the vertex...  I have the high resolution  (6 millimeter
> increment, .05 mm accuracy) precision file, 2 D, coordinates,  of the
> mathematical  surface profiles of Jamesburg, both the main reflector, and the
> secondary reflector,   and would be interested in studying the affects of
> errors vs frequency,  on the beam pattern. 
> If I could write a realistic program that modifies the profile data set, 
> to simulate a known RMS surface error, then run in a standard parabolic
> modelng program to see how a real dish responds compared with the theoretical
> surface.  We have about .050" RMS errors on the dish.
> 
> (By the way, I am very new to this... a good technical background, but no
> in-depth experience in this application...) 
> 
> Thanks, 
> 
> 
> 
> 
> All the best,
> 
> Pat Barthelow   (916) 315-9271; aa6eg at k6bj.org
> 
> Jamesburg Moonbounce Team
> 
> http://www.jamesburgdish.org
> 
> http://www.greatdanepro.com/Blue%20Bueaty/index.htm
> 
> 
> 
> 
> 
> > Date: Sat, 27 Sep 2008 08:34:53 -0700
> > From: om6aa at yahoo.com
> > To: aa6eg at k6bj.org
> > CC: moon-net at list-serv.davidv.net
> > Subject: Re: [Moon-net] Parabola Model, Pattern generation;	with Random
> surface errors
> > 
> > Hello Pat,
> > 
> >    I am afraid, that it is not possible to calculate such electrically
> large structure as is Jamesburg dish antenna (involving also it`s surface
> deviation) with professional el.mag. modelling softwares as are CST MW Studio
> , FEKO or HFSS due to huge amount of data. This task must be calculated
> separately as an independed efficiency coefficient of whole antenna, or must
> be involved in illumination efficiency coefficient. More about this problem
> you can find in the book " THE PARABOLOIDAL REFLECTOR ANTENNA IN RADIO
> ASTRONOMY AND COMMUNICATION" by JACOB W.M. BAARS,  ISBN-10: 0-387-69733-0. In
> the book are also  some routines written in software package "Mathematica"
> which you may find useful for your purposes.
> > 
> >    Sincerely, 
> >                      Rasto OM6AA
> > 
> > --- On Fri, 9/26/08, Pat Barthelow <aa6eg at hotmail.com> wrote:
> > From: Pat Barthelow
> >  <aa6eg at hotmail.com>
> > Subject: [Moon-net]
> >  Parabola Model, Pattern
> >  generation; with Random surface errors
> > To: moon-net at list-serv.davidv.net
> > Date: Friday, September 26,
> >  2008, 12:46 PM
> > 
> > Folks: 
> > 
> > Are there any Antenna modeling programs, within reach of EME enthusiasts
> ( If
> > not them, then avaailable for professional work)  that would allow you
> to enter
> > a fine detail theoretical  surface coordinate database, then have the
> user apply
> > an empirically derived RMS error metric, to the perfect coordinate
> database,
> > generating a representative, as found, surface?   
> > 
> > Then, based on the real, "lumpy" surface, generate beam patterns, for
> > a given frequency.  It would be interesting to see Jamesburg's real
> pattern,
> > both at C band, it's designed frequency, and then pushing it to higher
> > frequencies, say, X band, and see how the pattern degrades, based on an
> RMS
> > error which we determined to be in the vicinity of .050"
> > 
> > 
> > 
> > 
> > 
> > All the best,
> > 
> > Pat Barthelow   (916) 315-9271; aa6eg at k6bj.org
> > 
> > Jamesburg Moonbounce
> > 
> > 
> > 
> > 
> >  Team
> > 
> > http://www.jamesburgdish.org
> > 
> > http://www.greatdanepro.com/Blue%20Bueaty/index.htm
> > 
> > 
> > 
> > _______________________________________________
> > Moon-Net posting and subscription instructions are at
> > http://www.nlsa.com/nets/moon-net-help.html
> > 
> > 
> > 
> > 
> >       
> > _______________________________________________
> > Moon-Net posting and subscription instructions are at
> http://www.nlsa.com/nets/moon-net-help.html
> > 
> _______________________________________________
> Moon-Net posting and subscription instructions are at
> http://www.nlsa.com/nets/moon-net-help.html

-- 
Psssst! Schon vom neuen GMX MultiMessenger gehört? Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger

_______________________________________________
Moon-Net posting and subscription instructions are at http://www.nlsa.com/nets/moon-net-help.html




More information about the Moon-net mailing list