First light from multi-mode horn for 1420 MHz

[David L, N5OIQ]

Hello Everyone,

After a lot of frustration trying to work at lower frequencies, I have returned to 1420 MHz where I had some success using a SETI 'Horn of Plenty' antenna, even in my RFI-hostile suburban location.  The SETI 'Horn of Plenty' is very easy to build, but its design is far from optimal.  Its side lobe performance is likely not that good, letting a lot of weak carriers and some broadband RFI into receiver, which was a headache, especially when trying to observe quiet regions of the sky such as in Leo.  The way the probe is situated in the SETI 'Horn of Plenty' probably also leads to the creation of other propagation modes in the waveguide.

To improve my system, I opted to build a mulit-mode 'Potter' horn antenna.  An easier-to-build version of such an antenna was described by Skobelev et al. in an IEEE article about ten years ago and has been re-modeled by W1GHZ as a high performance feed for dish antennas.  I opted to build one with a horn length of 6.5 wavelengths and use it on its own.  It is very close in design to that described in this article:

http://www.vhfdx.ru/apparatura/accurate_noise_figure_measurements_1296_mhz

Side lobes are predicted to be roughly ~15-20 dB below that of a classic pyramidal horn antenna.  It is harder to build, but it is worth the effort (see the attached file).  RFI is reduced considerably and I'm getting some clean looking spectrograms.  In particular, when observing in a quiet part of the sky in Leo, I can still get a good detection of neutral hydrogen. I haven't done any tuning of the probe yet and I'm getting a Leo/Ground Y factor of 7 dB.  This isn't so good, but I think this is due to an untuned probe.

I've got a some more work to do on compensating for gain flatness in the system.  Right now, I'm am flattening the spectrum based on spectra obtained at 1418 MHz, but due to some ripple in my interdigital filter, compensation is still off a little (more on this later).  

I've also included some of the rough code that I used to generate the spectrograms.  The plt.specgram routine that is a part of the matplotlib library is a fast way to calculate spectrograms that should allow for real time calculations.

David

[Marcus D. Leech]

What is the aperture of your horn?

Those are very nice results.

--
Marcus Leech
Principal Investigator
Shirleys Bay Radio Astronomy Consortium
http://www.sbrac.org

[David L, N5OIQ]

The diameter of mine is 4 wavelengths.  W1GHZ has got some nice charts on variations between length and aperture (-/+ 10% doesn't hurt side lobe performance too much).  I scaled the TM11 mode transition according to the RA3AQ design (from the G4DDK preamp tweaking link), since the length of this region is not specified in the Skobelev or W1GHZ descriptions.  

http://www.w1ghz.org/antbook/conf/optimized_dualmode_feedhorns.pdf

[Marcus D. Leech]

> The diameter of mine is 4 wavelengths. W1GHZ has got some nice charts
> on variations between length and aperture (-/+ 10% doesn't hurt side
> lobe performance too much). I scaled the TM11 mode transition
> according to the RA3AQ design (from the G4DDK preamp tweaking link),
> since the length of this region is not specified in the Skobelev or
> W1GHZ descriptions.
>
> http://www.w1ghz.org/antbook/conf/optimized_dualmode_feedhorns.pdf
>

Hmm, so slightly-less aperture than my 93cm dish, but likely much better
aperture efficiency, since you don't have to deal with the illumination
efficiency problem. Nicely done.

[Jan Lustrup]

Wonderfull results....good going..

Jan

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[David L, N5OIQ]

The prime goal I had for this antenna was to strive for the lowest possible amount of energy headed outside the main beam.  Some of this practical, just to avoid RFI.  From the literature, it looks like Potter horns can achieve ~80% "Aperture efficiency" which should probably best be thought of as "main beam efficiency".  A dish will be more complicated.  Even if the Aeff is on the low side, if it is underilluminated, it could still perform well.  I like the descriptions of these issues in the web article linked below.

By getting Tsys way down with the horn, deltaT is probably about as small as I can make it.  The downside is the large beam is seeing a lot of sky and with variable Doppler shifts from different sources changing as they cross the meridian, a single horn is going to have a lot of source confusion issues.  I'm hoping that once I get a second built, these problems can be solved in an interferometer configuration (if I ever find enough time and energy).

http://www.astron.nl/other/workshop/MCCT/TuesdayCappellen.pdf

[Marcus D. Leech]

> The prime goal I had for this antenna was to strive for the lowest
> possible amount of energy headed outside the main beam. Some of this
> practical, just to avoid RFI. From the literature, it looks like
> Potter horns can achieve ~80% "Aperture efficiency" which should
> probably best be thought of as "main beam efficiency". A dish will be
> more complicated. Even if the Aeff is on the low side, if it is
> underilluminated, it could still perform well. I like the
> descriptions of these issues in the web article linked below.

Well, it seems like a nice solution. You could just scale it up for a
larger aperture, and see how you do. Did you build the horn yourself, or
have a sheet-metal shop do most of it?

[David L, N5OIQ]

I built most of it myself out of aluminum sheet metal and it was a lot of work.  The input came from an old RAS 1420 MHz circular wave guide.  I'm worried about the transitions between different pieces of metal.  So far, I've avoided putting aluminum insulation tape inside the horn, I'm not sure what sort of effect it might have on performance (although that did seem to help the SETI Horn of Plenty).  Instead, I packed aluminum foil into the joints in the step transitions to provide continuity.