Patch antenna as a parabolic dish feed

834 views
Skip to first unread message

Thomas Consi

unread,
Jul 16, 2022, 8:51:34 AM7/16/22
to Society of Amateur Radio Astronomers
What are the advantages and disadvantages of a patch antenna as a feed for a parabolic dish compared to other types of feeds?

Thanks,

Tom

Marcus D. Leech

unread,
Jul 16, 2022, 9:01:58 AM7/16/22
to sara...@googlegroups.com
For our project, the situation required a very compact feed--we simply
didn't have room up at the prime focus for a waveguide ("can") feed--not
without removing a very-heavy
  feed support ring and replacing it with something different.

For deeper dishes the low-gain nature of a patch feed was ideal, and you
get a fairly-strong band-pass response "for free".  Now, the patch we
use is air-dielectric, which
  means that the dielectric losses are very very low.   Most "patch"
antennas you see out in the "wild" are PCB or "printed" patch antennas. 
They are necessarily somewhat
  lossy.  In fact I think the entire area of "patch" antennas developed
with PCB/printed patches, and only later were air-dielectric patches
considered for applications requiring
  very low loss, like EME, satellite, and radio astronomy.



Alex P

unread,
Jul 16, 2022, 9:47:01 AM7/16/22
to Society of Amateur Radio Astronomers
In response to Marcus' post of a few days ago on this topic,
Just this morning I began building a 'patch' in a cake pan.
The goal : to be able to quickly swap between a Loop and Patch feed and compare their performance with my configuration.
More iterations required before field testing....  ( and a few more cake pans )

Alex




PRT_patch01.jpg

djl

unread,
Jul 16, 2022, 11:39:38 AM7/16/22
to sara...@googlegroups.com, Marcus D. Leech
Just finished a 437 MHz patch feed to try on the DSES dish following an
excellent design by W5LUA (if I remember right!). Had a hard time
finding a pan big enough, finally found a pizza pan, nearly perfectly
right, aluminum and well made. The trusty HVAC end caps were far more
expensive at that size and of course more flimsy. Follow the pizza pan
track for other sizes and depths, very sturdy and easy to work. Use the
antioxidizing goop for aluminum wiring and aluminum soldering kits if
needed.
As a graduate of the New Mexico Tech culinary hardware construction
crew, I was very satisfied. Don't forget the always ready to serve
pressure cooker hermetically sealed housings, and other well constructed
Al and stainless stuff for commercial kitchens (e.g. hemispherical soup
ladles and the like)
Don
> --

------------
When in trouble, when in doubt,
Run in circles, scream and shout.
(Naval War College Football Team)
----------------------
Dr. Don Latham AJ7LL
PO Box 404, Frenchtown, MT, 59834
VOX: 406-626-4304

Marcus D. Leech

unread,
Jul 16, 2022, 12:07:11 PM7/16/22
to djl, sara...@googlegroups.com
On 2022-07-16 11:39, djl wrote:
> Just finished a 437 MHz patch feed to try on the DSES dish following
> an excellent design by W5LUA (if I remember right!). Had a hard time
> finding a pan big enough, finally found a pizza pan, nearly perfectly
> right, aluminum and well made. The trusty HVAC end caps were far more
> expensive at that size and of course more flimsy. Follow the pizza pan
> track for other sizes and depths, very sturdy and easy to work. Use
> the antioxidizing goop for aluminum wiring and aluminum soldering kits
> if needed.
> As a  graduate of the New Mexico Tech culinary hardware construction
> crew, I was very satisfied.  Don't forget the always ready to serve
> pressure cooker hermetically sealed housings, and other well
> constructed Al and stainless stuff for commercial kitchens (e.g.
> hemispherical soup ladles and the like)
> Don
Here's the sub-lunar patch feed, integrated into our feed-array:

https://photos.app.goo.gl/NVLYSseSC36ChTGMA


Marcus D. Leech

unread,
Jul 16, 2022, 12:15:21 PM7/16/22
to djl, sara...@googlegroups.com
On 2022-07-16 11:39, djl wrote:
> Just finished a 437 MHz patch feed to try on the DSES dish following
> an excellent design by W5LUA (if I remember right!). Had a hard time
> finding a pan big enough, finally found a pizza pan, nearly perfectly
> right, aluminum and well made. The trusty HVAC end caps were far more
> expensive at that size and of course more flimsy. Follow the pizza pan
> track for other sizes and depths, very sturdy and easy to work. Use
> the antioxidizing goop for aluminum wiring and aluminum soldering kits
> if needed.
> As a  graduate of the New Mexico Tech culinary hardware construction
> crew, I was very satisfied.  Don't forget the always ready to serve
> pressure cooker hermetically sealed housings, and other well
> constructed Al and stainless stuff for commercial kitchens (e.g.
> hemispherical soup ladles and the like)
> Don
>
I scored some 18" aluminum pizza pans a few years back to use with my
full-wave-loop feeds for 408MHz and 611MHz.

The main benefit for a patch for us was the low gain (useful for deeper
dishes in the 0.28-0.34 F/D range), and the "for free" filtering you get
with a patch.
  They are quite narrow-band--expect 5% at the most, which also means
they're a bit finicky to tune, but if you have a modern VNA, like the
NanoVNA, it's
  fairly straightforward...

They're also quite compact, because the radiator-to-ground spacing is
usually quite small, so if you, like us, have considerable constraints
in the amount of
  Z-axis dimension available, a patch is "just the thing"...

Hamish Barker

unread,
Jul 16, 2022, 7:09:47 PM7/16/22
to sara...@googlegroups.com
Is a patch essentially the electrical inverse of a loop ( but driven how? ), in a similar fashion to a slot is the inverse of a rod?

--
--
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.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/5efaafb9-db70-82f9-287e-5b9961f6becf%40gmail.com.

Marcus D. Leech

unread,
Jul 16, 2022, 7:13:55 PM7/16/22
to sara...@googlegroups.com
On 2022-07-16 19:09, Hamish Barker wrote:
Is a patch essentially the electrical inverse of a loop ( but driven how? ), in a similar fashion to a slot is the inverse of a rod?
Not sure.  The patch that I threw together was directly driven from the center conductor of a type-N connector, about 18mm in from one
 edge of a square 93mm piece of copper sheet.  Perhaps 6-7mm above a reflector surface.

There are also patches that are driven with a capacitive probe from underneath.

The patch that we're deploying, from sub-lunar is a circular patch and I'm not sure  whether it's direct-coupled or not.


djl

unread,
Jul 16, 2022, 7:38:42 PM7/16/22
to sara...@googlegroups.com

I found a clue in one of the patch presentations on line. Suppose the center of the circular patch is at zero ohms impedance at the desired frequency. The edge is close to the impedance of free space. So 50 ohms impedance is then calculated based on the radius?   Furthermore, if at the 50 ohm radius, one travels along a diameter 90 degrees, the signal there is at 90 degrees of phase wrt the starting point. Thus a 90 degree hybrid, properly applied, will give RH or LH circular polarization. That's rough thinking, but seems sort of right.  As with a loop in a can,  the patch to pan spacing should be the magic 1/8 wavelength?  Some of this may barely escape magic thinking?

Don

djl

unread,
Jul 16, 2022, 7:56:20 PM7/16/22
to sara...@googlegroups.com

  PY2BS Patch Feeds built by W5LUA , North Texas Microwave Society.

Look also for stuff by PY2BS on patch feeds.  This is older info, apparently from the days when sheet copper was available and not too expensive.

There are also some papers in the IEEE journals. They want to charge you some portions of your anatomy for copies. I got a couple of the papers.  Usually there is just a taste of what you want to know, with just enough withheld so you can't readily use the paper to build anything.  You will have to find the right Chinese database to get useful information, on the Chinese dark web...

Don

Jeff Kruth

unread,
Jul 16, 2022, 9:11:24 PM7/16/22
to sara...@googlegroups.com
A patch is an equivalent to a dipole (really a cavity resonator but E field pattern like a dipole over a ground plane).
Fun stuff! Air dielectric is the only way to go for R-A as efficiency is best. wide angle feed, and maybe not as symmetric as you would like in E & H planes, but OK enough. Single patch good for feeding a deep dish, while arrays can feed shallower ones.
Saw some air dielectric patch arrays on feed canisters for the 140 footer at NRAO Greenbank. Big boys have used 'em.
73
Jeff Kruth

Marcus D. Leech

unread,
Jul 16, 2022, 9:17:15 PM7/16/22
to sara...@googlegroups.com
On 2022-07-16 21:11, 'Jeff Kruth' via Society of Amateur Radio Astronomers wrote:
A patch is an equivalent to a dipole (really a cavity resonator but E field pattern like a dipole over a ground plane).
Fun stuff! Air dielectric is the only way to go for R-A as efficiency is best. wide angle feed, and maybe not as symmetric as you would like in E & H planes, but OK enough. Single patch good for feeding a deep dish, while arrays can feed
Apparently the choke/baffle improves E&H symmetry.

Our dish is F/D = 0.319, so having something with a wide illumination pattern is good, and the narrow-band nature is also of great benefit.   A bit like having
  a cavity-input LNA, but more modular and cheaper...




shallower ones.
Saw some air dielectric patch arrays on feed canisters for the 140 footer at NRAO Greenbank. Big boys have used 'em.
That's good to know.  I was skeptical of patch feeds for several years, because ALL the implementations I'd seen were *printed microstrip* patch
  antennas, which are somewhat lossy.   Although, as a counter-point the feeds on CHIME are basically printed patches, and they don't have much
  of a problem (granted, they're operating at 400-800MHz).




Wayne Hilliard

unread,
Jul 18, 2022, 5:43:11 PM7/18/22
to sara...@googlegroups.com
Where is some good design information about patch antennas. That 10 foot dish I'm piecing together has  .41 f/d so maybe a patch would be good? I'm aiming at initially hydrogen line but ultimately perhaps 1296 eme. My call Ka1cxd
Thanks a bundle!
Wayne H

Marcus D. Leech

unread,
Jul 18, 2022, 5:56:03 PM7/18/22
to sara...@googlegroups.com
On 2022-07-18 17:42, Wayne Hilliard wrote:
Where is some good design information about patch antennas. That 10 foot dish I'm piecing together has  .41 f/d so maybe a patch would be good? I'm aiming at initially hydrogen line but ultimately perhaps 1296 eme. My call Ka1cxd
Thanks a bundle!
Wayne H

The patch antenna that I use is made by Sub-Lunar systems, but it is in turn based on a design by SM6PGP.

In general, patch antennas are lowish gain (5-7dB), so they don't have quite enough gain to properly illuminate a shallower dish like F/D = 0.41.  SM6PGP seems to have
  come up with a "beam forming ring" which makes them higher gain for shallower dishes.    My take on said "BFR" is that it's like a parasitic element in a Yagi.

I actually built my own patch feed initially, and it worked OK -- but the commercial one from sub-lunar has better S11 (17dB on mine, vs 22dB on the sub-lunar), and it's
 just better thought out.

But what worked for me was a square of copper 95mm on a side, with the feed points in about 18mm from two adjacent edges, with the whole thing about 6mm off
  a circular reflector surface. The feeds were type-N and the patch was soldered directly to the pins on the type type N connectors, giving about 6mm clearance to
  the ground plane.   I would not recommend doing this without a VNA handy to check on things.  These patch antennas are notoriously narrow-band, so going in
  "blind" doesn't work very well, unlike other, more-forgiving antennas like dipoles and full-wave loops and the like.

If you don't need the filtering effects of a "patch", and you're only going to do a single polarization, a full-wave loop is MUCH more forgiving.  Basically a full-wavelength
  of thick wire formed into a loop, about 1/5 lambda above a suitably-sized ground plane.  Just like that is reasonably compact, and has enough gain to adequately
  illuminate a shallower dish.  Or, if you have the headroom, a "soup can" feed made from 15cm round duct.




b alex pettit jr

unread,
Jul 19, 2022, 9:30:52 AM7/19/22
to sara...@googlegroups.com
.Hello,

 I suggest starting with the Loop_Feed  .... testing it and then building the Patch_Feed.

With the Loop Feed, all dimensions but length of the 211 mm of wire can be cut to size with the Wire Length tweaked with a VNA..

With the Patch Feed,   the Disk Diameter, its Spacing from the back reflector, and the Feed position all interact ... much more Guess&Test required.
I will be adding a small choke at the center of the patch disk for ESD, it has No Effect on the VNA measured antenna parameters.

Although it will take on-parabola testing to compare beam widths & performance, their SWR bandwidths are about equal.

Patch ( Approx ): 5" Cake Pan cut to 31mm deep,  ~100mm dia copper Patch, spacing  5-8mm  ( TBD'd ) from 'pan' , Feed Point ~  19mm offset from Center

Loop Dimensions :


( adaptation of the OM6AA design for 1.42G )



Cheers,
Alex P   KK4VB

Patch_vs_Loop.jpg

b alex pettit jr

unread,
Jul 20, 2022, 5:59:11 AM7/20/22
to 'b alex pettit jr' via Society of Amateur Radio Astronomers
PATCH  Update

If you want to experiment, here are some dimensions

Patch Dia  103mm
FeedPoint 19mm from Center
Reflector<>Patch 8mm

Reflector : 5" cake pan, 31mm deep 

 I used 1mm copper sheet, but suggest 0.8mm ( 0.032" )

Data obtained With the Feed Mounted on the 1m dish.
( VNA calibrated @ the feed connector )

Z = 49.10 - j7.29 @ 1420 MHz


Alex P  KK4VB

Patch_Mtd_sm.jpg

Alex P

unread,
Aug 2, 2022, 5:53:42 PM8/2/22
to Society of Amateur Radio Astronomers
I finally did some sky  testing of the Loop and Patch Feeds as shown a few posts back.

The PATCH Feed works very well.
The LOOP Feed had performance issues. I just identified it seems related to the ARS 2"  Feed-Thru.

SO, for now, try the PATCH Feed design only.

Alex P

Jim Abshier

unread,
Aug 3, 2022, 12:01:27 PM8/3/22
to sara...@googlegroups.com
Alex,

My experience with loop feeds revealed that the feed impedance is very
sensitive to spacing between the loop and the reflector. If you have not
optimized this spacing, your loop feed may not be giving you the best
performance that it is capable of providing. I have attached a figure
from the Antenna Engineering Handbook that shows how impedance of a loop
feed varies with reflector spacing.

Jim Abshier

On 8/2/22 5:53 PM, 'Alex P' via Society of Amateur Radio Astronomers wrote:
> I finally did some sky  testing of the Loop and Patch Feeds as shown a
> few posts back.
>
> The PATCH Feed works very well.
> The LOOP Feed had performance issues. I just identified it seems
> related to the ARS 2"  Feed-Thru.
>
> SO, for now, try the*PATCH Feed design only.*
>
> Alex P
>
>
> --
> --
> 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
> <mailto:sara-list+...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sara-list/60286203-6848-4778-94df-981627331b49n%40googlegroups.com
> <https://groups.google.com/d/msgid/sara-list/60286203-6848-4778-94df-981627331b49n%40googlegroups.com?utm_medium=email&utm_source=footer>.
Loop Feed Impedance.pdf

Marcus D. Leech

unread,
Aug 3, 2022, 12:20:12 PM8/3/22
to sara...@googlegroups.com
On 2022-08-03 12:01, Jim Abshier wrote:
> Alex,
>
> My experience with loop feeds revealed that the feed impedance is very
> sensitive to spacing between the loop and the reflector. If you have
> not optimized this spacing, your loop feed may not be giving you the
> best performance that it is capable of providing. I have attached a
> figure from the Antenna Engineering Handbook that shows how impedance
> of a loop feed varies with reflector spacing.
>
> Jim Abshier
I have generally used about lambda/5 as the loop spacing with reasonable
results.

Over a 1 lambda reflector, a full-wave loop has a higher gain than a
patch feed.   Loops are often used for shallower dishes, although
  forward gain can be adjusted a little bit with reflector size.

Without a "BFR", patch feeds have fairly low gain, so they work well for
deeper dishes with F/D in the 0.28 to 0.32 range.  I rarely
  encounter dishes below F/D 0.3, at least at size ranges useful for
RA.   It is often the case that dishes in the 1-2m class that
  have been "pulled" from point-to-point microwave service are quite
deep--some of them even feed at the aperture plane!
  This is where a patch or a simple dipole might be the only reasonable
choice.

A full-wave loop has a bandwidth in the 8-10% region, whereas a patch
has bandwidth in the 3-4% region.   The reduced bandwidth
  can be helpful in doing some first-order RFI filtering to reduce the
chance of intermod and the like.  This and the low-gain are
  why we're using a pair of patch feeds at the CCERA Carp site--the
dish has an F/D of 0.319, and the observatory isn't exactly
  "in the middle of nowhere", so anything you can do to limit the
amount of "crap" seen by the first LNA stage is a blessing.


b alex pettit jr

unread,
Aug 3, 2022, 2:08:27 PM8/3/22
to sara...@googlegroups.com
Hello Jim and Marcus,

There may have been several issues during last week's sky tests.

I have rebuilt the Loop <> N interface and, as Jim suggested, optimized the Loop to Reflector distance ..
It is  26mm which is ~ 1 mm of the original copper/brass unit built last year ( and used for all data to date )

I will test this revised Loop tomorrow

LOOP   ( Built Today )
Inline image


For Ref,
 PATCH
Inline image

This is  an impedance analysis I performed a week ago .. using a 100 mm dia thin PC board with lots of holes over a 12" sq aluminum sheet.
in the 1.42GHz range
From it, I found the optimal feed point .. which confirmed Marcus' configuration.
Inline image

 I must have been very 'lucky' on the first Loop built last August .

Thanks for the Assistance !

Alex P




Marko Cebokli

unread,
Aug 4, 2022, 1:15:57 PM8/4/22
to sara...@googlegroups.com
Hello!

I am trying to see how the solar activity correlates with GPS errors. I
made a long recording of my position, from the eighth of July to the
second of august. Now I am looking for solar activity data for that
period. I've checked swpc.noaa.gov and www.spaceweatherlive, but I can
only find data for the last seven days. I guess there should be some
site somewhere with more archived data, I just can't find it.
Also, maybe somebody here, who is observing SIDs or similar, is making a
publicly accessible archive of his data?

Marko Cebokli

Rodney Howe

unread,
Aug 4, 2022, 1:51:06 PM8/4/22
to 'fasleitung3' via Society of Amateur Radio Astronomers
I think you can get recent data from here:  ftp://ftp.swpc.noaa.gov/pub/indices/events/


--
--
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.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/c13ad4bddb015eeaa1dea2f2222bef9a%40hamradio.si.

Marko Cebokli

unread,
Aug 5, 2022, 1:38:08 PM8/5/22
to sara...@googlegroups.com
Hello Rodney,

thanks for the link!

I was looking for graphs like those on the front page, but can compare
the events from the list with my curve anyway.

I have attached the graph of my GPS height data. There was a pronounced
day/night cycle until 20th of July, when the amplitude fell, and later
more or less completely vanished on 25th. Of course, that could be a bad
contact in some connector, or some local QRM coming up.

Marko Cebokli



On 04.08.2022 19:51, Rodney Howe wrote:
> I think you can get recent data from here:
> ftp://ftp.swpc.noaa.gov/pub/indices/events/ [2]
>
>> On Aug 4, 2022, at 11:15 AM, Marko Cebokli <s57...@hamradio.si>
>> wrote:
>>
>> Hello!
>>
>> I am trying to see how the solar activity correlates with GPS
>> errors. I made a long recording of my position, from the eighth of
>> July to the second of august. Now I am looking for solar activity
>> data for that period. I've checked swpc.noaa.gov [1] and
> http://groups.google.com/group/sara-list?hl=en [3]
> ---
> 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.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sara-list/BD202A1E-1500-4E11-97F9-ABCBE5B3DC40%40frii.com
> [4].
>
>
> Links:
> ------
> [1] http://swpc.noaa.gov
> [2] ftp://ftp.swpc.noaa.gov/pub/indices/events/
> [3] http://groups.google.com/group/sara-list?hl=en
> [4]
> https://groups.google.com/d/msgid/sara-list/BD202A1E-1500-4E11-97F9-ABCBE5B3DC40%40frii.com?utm_medium=email&utm_source=footer
GPS_H_08jul-02aug_e1.png

Dave Typinski

unread,
Aug 5, 2022, 1:43:57 PM8/5/22
to sara...@googlegroups.com
Hi Marko,

The SWPC does archive most of their plots/images. Just have to poke around
their FTP server to find them. There is a LOT of material there.
--
Dave
Reply all
Reply to author
Forward
0 new messages