It is well that the antenna plans are starting to congeal. I do have some comments upon the diagram that John presented, herewith attached.
1) There is an option, whether employed or not, that the U band could be the stacked CP patch design previously shown, with the L band array mounted thereon, also as previously shown.
2) The U band antenna currently shown in John's design could use a modification of the dipole design used for all of the AMSAT-DL P3 satellites. We call that "design adaptation" (not stealing!).
3) The V band antennas are of a great concern as shown in John's design. The P3D V antennas were more than 880mm end-to-end span, and a nominal 144mm spaced off of the spaceframe. Three of these on top of the Eagle seems to me to be of a gargantuan, unwieldy concept. These antennas would greatly impact the spacecraft envelop as mounting in a launcher. John has talked of a bent-end antenna design, is that really practical from an RF sense?
I am wondering if some form of dual-feed full-wave V band loop might not serve better in this application. Stan Wood had shown such a full-wave loop design for the Small Eagle that was only about 50mm high. Properly designed, it would not necessarily have to be 144mm high, so I am led to believe. Can some astute antenna designer look at this for us? I feel that we need to look at the options.
Dick Jansson --------------------------- mailto:[email protected] [email protected] ---------------------------
MessageThe U antenna can be anything that gives a 120-degree beamwidth and can be centered on the spin axis. There's no way to correct spin modulation on that band and the 30-50 bps uplinks will be more sensitive to phase pertubations than any other uplink. I'm assumng that there will be no class 1 uplinks on the L receiver. The advantage that I see in the crossed dipoles is that they leave more space for the S2 phased array.
The V antenna is the most problematic. I modelled the 3 dipole array (ends bent down to make each 600 mm long) six weeks ago and the main problem will be getting the antenna high enough off the satellite body to broaden the beam. I modelled a full-wavength loop today and it is almost the same. The 0.2h/0.5h/0.75h in the file names is the antenna height in meters.
73,
John KD6OZH ----- Original Message ----- From: Dick Jansson-rr To: AMSAT Eagle Sent: Monday, September 25, 2006 13:35 UTC Subject: [eagle] Antenna Plans
It is well that the antenna plans are starting to congeal. I do have some comments upon the diagram that John presented, herewith attached.
1) There is an option, whether employed or not, that the U band could be the stacked CP patch design previously shown, with the L band array mounted thereon, also as previously shown.
2) The U band antenna currently shown in John's design could use a modification of the dipole design used for all of the AMSAT-DL P3 satellites. We call that "design adaptation" (not stealing!).
3) The V band antennas are of a great concern as shown in John's design. The P3D V antennas were more than 880mm end-to-end span, and a nominal 144mm spaced off of the spaceframe. Three of these on top of the Eagle seems to me to be of a gargantuan, unwieldy concept. These antennas would greatly impact the spacecraft envelop as mounting in a launcher. John has talked of a bent-end antenna design, is that really practical from an RF sense?
I am wondering if some form of dual-feed full-wave V band loop might not serve better in this application. Stan Wood had shown such a full-wave loop design for the Small Eagle that was only about 50mm high. Properly designed, it would not necessarily have to be 144mm high, so I am led to believe. Can some astute antenna designer look at this for us? I feel that we need to look at the options.
Dick Jansson --------------------------- [email protected] ---------------------------
------------------------------------------------------------------------------
_______________________________________________ Via the Eagle mailing list courtesy of AMSAT-NA [email protected] http://amsat.org/mailman/listinfo/eagle
John:
So why not a full-wave loop at 100mm high? Indeed it might not be like your station antenna, but we are not asking a huge amount from that antenna. I am quite concerned for the potential s/c envelop. There are other possibilities such as the side-mounted antennas such as for P3E (AO-13-type), although sweeping shadow patterns will cross the solar panels, such momentary interruptions of cell strings has not yet been evaluated. Could it be acceptable? I don't know the answer to this, but I do know that I feel we must be perhaps a little less conventional on the solution for Eagle.
Dick Jansson --------------------------- mailto:[email protected] [email protected] --------------------------- -----Original Message----- From: John B. Stephensen [mailto:[email protected]] Sent: Monday, 25 September, 2006 1944 To: Dick Jansson-rr; AMSAT Eagle Subject: Re: [eagle] Antenna Plans
The U antenna can be anything that gives a 120-degree beamwidth and can be centered on the spin axis. There's no way to correct spin modulation on that band and the 30-50 bps uplinks will be more sensitive to phase pertubations than any other uplink. I'm assumng that there will be no class 1 uplinks on the L receiver. The advantage that I see in the crossed dipoles is that they leave more space for the S2 phased array.
The V antenna is the most problematic. I modelled the 3 dipole array (ends bent down to make each 600 mm long) six weeks ago and the main problem will be getting the antenna high enough off the satellite body to broaden the beam. I modelled a full-wavength loop today and it is almost the same. The 0.2h/0.5h/0.75h in the file names is the antenna height in meters.
73,
John KD6OZH
----- Original Message ----- From: Dick mailto:[email protected] Jansson-rr To: AMSAT Eagle mailto:[email protected] Sent: Monday, September 25, 2006 13:35 UTC Subject: [eagle] Antenna Plans
It is well that the antenna plans are starting to congeal. I do have some comments upon the diagram that John presented, herewith attached.
1) There is an option, whether employed or not, that the U band could be the stacked CP patch design previously shown, with the L band array mounted thereon, also as previously shown.
2) The U band antenna currently shown in John's design could use a modification of the dipole design used for all of the AMSAT-DL P3 satellites. We call that "design adaptation" (not stealing!).
3) The V band antennas are of a great concern as shown in John's design. The P3D V antennas were more than 880mm end-to-end span, and a nominal 144mm spaced off of the spaceframe. Three of these on top of the Eagle seems to me to be of a gargantuan, unwieldy concept. These antennas would greatly impact the spacecraft envelop as mounting in a launcher. John has talked of a bent-end antenna design, is that really practical from an RF sense?
I am wondering if some form of dual-feed full-wave V band loop might not serve better in this application. Stan Wood had shown such a full-wave loop design for the Small Eagle that was only about 50mm high. Properly designed, it would not necessarily have to be 144mm high, so I am led to believe. Can some astute antenna designer look at this for us? I feel that we need to look at the options.
Dick Jansson --------------------------- mailto:[email protected] [email protected] ---------------------------
_____
_______________________________________________ Via the Eagle mailing list courtesy of AMSAT-NA [email protected] http://amsat.org/mailman/listinfo/eagle
I'd like to make an appeal for (at least) 2 more S2 uplink patches. My plan is that we be able to process interferometric uplink data to serve as a "where is the earth?" attitude determination technique. Attached is a sketch showing where 2 elements might be located. The baselines Bx & By are the maximum baselines that these antennas offer. The sinusoidal interferometer fringes have a periodicity on the sky (fringe spacing) of B/(lambda) in radians. For a B ~ 500mm at 3.4 GHz=88mm, the fringe spacing ~0.17 radians or ~10 degrees. With a S/N ~ 10, the angles can be measured to ~0.05 fringes, i.e. about a half degree.
The way I have sketched the interferometer, the S2 array elements on shorter baselines can be used to quickly resolve the 2pi interferometer ambiguities.
73, Tom
Why not 4 entirely separate patches as far apart as possible or 3 patches next to the edges not occupied by the V antenna.
73,
John KD6OZH
----- Original Message ----- From: "Tom Clark, K3IO" [email protected] To: "AMSAT Eagle" [email protected] Sent: Tuesday, September 26, 2006 05:13 UTC Subject: [eagle] Re: Antenna Plans
I'd like to make an appeal for (at least) 2 more S2 uplink patches. My plan is that we be able to process interferometric uplink data to serve as a "where is the earth?" attitude determination technique. Attached is a sketch showing where 2 elements might be located. The baselines Bx & By are the maximum baselines that these antennas offer. The sinusoidal interferometer fringes have a periodicity on the sky (fringe spacing) of B/(lambda) in radians. For a B ~ 500mm at 3.4 GHz=88mm, the fringe spacing ~0.17 radians or ~10 degrees. With a S/N ~ 10, the angles can be measured to ~0.05 fringes, i.e. about a half degree.
The way I have sketched the interferometer, the S2 array elements on shorter baselines can be used to quickly resolve the 2pi interferometer ambiguities.
73, Tom
--------------------------------------------------------------------------------
Via the Eagle mailing list courtesy of AMSAT-NA [email protected] http://amsat.org/mailman/listinfo/eagle
John B. Stephensen wrote:
Why not 4 entirely separate patches as far apart as possible or 3 patches next to the edges not occupied by the V antenna.
73,
John KD6OZH
If you consider the "remote" element along with all 6 elements along the [Bx By] baselines, we get a really easy way to resolve ambiguities rapidly. If you consider the 6+1 elements as an array, the net power pattern is the Fourier transform of the autocorrelation function. The sampling function ends up nearly complete with the geometry I showed. If I've confused the issue too much, wait for my antenna arraying papers at the symposium.
Tom
Tom, Is there a special thing about having the 6 X 6 arrays square? Is the spacing between the element a specific number? Stan and I are going to look a making some patches in cavities like we have made for some of the small military satellites. We were able to obtain 8dB gain with those and they were circular polarized. Please give us the specific spacing and exact frequencies and we will see if we can make a prototype.
On Sep 26, 2006, at 2:09 AM, Tom Clark, K3IO wrote:
John B. Stephensen wrote:
Why not 4 entirely separate patches as far apart as possible or 3 patches next to the edges not occupied by the V antenna.
73,
John KD6OZH
If you consider the "remote" element along with all 6 elements along the [Bx By] baselines, we get a really easy way to resolve ambiguities rapidly. If you consider the 6+1 elements as an array, the net power pattern is the Fourier transform of the autocorrelation function. The sampling function ends up nearly complete with the geometry I showed. If I've confused the issue too much, wait for my antenna arraying papers at the symposium.
Tom
Via the Eagle mailing list courtesy of AMSAT-NA [email protected] http://amsat.org/mailman/listinfo/eagle
If the individual patches have too much gain it may restrict the offpointing angle. The beam needs to be steered 45 degrees off the spin axis. Doesn't this require that the individual patches have sufficient gain at that angle?
73,
John KD6OZH
----- Original Message ----- From: "Louis McFadin" [email protected] To: [email protected] Cc: "John B. Stephensen" [email protected]; "AMSAT Eagle" [email protected] Sent: Tuesday, September 26, 2006 15:28 UTC Subject: Re: [eagle] Re: Antenna Plans
Tom, Is there a special thing about having the 6 X 6 arrays square? Is the spacing between the element a specific number? Stan and I are going to look a making some patches in cavities like we have made for some of the small military satellites. We were able to obtain 8dB gain with those and they were circular polarized. Please give us the specific spacing and exact frequencies and we will see if we can make a prototype.
On Sep 26, 2006, at 2:09 AM, Tom Clark, K3IO wrote:
John B. Stephensen wrote:
Why not 4 entirely separate patches as far apart as possible or 3 patches next to the edges not occupied by the V antenna.
73,
John KD6OZH
If you consider the "remote" element along with all 6 elements along the [Bx By] baselines, we get a really easy way to resolve ambiguities rapidly. If you consider the 6+1 elements as an array, the net power pattern is the Fourier transform of the autocorrelation function. The sampling function ends up nearly complete with the geometry I showed. If I've confused the issue too much, wait for my antenna arraying papers at the symposium.
Tom
Via the Eagle mailing list courtesy of AMSAT-NA [email protected] http://amsat.org/mailman/listinfo/eagle
participants (4)
-
Dick Jansson-rr -
John B. Stephensen -
Louis McFadin -
Tom Clark, K3IO