Jump to content
happyelectron

High End Diversity Systems Theory & Architecture

Recommended Posts

Hi folks! It has been a while since I've posted - it turns out 3 kids are a bit of a handful. :rolleyes: It amazes me how much you all have figured out since I was active several years ago!!

In the meantime, between diaper changes, I've been pondering the general problem of getting realiable video. It amazes me that this is such a hard problem. In researching the topic on the Internet I ran across a company that had some intersting solutions/white papers that has really taught me a lot.

I am not affiliation with them, but I thought their approach could help inform and guide some of the stuff everbody is working on.

You can see a picture of their diversity approach at:

http://www.avalonrf.com/Products/dvreceive...ingantenna.html

A technical paper describing their system can be seen at:

http://www.avalonrf.com/Literature/applica...cationNote1.pdf

I must confess that when I saw the above system, I thought it was one of the ugliest things I've ever seen. It looks like he piled up a box of springs in a Rube Goldberg design to create an extremely inelegant solution to the problem.

However, I spoke with the chief engineer, and the more I spoke with him, the more I have grown to appreciate the difficult and subtle trade-offs that he chose that led to the design. The engineer claims they can get 20 miles without the need for physical tracking with this system. In particular, his use of a helical antenna has not been duplicated in many of the approaches I've seen here.

I've gotten a RF powermeter, and a bunch of different 900 Mhz antennas. I've been testing various combinations to see what works best. My initial results are that the helical has very good wide directional characteristics, with no nulls, as well as staying *really* circularly polarized at different transmitter/reciever orientations. I am currently looking at using multiple helicals to create a diversity receiver design similar to the above.

I'll post some of my experiences with the antennas later. In the meantime, thanks for helping keep my head in the clouds ... (and out of the diapers)!

Share this post


Link to post
Share on other sites

Happyelectron,

Interesting, any chance you can direct that engineer this way? I'm sure we can all learn from him expertise! He might even end up enjoying a new hobby?

Good luck with the diapers, you have my sympathies :D

Cheers,

Sander.

Share this post


Link to post
Share on other sites

Hi Happy,

I'm a bit keen on the 360 diversity systems since I buit my first one last year and have done a lot of testing and research along the way. So I feel I should comment :P

The engineer claims they can get 20 miles without the need for physical tracking with this system.

20 miles is a bold claim, what tx power and aerial were used ?

his use of a helical antenna has not been duplicated in many of the approaches I've seen here

I have considerd helicals as I make my own anyway, the reason I have not used them on a diversity 360 system is that you loose half the received power compared to a patch. Circularly polarized aerials are good for 3d flying and high flying over your ground station but are no benefit at long range touring.

Terry

Share this post


Link to post
Share on other sites
  20 miles is a bold claim, what tx power and aerial were used ?

He said they use a standard 1/2 wave dipole whip antenna. I can't remember the power, but I think he said it was less than 5 watts. These systems are used on real helicopters for TV news and police video surveilance. He said on the transmit side they have an arm that they lower down below the skids of the helicopter so their isn't interference with the helicopter. He also said they sometimes use a conical helical antenna on the xmit side.

With respect to the null on the whip, he said that when the helicopter is directly overhead, usually the distance is not that far and they don't need as high a sensitivity for the receive antenna. In fact, if you look closely at the picture, on top of the receiver, you will notice that they use a patch antenna to cover the area directly over the unit.

I have considerd helicals as I make my own anyway, the reason I have not used them on a diversity 360 system is that you loose half the received power compared to a patch. Circularly polarized aerials are good for 3d flying and high flying over your ground station but are no benefit at long range touring.

The above is true under ideal circumstances, i.e. - the transmitter whip oriented vertically with respect to a similarly oriented and polarized reciever antenna. However, if the plane changes orientation via turning and banking sharply, the transmit antenna will become more horizonitally polarized compared to the reciever antenna. This will cause a dramatic drop in the amount of power picked up by the recieve antenna - much more than the 3dB loss experienced by using a circularly polarized recieve antenna. With my powermeter, I've seen the signal easily go down by 10dB just by trying to feed a horizontially polarized recieve antenna with a vertically polarized transmit antenna.

Since the plane is constantly moving around I believe polarization mismatch contributes significantly more to intermittant video drop-outs than the half power loss from going to a circularly polarized receiver antenna.

In fact, the HAM radio guys use Helicals to track satellites. Apparently, the polarization changes caused by different sattellite orientations and attitudes creates a similar tracking problem to what we're experiencing here.

The other thing I should note is that I've had a hard time finding true circularly polarized receive antennas at 900 Mhz. The best I've been able to find is an "elliptical hemispheric" antenna made by Astron Wireless. I've taken the thing apart and it appears to be a turnstile-like antenna over a ground plane. The ellpitcal characteristics are due to a piece of phasing coax between two dipoles oriented at 90 degrees sitting over the ground plane.

The end result is that as the xmit signal moves closer to the horizen, it goes from being circularly polarized to horizontally polarized. This causes a significant drop in the received signal if I use a vertically polarized xmit antenna. I am not seeing as pronounced effect when I use Helicals.

Share this post


Link to post
Share on other sites
I think he said it was less than 5 watts

Thats a lot by our standards, only jet seems to use that sort of power :blink:

In that case though I see 20 miles is possable.

they have an arm that they lower down below the skids of the helicopter so their isn't interference with the helicopter.

I too have the aerial lower than the landing gear for the same reason.

With respect to the null on the whip, he said that when the helicopter is directly overhead, usually the distance is not that far and they don't need as high a sensitivity for the receive antenna.

Very true, I have reported this before, even with the aerials pointing out to the sides you have to overfly the 360 system very high to loose the picture.

However, if the plane changes orientation via turning and banking sharply, the transmit antenna will become more horizonitally polarized compared to the reciever antenna. This will cause a dramatic drop in the amount of power picked up by the recieve antenna - much more than the 3dB loss experienced by using a circularly polarized recieve antenna.

Thats why I fly with the co-pilot switched on and all turns are very flat and gentle. My flights are for exploring or still picture taking so there is no need for sharp turns.

The other thing I should note is that I've had a hard time finding true circularly polarized receive antennas at 900 Mhz.

Me too, not easy for 2.4Ghz either, even a helical is not 100% circular.

Keep your thoughts comming, I may yet take the patches off my old 360 system and try some helicals to see how it compares. :D

Terry

Edited by Terry

Share this post


Link to post
Share on other sites
Interesting, any chance you can direct that engineer this way? I'm sure we can all learn from him expertise! He might even end up enjoying a new hobby?

I forward him an email invitation to join ...

The satellite antenna is circular polarized too, so when it passes overhead, the polarizations are reversed. The simplest solution for our application is to not use a circular polarized antenna on the model's Tx. Otherwise, you would need to manually or electronically switch between left hand and right hand polarized Rx antenna sources.

I agree that it is not very useful to use a circularly polarized antenna on the transmitter. They tend to be large and cumbersome, and only are circularly polarized if you look at them on-axis, or head on.

The main reason to use a circularly polarized recieve antenna is to make sure that the recieve antenna can pick up both vertical and horizaontally polarized transmitted radiation. The price we pay for this is a 3dB loss in receive antenna sensitivity if it is fed by a linearly polarized source like a whip. A circularly polarized antenna fed by a circularly polarized transmit antenna, will not experience this 3dB loss penalty.

However, compared to the 10dB or larger losses one experiences trying to pick up a horizontally polarized signal with a vertically polarized antenna a circulary polarized antenna tends to provide a more consistant output - especially when the plane is moving around in a lot of different orientations.

However, as to how a helix switches from being left or right hand polarized, I'm a little confused as to how that might happen. I spoke with an Antenna engineer, and they said that a helix that doesn't have a ground plane, emits whatever handed polarization it produces out of both ends. For example, a right hand helix is right handed out of both the front and back. With a ground plane it produces whatever handed polarization mostly out of the front.

And if you look at a helix from the side, it goes from being circularly polarized to linearly polarized parallel to the dirction of the coil windings. So relative to the ground plane a helix is horizontally polarized when looked at from the side.

So, I can't figure out how a pair of satellite helix's changes their polarization. Any more details about the transmit/recievers system where this happens?

Share this post


Link to post
Share on other sites
So, I can't figure out how a pair of satellite helix's changes their polarization.
The polarization is configured by how the RF currents are stirred. On a circular patch antenna, this is influenced by the feedpoint location in relation to another RF current path that is out of phase. With a helix, it is basically a matter of the element's winding direction.

Any more details about the transmit/receiver's system where this happens?

The satellite communication hams that use yagis, flip their antenna 180 degrees (a simple solution since they often use a automatic tracking rotor). If this is not possible, they could switch to another antenna that has the correct polarization. Fixed position satellites (example GPS) are friendlier and don't need this trickery.

Share this post


Link to post
Share on other sites
a helix that doesn't have a ground plane, emits whatever handed polarization it produces out of both ends

That's odd I have never seen a helical aerial with no ground plane, the helicals I make all use a ground plane and the first 1/4 turn of the helix is also the matching section against the ground plane.

However, as to how a helix switches from being left or right hand polarized, I'm a little confused as to how that might happen.

I have only ever seen a helix wound either right hand or left hand.

Terry

Share this post


Link to post
Share on other sites
QUOTE 

a helix that doesn't have a ground plane, emits whatever handed polarization it produces out of both ends

That's odd I have never seen a helical aerial with no ground plane, the helicals I make all use a ground plane and the first 1/4 turn of the helix is also the matching section against the ground plane.

You are absolutely right, nobody makes or uses helicals without a ground plane. The ground plane is needed for impedance matching as well as to boost the signal gain from the antenna.

I brought up the no-groundplane thought exercise to try to understand, why in Mr. RC-cam's Satellite example, reciever antenna's handedness would need to be switchable to get a strong satellite signal. In particular, I was trying to understand how a satellite with a particular handed helical xmit antenna, can deliver opposite handed radiation to a recieve antenna. Since there is no way for a xmit helical's polariztion handedness to change, regardless of how you look at the antenna, it should always show up at the reciever with the same handedness. This would mean that you should not need a mechanical or electronic handedness switching system as the satellite moves across the sky when using a well pointed helical recieve antenna.

However, I suspect the reason for the confusion stems from the fact that most Yagi's are linearly polarized. In this case, the Yagi's orientation, or polarization, *would* need to be changeable. A linearly polarized Yagi would need to be rotatable 90 degrees to get it to line up with the satellites polariztion - even if the satellite had a circularly polarized helix xmit antenna - which produces both a vertical and horizontally output. But this is because a circularly polarized xmit helix will only emit horizontally polarized radiation from the side - not circularly polarized radiation. So if the Yagi was linearly vertical, and the satellites helical xmit antenna viewed from the side was horizontal, the signal would be very attenuated.

One should not have to change the orientation or handedness of a helical recieve antenna, if it is used instead of a Yagi, since a helical can recieve both vertical and horizontal components as long as it is pointed straight at the satellite - at the expense of a 3dB signal loss.

I sure hope the above makes sense, since this stuff is pretty hard to convey in words. :blink:

Share this post


Link to post
Share on other sites
In particular, I was trying to understand how a satellite with a particular handed helical xmit antenna, can deliver opposite handed radiation to a receive antenna.

Here is what the 19th edition of the ARRL Antenna Handbook says (Page 19-9):

"In working through a satellite with a circularly polarized antenna, it is necessary to have the capability of switching the polarization sense. This is because the sense of the received signal reverses when the satellite passes the nearest point to you. If the received signal has right hand circular polarization as the satellite approaches, it will have left hand circularity as the satellite recedes."

A popular web site that includes a note about flipping the antenna via the rotor: http://www.ultimatecharger.com/dish.html. Look for the section titled: Satellite Tracking with Yagis (then jump below that section title to the forth paragraph). BTW, from what I have been told, a lot of ham satellite communications are effectively done with only RHCP.

However, I suspect the reason for the confusion stems from the fact that most Yagi's are linearly polarized.

Generally speaking, a linear antenna can be coerced into a circular polarization. With the Yagi, the trick is to combine two designs together, with a phase delayed feedpoint on one of them. In the case of a Yagi for satellite apps, hams use a Crossed Yagi design, which utilizes this trickery. They are circular (otherwise satellite communication would not be effective). I suspect there will be a lot of construction projects on the web that describe how these are made. If you have the 19th edtion ARRL Antenna handbook, please see page 19-10, Figure 15 for details. Also, for a switchable RHCP/LHCP antenna example, please see page 19-12, figure 21.

I'm definitely not a practicing satellite ham hobbyist. What I know about it was derived from what I had been told by other hams and what I have read because of my interest in antenna design. So, if there are technical holes in my claims, I apologize in advance.

Share this post


Link to post
Share on other sites
  Here is what the 19th edition of the ARRL Antenna Handbook says (Page 19-9):

"In working through a satellite with a circularly polarized antenna, it is necessary to have the capability of switching the polarization sense. This is because the sense of the received signal reverses when the satellite passes the nearest point to you. If the received signal has right hand circular polarization as the satellite approaches, it will have left hand circularity as the satellite recedes."

Mr. RC-cam, all I can say to the above on this great Thanksgiving day is "gobble, gobble, gobble ..." As the big Thanksgiving Antenna Turkey, I am certainly confused how a certain handed xmitted polarized signal becomes the opposite handed polarized signal simply by having the Satellite move across the sky. :unsure: However, the day I start believing my own personal theoretical speculation over the 19th edition of the ARRL Antenna Handbook is the day I know I've changed one too many diapers ... and should start wearing the old ones on my head ... and taking a number of deep breaths, etc...

Also, in support of your point, if you look closely at the picture of the AvalonRF diversity reciever, http://www.avalonrf.com/Products/dvreceive...ingantenna.html, you will see that they have wound the Helical Antennas both left and righ handed on the receiver. This would hint at a kind of handedness polarization switching ability in the unit itself - although I see no description of it in the white paper - and the engineer didn't mention it when I spoke to him.

At any rate, I based most of my speculation on a number of conversations I've had with Dr. Dave Clingerman over at the Olde Antenna Labs - http://www.hamtv.com/oal.html. He is an old time HAM who has made almost every type of Antenna for everybody from SETI to NASA. If you look closely at the web site, he makes both Helical Antennas as well as phased Circularly polarized Yagi's of the type you mentioned in your last post.

He was the one who brought to my attention the certain handedness polarization emitted from the helical. So I am truely stumped as to why it is different as the satellite moves across the sky, and will have to wait and talk to him after the Holidays to find out what's going on.

Meanwhile, gobble, gobble, gobble ... :rolleyes:

Share this post


Link to post
Share on other sites
you will see that they have wound the Helical Antennas both left and righ handed on the receiver

It seems to me that thay are all the same handed but there is a groove so that they can be wound the other way.

Terry

Share this post


Link to post
Share on other sites
I am certainly confused how a certain handed xmitted polarized signal becomes the opposite handed polarized signal simply by having the Satellite move across the sky.

It may all be a wives tale. :)

Here is a simple Thanksgiving day experiment. Spin your hand/arm in front of you in a clockwise rotation. As you do this, move it over your head and continue until it is behind you. Keep it spinning. Now look over your shoulder at the rotation. It will now be counter-clockwise.

That analogy is the best I can do. But, it does seem to support the theory.

Share this post


Link to post
Share on other sites

OK, so here is an update from the Antenna Turkey. (BTW, Mr. RC-CAM I flapped my arms around on Thanksgiving day trying your polarization experinment until my wife threatened to put me in the oven and serve me for diner.) ;)

Also, before I get into this arcane, tedious, and probabally marginally useful technical communicade, I must say that it only goes to show that even the experts don't agree as to what is going on with Antennas. So, if they can't agree, we are all probabally doomed to long and painful conversations about the relative merits of different antenna configurations. :ph34r:

Anyway, I finally spoke with Dave over at the Olde Antenna Labs. His short answer was "Whoever wrote that must be smoking something." (His words, not mine.) I forwarded him a copy of the ARRL references, and we tried to look it up. Unfortunately, he only had the 20th version of the Antenna Handbook. Most of the section is the same, but we couldn't find the following in the 20th edition:

Here is what the 19th edition of the ARRL Antenna Handbook says (Page 19-9):

"In working through a satellite with a circularly polarized antenna, it is necessary to have the capability of switching the polarization sense. This is because the sense of the received signal reverses when the satellite passes the nearest point to you. If the received signal has right hand circular polarization as the satellite approaches, it will have left hand circularity as the satellite recedes."

So perhaps it was omitted in the most recent update because it was in error. Dave also felt strongly on theoretical grounds that it could not be accurate. He said that the only time he knew that polarization handedness changed was on moon bounce Radio experinments. (This is when they emit a pulse from the earth and then listen for it to bounce back from the moon.) He said in moon bounce experinments, it does happen that a certain handed polarized pulse would come back from the moon with the other handedness. He said that for hitting certain passively reflected satellites, ie - Sputnik, that the bounce handedness would come back different.

However, for any originally handed circularly polarized transmit antenna, ie - those on normal satellites, it would show up on the ground the same way it was emitted. This is true, regardless of how the satellite is oriented. (Thus, we don't have to flap our arms anymore.)

To help lend credence to his view he said he had used satellite links 10,000+ times and not had to switch between left and right hand circularly polarized antennas as the Satellite moved overhead. Dave also said he was on the antenna design comittee for the recently retired OSCAR Amateur satellite and personally knew most of the authors who wrote the Satellite section of the 20th edition of the Antenna Handbook.

He said that if we could provide a list of the authors, he said he would try and contact them personally and find what was going on.

Thats all for now ...

Antenna Turkey, signing off ... Gobble, Gobble, Gobble :)

Share this post


Link to post
Share on other sites

Thanks for the update. The ARRL antenna book doesn't list the authors. But the books provides a contact email: pubsfdbk(AT)arrl.org

In case your contact or arrl needs to see the passage, I have scanned the page and posted it below (the area I outlined in blue is the section of interest):

post-2-1196304810_thumb.jpg

Share this post


Link to post
Share on other sites

Not just your copy Thomas it's in my 19th copy too, I agree though it dose not make sense to me either. Dose this mean you are ready to build a helical diversity system happy ?

Terry

Share this post


Link to post
Share on other sites

Here is an email that I got from Dave today about the mysterious switching polarization. I kind of understand it ... Gobble!

BTW, Terry, I am trying out various combinations of 900 mhz antennas to see which ones work best. I'm going to start another thread with this information once I get results. I am still leaning towards using Helixes though ...

---------------------------------------------------------

I've been thinking about what you were told and what your friend related to you about circular polarization changing. In essence it does not. The article may have been referring to the Oscar VII satellite which has a canted turnstile antenna system on it for a 2M downlink. In this case the satellite antenna would have exhibited circular polarization - left, right indifferent, not important. This satellite spun on its "Z" axis in order to keep all the solar cells illuminated equally.This was accomplished using two methods; a black on one side white on the other carpenters tape 10M dipole and by magnetically stabilizing the spacecraft using a bar magnet aligned with the "Z" axis. The satellite was sun synchronous. The canted turnstile elements (monopoles) protruded 45° out of the bottom or "Y" axis of the spacecraft. After launch and in a couple of week the spacecraft aligned itself with the magnetic lines of force of the Earth. At the poles it would tumble according to simple laws of astromagnetics. Now,on the polarization of the canted turnstile. The spacecraft was going to be turning on its "X" axis. If looking at the spacecraft coming toward a receiving station the polarization could be RHCP or LHCP it didn't matter. Upon going away from a receiving station the polarization would yes, appear to have shifted by 180° and this is why we all used two antennas to receive the down link - one RHCP and one LHCP or one antenna on which we were able to shift the polarization (be it circular) by 180°. It all came back to me in a blinding flash. We designed and built that spacecraft to have an active life of 8 years. It lasted more than that and when it finally died it came back to life 27 years later. If you have a way of checking the Keplarian Element or "Keps" you will see listed once 'again' the effimerides for AO-7. This satellite was a special case (an exercise in astrophysics) where dual circular polarization at the ground station was required to have complete horizon to horizon reception of this 'bird'. We don't do things that way any more. If playing with the satellites it is always good to be ale to switch polarization as some things can happen that without the ability to switch one can be locked out. Sorry, but my instant recall isn't what it used to be.

Regards, Dave...

Share this post


Link to post
Share on other sites

To add more confusion to the mix...

My 19th edition ARRL book is © 2000. The 2005 release of the ARRL antenna handbook repeats the claim about the polarization reversal from moving satellites. But in this edition it limits the scope of the effect to some LEO satellites.

In Chapter 19 (2005 edition) it says:

"In working through a satellite with a circularly polarized

antenna, it is often convenient to have the capability

of switching polarization sense. This is because the sense

of the received signal of some of the LEO satellites reverses

when the satellite passes its nearest point to you. If the

received signal has right-hand circular polarization as the

satellite approaches, it may have left-hand circularity as

the satellite recedes. There is a sense reversal in EME work,

as well, because of a phase reversal of the signal as it is

reflected from the surface of the moon. A signal transmitted

with right-hand circularity will be returned to the Earth

with left-hand circularity. Similarly, the polarization is

reversed as it is reflected from a dish antenna, so that for

an overall RHCP performance, the feed antenna for the

dish needs to be LHCP."

To my surprise, I found the entire chapter 19 of the 2005 ARRL book online. Overall, there is some nice information on circular antennas, so it is worth looking at:

http://yb1zdx.arc.itb.ac.id/data/OWP/arrl-...ook-2005/19.pdf

From what I gathered by reading about the antennas on the LEO satellites (via Google), it appeared to me that they tend to be linear designs and not circular. So, the issue we are bantering about just won't occur in those cases. I saw that some amateurs that worked LEO satellites were also using linear antennas on their station too (to save $); they merely changed the antenna's orientation to maintain the RF link.

From what I am getting out of this is that with satellite work, it does not seem to be an issue on the practical level. In regards to the conflicting polarization reversal information, where the Tx is moving about, it probably deserves some bench experiments. But, I just don't have time to build the cross linear antennas to do that (especially since I have no need for one). But, for sake of knowing more, maybe one day ...

Edited by Mr.RC-Cam

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×