Diversity Operations

Bill in Georgia
Posts: 51
Joined: Sun Apr 09, 2017 8:03 pm

Diversity Operations

Postby Bill in Georgia » Sat Apr 15, 2017 11:45 am

In general QSO's with other Anan owners, the subject of the hows and whens of the Anan Diversity comes up.
and really unless I haven't done my homework, there is very little written about the 1-2-3-4's of setting up a system. Maybe whoever wrote the program has the instruction manual hidden off on a web site somewhere..
Ant body with a little help on the subject??
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Sat Apr 15, 2017 1:27 pm

Good idea, Bill! I will make this thread a "sticky". Here's what I've posted in the past, perhaps it will be of help...

Diversity can only be achieved using radios that have two separate ADCs. Currently this would be the 100D, 200D and 8000.

In order to achieve any kind of non-coherent diversity, be it amplitude, polarization, or spatial (multipath), or to achieve two element coherent beamforming, each receiver must use a distinct antenna. Therefore you need to assign RX2 to ADC2, and an antenna that is not the main antenna needs to be attached so that it feeds RX2. On the 100D and 200D that means the "RX2" connector on the rear panel. On the 8000 that means attached to the "ADC2" connector on the rear panel.

For non-coherent modes (spatial, polarization, or amplitude) one does not use the so-called "diversity" mode of the software. Instead you want to simply adjust the audio volume and pan controls as desired. A typical example would be using a Beverage antenna on RX2. Obviously this is more difficult when using VAC, as both VAC1 and VAC2 must be fed into an external audio mixing application like Voicemeeter Banana or a DAW application.

For coherent beamforming you would, of course, use the "diversity" mode of the software.

I have yet to find a reference that explains coherent, multi-element beamforming that does not involve piles of math. And I've also failed to find a good, simple reference that deals with only two antenna elements, which is how many we are restricted to with our radios. However, if you can stand a lot of math this is a good one:

https://scholar.valpo.edu/cgi/viewcontent.cgi?filename=32&article=1000&context=engineering_oer&type=additional

And it comes with an accompanying video:



This one is a bit more intense; remember that in our case N always equals 2. Remember also that very few of us have a matched set of antennas we are using for coherent beamforming and that will totally distort the resulting beam pattern:

https://courses.cit.cornell.edu/ece303/Lectures/lecture33.pdf

(Note: originally I had referenced this link, https://www.ece.nus.edu.sg/stfpage/eleh ... Arrays.pdf, and it was a very good one that dealt specifically with two element radio arrays. Sadly, the link is now dead.)

See also Page 149 of:

https://github.com/TAPR/OpenHPSDR-wdsp/blob/master/WDSP%20Guide.pdf

From a more qualitative perspective, consider the following:

1. You cannot null out noise because the phase relationship between noise at the two antennas is random.

2. You can null out an interfering signal, because the phase relationship of the interfering signal at the two antennas is NOT random.

3. There is often a language problem when discussing (1) and (2) above. Interfering signals are often called "noise", as in "The electrical noise from my air conditioner is awful." Understand that the so-called "noise" from the air conditioner (or a bad wall wart, or whatever) is an interfering signal that is emitted from what is essentially a point source. Do not confuse this with atmospheric noise, which is random and comes from a diffuse source.

The trick then is to design an antenna array such that you can a) steer a beam in the direction of the signal you want while b) simultaneously steering a null in the direction of the interfering signal. Designing such an array with only two elements that will work over all frequencies and at all combinations of angles is not possible.

In my case I happen to have a very noisy air conditioner. While I did not design an array, so to speak, I did erect an antenna close to the air conditioner such that it normally has a much larger response to the interfering signals than my main antenna does. Then, when using diversity (beamforming), this usually works quite well to allow me to null out the interferer at the frequency of the station I'm working.

Note also that, assuming you do have a good setup for nulling the interference, you will see on your spectral display that it generally only nulls around a small bandwidth at any given time. It is hard to get a good null to be many tens of kilohertz wide because the phase relationship changes by a substantial fraction of a wavelength vs. frequency. Hence you adjust to null the interferer at the frequency you are working, not across the entire spectrum.

Finally, there is definitely some technique to doing this efficiently. My workflow is as follows:

1. In the diversity control panel I first select RX1 and RX2 in turn to ensure that both receivers can see the interferer, and that the interfering signal is the same amplitude or larger in RX2 (this assumes RX1 is your main, desired receive antenna). If you do not see that this is the case, then you very likely will not be able to null the interferer.

2. While at a fairly large amplitude (near the outside of the circle) I adjust phase to find a minima on the interferer.

3. I adjust amplitude at that phase to further minimize the interferer. You usually want the smallest amplitude that works.

4. I make very fine adjustments around the gross set point found in (2) and (3) above.

73!

Scott/w-u-2-o
Bill in Georgia
Posts: 51
Joined: Sun Apr 09, 2017 8:03 pm

Re: Diversity Operations

Postby Bill in Georgia » Sun Apr 16, 2017 5:08 pm

Thanks, the best approach I have seen thus far...will give it a try..I have antenna galore so it should work..

-a Hytower..6 to 160, with over 100...,80 foot radials, sitting a-top a underground lake.
-a 5 element 20
-a 4 element 20
-a 12 element log
...am finishing up..(this afternoon), DX engineering, NCC-2 phase system. I set it up on 40, with two preamp'd 102 inch whips..1/4 wl separation, mounted ground level on a 10 foot ground rod and 8 radials each. and fed by RG6 at 1/2 wl drop.
So we shall see....
Thanks.
-Bill - KC4PE
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Mon Apr 17, 2017 12:15 am

Jealous of your antenna farm!
User avatar
n1eu
Posts: 2
Joined: Fri Mar 31, 2017 4:44 pm
Location: Albany NY
Contact:

Re: Diversity Operations

Postby n1eu » Mon Apr 17, 2017 2:30 pm

I'm not totally clear on Scott's description of non-coherent diversity. I think what we need to be talking about for the "other" diversity mode is "stereo diversity," which is what it's called on the PowerSDR front panel ("SD").

MultiRX is not the way to do stereo diversity. MultiRX controls are for multiple slice receivers off RX1 (ADC 1) and not for RX2. What you do for stereo diversity is push the RX2 button, and then a button "SD" appears at the bottom central of the main screen and you click "SD" to engage stereo diversity. This will run RX2 off of VFO A along with RX1.

I'm working off of memory, but I think this is right.

73, Barry N1EU
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Mon Apr 17, 2017 3:42 pm

Barry,

Non-coherent diversity basically is any method that causes one to choose between two (or more) signal sources based on some metric of signal quality, usually which one is the strongest. The S-meter, your ears and your brain make the choice. How you get that info to your ears/brain is up to you. You can use straight MultiRX, or you can use so-called stereo diversity.

When RX2 is active, it is controlled by VFO B, and that is the audio you hear on the RX2 audio slider when using MultiRX (or stereo diversity).

73,

Scott
User avatar
WA0VY
Posts: 83
Joined: Sun Apr 09, 2017 3:41 pm

Re: Diversity Operations

Postby WA0VY » Sun Jul 09, 2017 9:49 pm

Has anyone tried using a rotatable magnetic loop in a coherent beamforming diversity arrangement? I have some RFI sources that I simply cannot find, mostly because they are intermittent and there is only so much help I can expect from my local energy company. My current array is wholly inadequate for this. I have a very directional multi-band yagi (OB 13-6) and a vertical, located fairly close together. The vertical never receives the interfering signal at a greater amplitude than the beam except when the beam is pointed away from the source, in which case I don't need to null it. I am contemplating replacing the vertical with a rotatable loop but not sure that will yield better results. hopefully, someone has tried that and can report.
73 Brent WA0VY
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Mon Jul 31, 2017 2:15 am

Brent,

I think you are going to have to be the first one to try a rotatable magnetic loop ;)

The downside to that approach is that magnetic loops are very inefficient. Any antenna that is physically small compared to the intended wavelength is inefficient by definition.

The upside to that approach is that you can use the directivity of the loop to better select the interference in question.

Another upside is that you can probably put the loop closer to the local interferer than the tower mounted antenna, assuming the source of the interference is on your property.

The bottom line is that the signal level of the interference in question must be equal to or larger on the "noise source) antenna than that obtained on your main antenna.

73,

Scott
AB2EZ
Posts: 99
Joined: Sun Apr 09, 2017 2:29 pm
Location: Princeton, NJ

Re: Diversity Operations

Postby AB2EZ » Mon Jul 31, 2017 2:52 am

Scott

I respectfully disagree with the last sentence of your last post.

If the interference component of the signal at the output of the magnetic loop is smaller than the interference component of the signal at the output of the main antenna ... one can just amplify the signal at the output of the magnetic loop antenna and/or attenuate the signal at the output of the main antenna ... in order to match the levels of the interference components.

What is important is that the ratio of interference to signal at the output of the magnetic loop be larger than the ratio of interference to signal at the output of the main antenna.

If the magnetic loop antenna is near the source of the interference... and oriented to pick up the interference... then it should work fine ... assuming the interference source is generating a magnetic field in addition to an electric field.

Note that an electric arc may not generate much of an RF magnetic field... even though it generates a local RF electric field.

Stu
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Mon Jul 31, 2017 12:59 pm

Stu,

I agree, thanks for adding that refinement: it is the signal to noise ratio that matters, and not just absolute signal amplitude, as you say.

73,

Scott
User avatar
WA0VY
Posts: 83
Joined: Sun Apr 09, 2017 3:41 pm

Re: Diversity Operations

Postby WA0VY » Mon Jul 31, 2017 2:16 pm

Thanks for the thoughts. It is difference in the magnetic and electric components of a signal that gives me pause on this antenna. All sources of truly annoying interference appear to be electric (power line or similar). I have an 80-10 OCF dipole that is still up in the trees but has been moth-balled since I installed my tower and yagi. It is also aligned in the direction of the interference. I will give that a try first. I tried the vertical, but as noted, it is worthless.
73 Brent WA0VY
User avatar
EA1DDO
Posts: 5
Joined: Sun Apr 09, 2017 4:21 pm
Contact:

Re: Diversity Operations

Postby EA1DDO » Tue Aug 01, 2017 6:24 am

Hi,

I think the non-coherent diversity system is similar the one called "voter" used in legacy VHF repeaters.

73, Maximo
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Tue Aug 01, 2017 12:10 pm

Probably the most prevalent use of non-coherent diversity is in car radios. It is easy to see the multiple antennas and antenna connections printed on the rear windows of most modern vehicles alongside the normal defroster printed wiring.

73,

Scott
VK3HJ
Posts: 2
Joined: Sat May 27, 2017 1:28 pm

Re: Diversity Operations

Postby VK3HJ » Mon Aug 21, 2017 3:51 am

Having moved house early this year, I am still setting up my antenna farm. HF is pretty well established with beams for 40-10 m up and running.

Low Bands antennas are still to launch. I will be using a top loaded vertical to transmit, but my choice of receiving antenna is still to be made.

I can set up several Beverage antennas in favoured directions, but having a newish ANAN 200D, I would like to explore the possibilities of using the two phase-coherent receivers to my advantage. Local noise is not a problem. My nearest neighbours are several hundred metres away.

What practical electronic beam-steering antenna elements are being used with success?

Luke VK3HJ
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Tue Aug 22, 2017 2:26 am

VK3HJ wrote:What practical electronic beam-steering antenna elements are being used with success?

Luke,

I think most people are using beamforming (inappropriately named "diversity") primarily for noise cancellation from such point sources as internet devices, air conditioners, etc., myself included. To that end, most people are simply erecting an antenna that receives the noise at a higher level than their main antenna.

To use the beamforming capabilities of our radios and software for, well, beamforming, in a more precise fashion, I'd suggest that two, identical verticals are the easiest to model and control.

Again, I'd point you to this reference:

https://www.ece.nus.edu.sg/stfpage/elehht/Teaching/EE4101/Lecture%20Notes/Antenna%20Arrays.pdf

With two verticals you'd need to determine the optimum spacing and orientation for your purposes.

73,

Scott
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Diversity Operations

Postby K9RX » Wed Nov 08, 2017 8:39 pm

This is a late reply - I have tried using SD, Stereo Diversity and haven't been able to hear a difference using it vs. just using the individual antennas.

I have RX2 on using a separate RX array (4 element Hi-Z vertical array) as well as the xmit antenna, a rotary 144' high dipole on 80/75 or a vertical "T" on 160 .... I turn on SD .. I've played with the pan controls but just don't see any enhancement or actually any change at all in what is evident in the headphones (stereo of course) ... the signals, especially on 80, are similar in level (maybe 5 db difference overall) ... help?

Gary
K9RX
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Wed Nov 08, 2017 8:52 pm

Gary,

Stereo diversity is not a mechanism that results in any intrinsic improvement in the ability to receive signals. Stereo diversity is merely a mechanism that allows one to simultaneously compare a signal from RX1/Antenna A and RX2/Antenna B using the acoustical properties associated with putting one signal in the left ear and the other signal in the right ear. One then chooses the signal one likes the best and uses that. If both signals are essentially the same, then there is no clear winner. If the signals are different, you can't do any better than the signal which is the better of the two.

You can compare the signals in a less complex manner using RX1 alone by simply switching between ANT1 and ANT2 (for example). You are just not comparing them simultaneously, but sequentially. Personally I find this method more than adequate, and never bother with the complexity of using stereo diversity.

Finally, this may be obvious, when using stereo diversity it is important to groom the settings of both receivers such that they are identical, e.g. attenuation, AGC mode, AGC level, audio gain levels, noise reduction and blanking settings, etc.

73,

Scott
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Diversity Operations

Postby K9RX » Wed Nov 08, 2017 9:23 pm

Scott,

thanks for that. Yes I am aware of the principals and the expectation of results. I had a Flex6700 here for a while and I tried their SD and it had a noticeable difference to the point where it was a big distraction. Literally signals swinging left to right - which is what I had expected and what it delivered ... on CW it was too much to handle and still copy high speed code. BUT it did help at times. On the 6700 it was obvious ... here - nothing.

And that was using the exact same antennas I have now.

That stated possibly indeed its conditions and the signals aren't doing what they need to do to make them stand out as 'different'. I'll keep checking. And possibly I didn't adjust the pan quite right ... although it seems like I did play around with it. I was hoping there was something that I had missed in 'selecting' it.

thanks

Gary
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Diversity Operations

Postby K9RX » Sat Nov 11, 2017 1:58 am

Scott (et al),

So I found out what I was missing regarding Stereo Diversity! I have the, well A pan function mapped to the MIDI panel ... but its the one that when you slide it from one side to the other on the panel the result on the screen is the pan functions shown together one above another - one slides to the left - one slides to the right ... but this appears now to be only for RX B when using ADC1 ... i.e. its not SD (which I knew) but, and this is the important thing - it ISN'T how you pan for SD.

In Rob's video ... or was it N1EU ... anyway I noticed that of those dual sliders he had only the top one slid to one side, I think the left ... and then for RX2 he has that slider slid to the other side... when I do that well bingo bingo! DIVERSITY. I used it just now on the 3C1 on 160 meters and it definitely fills in that small amount needed to get a contact completed when the signal is marginal but seems to be 'swinging' from one antenna to another. I also went up to 75 meters and managed to work a very weak but readable, only with SD, 9U4M there as well. So it DOES work and does a good job.

Thanks - and as always thanks to the developers (this one goes WAY back I'm sure) for this functionality.

Gary
K9RX
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Sat Nov 11, 2017 2:31 am

I never assumed it was operator error ;)
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Diversity Operations

Postby K9RX » Sun Nov 19, 2017 4:12 pm

A follow up on my experience using Stereo Diversity (SD). It works very well. I use it all the time now on 80 and 160. I have on 80 a rotary dipole at 144' as the transmit antenna ... and on 160 a 90' vertical T with 36 120' radials. For RX I use a Hi-Z 8/4 vertical array which consists of 2 4 element arrays on a common center one turned 45° to the other and thus able to point within 45° and have no more than about 1 db drop at the cross over ... it works quite well on 80 ... not as much on 160 where I think I have interaction (re-radiation) from the rotary tower nearby or even the 160 TX antenna.

So the one issue is this. I can't find a way to both use SD and work split. It SEEMS to be there - if you enable Split/MutiRX with SD on you do get the second "blue" passband as you would normally get with SPLIT/MULTIRX on (RX2 off, I..e no SD). [see note below on how to do this]. I can then listen with SD on the slaved frequency (VFO's are synced when doing SD) ... and hear what would be my TX frequency ... the one thing that doesn't work is that I can't TRANSMIT on the TX frequency! No matter what I always TX on my synced frequency. I have sent off a note on this issue which makes working split and using SD impossible - at least in a manner that is transparent and what should be easy to use .... and I've offered a suggestion on an alternative way to implement this functionality.

Note I can use SD to receive and work split - but its cumbersome - the only way I've found to do it is to use XIT. The issue here is listening to the TX frequency in trying to find where the DX is listening ... since you can't SEE where you are it isn't readily apparent where you are ...using this to listen to the TX frequency makes a near impossible task to find where the DX is listening. Ideally the way it seems set up - i.e. being able to have SD AND MutlRX would be great...

if anyone has any questions I'm happy to offer what I can.

Gary
K9RX

USING SPLIT/MultiRX with SD as currently implemented (3.4.3): Turn on SD, VFO SYNC. Tune in station you want to RX. Set filters accordingly (I always have them set the same - don't know if there's any benefit to having them different and if so it would be nice when doing SD that 1) VFO-SYNC automatically gets turned on and 2) the filter width is the same on both VFO's. Now notice that below the VFO-A window there is a subwindow that normally says what this frequency is "selectively" defined as - say 80 Meter CW - now has a frequency in it! It is often way off of where you are which is why it always confused me when I saw it before - BUT if you right click the mouse, twice and get the red cross hairs - and now click in the panadapter where you want to TX - BINGO - that is the frequency it is displaying!! So you can't "tune" the TX frequency, you have to use this method to select it - but now being able to readily click on a signal that you think is working the DX allows you to hear them and confirm this. Again: the only problem, a killer obviously, is that it doesn't TX here!
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Sun Nov 19, 2017 4:18 pm

Just peeked at qrz: that's a hell of an antenna farm you've got growing there, Gary!

73!

Scott
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Diversity Operations

Postby K9RX » Mon Dec 04, 2017 1:32 pm

Thanks Scott.... a second tower yet to go if I ever get my home designed/built SteppIR DB30 done ... parts on order ...

So on Diversity - it works fantastic guys! I had a Flex6700 here for awhile and couldn't get their diversity to work were it seemed to help ... it was difficult to 'follow' as it would change from side to side too fast ... especially on SSB. But the PSDR mRX diversity seems to - hell it DOES work great. Note one change to my description of use above - you don't need, probably don't want "VFO SYNC" on! Only SD need be on. Be careful with the AGC - it works best when they're the same relative to that antennas baseline RX level ... and don't hesitate to bring the LEFT/RIGHT controls in towards center if one antenna dominates a little more than the other - take its level in towards the center, you can hear the effects in the headphone as the signal will move from that ear more towards center-head. No need to center it - just bring it in a little bit.

Note additionally the filtering in this puppy is incredible. I have, several times now, had DX stations on CW that were weak to extremely weak - and some idiot decides if he can't work them no one will and starts to tune on the DX frequency ... HOWEVER in some of these cases these guys are actually slightly off frequency - in a "normal" radio this wouldn't matter if they were off by 12 Hz or so ... but with THIS radio, actually software, it works a treat. I have my CW set to 8192 on the buffers to keep a very tight skirt .... and I can literally, using the SPECTRUM display, tune such that the offending station, even if they are 20 db stronger, JUST slides off the edge while the DX is still in the passband and BAMM! There they are again (near) perfect copy! Or at least unhindered by the offender. This is using a 33Hz filter setting, APF set to 7 and APF width set to 115 (any less and it starts to be counterproductive).

There's still things that need to be done when one turns on RX2. It is cumbersome for split operation ... and there are design flaws in how it is set up ... hopefully these can be addressed (they're on my list of items: Bugs, suggestions, anomalies)... take for example the yellow cross hair - it doesn't go across the barrier from RX2 to 1 ... so if you're listening primarily on RX2 using your dedicated RX antenna, which is the whole purpose here, and there is a signal seen there on RX2 but not on RX1 you can't hover over the signal over RX2 as it won't respond to a click to move the frequency here - you have to instead guess where it is on RX1 (likewise the cross hairs don't 'drop down' to RX2) and get it 'close enough'. Also the AGC line jumps where you go to grab it - there is an offset when you click on it. And a good number of other things - but not being able to easily do split is one that I hope can get fixed soon.

Anyway - anyone wanting to talk through some of these things please write me direct at the ARRL address.

Gary
K9RX
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Diversity Operations

Postby K9RX » Mon Dec 04, 2017 1:37 pm

Stereo Diversity:

I'd love to know what this is doing exactly ... I know the principal - phase locked so that you have signal phase coherence and thus "overlap" ... if using antennas that are different polarization, as an example, one can literally 'see' (hear) the signal as it curls in the 'ether' going from one antenna to the other ...

however I find there are times when I can mute RX1 ... often actually especially if the RX antenna is that much better than the TX, and I have done this test before - I can listen to a weak station and I swear I can hear that station jump out of the noise, or back in to it again, by hitting/unselecting the SD button. So this would imply there is something additionally going on since I'm only listening to one RX when doing this ... there is a - seems to be a definite (seems - definite?) improvement in S/N when SD is selected.

explanation please?

Gary
K9RX
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Mon Dec 04, 2017 3:55 pm

Stereo diversity is a non-coherent, non-beamforming mode of diversity. It cannot result in better performance than the better of the two antennas attached. No SNR improvement is possible over the better of the two antennas in this mode of operation.

From an old Flex Radio paper on the topic:

"To accomplish stereo or dual diversity reception, it requires two receivers that are frequency and phased locked on the same signal outputting the demodulated signal to two discrete audio channels. Optimally both receivers need to be virtually identical in hardware and configuration (demodulation mode, filter characteristics, preamp, AGC setting, etc) in order to create a sense of "stereo spatial depth" for the listener. Humans are very good at localizing sounds and quickly estimating their direction and distance because we have two ears that allow for multi-path reception of audio frequencies. The end result is by being able to hear stereophonically, our brains can use this information to extract intelligibility from weak and noisy signals."

That description is not entirely correct. It would be better to say that It is merely a way of combining the audio from two antennas in a way that allows your brain to automatically choose which of the two signals to listen to (which has the best SNR) at any individual point in time. Stereo diversity is best used to achieve an improvement in average SNR given time varying conditions that cause the best performing antenna to be continuously changing. Unfortunately, such conditions rarely obtain in amateur radio HF operations and with the typical sort of antenna combinations used by amateur radio operators for HF. Most antenna combinations exhibit gross differences in performance, e.g. horizontal vs. vertical, or dipole vs. Beverage, etc. However, if you had something like a turnstile array that provided both X and O mode reception on separate feeds then SD would really shine!

I suspect SD is far more valuable for those who are working at VHF and above with their radios. For HF it will generally not improve performance over simply choosing the best antenna. However, if you find the psycho-acoustics of it to your liking (because things do sound different when activated) then by all means continue to enjoy it!

73,

Scott
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Diversity Operations

Postby K9RX » Mon Dec 04, 2017 4:50 pm

as I had said I understand how it is suppose to work - and as such it most definitely could improve things if both antennas are used and say the polarization is changing and i have 2 antennas to take advantage of that - I can add to that contrary to your comment and suggestion it DOES work on HF - I know - I have experienced it nightly on 160 and 80 using both antennas ... I would however wholeheartedly agree that the antennas used matter a great deal.

BUT what I have also seen is improvement with RX1 muted .... so please - don't just say "its all in your head" ... implying some lack of objectivity on my part ... I can honestly say there ARE bugs in the software - there ARE anomalies in the software - I've had strange things happen ... I'm being open minded and working with the developers as I can and as they allow me (some of the recent additions/corrections/changes were from me) ... and I'm wondering IF there is something else going on. I AM trying to be objective. Lets try to stay on the same page - a better answer might have been "if indeed you are hearing something then I don't know what is causing it!" ...

Gary

what antennas have you used for SD on HF, specifically 80 or 160 where you've drawn this conclusion that it isn't effective ?
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Mon Dec 04, 2017 5:32 pm

OK, Gary: if indeed you are hearing something then I don't know what is causing it! ;)

Because there is no magic in SD. It's exactly the same as MultiRX except the receivers are phase locked.

I've never used SD. There is no point for me. My antennas are so radically different that there is no comparison, one is always substantially better than another. On 160, 80 and 60 I use an inverted L. On 40M a wire half square. On 30 and up a multi-band dipole. Although I generally listen to 40 and below on the half square as it is much quieter with superior SNR to the L. Once in a while on 60M the L will work well, but again it's glaringly evident, no SD required.

73,

Scott
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Diversity Operations

Postby K9RX » Tue Dec 05, 2017 12:07 am

My point was that I was trying to find out what it did - when one invoked SD, what does it do other than phase lock the receivers? Does it do anything to the AGC for example as that could account for the change. Note although the perception is I'm hearing an increase in S/N, albeit very small, it might actually be that when it is turned on its "right" and when off something else is different causing a loss of S/N! Thus my question on what is going on - I DO know that there's nothing magic to SD and you can't get any more than the max of any one antenna at any time...

and it might well be a coincidence relative to QSB - although I've 'heard' it a few times now prompting the question.

thanks for the reply.

Gary
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Diversity Operations

Postby w-u-2-o » Tue Dec 05, 2017 1:42 am

AFAIK it does nothing but phase lock the receivers. Everything else remains under the operator's manual control.
User avatar
EA1DDO
Posts: 5
Joined: Sun Apr 09, 2017 4:21 pm
Contact:

Re: Diversity Operations

Postby EA1DDO » Fri Jan 26, 2018 5:12 pm

Hi Scott, all,

I remember when you said:

As a group we are still not fully exploiting the synchronous receiver capability inherent in these radios. Instead we continue to rely on constructs such as so-called "phasing harnesses" or "array controllers", all of which use coaxial delay lines of fixed values to obtain a fairly limited set of phase relationships and therefore a fairly limited set of beam (antenna pattern) configurations. Examples include the venerable four-square array, as well as the stacked yagis of those seeking to control elevation angles. I look forward to a future where such "array controllers" are replaced by multi-channel amplifiers backed by SDR-based synchronous transmitters and receivers.


If we have, let´s say, two antenna (or more), and we place a SDR at his feed point. I can see one of the key points is to get them working in a coherent way.
Sharing the same oscillator is not doable because they will be located at some distance. So I wonder if we can get a central timing system, then feeding this sync to every SDR somehow.

I have no idea if something like "IEEE 1588v2 Precision Time Protocol" could be enought to achieve the target.

https://en.wikipedia.org/wiki/Precision_Time_Protocol

I am thinking in a central GPSDO connected to a IEEE1588v2 ethernet switch, then fibreoptic LAN to the SDR located at the antennas.

Check this small article:

https://turbofuture.com/industrial/An-Introduction-to-Software-Defined-Antennas

What are your thoughts?

Thanks

73, Maximo

Return to “Diversity Operations (PowerSDR & Thetis)”