# DX reciprocity?

Discussion in 'General Technical Questions and Answers' started by Pushraft, Nov 20, 2008.

Not open for further replies.
1. ### G3TXQHam MemberQRZ Page

You seem to me making the mistake of thinking you can change the launch angle by tilting the beam - you can't! The height of the antenna determines the launch angle.

If you are prepared to do some reading take a look at this page on my web site - it should give you an insight into the "mapping" between antennas at different heights and typical angle-of-arrival statistics:

http://www.karinya.net/g3txq/hexbeam/height_2/

73,
Steve

2. ### PushraftBanned

I dont understand how anyone can say for sure what the DX path is and that it is identical in both directions. How would you know that? There are many paths it can and does take so why would it be identical both ways? What about ducting? Is it possible that ducting occurs moreso in 1 direction than in the other because they have different entry, exit points? I am rather new to propagation but I would not expect the paths to be identical thru an atmosphere that is constantly changing.

Also, saying that electricity takes the path of least resistance is misleading and incomplete. A simple example is if you have a 12V battery and put a 100 ohm resistor and a 450 ohm resistor in parallel across the battery terminals. Using that wording, one might think current will only flow across the 100 ohm resistor because that is the path of least resistance. However it is clear that current will flow across both resistors just in different amounts.

It would be better to say electricity takes ALL paths of resistance but even that is misleading and probably wrong. That is a different discussion I dont want to get into here. This thread is about DX reciprocity.

3. ### PushraftBanned

A couple more things here. The 20dB hop loss was just an example, it could be much less such as 4dB if it was over ideal "ground". Actually it could change so those simple formulas using constants wont hold. If it was 4dB then your formula would favor the double skip, not the single. Also, I could "play" with the antenna loss numbers so it would favor the double skip. We are just guessing here. I would like to know what really happens.

Also, you can change the angle of maximum radiation of an antenna by tipping it. This can be modeled too. Many people just point their beam along the horizon but tipping it up or down does change the pattern. Why do you think some antennas come with a downtilt adjustment? Why do they sell tilt adjusters for antennas? In one of my ARRL books, they even did a test for HTs in which tilt was considered. I think the conclusion was if you know the location of the target you are communicating with, it is better to face them and slightly tip the transceiver in that direction (in other words, down towards the ground in the azimuth direction of the target).

The antenna tilt angle does not exactly relate to the tilt of the pattern but it is somewhat related. For example, if you tilt a little bit down the pattern should be lower but if you tilt too much down, too much of the signal will bounce off the ground and be reflected upward therefore raising the pattern.

4. ### G3TXQHam MemberQRZ Page

Push, I don't wonder that folk get frustrated trying to help you! It seems that, once you get an answer you don't like, you "change the rules"!

The comment you originally made which I tried to reply to was:

There may be non-reciprocity, but your take-off angle/geometry argument does not explain it, for reasons I've already outlined.

Muddled thinking again. I have never said the signal follows only one path - I have been careful to say that there is a "dominant path". Just like in your example the dominant current is flowing through the 100 Ohm resistor.

And now you've introduced this example, try reversing the battery and see if the electrons prefer to take a different path

73,
Steve

5. ### G3TXQHam MemberQRZ Page

Push, you're missing the point again. Of course the figures can favour the double-hop path! But if they favour it from A>B they will also favour it from B>A - in other words the paths are reciprocal.

I'll set you a challenge. Take the simple model we have just been using - don't introduce any furher "complications" - and give me a set of numbers that will make the A>B attenuation greater than the B>A attenuation.

Steve

6. ### G3TXQHam MemberQRZ Page

Push, I'm learning a lot this afternoon. Can you give me the URL for an HF antenna with a downtilt adjustment - I don't think I've ever seen one. We are talking about HF ionospheric propagation aren't we?

Steve

7. ### PushraftBanned

I'm not going to do that because the simple model doesnt seem like a good model of what is really happening. I just have a lot of trouble accepting that for say a 5000 mile total DX "trip", that the path in both directions is identical. I remember reading about this a long time ago. I suppose it can be tested by taking 2 nearly identical setups on each end of the link and comparing results. If one fades a lot more than the other or is much weaker on average... then something "fishy" is going on. That is, if you eliminate all possible differences except propagation and just observe the results, some conclusions might be drawn from that.

Some people are probably reading this thread and thinking "Push, who gives a crap we cannot control the atmosphere". Although DX works and quite well at times, I would like to know more about the actual 2 way paths just to satisfy my curiosity.

Also I didn't state who stated electricity takes the path of least resistance so I wasn't pointing fingers. I was just saying the wording is not good (not descriptive enough and is misleading).

I noticed you didn't "correct" me about the tilt so are you now agreeing with me about that?

8. ### G3TXQHam MemberQRZ Page

Push - you see what I mean about changing the rules? You proposed a simple model to illustrate a point - I pointed out a flaw in your logic (not the model) - now you say you don't like the model. Are you incapable of admitting your logic was wrong?

At this stage I give up - my head hurts and there are dents in the wall. Over to someone else

73,
Steve

9. ### AB1GAHam MemberQRZ Page

The comparison to electricity is actually not that misleading and yes, you need to go there, because the effect is similar.

If you place two electrodes in salt water the conductive path is not a narrow straight path between the electrodes; rather, the current density tapers off the more you leave the straight line, and there is no boundary where the current goes to zero.

The ionosphere is a similar animal, but for the first order analysis you have to assume time-invariance, i.e., let's ignore fading and scintillation for now. You can look at the ionosphere as a transmission medium which is coupled to the transmitting and receiving antennas. This coupling depends on the geometry of the antennas and the composition and component distribution of the ionosphere.

The antennas at each end of the path do not emit infinitely narrow beams, but rather an angular distribution of power into space, but you can imagine the emission as being made up of an infinite number of infinitesimally small rays, each with an intensity that depends on the antenna pattern.

Let's follow one ray. It carries a fraction of the total emitted power as it leaves the antenna. This ray is then scattered by the ionosphere; if the ionosphere is behaving, the scattering isn't random, and most of the power is reflected (or more precisely, refracted) through some angle. At some point the ray either misses the receiving antenna or interacts with it. If it interacts with it, the amount of power collected by the receiving depends on its pattern.

What you can do for one ray you can do for all of them, so by summing up the effects of all the rays you can, at least theoretically, calculate the expected signal given the power, antenna patterns, and mathematical model of the ionosphere. (OK, practically it's impossible, but so what?)

Sorry for the long lead-in, but here's the crux of the argument: All of the processes described above are reciprocal. The coupling between the antenna and free space is reciprocal, as are the scattering processes that lead to refraction and reflection. That means that each pair of rays, from point A to B, and point B to A, follow the same path. Which means that the sum of the effect of all rays is also reciprocal.

Another thought experiment. You have two mirror and a friend handy. You set up the mirrors so that you can see your friend and he can see you. Now move so that you can't see your friend any more. Is it possible for your friend to still see you?

10. ### PushraftBanned

I'm trying to find the post that someone made where they describe how they were working a sporadic E layer and the one person had a tilt adjustment. He tweaked it until the signal was peaked and the angle of tilt was 27 degrees. That is, with the beam pointed up 27 degrees from the horizon, the signal was best between them. I dont know if I interpreted this correctly but it seems like experimenting with this could yield some very interesting results, especially if someone is normally in too close to you for DX or between hops. Of course with large beams this is very impractical but it should change the skip angle... why wouldn't it?

To say that tilt angle on an antenna has absolutely no effect on the pattern seems preposterous. For example, if you had a vertical antenna on a mast on your house and the wind knocked it so it was diagonal, are you saying the elevation pattern of that antenna would be the same as the vertical one?