Polarization of Signals and Ground Gain

The following discussion about the influence of the polarization of signals and ground gain took place in the moon-net. I found the subject so interesting that I decided to put together all the messages in this page.

What causes the so called “ground boost” when moon is low above horizon? Is this effect frequency and/or polarization dependent?

Ground boost, or ground-reflection gain, is caused by reflection from the earth. It is polarization-dependent in that the ground tends to reflect horizontal waves better than vertical. At VHF/UHF, it is not very frequency-dependent, but on receive the earth is noisy so at low elevation angles, you’ll pay a price in noise at 70 cm. At 2m, whether the noise increases at low angles depends upon the man-made noise level at your QTH.

On 5-Mar-2002 OZ1RH wrote:
Answered in my lecture on Ground Gain for eme, link from www.oz1rh.com

My findings are that the ground gain is not dependant on polarization. I have yet to find a time where there is a difference between horizontal and vertical and I do make it a point to check this every time I get ground gain (=very often on my setting moon). The same enhancement is always present in both planes.

Peter, SM2CEW wrote:
> My findings are that the ground gain is not dependant on polarization. I 
> have yet to find a time where there is a difference between horizontal and 
> vertical and I do make it a point to check this every time I get ground 
> gain (=very often on my setting moon). The same enhancement is always 
> present in both planes.

I agree that ground gain is not different for V or H polarisation. I afraid Peter’s statement can be misunderstood so I will try to make it very clear:

I have studied what happens when the total Faraday on my own signal is 0 (or 180) degrees. This means I have made observations when my own echoes are strong in the same polarisation that I use for tx and very weak or absent in the orthogonal polarisation.

What I observe is that if the moon is high, the echo H to H is exactly the same as the echo V to V. (strongest “O” in 2 minutes about +/- 1dB accuracy)

BUT, if the moon is low, so I send (and receive) well to an angle downwards that fits the mirror image of the moon on the ground I observe sometimes that the H to H signal is much stronger than the V to V signal. The difference is in the order of 10dB or more. About 5 to 10 minutes later the relative levels has reversed, V to V is very much stronger than H to H. ( Halfway between these two extremes the two polarisations are equal)

What happens is that the signal from the mirror image can equaly well enhance or degrade. When it is in phase it will add, but when it is at 180 degrees it will subtract!! (And cause ground loss)

When a H wave is reflected, the phase is changed by 180 degrees, but when a V wave is reflected, the phase is unchanged. Therefore, at the moment of max ground gain for H pol, there is maximum ground loss in the V pol and vice versa.

There is (theoretically) some difference between the ground reflectivity for H and for V. Practically I have not observed that. The difference is small at angles below 10 degrees with the type of ground I have around here.

Theoretically there is no limit for how much ground gain one can get. (Imagine a valley that happens to have parabolic shape!) Practically one can expect 3 to 5 dB which makes a really dramatic difference on own echoes for which the gain comes twice!

Ground boost is extremely useful for small antennas but already for a 4-yagi array the time it lasts is too short for it to be of much practical use. With a small antenna and variable height or variable polarisation ground boost can be used for an hour or two at a time.

Peter is right. There is no theoretical reason for difference in polarization. Simple question is: What will happen to the circular polarization –> decomposition to the linear vector ? how the phase shift will happened?

Thanks to Palle and Peter for their insights and further information. It’s interesting that results obtained using vertical polarity seem to differ. Probably that has something to do with differences in terrain and ground conductivity at the various stations. We can’t all have flat sea water (6 dB ground gain)!

There is no question that Peter and Paul are misleading the members of this list.

The theory is as follows:

“mirroring” (From standard textbook on electricity)

Assume a small charge Q at a small distance above a very large flat conducting surface. To solve Maxwell’s equations for the electric field above the surface one can place a mirror charge -Q (opposite charge) below the conducting surface and then solve Maxwell’s equations for the two charges with the conducting surface removed. The solution one gets above the plane from this method is the correct electric field from the single charge above the plane. Below the plane the field is zero of course. The mirror charge is a tool to solve the equations easier.

Now apply this theory to a horizontal antenna. Let us look at one of the electrons that oscillates back and forth 144 million times a second in a dipole that is at a height H above a very large flat conducting surface. To find the far field we replace the conducting surface with a mirror dipole placed at -H (below the ground) and we put a positron in the mirror dipole that moves synchronous with the electron. When we analyse the radiation from the two dipoles we find that the electromagnetic field has opposite sign because the sign of the charges is opposite. An electron moving to the left or a positron moving to the right both mean current flowing from left to right!

For a vertical dipole we do exactly the same. This time the result differs because when the electron moves downwards, the positron moves upwards. When looking at the dipoles from far away one will observe the particles move in antiphase so because of their opposite charges the corresponding currents are in phase.

When observing a dipole above ground at an angle that makes the distance difference to it’s mirror image zero or any integer number of wavelengths, the radiation from vertical dipoles will add while the radiation from horizontal dipoles will cancel.

This is not “just theory” I can not say for sure what “mistake” Peter did, but I have studied the phenomenon very carefully here. The ground where the reflections happen here is flat farmland. And I can very clearly observe a minimum (really weak signals) in one polarisation while there is a ground gain maximum in the other polarisation.

Maybe, for example, if there are many threes growing on the ground, the refractive index will be different for horizontal and for vertical waves. This means that the wave speed is lower for vertically polarised radiation. This could be the explanation for Peter’s findings.

I have offered to give my lecture on ‘Ground gain and radiation angle at VHF’ in Prague, but no answer yet from the organisers. The lecture might be a starter for the discussion, as I do not know too much about the Brewster angle. See www.oz1rh.com , please read the Word-document as the html version lacks some formatting.

Quote from Chapter 5 Polarization and ground effects: 
“For DX-ers on HF it is well known that the radiation angle of a horizontal polarized beam is a function of its height over the ground and ground conductivity has little influence on low angle of radiation. Things are different for vertical polarization. A vertical antenna needs perfect ground to give low angle of radiation. In fact is does not radiate at all below an angle called the Brewster angle. This angle is dependent of the conductivity of the reflecting ground. Check an antenna book on verticals for 80 m, the 80 m DX’ers know all about this. Does low angle of radiation interest us for VHF? Yes ground gain is happening in the range of 0-10 degrees.

Horizontal polarization
 Ground gain for one reflection 5-6 dB
 Ground conductivity Little influence at low angles
 Lossy ground Little less ground gain and not so deep nulls in pattern
 Good ground conductivity (salt water) Deep nulls in pattern, but peak almost uninfluenced

Vertical polarization
 Ground gain for one reflection Perhaps 3 dB
 Ground conductivity No radiation below Brewster angle 5-20 degrees
 Lossy ground High Brewster angle 20 degrees?
 Good ground conductivity (salt water) Low Brewster angle < 5 degrees?

It seems a good idea to use horizontal polarization if one wants low angle of radiation. Vertical polarization has less ground gain. A circular polarized signal contains two components: a horizontal polarized and a vertical polarized one. Ground reflection influence the two components differently as describes above and after ground reflection the signal is no longer circular but somewhat elliptical polarized at low radiation angles.”

Hello Palle,

I am certainly no expert in any of this, but I encourage you to spread the information about ground gain. Many folks seem to be confused about it. All I can tell you, is that on 6m EME, everything is so marginal to begin with, that at least one station MUST have ground gain, and of course, if both stations have ground gain, it is all the better.

I still receive inquiries from 6m stations, for example, who ask for suggestions about elevating their one or two small yagi arrays. Their single yagi pointed at the horizon acts like 4 yagis, and 75% of the time cndx are not favorable enough for a contact even then. Such stations certainly cannot afford to “throw away” such ground gain and elevate their antennas! In most cases, a single 9 element yagi pointed at the horizon has the effective gain of my 4 yagi 6m array, and since I hear my own echoes over half the time while pointed up at the moon, it should be just a matter of time until I can contact anybody who can hear their own echoes on the horizon.

Ground gain with a single yagi can be a wonderful thing! As so nicely illustrated by the charts on your website, a single yagi has muliple ground gain lobes, providing muliple chances at EME contacts as the moon moves in front of the different lobes. At least on 6m, it seems we need as many chances as we can get – HI! And I have been simply AMAZED at the difference it makes when an antenna points at a salt water horizon! The smallest stations I have worked on both 2m and 6m EME have had single yagis looking over a salt water horizon. The books may say that there are only a couple dB to be gained from salt water compared to average ground….but I am here to tell you that every last dB makes a HUGE difference when you are involved with a very marginal EME situation.

I found on 6m EME with my 21m long single yagi 21m above ground, that my third lobe up at 18 degrees was the same gain as my free space gain (when it was elevated). By leaving the antenna pointed at the horizon, I had significantly more gain in my first (4 degrees) and second (11 degrees) ground gain lobes, and I was able to make a number of contacts using the second lobe. I might also point out that I have noticed that, especially on 6m, there are many factors that can “interfere” with the propagation of the signal to the moon and back. Most of these factors (ionospheric scattering and refraction, tropospheric ducting, etc.) are more of a factor the lower the angle of radiation (and the more atmosphere/ionosphere the signal has to pass through). On a number of occassions I have noticed good signals from horizon stations when the moon was high enough to be up in front of their second ground gain lobe, but I have heard NOTHING when it passed in front of their main antenna lobe.

Long live the big single yagi pointed on the horizon, with its higher ground gain lobes! Good luck in spreading more information about this very important subject! MNI TNX and VY 73, Lance

Leif, When I get groundgain I beam out over water, a very far stretched lake, nothing is obstructing the path. I do not se the cancellation you see in the opposite plane. When I have GG in one plane I see the same thing in the opposite. Ever since you mentioned this, quite a long time ago, I have made it a point to check as I had not seen the same results. Never have I seen the effect you see. It could be due to the fact that I use 3 yagis high in my stack and have a narrower lobe than you, hence see ground gain at a lower elevation (typically start seeing it at 3 deg and lower).

So, if I am misleading the group because of the real world not matching the text book, I am sorry to say that I will stick to my story.. Hi !

Hej Leif,

Thank you for very good explanation. There is still something I do not understand.

As far as I understand your explanation lead to the conclusion that interference pattern for horizontal and vertical polarization will be different. This is for the case when one dipole is excitating electromagnetic field. One dipole has omnidirectional radiation pattern. For large number of dipoles it will be superposition of the EM fields and the pattern will be more complicated depending on the distances between dipoles ( field sources ) versus wavelength. It is even more different if the magnitudes and phases of those sources are not equal.

In case of ground gain I am thinking about reflection of the wave radiated from high gain antenna having almost plane phase front. Reflection occurs at the large distance from the antenna and the surface of the reflector ( ground ) is also large versus the wavelength. The antenna can be very low above the ground or very high above the ground. There will be also space attenuation in near field between antenna and reflector ( ground) which has to be considered. This will generate some additional sidelobes which levels depends on the height of the antenna above the ground. In this case I would rather apply optics law. Assuming ideal reflector there, will be no difference for polarization. But the ground – soil is not a perfect conductor, so it could be some difference depending on the ground properties. Salt water is much better almost ideal for this case.

Long time a go at my first work ( back in Poland ) I had to do with the satellite C-band feed systems for 32 m dish using so called light-guide which is similar to the waveguide, but the dimensions are much larger then wavelength, about 2 by 2 meters. This arrangement was at the base of the antenna , guiding waves vertically. There was one flat mirror at the top, changing angle with elevation of the dish. This reflector fed tapered circular waveguide which ended with corrugated feed horn illuminating Cassegrain subreflector. The primary feed at the basement was radiating on flat metal surface with very small angle causing reflection in some degree similar to the case of the ground gain. The feed had two linear polarizations one parallel with the light-guide surface and second orthogonal. Those two signals were amplified and combined in order to obtain 4 different polarizations: Linear horizontal, linear vertical, circular left and right.

Ohio State University Radio Astronomy Observatory had large antenna ( Big Ear built by Prof. John Kraus W8JK ) which consists of the feed system illuminating fixed mounted sector of the parabolic reflector ( looks like a wall ) about 250m wide and 40 m high ( I do not remember exact dimensions). This reflector focused the beam and next illuminate large flat reflector which was tilted in order to obtain required elevation angle. From the beginning the main feed system consists of the pyramidal feed horns. But the problem was the noise temperature picked up from the ground. Then they modified the antenna: the surface between feed and parabolic reflector was covered with sheet metal and what is important here they modified the feedhorns in order to accommodate radiation pattern to the ground gain. And again they use all four polarizations: linear horizontal, vertical, right and left hand circular. Unfortunately I learn that this instrument does not exist any more and there is a golf corse now at this place ( generating income).

Similar radio telescope exists in Nancay France built by Hevre’s F5HRY father. The difference was in ground reflector, Nancay instrument has some metal mesh screens at he certain angle to the ground in order to lower ground noise contribution, but not use beam forming obtained the same way as in Ohio.

My own observations are based on 432 MHz EME experience. I never could figure out the difference in ground reflection for linear polarization horizontal or vertical on 432. The difference was caused by the Faraday rotation. Some times I had echoes on opposite polarization or distorted to elliptical, but I never detected weaker echoes on transmitted vertical compare to the horizontal polarizations. Ground reflection has to do with the reflection properties of the ground, which might be different for different frequencies..

I am maybe wrong, this is very interesting subject. All comments are very much appreciated.

I wonder if the ‘ground gain’ you are seeing Peter is really the ground or just ionisperic focusing (scintillation). My findings indicate that single yagis or equivalent get good ground gain but once you stack except in rare terrain situations, you won;t get ‘ground gain’ because the H plane pattern is so narrow that the antenna pattern never gets to the ground at a point where ground gain is the result. i have never seen ground gain with any eme array and that includes my 6m array of 4 x 6m7JHV yagis wher the bottom yagis are just 10 feet ( half wave) over ground and the top pair were at 35 feet. echoss are the same from straignt up all the way down to moon dissappearance over the horizon. my 2 wavelentgh Yagis at 45 feet high however gets good ground gainat 4 to 6 degrees and never lower. i have repeated this 6m result at three different locations.

Polarization is defiinitely a factor in ground gain. Horizontal antennas see the ground as a reflecting surface paralell to their polaration. Vertical antennas see the earth as a cross polarized element so ground reflection is minimal. Brewster angle effects may cause some vertical polarization enhancement.

These effects are what I and others have noticed and do not necessarily agree with the computer analysis of the system. However, moon echos are a pretty good sign of where your radiation angle is. this is one area where I feel the computer simulation falls apart.

Hi Mike,

Remember my old 4×14 at 5.2m stacking distance?

Sidelobe level was only -8 or -9 dB. I had very good ground gain when the sidelobe hit the moon via ground reflection while the main lobe hit the moon directly:-)

Overstacking does not reduce the gain, it creates strong sidelobes that can be useful some times….

Hi All,

I do not have any fancy formulas or even real world ground gain experience, but I read a lot and figure things out. I suspect that both Leif’s formulas and Peters experience can be reconciled. For the low yagi near flat land, it seems obvious that distinct lobes will be present for various yagi heights and Leifs assesment makes perfect sense even though I do not FULLY understand it. What also seems logical to me is that an array with a very narrow vertical beamwidth looking way out on to a lake will have many splinter sized lobes due the great number of wavelengths distant to the reflection point. These lobes, logically would tend to converge so there is effectively no difference with polarization. Since water is a great reflector, the brester effect is no factor for vertical polarization. Both Leif and Peter makes perfect sense. Hey Leif, what happens to 45 degree polarization signals? Is the peak between vertical and horizontal? Peter, I wonder how high your array is and how far the lakes edge is from your array?

For Mikes 4 yagi example, I think since the upper two yagis are at a very different height compared to the lower two yagis, the resulting maximum lobe is very different for the two sets, so there is never a chance for the two lobes to converge. I suspect that very high 4 yagi arrays can show ground gain, but with very narrow elevation lobes.

I’ll be QRT now in search of a country home on the west side of a lake.

Jim, to answer your questions:

The bottom two antennas of my 6 stack are 5 mtrs over ground and the top two are at 14 mtrs. Distance to the lake is about 200 mtrs. I wish everyone could have a moonset or moonrise lake..

My secret is that I go out there and salt the water before every moon session.. :-))

Having heard that using vertical polarization results in reduced ground gain, I would be interested in hearing comments from those of you that can switch your polarization between v&h.

I just put up a small 6m eme array: http://www.mtaonline.net/~nl7z/6m%20eme.html

The reason for the vertical polarization is to reduce the spurs I get from local tv ch2.


I have compared horizontal vs vertical echoes on 144 MHz when ground gain has been present hundreds of times and I could never see any difference. Both polarities have been enhanced equally. I know that some will air different opinions, but on this I am very sure. My setup is 6 x 19el xpol v/h. Same thing when I was using my 8 mtr dish and a rotable feed for 144 MHz.

Hi Peter and Kevin,

I have studied the ground gain on 144 MHz and found that it is similar for V and H polarisation. Ground gain does (of course) not occur simultaneously on H and V, when one has a maximum, the other has a minimum. Using the correct polarisation while the moon is rising makes a very large difference 🙂 (I am sure Peter can see the difference time-wise. When one polarisation has ground loss so it nearly disappears, the other polarisation is strong.)

If V pol helps against QRM I think it is a good idea to make use of that. Maybe it will be clever to mount the antenna nearer to ground to make the ground gain last longer while loosing the second lobe. Guess it depends on the local ground properties.

On 17-Sep-2004 SM2CEW wrote:
Leif, we have been over this before, ground gain IS present at the same time in both polarities, absolutely so!

And since it was up for debate I have been checking even more closely, and the ground gain pattern is exactly the same for both vertical and horizontal over time, or call it moon position over the horizon.

I suspect that may be due to the polarity rarely being all one way or the other, but some of both almost all the time, as Lionel has observed over the years with his rotatable (not switched) array.

Just a thought.

To be sure about what is going on one has to check the reflections during many moon-rises and moon-sets. I have been doing so and waited for the rare occasions when a transmitted H-pol signal comes back in pure horisontal polarisation. (Then a transmitted V-pol comes back vertical as one should expect.) My two-channel system measures polarisation angle accurately so I know when Faraday rotation is right for the experiment. The difference in echo levels between H and V is absolutely remarkable beyond any doubt, well above 10 dB at the point of maximum ground gain for one and maximum ground loss for the other.

There are of course points in between when the ground reflection arrives at a phase angle of 90 degrees. Then there is similar (but modest) amounts of ground gain both for H and V.

I do not know what is behind Peters observations, I doubt they were done during times with zero Faraday rotation.

So if there is a difference between Leifs an my findings they are certainly not because of me making observations only at Faraday offset greater than zero as implied below. And since this was up for debate a few years ago on moon-net I have made it a point to check this very closely.

Could the different vertical pattern between his single yagi/4 yagi and my 6 yagi (3 high stack) play a role? My ground gain (over a lake) sets in at very low elevation, 3deg or less.

Hi fellow hams, please allow me to add a side note on this ground gain issue.

I hear what EME operators say about ground gain but I still have some stomach-ache with this concept. If ground gain is effective in EME, it has to be effective in, say Aurora, MS and sporadic E dx propagation too, i.e. in all dx modes associated with nonzero elevation angles. I am actually surprised that this effect isn’t considered in terrestrial dx propagation at all. Perhaps the effect isn’t that obvious here compared to EME operations, nevertheless a consistent picture would be highly desirable.

I appreciate your discussion very much indeed. Learning and understanding ground gain in EME operations, we may possibly draw conclusions from it with respect to double-hop sporadic E propagation. Contrary to the EME community, ground reflection of 2m radiowaves isn’t generally accepted in the terrestrial dx community – and we indeed found potential examples of ground reflection in forward scatter which are difficult to explain and difficult to accept.


Ground reflection in the near field is well documented in the antenna literature and should not be controversial except, perhaps, for the reflection characteristics of actual (non-perfect) ground in real-world (generally cluttered) locations. Those are notoriously difficult to measure, ergo some of the thread we’ve been seeing on moon-net. Double-hop sporadic-E raises a separate issue, namely ground reflection at long distances from the transmitting station. That is one possible mechanism for the long-distance propagation that has been observed, but there are others, such as chordal hop from one Es cloud to another, that could also explain it.

I don’t have a good explanation for the long-distance Es, but I have had a lot of experience with near-field ground reflection in EME and have no doubt at all of its existence at my QTH.

As for long-distance Es, I made one QSO, many years ago, with K0WLU/7 at a distance of 2585 km (1605 mi), at a time when I was also hearing Es signals from W8 stations approximately halfway between us. That would suggest double-hop Es, but of course does not prove it.

Here are some practical examples that the grond gain effect isn’t only noticeable in EME communications. I measured from echoing that the ground gain effect max. when the Moon is at 10 degrees elevation at my qth. At the other hand from my logs the most meteor scatter qso’s are made around 1000Km. At this distance the reflection area is about 10 degrees elevation seen from my qth. Coincidence? Also in satellite transponder communication the effect noticeable.

HI VOLKER, All EMErs as well as DXers

First of all – I am NOPT an expert as well but : Its my general observation through mni Yrs, that has brought me to this conclusionabout the point ground reflection gaian, and why its NOT is being useble for other reflections

Lets make a throught-experiment.

When the signal is being reflected from a object (moon) at a low angle you must imagine standing and looking a night – looking at the moon. You will see the direct moonlight and the reflected . from the ground ( play its water to see this.

This is what happens – and the “gain occur” when the phases of the two sources is correct. Seen from this point, it should be rather easy to understand the “gain” now and then. Now At Au the Freq is changing fast and simultainsly so YOU will NOT see a phaseaddition of gain here.the

This will eigher not be actual at reflections at higher angles why the phasedifferance will be too large.

This is my understanding of the phenomen .

From a mathematical point of view, the elevation angle of maximum and minimum ground gain depends on the actual terrain slope and, of course, on the actual antenna height above ground. Thus, changing the actual antenna height you may convert a field strength minimum into a maximum. With other words: with variable antenna height, you may permanently operate at maximum ground gain in all EME QSOs.

In accordance to a brief calculation on my blackboard, you must double the antenna height in order to convert the first minimum into the first maximum. For example: if your actual antenna height is, say 7 meters above ground and your EME QSO is suffering from the first field strength minimum, push your array up to 14 meters and this minimum converts into a ground gain maximum. Another example: assuming your antenna height is 18 meters and you are entering the third null at high elevation, you may do the following:

i) to address the third ground gain maximum: increase your actual antenna height to 120 percent of the original height, i.e. 21.6m ii) to address the second ground gain maximum: reduce your actual antenna height to 80 percent of the original height, i.e. 14.4m iii) to address the first ground gain maximum: reduce your actual antenna height to 40 percent of the original height, i.e. 7.2 m

Thus, we may identify various optimum antenna heights as a function of the actual moon elevation. This will involve some sort of mechanical engineering, of course, but permanently operating at optimum ground gain in all EME QSOs appears quite an attractive opportunity. In my view, this method may in particular boost the EME capability of small arrays – compared to the big guns in moonbounce communication, the mechanical effort is much smaller here.

On 20-Sep-2004 I3DLI wrote:
Volker: From a practical standpoint, you are dreaming. Believe me.

> Having heard that using vertical polarization results in reduced ground 
> gain, I would be interested in hearing comments from those of you that can 
> switch your polarization between v&h.

The reason it is explained in my article on ground gain on www.oz1rh.com Pls read the original text as the table is not displayed correct in this email.

For DX-ers on HF it is well known that the radiation angle of a horizontal polarized beam is a function of its height over the ground and ground conductivity has little influence on low angle of radiation. Things are different for vertical polarization. A vertical antenna needs perfect ground to give low angle of radiation. In fact is does not radiate at all below an angle called the Brewster angle. This angle is dependent of the conductivity of the reflecting ground. Check an antenna book on verticals for 80 m, the 80 m DX’ers know all about this. Does low angle of radiation interest us for VHF? Yes ground gain is happening in the range of 0-10 degrees.

 Horizontal polarizationVertical polarization
Ground gain for one reflection5-6 dBPerhaps 3 dB
Ground conductivityLittle influence at low anglesNo radiation below Brewster angle 5-20 degrees
Lossy groundLittle less ground gain and not so deep nulls in patternHigh Brewster angle 20 degrees?
Good ground conductivity (salt water)Deep nulls in pattern, but peak almost uninfluencedLow Brewster angle < 5 degrees?

It seems a good idea to use horizontal polarization if one wants low angle of radiation. Vertical polarization has less ground gain. A circular polarized signal contains two components: a horizontal polarized and a vertical polarized one. Ground reflection influence the two components differently as describes above and after ground reflection the signal is no longer circular but somewhat elliptical polarized at low radiation angles.

Having a wind-up tower, I’ve tried this on 144, and it works. But fitting an electrically operated winch and height readout is (like 144!) a long way down my list of EME priorities…

Of course we can also make use of ground-gain effectively for long periods without varying the height of the antenna, if we have suitable local topographies. Having a small ridge which rises from the south-east to the south of my house, I can enjoy ground gain, and surprisingly frequent audible echoes with my single (h-pol) yagi, for several hours at low southern declinations. The single biggest EME signal I’ve ever heard on 144 was from a VK running 400W and 4 (v-pol) yagis, when we were both able to exploit ground gain. It’s a pity that there aren’t more people active at those times.

Hi Peter, Ray and all,

> Could the different vertical pattern between his single yagi/4 
> yagi and my 6 > yagi (3 high stack) play a role? My ground gain (over a lake) 
> sets in at very low elevation, 3deg or less.

Confronted with the details of what Peter has been doing I decided to have a look in a text-book. Yes, ground reflections may behave differently and it depends on the angle and the properties of the reflecting surface. There is something called the Brewster angle. I think it is about 8 degrees for water but I do not know for sure. In a loss-less dielectric there is no reflection at all for vertical polarisation. The reflection changes sign at the zero-crossing. I must conclude that sweet water is not very lossy (compared to the high dielectric constant) and therefore Peter observes the same phase in his ground reflections from both H and V.

Over farm-land and at higher elevations the phase is opposite between H and V ground reflections. According to my book, lossy media may give any phase shift near the Brewster angle and a wave that comes in with linear polarisation may arrive as an elliptic wave at the antenna. I have never observed it, seems I have good conducting ground around here:-)

I guess others on this list know it all much better. One could compute it with NEC or look in a better text-book than the one I have. I tried to search the Internet but found nothing. It would be nice if someone could compute and upload the theoritical reflectivity and phase shift for reflections on salt and sweet water as well as some types of land.

To All of intrest All having been said. I worked NL7Z on 6 EME with his vertical pol, he showed a pronounced peak at 4 deg, nothing was heard from 0-3.5 degrees, others are in the process of scheds with him . I think he will do well with this setup, the only reason for his vertical polarity was to reduce the local spurs from a horizontal TV TX on his receive . Don’t knock success

Hello Chris and all I would like to point out that as we try to lower the take off angle by increasing it’s height we also introduce a larger field of local noise. The closer we can keep the eme array to the ground the quieter it will be , of course this is the norm, there is always someone somewhere that this will not be true. The most we can get is about 6 dB and you are still limited to the relative short window at the horizon. With the expense of tall towers, long low loss feedlines and control cables, preamps out of reach, extra windload at these heights, hard work to get to them, etc, etc, just place it at a good low level and work them all, just put up 4 or more medium size horizontal antennas and enjoy the whole sky. Make it simple , if it is down near the ground you don’t even need motor drives, It was several years before I installed motors on my array and I worked all 50 before then. once each half hour just reset it and enjoy.. On 50 and 144 never use any polarity control other than patience and next time maybe…… I’ll get off my soap box now. just enjoy this great hobby

On 21-Sep-2004 AL7EB wrote:
I use four M2-xpol-20 antennas on 2m and normally run V-pol to the lower-48 states as spacial polarity shift dictates. I have run both polarities with smaller stations in Europe at my Moon-rise (their Moon-set) and see slight enhancement from 4-7 degrees. Of course Faraday can influence polarity choice a lot. There seems to be so many variables that can affect signal level that it is hard to make any hard fast statements. I do find myself in V-pol a lot, though.

Boy you all read things close, I had to go back and reread what I had written to find this one… “on 50 and 144 never use any vertical pol” Now that is a good one , Sorry, what I was trying to say, was that I have never used polarity control on 50 and 144, sometimes it would have been nice I guess. but with 5 EME stations under one roof I try to make things simple. I am for anything that works, Just be active and enjoy this great hobby of ours.

During WWII the Germans used ground reflections and variable antenna height to measure the angle to enemy aircrafts – around 220 MHz I think.

Some hams use variable antenna height to get ground gain at the desired moon angle. According to what I have been told the theory described on www.oz1rh.com works in practise though my theory (which I did not invent but gathered from the referenced sources) may not be complete. I know my text on vertical polarisation is short, but I still think ground gain from vertical polarisation needs good ground conductivity and that it will not go below the Brewster angle – which is low (>1 deg?) over good ground = salt water and several degrees (5-10?) if ground conductivity is bad.

I guess I should have commented on the fact my antennas have to look at a +2 degree slope to the east at Moon-rise and thru a mixed forest of 45-foot birch, spruce, and cottonwood which fortunately do not totally attenuate the signal (leaves drop by late Sept. helping). This conspires to limit my lower elevation limit probably due as much to ground noise as any other effect. I find I don’t begin to hear until the Moon rises over 2.5 degrees.

One factor in the raising of the tower from 40 to 50 feet (to clear most of the trees at Moon-rise). The other was deterioration of the concrete tower base which required replacement (used better materials this time and used self-supporting base). The spring storm (65 mph wind sustained for 30-hours) of 2003 killed my azimuth rotator and put me out of business until last winter (then discovered the elevation did not work too late into winter to fix). So taking the opportunity to re-build better including burial of all feed and new control lines in pvc conduits…looking forward to Oct. 9-10!

Long time since it was so good reading here!

One question! how far from antenna does the ground affect the pattern? (i hope i used the right word) i guess it have with the elevation angle to do and also antenna height.

Peter, Leif, Palle, et al. —

Wave reflections from the ground or any other surface involve a “reflection coefficient” with both magnitude and phase. For real surfaces the magnitude of the reflected wave (relative to the incident wave) is less than 1.0. The phase shift depends on polarization and on the angle of incidence.

For ground reflections, horizontal polarization leads to a reflection magnitude close to 1 and a phase shift of nearly 180 degrees. Everything is relatively simple, and the expected elevations of maxima and minima in your antenna pattern depend only on antenna height.

For vertical polarization, things are different. Both magnitude and phase of the reflected wave depend strongly on the wave angle. The phase shift will be close to 180 deg at low elevations, decreasing to near zero at high elevations. The “Brewster angle” that Palle mentions is where the excess phase of the reflected wave goes through 90 degrees, and the magnitude goes through a minimum. As Palle says, at radio frequencies the Brewster angle is < 1 deg over salt water, about 5 deg over fresh water, and around 15 deg over average ground.

The upshot is that at elevations *below* the Brewster angle, vertical and horizontal polarizations behave similarly with respect to the “ground gain” that we are discussing. Because the magnitude of the reflection coefficient is smaller for vertical polarization, the pattern maxima and minima will be less pronounced; but the elevation angles of extrema will be nearly the same. At elevations *above* the Brewster angle, situation is reversed: maxima in the Vpol pattern will coincide with minima in the Hpol pattern.

Peter says “My ground gain (over a lake) sets in at very low elevation, 3 deg or less.” His Brewster angle is presumably that for fresh water, about 5 degrees, so Peter’s conclusion makes sense. Over a perfectly flat lake, large in extent, I would expect his Hpol ground gain to be close to 6 dB and his Vpol ground gain to be more like 4 dB, but the 2 dB difference would be hard to measure. Moreover, “real” conditions (a somewhat smaller lake having surface waves, etc., etc.) would probably reduce this difference.

If Peter did the same experiments looking over salt water, where the Brewster angle is < 1 deg, he would reach the opposite conclusion. Vpol maxima would then occur at the same elevations as the Hpol minima.

Those of us with antennas looking over normal ground, rather than water, will nearly always see our ground gain below the Brewster angle. We will therefore get slightly (~1 or 2 dB) less ground gain with Vpol than with Hpol; but we should always see the pattern maxima at the same elevations, regardless of polarization.

Hakan asked “how far from antenna does the ground affect the pattern?” I believe the correct scale is that of the first Fresnel zone. Figure on a few hundred wavelengths, at least — and more if your antenna is higher, and the ground-gain lobes therefore lower.

Joe and all!

I have wrote down a date where i found realley strong signals from VK7MO, is it possible to calculate what should be a normal signal level? i decoded him on -14db in JT65B and degradation was 2,7db.

For VK7MO I used P=400 W, Tx feedline loss = 1 dB, antenna gain 20 dBi. For your station I used NF=1 dB, Antenna gain = 22 dBi, ground gain 4 dB, and total sky temperature 700 K. These numbers produced an estimated signal strength of -14.3 dB in WSJT units, just about what you observed.

As everyone on Moon-Net knows, it won’t always work out this nicely. Many things affect signal strength at a particular time, and it is *very* hard to predict all of them to within +/- a few dB.

I think what Joe states below applies to the typical high yagi, perhaps 20 or 30 meters high. In that situation in an Urban area, you would almost certainly have a house nearby that would break up the reflection. I never see ground gain on my 30 meter high yagi.

An interesting situation can happen even on a small Urban lot when the yagi height is limited to only a few meters. The yagi is much closer to the ground, so the reflection point is much closer to the yagi and the fresnel zone is much smaller at the reflection point. I think you could use some simple geometry to figure out about where the reflection point is. When the yagi is this close to the ground, the peak elevation for maximum ground gain can be in the 10 to 15 degree range. If you draw a line from the yagi toward the ground at the same angle as your elevation lobe, you should be very near the mid point of the reflecting ground. Some sources of inaccuracy would result from where exactly on the yagi you drew your line from as well as not accounting for the need of an additional 180 degree phase shift on the reflection path to bring the reflected signal into phase with the direct signal. I do not have any math to back this up, but practical experience suggest that my statement must be fairly good aproximation. Perhaps someone else has better math that can get closer. I did some experiments when my horizontal array was down for upgrade.

When only a 4 element yagi is used, the yagi can be placed very close to the ground (perhaps 1 meter) and it is easy to see that the reflection point must be very close in to the yagi. One important consideration is that if a very long yagi is placed too low, the ground gain lobe will want to be too high and both the direct and reflected signal will be well outside of the main lobe of the yagi. Remember, the direct ray is happening at the maximum signal elevation angle in a positive direction as well as a reflected ray at the same negative elevation angle. The yagi must be placed high enough that the direct and reflected component of the signal is within the -1db point of the free space elevation pattern of the yagi. This would give maximum moon time with good gain for any specific yagi. This would be a major concern for a 100 foot boom yagi and would tend to suggest better luck with 4 or more 3wl horizontal yagis side by side only a few meters off the ground. During moon-rise one might start out with the array a bit higher off the ground to get an early start on moon-rise, then slowly lower the array as the moon rises to track the moon elevation. When the array is too low, gain falls off because the rays are outside of the beamwidth of the yagi. This might work surprisingly well even at somewhat higher moon elevations because with 4 x 3wl yagis there would be signal to spare even with modest ground gain. Since the yagis are operated so close to the ground, they can be maximum gain designs without suffering from too much local noise pickup.

This is something I would consider when I get my large antenna farm. Perhaps 16, 3wl yagis would be managable with a W5UN type mechanical design 50 meters wide and 2 – 3 meters high. Perhaps this can be done with standard field irrigation hardware if variable elevation was not an objective. Hmmmm, 13dbd gain for 3wl yagi, 12db stacking gain and 5 db ground gain….. 30dbd!

OZ1RH published in the Prague EME conf proceedings page 99, the formulae which enable you to calculate the position and dimensions of the elliptical Fresnel zone, where the reflection takes place. I calculated the numbers for a horizontal( 0 degrees elevation )144MHz antenna height of 20 m . The results are. distance to the reflection point 800m distance to the nearest point of the Fresnel zone 136m distance to the furthest edge 1,043m max width of the fresnel zone 113m Quite a substantial area, to get full effect one would need this to be flat to (I suspect) a few wavelengths. At higher elevation angles the zone will be further away. Another factor is slope of the ground accross the reflection zone. I can assure you, from work done in the MLS programme in the 1970s, that a uniform cross slope accross the fresnel zone will introduce a azimuth pointing error, so if you have that condition it is probably worthwhile exploring the azimuth direction for best results.

On 22-Sep-2004 GM4JJJ wrote:
Been following the interesting discussions on reflections and Brewster angle, the same type of effects are visible to us with our eyes and Polaroid sunglasses:

When light reflects from a horizontal surface at an angle, the reflected light tends to be polarised horizontally. At a specific angle, the light is completely horizontally polarised because any vertically polarised light that hits the surface at this angle is allowed to enter the surface without reflection. Since reflections from horizontal surfaces are mostly horizontally polarised, glare is mostly horizontally polarised. Polaroid sunglasses are designed to filter out the horizontally polarised light partially eliminating the reflections.

Leave a Reply

Your email address will not be published. Required fields are marked *