9 Jan 2013

Earth electrode antenna conclusions (472kHz)

With some further tests today I am now able to arrive at some conclusions from my experiments with an earth electrode antenna at 472kHz. Many people including Jim M0BMU and Rik OR7T have been particularly helpful in analysing the data.
My current 472kHz earth electrode "antenna" - can you see it in the grass?
These are my conclusions:
  1. The earth-electrode antenna at 472kHz (2 earth rods in the soil 15-20m apart fed from the TX output) behaves like an H-field loop transmitting antenna.
  2. It has directivity, with strongest signals in the line of the loop and weakest signals at right angles to it.
  3. It works as an effective RX antenna too.
  4. In my environment the loop looks like about 50-60 ohms resistive, so a good match to my transverter directly without matching.
  5. The structure works because much of the return current flows deep within the soil and rock beneath the earth-electrode antenna. In my case Rik OR7T calculated that the loop area in the ground is effectively 290m sq with a radiation resistance of 0.017 ohm and a loss resistance of 66 ohms.This is a BIG loop!
  6. Performance compared with my 9m high top loaded Marconi antenna averages only around 8dB down, not a bad figure at all, even with the connecting wire on the ground and not elevated at all.
  7. Where no other antenna option is available, the earth-electrode antenna is well worth trying both on 472kHz and on 136kHz both for RX and TX. Although it works well here, your geology may be different and results not the same.
These last few days have been most interesting and, with the help of many people - listeners giving me WSPR reports, advice from others on the Internet forums, suggestions by email - I have been able to do some quite useful science. Thank you one and all.

18 comments:

Bernie said...

Must try 472 kHz across the two earth rods already fitted at the G0VQH QTH, but never used.

Bernie
G4HJW

Roger G3XBM said...

Pretty sure a 2-way CW QSO on the earth electrodes each end would be easy Bernie.

Bernie said...

Also, because the current antenna (7m centre loaded verticle with four 7m top-loading wires) is so ugly, using earth rods would be way of continuing operation on 472 kHz.
The original idea was to have only 2-4 weeks operational.

I am using 50W RF into this antenna.

Bernie
G4HJW

Bernie said...

Hi Roger,

I obviously keep hearing you!

Look also for Phil GW8MLA near Welchpool - I-m sure you must know him.

Roger G3XBM said...

Bernie,I have already copied Phil GW8MLA, and he has copied me, using WSPR, on the earth electrodes.

Roger G3XBM said...

Forgot to mention that the 2 earth rods need to be as far apart as possible and free from obvious grounded things between them so that the current in the ground is able to flow into the soil/rock as a large unimpeded loop. Best thing is to try it and see how well (or otherwise) it works for you. I was amazed how effective it was considering the simplicity of the whole set-up. The output from my transverter is just over 10W.

Bernie said...

Well done with Phil! - I'll call him later to see if he realises what you were using.

Although I have radials below the lawn here, I believe that the ground is free from wires/pipes at Jenny's QTH, so I am hopeful of good performance there. The spacing is about 20m, as best as I recall.

As for here, I was thinking of ground rods in the field beyond my garden. This would be relatively easy to set up and try.

ttfn,

Bernie

Bernie

Anonymous said...

Interesting. I have a 300ft Beverage with earth rods at either end. Worth removing the transformer and load resitor and giving that shot do you think?

John
G3TYB


Roger G3XBM said...

Yes John. I cannot guarantee success (it depends on many factors) it is certainly worth a go,

Todd Dugdale said...

I wonder if you could orient the rods to take advantage of the telluric current. If so, the time of day would be a heavily influential factor.

Anonymous said...

Hi Roger,

Interesting experiments !

Just a few thoughts.

Can you put in two more earth rods at opposite quadrants to the existing rods. So that you can 'steer' the radiation pattern. It would be interesting to set this up with two RX's running concurrently (no pun intended). To better validate your observations WRT directivity.

Could you sink say 10ft of 4" plastic pipe. Then set the earth electrodes into the ground below the pipe. So that the first 10ft is insulated and spaced away from the soil. I suspect that if the antenna is indeed working as a loop. The first few feet of relatively dry soil may just be adding to the losses. It's probably the wet stuff lower down that may be playing a more significant part.

Looks like I'll have to get some copper pipe connected to the garden hose and start setting down some long earth rods to play with.

Regards,

Martin - G8JNJ

www.g8jnj.webs.com

Anonymous said...

Hi Roger (again),

One further thought.

Would it be possible to change your WSPR call sign to /A or /P when using the ground electrodes ? So that it's easier to tell from the logs which antenna you are using.

Regards,

Martin - G8JNJ

www.g8jnj.webs.com

Anonymous said...

>Could you sink say 10ft of 4" plastic pipe.

Yes.. I had the same thoughts about avoiding the first few inches of wet topsoil.

A 2nd advantage would be you could pour nice salt water down the pipe when you really wanted to flush-out the DX :-) G6AIG

Anonymous said...

IF you can believe the EZNEC model it suggests that my hypothetical 300ft ground antenna would have a significant low angle broadside null and become Omni directional as the radiation angle increase with a max directly vertical. This is in contrast to the same length antenna at 40 ft high and end fed which has some directivity along its length and a bit of a null overhead. All theoretical so far, to busy building the Tx to mess about in the freezing cold wet outdoors!
John
G3TYB

Roger G3XBM said...

John, how did you model the ground conductivity at 472kHz?

Bernie said...

Have been running two earth rods since mid-morning, and immediately resulted in a spot from Harrogate.

The rods are standard 4ft items and are spaced by 30m. The resistance was about 120 - 150 ohms, so way above what you measured. A ferrite cored transformer soon had that looking like 50 ohms.

It runs nearly N-S, and I notice that most spots are in-line, as you suggested. Phil over in Wales receives nothing and vice-verca.

Bernie
G4HJW

Anonymous said...
This comment has been removed by a blog administrator.
Anonymous said...

Now that is the dodgy bit and I am not at all confident about. I used MIMEC ground with average figures for conductivity, reasonable as my garden is alluvial silt/loam on a bed of chalk with lots of water underneath, it is also rated as good agricultural land. For the bulk earth resistance I simply put it in series with the antenna at its connection to ground as a load. I played with figures between 50 and 12 ohms. 50 because it is the ‘standard’ figure tossed about and 12 because it is what
I optimistically hope my earth mat will be – 24 earth rods over 160 square metres – dividing 50 by the square root of 24 is the basis of my guess. I have some justification for that from the estimated system Q inferred from the receiver noise. That is the noise bandwidth for the LW is approximately 30 kHz, giving a Q of 15 which is what the matching network calculator suggests will be obtained with 12 ohms or so. Until I can inject some real RF at 475 KHz and measure power and current this is all speculation of course.
John