45 - Using MMANA-GAL

 

44 - Choosing between a multi-band EFHW and a single band OCFD Vertical

 

Choosing between a multi-band EFHW and a single band OCFD vertical

Last Month I reported that my 40m long 80m EFHW was really convenient in that it allowed me to operate on all HF bands but suffered from “nulls” in its propagation patterns. I have been gathering bits to make my own cobweb and then along came December’s Radcom article (2023) about a Halfwave Vertical antenna fed off centre (OCFV) to give good gain without needing those tiresome radials. Tim Hier, M5TM wrote the article on pages 18-20. It covers a single band antenna but it is easy to build – if you have a suitable fishing pole and a transformer plus a couple of chokes.

I don’t have room for radials in the corner of the garden that I operate from otherwise I would have built a DXCommander vertical, Callum has all the constructional details on his web site http://M0MCX.co.uk or http://DXCommander.com  You do need to lay down maybe 16 radials on the ground for it, each maybe 12 to 16 feet long. I do recommend trying this antenna if you do have room for radials, but I don’t.

Back to the OCFV that needs no radials. I began my ham radio using a halfwave CB antenna shortened a bit for 10m and had been able to tune it up on all bands from 20m and up and got good use out of it, when it was nearly forty feet of the ground and I also had good experience with my 20m long horizontal Offset Centre fed dipole (OCFD) for 40m and its odd harmonics that I reported in CONTACT in April, 2020.

One question not answered in Tim’s December article was what effect raising the base of the vertical would have. This is not easy to do with a conventional vertical because of the need for radials – either a large set of untuned radials lying on the ground or a small set of tuned radials elevated along with the base but a halfwave vertical does not need radials and can be raised easily. Tim only discusses using his vertical low to the ground (28cm) so I wondered how it would fare if raised high up.

How to choose what height was best? I have a two story garage with its apex at 20 feet so it is relatively easy for me to get above 7 or 8 metres of height, I have two or three old aluminium sailing masts that are 25 feet long and I often lash 38mm thick broomsticks to these to get me near 10m so gaining height is not too much of a problem.

Ordinary half wave long dipoles have an impedance of 50 to 70 Ohms at their centre, depending on height, but the patterns of current and voltage as you move away from the centre are such that you can pick any impedance you want. Windom antennas use this by going to a point 33% along the half wavelength long wire to give 200 Ohms and use a 4:1 impedance transformer, (a 2:1 Voltage transformer, i.e with a turns ration of 2:1 ) this will convert 200 Ohms to 50 (if you wire it the right way around!).

Tim opted for a 14% feed point and this gives 450 Ohms and therefore needs a 9:1 impedance transformer (3:1 voltage or turns ratio).  This was a good starting point and Tim’s article describes the appropriate lengths needed, he notes that real life gave him lengths 7.4% less than theory and you must always expect some experimentation when building real antennas, unless you can get into outer space and use infinitely thin wire that is a perfect conductor.

One thing Tim does not stress (although he does mention it) is that you must use chokes when feeding unbalanced antennas, best practice is to have a choke right at the transformer unlike an EFHW which needs a choke 8 feet (1/20^th wavelength) or so down the feeder and a second choke down the feedline, where your coax cable enters your shack.

If you omit these chokes the antenna will work. Antennas always work but there may be strange goings on in your shack that go un-noticed but affect performance. A raised noise level and a susceptibility to interference or you may generate interference that creates problems in your shack, or worse, creates problems outside your shack. It is not good to activate your neighbours burglar alarm or coffee machine every time you transmit. It is not good to have your neighbour’s Wifi drop put either, even if no-one notices.

So, Tim covers the practical aspects of making it but only gives results for one particular height. How does the OCFV behave at different heights.

MMANA-Gal is a free antenna modelling package that can plot propagation patterns and compute SWRs across different frequencies, you can vary the height of an antenna and try different ground configurations. Here I just use it at different heights. I covered its use in CONTACT March, 2020.

When a signal leaves an antenna it matters what the vertical angle pattern is. If you have a 100W transmitter and the antenna’s peak output is straight up in the air you will not get very far. The signal might bounce back to Earth if the operating frequency is in the right range of frequencies (below the maximum useable frequency for the particular take-off angle) and high enough so that the D layer does not absorb it.

If it does bounce back you will be able to talk to people in Northern Ireland and maybe the Republic of Ireland, and the rest of the UK. It is described as “cloud warmer” and an antenna with NVIS properties. (Near Vertical Incidence Skywave).

When the peak is between 5 and 10 degrees off the horizon you can expect a radio wave in the correct range of frequencies to bounce off the F layers and give you a contact quite far away, if the angle is 45 degrees then not so far away. If you have a directional antenna, horizontally and/or vertically you need to work out who do you want to talk to, or at least who are you most likely to talk to, where does your strongest signal land.

For DX contacts we want very low angles but not zero, unless your terrain falls away significantly. We typically look at the 5 degrees take-off angle. Of course to cover all areas up to the furthest DX you can work we need some energy to go out on all the different angles so you want a response that is not too tightly focussed.

It is a compromise. A tightly focussed beam gives you gain, more of your 100W goes in that direction but less in other directions, you pick up strong stations from the bigger lobes of your antenna pattern and weaker signals from the smaller lobes of the pattern. Transmit and receive paths are essentially identical, although some receive antennas have more noise. A tightly focussed beam suits particular distances, if you want to talk to all stations up to that distance you need a pattern with energies from 50 degrees down to 5 degrees. It is the same problem as the horizontal pattern of lobes, peaks and nulls. It depends who you want to talk to, or how many people you want to talk to.

Working out what angle gives what distance is an interesting exercise in mathematics, about A-level standard. I will write another short article about that or see my blog.

A useful graph from Laporte (Radio Antenna Engineering, McGraw-Hill 1952 - out of copyright) is in figure 1 below

Figure 1 Distance reached by various Take off angles, for differing F layer heights

As you can see the distance reached depends on what is called the virtual height of the ionosphere. For a layer height of 150 miles, you’ll get over 1,600 miles at 5 degrees, 1200 miles at 10 degrees and only 200 miles at 45 degrees. Of course what really happens is that you reach all these stations but you are much stronger at 1,600 miles than 300 miles.

It is hard to read the graph’s horizontal scale so I decided to work out my own spreadsheet, it’s on my blog http://MI5AFL.Blogspot.com and how I derived it is in another article.

Back to the OCFV; Tim gives a MMANA-GAL design in the article and I modified this slightly; Remember that MMANA-GAL is quite simple to use, the main screen has four tabs, “Geometry”, “View”, “Calculate” and “Far field Plots”. Figure 2 give the contents of the Geometry tab

 

                                 Figure 2 Part of the geometry tab shown in MMANA-Gal

If you want to paste the file in figure 3 into a text editor and save is as OCFV.maa you can run your own copy of MMANA-Gal on it. (or download it from my blog )

This antenna is placed on the ground (wire 1’s Z1 value is zero) – in real life you should lift it even a bit. Tim had his 28cm up. Better is 0.5m as a minimum. You can specify a height correction on the “Calculate” tab before starting the analysis. I did this for a range of heights to see what changed and plotted the table below. The first column gives the height of the base of the vertical and the 5^th column gives the gain at a take-off angle of five degrees.

Interestingly at about 7m you get two lobes, as seen in figure 5 below. The secondary lobe is at a lower angle and is better for DX. I asked MMANA-Gal to calculate the gain at maximum, (or the two maximums once the height exceeded 7m and also the gain at 10 degrees and 5 degrees from the horizon.

I also noted the SWR and the actual impedance, in practice these don’t matter too much, they are low enough to not need matching beyond the 9:1 transformer at the feed point. Do arrange the feedline to sit at a big angle from the wire (greater than 45 degrees) for maybe ten feet or more and as I said above do add at least one choke down the coax, preferably two. The chokes could be 12 to 14 turns of coax would around an FT240-43 or FT240-31.

Alternatively run 450 Ohm twin-lead back to the shack and convert it there. I have a balanced ATU (MFJ-974B) as well as a conventional MFJ-969 so I may experiment with both options.

 

OCFV at different heights from RADCOM Dec 2023, Tim M5TM design, modelling (different heights) by Ian MI5AFL
Height/m	Max gain	Angle	Gain10®	Gain5®	SWR	Real Z	Imag Z
0.0	0.20	20.00	-1.40	-5.10	6.50	773	1229
0.5	0.40	18.40	-1.00	-4.70	1.14	506	-30
1.0	0.50	17.50	0.00	-4.30	1.10	464	-44
2.0	0.90	16.00	0.00	-3.50	1.10	417	-33
2.5	1.00	15.40	0.20	-3.20	1.10	406	-20
3.0	1.20	14.90	0.40	-2.90	1.13	400	-8
4.00	1.30	14.00	0.70	-2.50	1.10	404	13
5.00	1.30	13.30	0.90	-2.20	1.10	417	22
6.00	1.30	12.70	1.00	-1.90	1.1	429	20
7.00	1.60	40.60	Note for the secondary peak at a lower angle see the line below
	1.30	12.00	1.10	-1.70	1	434	13
8.00	2.10	37.80	Note for the secondary peak at a lower angle see the line below
	1.50	11.00	1.40	-1.30	1	434	5
9.00	2.40	34.00	Note for the secondary peak at a lower angle see the line below
	1.70	11.00	1.60	-0.90	1	429	1
10.00	2.60	31.00	Note for the secondary peak at a lower angle see the line below
	2.00	11.00	1.90	-0.50	1.1	424	1

Table 1: Table of relative gains at different heights at the maximum and at 5 and 10 degrees

Things to note from table 1. The important figures are highlighted in yellow, underlined and in italics,

At a height of 0.5m off the ground the maximum gain is 0.4 dB and the gain at 5 degrees of take-off angle is -4.70db. We could use dBi here but the important thing is the difference in dB between different arrangements of the antenna.

At a height of 7m off the ground the maximum gain is 1.3 dB and the gain at 5 degrees of take-off angle is -1.70db.

In other words, at 7m there is 3dB more energy going out at five degrees, twice as much power for working long distance DX. To be sure 3dB is only half an S point, but it is an important half S point. It turns a 100W radio into a 200W radio! It is also better than the 0.5m configuration from 30 to 50 degrees (the second lobe up in figure 5 below)

Figure 4 Vertical Pattern at 0.5 Metres

The actual patterns from MMANA-GAL are shown in figures 4 and 5 below. I only give the vertical patterns, these antennas are omnidirectional in the horizontal plane, their patterns are just circles, they cover all points of the compass equally. I have also added a red line to the polar plots at 5 degrees.

Results will not be the same with a quarter wave vertical as the role of the radials is very important. You could use raised tuned radials, this is a different antenna.

You can get MMANA-Gal from the download button at http://gal-ana.de/basicmm/en/

And you can get the design files via my blog at https://MI5AFL.blogspot.com

Now to build it and compare it to my EFHW on 17m. I still want to build a cobweb as it is a multiband omnidirectional antenna for all bands from 20m and up and I will still need the EFHW for 40m and 80m.

P.S. After writing this whilst on holiday I came home to discover Tim had written another article in February’s RADCOM, He was exploring a single band delta loop and comparing it to a vertical and he has some of the height modelling in his article. I have covered a larger range of heights in Table 1 above as suits my own installation. He makes a strong case for using a Delta but I personally prefer an OCFV because of its simplicity and small footprint. Read both carefully and decide for yourself.

Any antenna works, what makes one better? Many factors matter, not just technical parameters. Cost, convenience, easy of making, ease of installation, robustness and longevity matter to.

73 de Ian / MI5AFL