Dipole antenna construction

[crackle]

There are many posts and ideas on the internet on how to construct a dipole for HF use on the amateur bands.
My original plan was to make a half wave 40m dipole, 10m each leg supplied from 300 ohm ladder line.
But I came across the design for the G5RV which every one seems to rave about.
So I ended up converting my 15m long wire to the dimensions of a half wave G5RV for 40m. The results are very disappointing.
It works better with one half of the dipole disconnected from the balun, and I am a little mystified as to why. My long wire antenna worked better as an all round antenna, tuning up on nearly all bands except for 160m band.

The layout is similar to that seen in my original drawing below, except the dimensions of the half wave dipole are cut to those given for the G5RV 40m dipole, which is 15.55m. The 300 ohm feed line goes direct into the balun in my MFJ 949E ATU.


The antenna does just about tune up on 40m but is hopeless on any other band.
I have been wondering if the balun fitted to the MFJ atu is suitable both in size and the way it is connected in to the circuit.
The balun is barely 1 inch diameter. There is a wire connecting from the upper binding post (just visible in the upper left corner of photo) to the centre binding post (black wire), to connect the "wire antenna" to the balun black wire.
The 300 ohm balanced feeder connects to the binding posts with the black and white wires. Please see diagrams below.



I have no idea how the dimensions of the G5RV have been arrived at, as it does not correspond to anywhere near a half wave of 40m, I am surprised it works on 40m at all.
I am wondering if it would be better to revert to my original plan of making each half of the dipole 10m long with a slight overlap at the far end.

Advice and comments would be very welcome. I have also posted this on Golborne

Thanks
Mike

[Radio Wrangler]

The full size G5RV is a reasonable compromise antenna intended to allow a length of coax feeder for convenience as a lead-in and to produce impedances within the reach of a period valve TX's pi tank. Part of the trickery is to get it to work on 15m which is the band which breaks the 2:1 harmonic sequence of tho other HF bands of the period.

The half-size G5RV doesn't really scale because of the clever trickery in the original, and the half-sized job is much more compromised than the original.

If you don't have to have a coax lead-in, then you can bring ladder line right into the shack to a balanced ATU and you have a much more versatile setup. It, too has compromises. There will be a few frequencies where the impedance goes outside the range of your ATU, but if you find one in a band of interest, a bit of pruning will move them around and you may be able to get all bands OK. Secondly, resonating an antenna with an ATU is lossy. The ATU and feeder carry the resonant current of the antenna so the losses of the ATU and feeder are multiplied. This sounds horrible, but if you use a low loss feeder and a low loss ATU, it can be made tolerable.

Low loss feeder would be open wire line spaced a few inches apart.

The low loss ATU is a more interesting beastie. The normal American design is the 'Transmatch' or 'SPC Transmatch' these are circuits dedicated to unbalanced antennae and transmitters. Where one offers an balanced antenna connection, you'll find a ferrite toroid balun. This is where things go wrong. It takes some care to design a balun to work oved a range of several octaves, like the HF bands. It is difficult to make such a balun work at high impedances. It is VERY difficult to make it work over a range of impedances which span low to high. It is pretty much impossible to make a balun which can both span the HF bands and span the range of impedances of a real world antenna at the same time.

So that balun is really just junk.

I built a big transmatch and went over the top on it... if a job's worth doing, it's worth overdoing! Fancy logarithmic directional power metering with a phase meter! Built to take silly power. Metal well clear of the big roller. Measured quite low loss.

But my antennae are balanced.

I found doing a balun to be the killer problem. My fancy ATU will win construction competitions, but really, I made a fool of myself.

The way out of the problem is to use a resonant balun and to tweak resonance to suit your operating frequency.

If it's resonant, it's an ATU. So the ideal solution is to amalgamate the ATU and balun functions.

re-enter the somewhat forgotten Z-Match. There was also a balun ATU by Alan Chesters in radcom in the 80's.

I use a dipole sized to my property with open feeder into a KW 109 ATU which is a Z-match. It doesn't equal the big dipoles and the triband beam up at the club, but it's the best I can fit in at home.

David

[crackle]

Originally Posted by Radio Wrangler

So that balun is really just junk.
David

Hi David
So what would you recommend? Coax to the dipole centre or another balun?
Would coax to a point on the fence opposite the centre of the dipole and then a balun to the 300 ohm feed line?
What type of balun, have you construction details?

I have been looking at this type.
https://ham.stackexchange.com/questi...ated-as-others

Thanks
Mike

[GMB]

Coax is a really bad idea if not being used in a correctly terminated and unbalanced mode. So no coax beyond an ATU or it will eat the signal.

Ladder line can be totally abused with little loss - but bear in mind that in doing this the feeder is radiating, so consider what it does to the overall radiation pattern.

My handy definitions:
"Aerial" - the structure that looks like it ought to be radiating the signal.
"Antenna" - the structure that actually does the radiating (often includes lots of cables and at higher frequences the operator's body too).

[crackle]

Originally Posted by GMB

but bear in mind that in doing this the feeder is radiating,

I am a little confused now, I had read that ladder lines in a balanced antenna do not radiate as the signal on each wire is equal and opposite.

Thanks
Mike

[Radio Wrangler]

In a ladder line, or a coax, there are two conductors. They can carry two signals at once. The differential signal has current along one conductor and back on the other. The common mode signal has current along both conductors in the same direction at the same time. Common mode currents either go to a path to earth at the far end, or are charging something up, possibly the capacitance to earth of the far part of the line.

Open line feeders and coaxes all radiate any common mode signal and also will receive any appropriately polarised incoming signal as a common mode electrical signal.

A good coax system has the radio equipment shielded and the common mode signal is kept on the outside. This prevents pick-up of domestic noise.

A good open wire system has a balanced ATU or balun which rejects common mode signals and this keeps domestic noise out of your receiver.

Just because each type of feedline can receive common mode signals doesn't mean they have to wind up at the input of the receiver, thankfully!

The differential signal in an open line feeder will radiate. The fields from the two conductors will be equal and opposite, but they are spaced apart and so the amount of cancellation is limited. The wider spacing, the more radiation, but the spacing is a small fraction of a wavelength, so what the hell. There's some, but it's small. Coax is a clever solution to the spacing problem. Having one conductor up the inside of the other, and accurately centred, the spacing is zero and no differential mode signal is radiated or received. Sounds good, but the structure is inherently unbalanced, and the capacitance per unit length is high which restricts it to low values of characteristic impedance. When run in high impedance situations, coax is pretty lossy. It isn't a good choice on the variable-Z side of an ATU.

I wouldn't recommend ANY balun to go between an unbalanced ATU and a wire antenna. The balun won't be able to cover much of the frequency and impedance range an ATU is useable over.

I'd recommend changing to a balanced output ATU where the balun function is implicit in the ATUs resonator. Look up Z-match or KW107 or KW109. There's a KW109 running by my right elbow right now.

David GM4ZNX

[GMB]

I am a little confused now, I had read that ladder lines in a balanced antenna do not radiate as the signal on each wire is equal and opposite.

That is the case *IF* they are being used in a balanced fashion.

What most people forget is that this is never the case in real life, or at least is hard to achieve. The classic "voltage balun" autotransformer is a much misunderstood thing - it only works if the input and output are correct, but they never are in real life. The "current balun" or common-mode choke is way better at forcing a balanced current situation upon a line but it cannot achieve perfection.

So feeders to aerials will be radiating, the question is how much.
Also, there will be standing waves in the feeder unless a very perfect aerial is attached. This only matters for coax that does a really bad job in that case - the open wire feeder is way more efficient so typically it doesn't matter (so it is not so important about the impedances at each end of it).

If you want to know what a pile of wire will do, try using aerial modelling software. It will give you clues - but don't beleive it too much because the great unknowns are things like the capacitance to nearby structures etc. and especially the ground conditions which is what makes an apparently ideal aerial go wrong.

[crackle]

Ok thanks for your answers.
Please can someone explain why 15.5m for a half wave 40m dipole and not close to 20m.

Thanks
Mike

There a few G5RV like antennas one is the ZS6BKW and another I built but can't remember the name.

[mickm3for]

Hi i can vouch for the z match i run a kw ezee match this has no balun, i feed with 450 twin feeder it has less loss than coax also gives a lower noise level than coax on rx . Being a ballanced system the signals in the two wires being equal and oposit they do not radiate and do not pick up local interferance i would not go back to coax Mick

The thing about "450 ohm" (or whatever) used in this sort of aerial is that it isn't matched at all. It does however get the aerial down to the tuner with much lower loss than co-ax. Even your bog standard 300 ohm ribbon is better than co-ax and good for a few hundred watts

[crackle]

Thanks for your replies.
I have been looking at the various articles on the G5RV again and I now know I misunderstood the requirements for the balanced line feeder. I now understand that this forms a critical part of the tuning of the antenna, the length of the line is also critical. So I have ordered a 7m length of 450 ohm feed line and plan to take this from the dipole centre, down to a 6 foot high fence and from there join it to a length of RG8X coax. The feed line needs to be about 16 foot and the coax to the shack will be about 12m. I will experiment with using a 1:1 current balun at the join to coax at the fence.
Doing things this way I hope will improve the antenna as at the moment the balanced line feeding into the MFJ ATU is running in a long curve under one of the arms of the dipole. Taking the feed line to the fence will take it away a right angled to the dipole.
Does this sound sensible?

Mike

[Keith]

Mike asked:

"Please can someone explain why 15.5m for a half wave 40m dipole and not close to 20m."

It's not just 15.5m (51ft). The original G5RV design specified 34ft of open wire feeder from the centre as well, before going into (75 ohm) coax. Thus the total length of each leg is really ~85ft. As I understand it, this length was a compromise to give a low impedance on the (then) five bands (80m through 10m) suitable to match the co-ax with a "reasonable" SWR. I eventually dispensed with the co-ax and fed whatever length of balanced feeder I could get in into a link-coupled balanced tuner. Never did get round to putting it into a box...

[crackle]

Hi Keith
A circuit diagram of your tuner would be nice.
Also any one else who has successfully got a G5RV Junior to work, it would be good to see the schematic of your ATU, and if you used balanced 300 or 450 ohm feeder all the way.
Thanks
Mike

[G6Tanuki]

I've played around in the past with both the full-size G5RV and the G5RV-junior, never with much success.

(both were the type with a specific length of ladder-line from the dipole-feedpoint to the point where it transitioned to coax). Personally I don't like having to use an ATU to coerce a non-resonant antenna into presenting a decent match to its transmitter; I've had a lot more success with coax-fed paralleled/fan-dipoles; currently I've got such a beastie for 7, 10 and 14MHz.

Given where we are in the current sunspot-cycle, 21/28MHz coverage in an antenna is not really something I would consider putting effort into. Your antenna will probably have fallen down long before the sunspot-cycle makes operating on those bands a prospect again.

[CarbonMike]

Have a look at the Carolina Windom. I have had a lot more success with this (especially QRP on 40M) than a full size G5RV. It deliberately radiates off the feeder. You will need to make a 4:1 weatherproof balun but apart from that it's quite simple. Lots of info on the web.
Mike
ZL2BBS

[ex seismic]

Don't bother with G5RVs of either size. Just put up a balanced doublet of the maximum size that will fit your space, feed it with ladder line straight to a balanced tuner as suggested already. You can use the G5RV wire if you already have one. keep the ladder line in free space as much as practical, i.e. not pinned to the wall/fence or whatever.

[crackle]

Originally Posted by ex seismic

Don't bother with G5RVs of either size. Just put up a balanced doublet of the maximum size that will fit your space, feed it with ladder line straight to a balanced tuner as suggested already. You can use the G5RV wire if you already have one. keep the ladder line in free space as much as practical, i.e. not pinned to the wall/fence or whatever.

Hi
Thanks for the suggestion, but it may be a little impractical as my shack is not immediately under the centre of the antenna.

Do you have a circuit for a balanced ATU?

Thanks
Mike

[GMB]

The so-called balanced ATU is not that really, it transformer couples to the output so the voltages are isolated from the input so can float to be anything. This should mean it will work wherever the feed point is (just do not use coax).

There are lots of articles on the "z-match ATU", just google that.

[Radio Wrangler]

The tuned balun atu by Alan Chesters in radcom in the early 80's really is a balanced ATU.

It can do balanced or unbalanced in, and balanced or unbalanced out.... rather versatile!

David