Hmmm.... a "remote matcher" that works by switching in lengths of ladder line via relays.... I like your thinking! Thanks for the tip, mate. I might just give this a try and see what it does for me!

73 de Frank PA3GMP /ZS6TMV

As long as I'm here anyway, let me ask yet another awkward question (and possibly spark a religious debate): what is, in your opinion, the best antenna tuner?

Yes, I know, such a beast does not exist. The perfect antenna tuner would be simple in design; match everything to 50 ohms; be suitable for both balanced and unbalanced loads, and have a loss so close to zero as to make no difference. But that's in an ideal world. Meanwhile, back on planet earth, we need to compromise. A tuner will either have a limited matching capability, or a limited efficiency, or require 16 roller coils for starters. 

So. Which type of antenna tuner comes as close to the ideal as possible?

73 de Frank ZS6TMV / PA3GMP

Dan,

I couldn't agree with you more. Something that works in one situation should not be assumed to work similarly (or at all) in another situation, especially not when antenna's are concerned. Also, SWR is incidentally related at best to overall antenna performance, yet many hams judge the quality and/or performance of an antenna solely by how it loads their rig. I've got a T-match here that will match my radio to a door knob if need be, but a 1:1 SWR does not a good antenna make. And my embarrassingly modest folded dipole for the 20m band made out of some wire and a ferrite toroid runs rings around my friend's terminated folded dipole... Which gives him an almost 1:1 match on all bands from 160 to 50m, but on 20m it burns up 3-6dB in the terminating resistor. Of course that makes little difference when working the next door neighbours, but also means that I can hear weak stations that he completely misses out on. Yet he is convinced that he's got the better antenna. Not only does it not need a tuner, but it was designed by experts, built by pro's, and he's paid around $600 for it, after all, so it has to be good. :-)

That said: reading up on the material suggested to me earlier in this thread, I see that in one of OM Varney's later publications on the G5RV antenna he apparently stated that his earlier inclusion of a balun might not be such a good idea, given the sometimes severe mismatch between the G5RV's feed point and the coaxial cable which caused the balun to saturate and overheat.

We should also keep in mind that Louis Varney grew up in an era when both hams and radio's were made of sterner stuff than what we find these days. You built a rig, you connected an antenna, and on the air you went. Only when the valves in the finals started to glow red and bend over there was reason to have a look at how the antenna loaded the transmitter. Feed lines were chunky ladder lines more often than not (none of the whimpy thin wire stuff we use now) and an SWR meter was a beast that few had heard of and fewer had seen. And being unaware of all the "issues" that we are concerned about now, hams blithely went ahead and made great QSO's in spite of all the matching issues that trouble us in this more "enlightened" era. As long as the valves in the finals were happy, everything was all right. Let's call CQ and have fun, 10:1 SWR or not - what's that SWR thing about, again?

There's progress for you. :-)

Excuse me for moment while I put my soap box away again. I've used it enough for one day.

73 de Frank ZS6TMV / PA3GMP

Eh? How many _degrees_ in length? (Keeping in mind that the G5RV is a multi-bander, so for which frequency should I tune the ladder line?)

Yes, I know, it doesn't take much to confuse me... :-)

73 de Frank ZS6TMV / PA3GMP

Hi, Cecil,
That is what I have always believed - and still believe. It's just that with the G5RV many designs explicitly state that you should not use a balun, and that's what I was wondering about.

Something I've been wondering about - is there a way to replace a coaxial feeder with a ladderline? I mean, without having the ladder line de-tune the matching section?

73 de Frank ZS6TMV / PA3GMP

Hi, Steve,

Indeed! What I read there is exactly what I was thinking. It also explains where the "don't use a balun" think came from. Very enlightening - thank you!

73 de Frank ZS6TMV / PA3GMP

I agree. Here in the Highveld region (around Johannesburg, South Africa) lighting is a huge problem as well, and so far the best know-how I've seen has always been from Polyphaser. Study it. You'll come away with a fairly good idea about what it takes to implement proper lightning protection. It will also show you that most of the things that most people believe about what is adequate protection, is nowhere near sufficient.

73 de Frank ZS6TMV / PA3GMP

I've been experimenting with a G5RV lately. There's one thing about the design, though, that I don't understand. All designs that I've seen state that you should connect the (unbalanced) coaxial feed line directly to the (balanced) feed point of the matching stub. Some designs state that a balun is not necessary, others specifically state that you should not use one.

Why?

I'd expect a balun between an unbalanced coax feeder and a balanced antenna to improve things - provided, of course, that the balun has been properly designed and built for the job and that it does not introduce losses high enough to deteriorate the performance of the antenna system as a whole.

So far the best argument for not using a balun is that "OM Varney himself never used one, either, and he did just fine" but for me that's not good enough. Call me overly critical. :-)

All suggestions are appreciated!

73 de Frank ZS6TMV / PA3GMP

Not necessarily... But you do raise an interesting point here, Dan: the reliability (or lack thereof) of the reports that we hear these days. There used to be a time (or so the real OM's tell this youngster) when a signal report was given on the basis of the operator's honest assessment of the signal received. These days most hams just read their S-meters. On my Yaesu FT-100 the digital S-meter has two positions: S1 and S9. Yes, there is a scale in between, but the darn thing never seems to use it. So on the basis of that S-meter (I call it a lie-detector) my S-reports are either 51 or 59... because being a lazy, modern-day ham who grew up with music on FM and is now used to 3G cellphone quality, the readability of a station is either very good, or I don't bother trying to work it! (OK, I'm exaggerating here: I'm not that bad... But you probably get my point.)

So. Having established that S-reports are often to be taken with a grain of salt, the next question is how hams quantify the performance of their antenna. Hearing a ham gushing about how"This antenna is great, the best I've ever had" sounds wonderful... but why does he feel it's a "great" antenna? Does it really outperform the proverbial random length of wire when that wire is installed, tuned and loaded up properly? Does it really pull in more DX than other antenna's that are a lot simpler or cheaper? Does it consistently perform spectacularly, or just when activity and conditions happen to come together, at which time owners of other antenna's will be raving about how much fun they are having, too?

Yes, you are right in that expectations are a major factor in subjective judgements. But perhaps more important is the totally uncritical attitude with which many modern-day hams shower down their praise of any equipment (not just antenna's) down onto the masses. Is it a great antenna just because the first time you tried it you immediately had three DX contacts with arm-chair quality copy? Well... I had that with the first piece of wire I hung from a tree (and I'm not joking here). Simply because I happened to try it at the right time, on the right bands, and because I had been living in apartment buildings all my life, where a 6 meter piece of wire hanging from the balcony was the best I could do, only to be "rewarded" with S9+20dB of QRM - on a good day.

Let me give you a practical example. I have here one of the most praised and maligned antenna's, a G5RV. I also have a simple folded dipole for the 20m band. I find that the FD outperforms the G5RV. So the easy conclusion (which indeed many hams have made and propagated over the years) is that the G5RV stinks. However... The FD is strung up tight between the edge of the house and a tree from east to west; fed through a toroid Guanella balun, and needs no tuning. The G5RV is strung between two trees at opposite ends of the garden in a north/south direction, and due to its length sags quite a bit so that the feed point is much lower than I'd like (less than 7 meters, I'd guesstimate). At the feed point of the tuned ladder line it's fed via a coaxial cable that is buried, has a choke in it to keep shield currents in check, and the whole thing is matched to the radio via a Kenwood AT-230 tuner.

Comparing these two antenna's "as is" would be foolish. One is a monobander, the other a multi-bander. One is a folded dipole, the other an open dipole. One has a matching section, the other is resonant. One has a choke, a buried coaxial feed line and a rather elaborate antenna matcher between it and the radio, the other has a toroid guanella balun and a free-hanging coaxial feedline. Apples, oranges... fill in the missing words.

Also, a problem that I keep running into is that I can hear lots of stations - but they can't hear me. Surely my antenna's must be terrible! Or not? Consider: I've got a Yeasu FT-100 which is known for its excellent receiver. I'm also working "barefoot" with a 100W output - 100W on paper that is; in practice it'll be closer to 80W, I suppose. The guy on the other side may be in the US or Canada where the legal limit is in the kilowatt range, or in Eastern Europe where amateurs use two and a half kilowatts just to tune up their antenna's. And he or she is probably surrounded by other hams who are withing shouting distance and can generate quite a pile-up that I will never be able to punch through... because I'm out here in South Africa, and I can hear the other station only because there's very little else going on in these parts.

So. Before blindly accepting the rave reports or tar-and-feathers treatments that some hams give some antenna's, it's probably best to keep in mind that a lot of factors come into play, and unless you know them and take them into account, you simply, and sadly, know nothing.

73 de Frank ZS6TMV / PA3GMP

Frankly, Pete, I don't think so.

The main thing to keep in mind (and the main reason for the endless debates about antenna's and how they compare in terms of performace) is that any antenna is part of a system, that does not only include feedlines, matchers, radio's and what not, but also the environment. Antenna X may work great for one ham (who lives in a pastoral area with excellent soil and few hills) but be a major disappointment for another ham (who lives in a rocky part of the world with poor soil and lots of hills). So comparing antenna A to antenna B is utterly useless. The only comparison that makes sense is the one between situation A and situation B, which not only includes antenna's A and B, but also all the other factors. Antenna specs look great on paper, but in the real world there is no guarantee whatsoever that a $1000 factory-built miracle will give you better results than a piece of wire hung from a tree - it depends on so many other things as well.

Having said that, I also believe that you have answered your own question at least in part: you have worked DX with a wide range of antenna's, and you've had fun doing so. Honestly, I don't see how spending a load of money on a "miracle" antenna is going to significantly improve on this, at least not in terms of enjoying the hobby. At least, that's my opinion - for what it's worth. ;-)

73 de Frank ZS6TMV / PA3GMP

IMO: bad idea. The currents that occur in the loop even at moderate power levels are huge. N-connectors are not meant to handle this current.

That current is the reason why, when you build a magloop and you don't have a vacuum cap, you MUST use either a butterfly or a split-stator capacitor. A "regular" (for lack of a better term) tuning capacitor has slide contacts on the rotor, and due to the large current they burn out.

It is also the reason why the loop benefits from a thick tube diameter. Due to skin effect only the outer layer of the tube is used, and if you use thin tube (say, like the shield of a piece of RG58 coax) the resistance becomes a huge issue. (That's not the only factor of course, but let's focus on the issue of loop current for now.)

I use 7/8" cellflex coax for magloop antenna's and that works fine. The ends of the loop are soldered to brackets made out of copper, which connect to the tuning cap (a butterfly type, because I have been unable, sadly, to get hold of a good vacuum cap that I can afford) by means of four stainless steel M6 bolts. I have found that to be a reasonable compromise between portability and efficiency. YMMV.

You'd better believe it.


There are several public domain software tools on the web (Google is your friend) that will allow you to plug in your loop diameter, tuning cap value and power level etc. and then calculate voltages, currents, efficiency, resonance frequency etc. for you with one click. I use a small program written by KI6GD. It tells me that, say, a 10 foot loop circumference, operated at 100W at 40m will put 4.3kV across a 188.6pF tuning cap. Now, horribly oversimplifying the matter by ignoring all kinds of relevant things like induction, resistive losses, phase, radiation and other things that matter greatly, let's just say that the current through the cap (i.e. the current through the point where you want to introduce your N-connectors) equals voltage divided by the reactance of the capacitor. Xc = 1/(2*pi*f*C) and I=U/Xc... which gives us an RF current through the N-connectors of roughly 35.7 amps. Even taking into account that this gross oversimplification is wildly off (which it is) you can see that no matter what you're going to be dealing with considerable current here. Which is the reason why you need a chunky cap, a thick tube and solid, low-resistance connections between the two. You need a high voltage cap to prevent arcing... but a high RF voltage across a sizeable capacitor means a high RF current going through it - and through the connectors that attach it to the loop.

So I wouldn't use N-connectors here. A trip to the hardware store for some 7/8" plumbing parts might be your best option.

73 de Frank PA3GMP/ZS6TMV

OK - time for a few random practical thougths here. :-)

First off, going through any trouble to achieve a boost from 5 to 7 Watts is ludicrous. A boost from 5 to 10 Watts is slightly less ludicrous, but still does not make much sense, imho. I do not agree with the often-heard statement that it won't make any difference - in marginal conditions, it will. If you're just unable to make a QSO at 5W, a boost to 10W can often get you just above the receiver's squelch level and/or noise floor. However, it will only make a difference in those cases. In all other cases (i.e. when you are able to make the QSO at 5W) you might achieve a slightly better signal-to-noise ratio but that will be it. Hardly worth much effort or money.

So. As long as you're thinking of an amp anyway, make it count - I would say that 20 or 25W is a minimum to make it worth your while.

Then a word about power supply. Most (all?) amps use the input power to drive the amp, and very little (if any at all) ends up in the antenna as output power. In other words, an amp that boosts your power from 5 to 10W has to PROVIDE the full 10W, and not 5W. Taking into account that your amp has an efficiency far less than 100% (depending on its design, an efficiency of 50 to 80% is usually a nice ballpark value to start with) your power supply will have to provide between 20 and 12.5W of power, respectively. For 25W i'm sure you can do the math yourself. So powering it with a small battery is right out of the question - also taking into account that the losses created by a battery's internal resistance (which get higher when the battery gets smaller) become considerable at higher currents.

I know you said you didn't want to hear that it cannot be done... but the fact of the matter is that you're limited to what physics will allow. So if you're hoping to get something for nothing here, I'm afraid that that will have to be the answer. :-)

73 de Frank PA3GMP/ZS6TMV

Hi, Bob,
I know them. They've been around for ages, and from what I understand they're not exactly the cutting edge of technology. I could be wrong of course, since I am far from knowledgable in this field, but from what I read I believe that there are more powerful and more efficient options available for far less, as the Parallax products are not cheap. Also the thought of having to use Basic doesn't fill me with joy, having worked with C and Assembly for several years. Basic is indeed super easy, but that simplicity becomes a drawback if you're outgrown it. :-)

But thanks for the suggestion - I appreciate it!

73 de Frank PA3GMP / ZS6TMV

That is indeed the first thing I thought of. :-) The long and the short of it is that it's not accurate enough. That may be due to the fact that I used servo's from an R/C model, though. These are controlled by pulse width (typically something like 0.6 - 2.4mSec for either a 90 or 180 degree rotation (i.e. 20 or 10nSec per degree, respectively) depending on make and model). Very simple to implement, but due to drift, the servo's lack of accuracy inherent with gear play, feedback pots, noise, and the simple fact that they weren't designed for this in the first place, it is almost impossible to accurately tune to a sharp dip. Not to mention that with analog components a timing accuracy of 10nSec is not all that easy to achieve, what with drifting and all that.

73 de Frank PA3GMP/ZS6TMV

Hi, Dan,
Of course not! :-) If I had, I wouldn't be asking all these questions, now would I?

The whole tuner thing is still in the breadboard stage, where it will remain until all the experimenting is done and everything is completely working and finished, and only then I'll be more or less committed to a certain design and a nice box with shiny knobs to put it into. The idea here is to design and develop, not just to assemble. :-)

Hang on - you may just have solved all my problems there. I see it's available from RS components, which has branches here in South Africa. Interesting!

Well... I see that it's called an "open source prototyping platform". I haven't read all of it yet (I have to work today and I'm currently deadlined so I shouldn't be thinking about this at all <g>) but does the open source aspect also apply to the design and software of the boards? In that case I should be able to make my own boards for personal, non-commercial use, right?