Other SOTA sites: SOTAwatch | SOTA Home | Database | Video | Photos | Shop | Mapping | FAQs | Facebook | Contact SOTA

Impedance at the end of a wire?


#1

Hello all,

Following some success with a W3EDP I am playing with End Fed Half Wave antennas, at the moment for 80m. I have on the bench a link coupled parallel tuned circuit, and its doing a lovely job of transforming a resistive load of between about 3k ohms and 5k ohms into a 50R resistive load. Its almost like magic to an electronic simpleton such as myself.

This is all fine, and I have a reasonable understanding of how an EFHWA works according to AA5TB at his excellent page on these matters:

http://www.aa5tb.com/efha.html

The thing is, how do I calculate the impedance at the end of a 1/2 wave bit of wire cut for 80m? What will actually happen is I’ll go out now with some wire and see what happens. But I’d like to be able to do the whole thing “by the book” - I’ve figured out the values of my capacitor and inductor, calculated the amount of turns on the toroid etc etc, but it would be better if I knew how to calculate the impedance of the end of a bit of wire!

Can anyone help?

Thanks,

Dave M0MYA.

PS. I have tried modelling such a thing in MMANA, but the output just isn’'t making much sense to me!


#2

In reply to M0MYA:

Dave

There is no easy way of calculating the impedance manually and so software modelling is the only way. Can you explain what is confusing about the MMANA figures?

The end impedance in real situations will be very dependent on lots of environmental factors and could vary wildly (due to all sorts of unknown variables) so the idea of simply calculating it and plugging in a matching network and expecting it to work is probably a non-starter.

73

Richard
G3CWI


#3

In reply to G3CWI:

Thanks, Richard.

If I model a 1/2 wave wire for 80m at 5m above the default “Real” ground it shows an impedance (at the source, I am presuming - I’m very green with this modelling lot) of 818.3 -j40586

The resistive part of that figure seems a little low to me, based on my (limited) understanding of the content of AA5TB’s stuff.

73,

Dave M0MYA (logged in as Geoff)


#4

In reply to 2E0BTR:

818.3 - j40586 Yuk! That’s a horrible impedance to match. It looks plausible though.

To get a feel for how quickly it might vary try reducing the aerial height by 50cm or adding 50cm of wire to the end (in the model). You might also try changing the ground types.

The MF aerials that I use have the same problems and no-one has found a way to reliably predict the impedance of them, thus to operate portable requires a box of ATU components!

Good luck.

73

Richard
G3CWI


#5

In reply to 2E0BTR:

"818.3 -j40586"
I think your model is not working correctly. I would try adding a small counterpoise and change the source type / placement. Your tuner is going to act as some sort of counterpoise, so its not unrealistic.

Modelling with eznec (split source) a 40m long vertical, feedpoint 1m above ground and 1m counterpoise I get approx 4k+j0

Regards,
Nigel. F/G6SFP/P.


#6

In reply to G3CWI:

Well, I’m going to tentatively say that the experiment is a success!

The setup is as follows:

Old variable capacitor, 2x 400pF in parallel to make 800pF (ish) at 3.4MHz.
A T50-2 with 24 turn primary, in parallel with cap.
4 turn secondary fed to a bit of 50 ohm coax with a lousy BNC on the end (6:1 ratio - pretty standard as I understand it)
1/2 wave wire cut for 3.666MHz on one side of the cap
1/20 wave “counterpoise” on the other side of the cap

I set it up on the analyser, dead easy. Set the '259 for 3.666, twizzle the variable cap until a 1:1 match is found and go. RX very good, nobody listening to 3.666 for the couple of CQs I put out. Following the advice of AA5TB, I prodded and poked at the 817 and the anaalyser and it didn’t alter SWR, so seemingly no hand effects, which is apparently indicative of a decent match.

It all seems to work. I’m just baffled that it does. It all seems fine in theory too, according to AA5TB, I just can’t believe my luck! I’ve been considering building this for yonks, but the fiddlyness of it put me off. Once I’d done the calculations and empirically derived a good turns ratio, it seeems to be OK. I am wary because it was too easy.

Unless I’ve misunderstood, the 1/20 wave counterpoise does away with the nasty reactances in the system.

I’ll put the components in a little box with some binding posts and take it out on the hill, see what happens.

73,

Dave M0MYA.


#7

In reply to M0MYA:

Dave

Don’t fall into the trap of thinking that low VSWR means an aerial is radiating efficiently - it doesn’t. But I’m sure that you know that!

73

Richard
G3CWI


#8

In reply to G3CWI:

Don’t fall into the trap of thinking that low VSWR means an aerial is
radiating efficiently - it doesn’t.

Exactly. My dummy load has a (measured) VSWR of less 1.1:1 from 100kHz to 1GHz. No matter how hard I try I can’t work anyone when I use it as an antenna. :wink:

Andy
MM0FMF


#9

In reply to G3CWI:

In reply to M0MYA:

Dave

Don’t fall into the trap of thinking that low VSWR means an aerial is
radiating efficiently - it doesn’t. But I’m sure that you know that!

Indeed I do. A good on air testing will be in order ASAP. I could really do with a field strength meter around here, I think the MFJ-259 “kind of” functions as one in freq counter mode, by indicating relative field strength on the impedance meter. Ah well, an activation would be more fun.

Andy: You’ll be telling us how you worked VK on 80m while tuning up into a lightbulb next, hi!

73,

Dave M0MYA.


#10

In reply to M0MYA:

If you’re still set up I will take a listen on 3.666 for you now Dave. I’m in Blackburn Lancs. IO83SS

Good luck,

73, Mark


#11

In reply to G0VOF:

Thanks so much for that, but its been wrapped up for a little while now. In order to set it up I’d have to wade through a field of sheep droppings and knee-deep grass in the dark!

Again, thank you, its most kind.

73,

Dave M0MYA


#12

In reply to M0MYA:

No problem Dave, I know just what you mean about sheep droppings, the convenient field I used to set up my 80m /60m link dipole was full of them! As well as a very suspicious Ram HI!

73,

Mark


#13

In reply to M0MYA:

I could really do with a field strength meter around here

http://www.qsl.net/vr2zxp/project/fieldmet/RF%20Field%20Strength%20Meter.jpg

VK on a lightbulb… no! But I can tell you a true story about someone using a MM1296-144 transverter with an FT290 once. They heard someone who was a known local 1296 DXer operating on 1297 FM. Strange they thought an asked why 1297 FM instead of 1296 SSB/CW. The reply was “I’m on 145FM… you’ve got a leaky attenuator/feeder between the transverter and FT290!”

Andy
MM0FMF


#14

In reply to MM0FMF:

I could really do with a field strength meter around here

http://www.qsl.net/vr2zxp/project/fieldmet/RF%20Field%20Strength%20Meter.jpg

Thanks, Andy, looks simple and useful.

73,

Dave M0MYA.


#15

In reply to G6SFP:

Hi Nigel,

Sorry, I missed your post in the flurry of other posts!

I have tried adding a counterpoise, and I can add a bit of wire who’s one end is at the same point in space as the main element, but I can’t figure out how to “connect” the source to it. If thats even what I’m supposed to do!

I must confess, I practical terms, I don’t have a clue how to use antenna modelling software. I use MMANA as its the easiest I’ve found thus far and I’m still struggling! A good howto would go a long way!

Your figures sound a lot more like the kind of thing I was expecting, I just wish I had the ability to replicate them!

73,

Dave M0MYA


#16

In reply to M0MYA:

Hi, Dave.

I’ve only just read your thread so sorry if I’m duplicating something in the e-mails.

As Richard says the end impedance of a halfwave is very high and very unpredictable in the real world. It’s also possible to match it (at some moment in time) but actually achieve very little radiation.

Can I suggest shortening it a bit (say 3/8 lambda). You should find it more predicatble, easier to match and that it will actually radiate better than an end-fed halfwave.

73, Richard (off to work …)


#17

In reply to M0MYA:

In eznec you can select a split source ( SI or SV ) and then locate it to the end of one of the wires at the junction i.e. 0% or 100% along the wire.

The other way of doing it is to put a small connecting wire between the two wires and locate the source on the connecting wire.

I haven’t used MMANA for a while so I am not sure if it has the split source type but the second method should work.

I have been using a half wave end fed wire as a kite antenna on 80m and it works really well, even on the ground :wink:

Good luck with the experiments.

Nigel.


#18

In reply to G3CWI:

To get a feel for how quickly it might vary try reducing the aerial
height by 50cm or adding 50cm of wire to the end (in the model). You
might also try changing the ground types.

The MF aerials that I use have the same problems and no-one has found
a way to reliably predict the impedance of them, thus to operate
portable requires a box of ATU components!

I have fitted the above components into a box - same setup, just more mechanically robust, which has allowed me to move the feedpoint around a lot more, and thus adjust the antenna’s height above the deck.

AA5TB says on his website:

“In practice I’ve found that I can set things up in the backyard and fine tune the system and when I get out into the field things don’t hardly change.”

Lucky AA5TB! Richard (CWI) is quite right - impedance alters to much with varied proximity to the ground. The ability to move the antenna about a bit shows this clearly. Pity. Trying a 3/8 wave wire did make it less twitchy, but would have needed a bit of fiddling to find a good turns ratio.

However, I have learned some things about feeding 1/2 wave end feds, so its been a good exercise. One particular point of interest is that you can do away with the 1/20 wave return by tying the cold end of the Hi-Z end of things to your radio/feeder outer etc. There is enough metalwork there to provide what little of a return the system needs. Otherwise, leave the Hi-Z end floating and use a 1/20 wave “counterpoise”.

I think the EFHWA would be fine for fixed use. As I already have a well made coupler, I’ll swap the core for one made of Type 6 and leave it set up in my student house for a bit of 20m PSK.

So currently, the W3EDP is winning the quest for an end-fed /P antenna. Back to the drawing board!

73 and thanks for the ideas.

Dave M0MYA.


#19

In reply to M0MYA:

I sometimes use, what is in effect, a half size W3EDP for 40m portable. Mine is 42ft long and is fed against 2 x 5m radials. I arrived at that length from a slightly different direction than W3EDP but the end result is the same. I often use mine without an ATU as the VSWR is sufficiently low on 40m to be no problem. Like Richard G4ERP, I would tend to steer clear of end-fed half waves. Feeding an aerial at a voltage node is a recipe for problems in my opinion.

Good luck with the experiments - as you mentioned earlier, they provide yet another excuse for an activation.

73

Richard
G3CWI


#20

In reply to MM0FMF:

http://www.qsl.net/vr2zxp/project/fieldmet/RF%20Field%20Strength%20Meter.jpg

Right then. We’ll have no talk of licenses and corn-flake packets, please!

Clearing out the shed the other day I found an old Micronta analogue multimeter. After a twiddle of the range selector, the shaft sheared, and having enough meters already, Geoff and I made the executive decision to wang it away. However, I spotted the moving coil meter and generous housing and thought that it might make a good basis for a field strength meter. Basically, I have a moving coil meter of unknown specification.

My first attempt to figure things out was bung a DMM across it and read off its internal resistance: ~4.kOhm

I moved on to a slightly “better” test. I connected the DMM in series with the unknown meter and the variable bench supply and adjusted the supply for FSD. A supply of 0.11V and a current of 21.9 microamps was observed. Ohm’s law tells me that the meter has an internal resistance of ~5kOhm. So, 1k adrift from the previous test.

Then I figure that maybe I could connect the unknown meter across the bench supply and adjust for FSD. This turned out to be 0.87V, I thought that this may be due to the internal resistance of the DMM used in test 2 (!?!?). I then put a variable resistor in parallel with the unknown meter, which I was going to adjust for 1/2 scale deflection, remove and measure, thinking that the value of the variable resistor = internal resistance of the unknown meter. Not so. I can’t get it to 1/2 deflection, not even with a dead short across the meter!

So, where am I going wrong, and how do I get nice concrete measurements (internal resistance, current required for FSD) for the unknown meter? And can I jiggery-poke it with resistors such that it will be useable in the circuit at the above link?

With confusion and embarrassment,

Dave M0MYA.