Elecraft

how to optimize end-fed?

Classic

List

Threaded

21 messages Options

Holger Schurig-2

how to optimize end-fed?

Hi all,

I have an end-fed antenna with some random wire. The UNUN at one end of it
has three sockets to plug the random wire in: 1:4, 1:9 and 1:16.

My KX3 has the built-in ATU.

I now want to find out on which band I best use which one of the sockets.
As a first step, I wrote a simple program kx3lc.py (see
https://gist.github.com/anonymous/5d53f5bdbc50782a9d5e2c8d7062be69) that
can give me an output like this:

holger@laptop:/usr/src/afu/kx3/swr$ ./kx3lc.py
L: 0.12 mH, C: 203.0 pF on transmitter side

Am I right to assume that the ATU settings with the lowest L is always the
best? So when I have (for the three sockets), these values,

L: 0.12 mH, C: 203.0 pF on transmitter side
L: 0.0 mH, C: 246.0 pF on transmitter side
L: 0.0 mH, C: 256.0 pF on antenna side (but lowest SWR 1.2-1)

... that the middle socket is the best?

73
Holger, DH3HS

PS: those values are all bogus, I measured when when my end-fed was in a
big curl inside my shack ...
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

k6dgw

Re: how to optimize end-fed?

> Am I right to assume that the ATU settings with the lowest L is
> always the best?

I don't know exactly what you mean by "best," but you said "always" so
I'd say no. If the length of your wire is in the vicinity of n * 90
degrees where n is any odd integer, it will have a low-ish impedance at
the end and 1:1 would be appropriate [if your UNUN had it]. If the
length is around 180 degrees [or any integer multiple thereof] it will
exhibit a fairly high impedance.

How long is the wire?
What band?

If the 1:4, 1:9, and 1:16 that you quote are turns ratios, the impedance
transformation equals the square of them. 12,800 ohms and a 16:1 turns
ratio yields 50 ohms. 12 Kohms likely exceeds the impedance at the end
of a half-wave by quite a bit.

The Elecraft ATU losses are very low unless you're trying to match some
outrageous impedance at the end of or beyond it's useful range.

73,

Fred K6DGW
Sparks NV USA
Washoe County DM09dn

On 9/29/2016 12:50 PM, Holger Schurig wrote:

> Hi all,
>
> I have an end-fed antenna with some random wire. The UNUN at one end of it
> has three sockets to plug the random wire in: 1:4, 1:9 and 1:16.
>
> My KX3 has the built-in ATU.
>
> I now want to find out on which band I best use which one of the sockets.
> As a first step, I wrote a simple program kx3lc.py (see
> https://gist.github.com/anonymous/5d53f5bdbc50782a9d5e2c8d7062be69) that
> can give me an output like this:
>
> holger@laptop:/usr/src/afu/kx3/swr$ ./kx3lc.py
> L: 0.12 mH, C: 203.0 pF on transmitter side
>
>
> Am I right to assume that the ATU settings with the lowest L is always the
> best? So when I have (for the three sockets), these values,
>
> L: 0.12 mH, C: 203.0 pF on transmitter side
> L: 0.0 mH, C: 246.0 pF on transmitter side
> L: 0.0 mH, C: 256.0 pF on antenna side (but lowest SWR 1.2-1)
>
> ... that the middle socket is the best?
>
> 73
> Holger, DH3HS

Elecraft mailing list

Re: how to optimize end-fed?

Fred said it right,

Let me go to another side of the question. No auto-transformer can match all of the antenna reflected impedances, but using a parallel tank circuit to ground CAN. The beauty of this arrangement is that you can tap the coil on the input of the coil and tap the output for a VERY wide range of impedance. R and j . What seems to have been lost in the transition from ancient and now is that we did not have ATU's. All of the PRE now used tuned circuit match boxes. Now you have to think about it for a few minutes.
Let me explain, the tuned parallel tank circuit can do a wide range of matching BECAUSE, If you tun the tank to one side of resonance your get a negative reactance, if you tune it to the other side you get positive reactance. HUMMMMMMM.
If the tank circuit is a true resonant one, the impedance across the coil from top to bottom is a range of the impedance available from HIZ to ZERO. This is the beauty of the parallel tuned circuit over a auto transformer. Oh well at least I remember it.....
Mel, K6KBE

From: Fred Jensen <[hidden email]>
To: [hidden email]
Sent: Thursday, September 29, 2016 1:57 PM
Subject: Re: [Elecraft] how to optimize end-fed?

> Am I right to assume that the ATU settings with the lowest L is
> always the best?

I don't know exactly what you mean by "best," but you said "always" so
I'd say no. If the length of your wire is in the vicinity of n * 90
degrees where n is any odd integer, it will have a low-ish impedance at
the end and 1:1 would be appropriate [if your UNUN had it]. If the
length is around 180 degrees [or any integer multiple thereof] it will
exhibit a fairly high impedance.

How long is the wire?
What band?

If the 1:4, 1:9, and 1:16 that you quote are turns ratios, the impedance
transformation equals the square of them. 12,800 ohms and a 16:1 turns
ratio yields 50 ohms. 12 Kohms likely exceeds the impedance at the end
of a half-wave by quite a bit.

The Elecraft ATU losses are very low unless you're trying to match some
outrageous impedance at the end of or beyond it's useful range.

73,

Fred K6DGW
Sparks NV USA
Washoe County DM09dn

On 9/29/2016 12:50 PM, Holger Schurig wrote:

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Don Wilhelm

Re: how to optimize end-fed?

Mel,

Your points are well taken when you are feeding an end-fed half wave or
a radiator of a multiple of a half wave.

The ideal parallel tank circuit should be fed with a coupling coil that
is isolated from the larger secondary coil - that reduces or eliminates
common mode currents in the shack.

That solution is practical for open coil configurations where the user
can tap the antenna to any point on the coil of the tuned circuit.
However, the tapping choices are limited in a compact, portable
implementation for use in portable operation which use toroids for the
inductor.

We used to use those isolated link coupled ATUs constructed of open coil
inductors where it was practical to tap the coil at any point, but
today's world of toroid wound inductors, that is not as easy.

If one has a resonant parallel tuned circuit, it will match very high
impedance, and a series tuned circuit will match very low impedance.
The link coupling will provide isolation from common mode currents.

Those type of ATU's work very well with a wide range of antennas with
varying feedpoint impedance.

However, the physical implementation of the ability to tap the antenna
to any turn of the high impedance tuned resonant inductor requires a
physically large coil. While it will work *very* well, it is not
consistent with small ATUs used for portable operation.

If you have a fixed length radiator and work only a single band, you can
devise a link coupled tuner that will do a great job, but if you are
multiband, and do not want to fiddle with coil taps, the
auto-transformer is a good compromise.

Yes, I still have my link coupled ATUs with plug-in coils for each band
and also have a Johnson Matchbox which is also link coupled. They do the
job well, but the convenience of toroid wound ATUs is an advantage in
simplicity and convenience.

73,
Don W3FPR

On 9/29/2016 8:04 PM, Mel Farrer via Elecraft wrote:
> Fred said it right,
>
> Let me go to another side of the question. No auto-transformer can match all of the antenna reflected impedances, but using a parallel tank circuit to ground CAN. The beauty of this arrangement is that you can tap the coil on the input of the coil and tap the output for a VERY wide range of impedance. R and j . What seems to have been lost in the transition from ancient and now is that we did not have ATU's. All of the PRE now used tuned circuit match boxes. Now you have to think about it for a few minutes.
> Let me explain, the tuned parallel tank circuit can do a wide range of matching BECAUSE, If you tun the tank to one side of resonance your get a negative reactance, if you tune it to the other side you get positive reactance. HUMMMMMMM.
> If the tank circuit is a true resonant one, the impedance across the coil from top to bottom is a range of the impedance available from HIZ to ZERO. This is the beauty of the parallel tuned circuit over a auto transformer. Oh well at least I remember it.....
> Mel, K6KBE
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Jim Brown-10

Re: how to optimize end-fed?

In reply to this post by Holger Schurig-2

This discussion makes me wonder if the ARRL Handbook and the ARRL
Antenna Book have somehow been banned! This is BASIC antennas, guys.
This is not CB radio, it is ham radio, and we're supposed to be able to
study this stuff and figure it out! Building antennas is one the
easiest parts of ham radio to do yourself, and once you've invested the
time to learn the fundamentals, it's a lot of fun!

New copies of these books can be bought for less than the cost of the
overpriced antennas mentioned here that mostly don't work very well,
while the cost of the wire to build much better antennas can be had for
a few dollars! Wire as small as #22 works fine for temporary wire
antennas for QRP operation. Insulators are easy to improvise with piece
of plastic with holes drilled in them.

While we can make a wire of random length load (take power) and radiate,
it's far easier if we make it a quarter wave. "One size fits all"
antennas are like "one size fits all" clothing. They work, sort of, but
FITS (and LOOKS) a heckuva lot better.

The most expensive part of a vertical for portable or hiking operation
is whatever you use to hold it up. I paid about $100 for a 10M long
telescoping fiberglass pole that collapses to 1 M. Tape the wire to it,
run out a couple of radials, and you've got a great antenna for any band
between 40M and 10M (just trim the wires to a quarter wave on the band
you want to work). For less than $10, you could do the same with 2-3 10
ft lengths of 1/2-in PVC conduit.

The picture of me on my qrz.com page is from our Chicago club's annual
QRP night in a local park My antenna was #22 wire wound on that 10M
pole, the pole was wedged between the seat and the top of the picnic
bench so it was at roughly 45 degrees to vertical, and I had one or two
radials laying on the ground. In a few hours, I made a half dozen QSOs
on 30M, including busting a pileup. A few years ago, W6GJB and I set up
on a peak near Sacramento with a KX3 running on internal AAs into a
similar antenna for 15M. We worked three continents in about 10 minutes.
There were two radials laying on top of low brush. The vertical element
was held up by a small tripod intended for a small camera.

Nothing personal intended, but my opinion of virtually all of these
antennas you buy comes under the heading of "a fool and his money."

73, Jim K9YC

On Thu,9/29/2016 12:50 PM, Holger Schurig wrote:

Holger Schurig-2

Re: how to optimize end-fed?

In reply to this post by k6dgw

Fred Jensen <[hidden email]> writes:

> How long is the wire?
> What band?

Who knowns? I'm not asking for a *specific* optimization, I'm asking
for a general rule of thumb.

I'm not talking about a permanent installation, I'm talking about going
to some field, or hill. I don't know the band in advance, I don't know
the propagation conditions. And I have several lengths of wire with me,
I can adjust the wire length. And given the circumstances of the
location, even the layout of my wire (only on fiber glass post, between
tree and fiber glass post, etc) will change (e.g. sloper, vertical,
etc). Even the high will be undefined and/or changeable.

That is actually to fun (for me): to change things and try out.

But then arises the question: if I change this or that ... how can I
find out what setting is "good" or maybe even "best". Assuming I have
only the things with me that I normally have with me when I'm in the
field: KX3 and laptop.

The most easy thing to change (because it's often near the bottom) is
the UNUN tap I'm using. And I can check the LC-setting in the ATU.

And now I would like to know if there is a method --- in the field ---
where I can optimize wire and tap. And if a low inductance of the ATU
could be measure for it. And if this is true in both setings (the
inducator can be at the antenna side or at the radio side).

So either this question has never been solved amongst the readers here
... or I'm completely on the wrong track and shouldn't care at all what
the ATU finds as a match.

So far Ron's answer come the farthest in answering the question, he
optimizes for a high impedance. Then I should go for the highest
impedance and select 1:16, should I?

73,
Holger DHSHS
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Holger Schurig-2

Re: how to optimize end-fed?

> Then I should go for the highest impedance and select 1:16, should I?

I meant antenna impedance, not the impedance of the internal ATU
inductor.
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Deni F5VJC

Re: how to optimize end-fed?

If you are working QRP, take a look at
http://www.qrpproject.de/UK/atu.htm

This describes the 'Fuchs' half wave antenna and a matching unit in kit
form. I have built this, very compact, it works well, gives a good
indication of optimum tune adjustment and is not totally dependent on a
good radial ground system.

73, Denis F5VJC

On 30 September 2016 at 08:29, Holger Schurig <[hidden email]>
wrote:

> > Then I should go for the highest impedance and select 1:16, should I?
>
> I meant antenna impedance, not the impedance of the internal ATU
> inductor.
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

Don Wilhelm

Re: how to optimize end-fed?

In reply to this post by Holger Schurig-2

Holger,

There are many answers to your questions. You will have to get more
specific about the wire lengths and the band(s) of operation for
specific answers.

If you want an easy antenna that will work for 40 thru 10 meters with no
feedline, download the KXAT1 manual from Elecraft and look at page 10.
It is just two wires and a BNC to binding post adapter, you can't get
much simpler than that.

If you just want to experiment with antennas, feedlines and matching
networks, do it with some education - get the ARRL Handbook and the ARRL
Antenna Book ans spend some time studying while doing your experiments.

73,
Don W3FPR

On 9/30/2016 2:24 AM, Holger Schurig wrote:

> Fred Jensen <[hidden email]> writes:
>
>> How long is the wire?
>> What band?
> Who knowns? I'm not asking for a *specific* optimization, I'm asking
> for a general rule of thumb.
>
> I'm not talking about a permanent installation, I'm talking about going
> to some field, or hill. I don't know the band in advance, I don't know
> the propagation conditions. And I have several lengths of wire with me,
> I can adjust the wire length. And given the circumstances of the
> location, even the layout of my wire (only on fiber glass post, between
> tree and fiber glass post, etc) will change (e.g. sloper, vertical,
> etc). Even the high will be undefined and/or changeable.
>
> That is actually to fun (for me): to change things and try out.
>
>

Lynn W. Taylor, WB6UUT

Re: how to optimize end-fed?

In reply to this post by Holger Schurig-2

Maybe I'm missing something, but you have a selection of taps, the
highest is 16:1.

If the wire is exactly 1/2 wave, then the impedance is going to be very
high, and the 16:1 tap will come closest to matching a high impedance load.

If the wire is some multiple of 1/4 wave but not a an even multiple (not
an multiple of a half wave) then the lowest tap is probably best.

I'd disable the tuner and check the SWR on each tap for each band -- if
I wanted to verify the (very simple) math.

73 -- Lynn

On 9/29/2016 11:24 PM, Holger Schurig wrote:
> The most easy thing to change (because it's often near the bottom) is
> the UNUN tap I'm using.

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

k6dgw

Re: how to optimize end-fed?

In reply to this post by Holger Schurig-2

Let me see if I understand the question: You're looking for a simple
algorithm to tell you how to "best" deploy a wire of indeterminate
length in a configuration yet to be specified using a transformer of
indefinite impedance ratio on any of a number of unnamed bands.

This is reminiscent of one of the five volumes in Douglas Adams'
trilogy, "The Hitchhikers Guide to the Galaxy." The people build a very
powerful computer and then asked it, "What is the meaning of the
Universe?" The computer begins to compute, as is the wont of computers,
and worked on the problem for a very long time. Finally, it announced
that it had determined the answer. By this time, there was no one in
the Universe that remembered the original question. When asked for the
answer, the computer replied, "Forty Two."

Possibly "forty two" would work for you as a rule of thumb? :-))

73,

Fred K6DGW
- Sparks NV DM09dn

- Northern California Contest Club
- CU in the Cal QSO Party 1-2 Oct 2016
- www.cqp.org

On 9/29/2016 11:24 PM, Holger Schurig wrote:

> Fred Jensen <[hidden email]> writes:
>
>> How long is the wire?
>> What band?
>
> Who knowns? I'm not asking for a *specific* optimization, I'm asking
> for a general rule of thumb.
>
> I'm not talking about a permanent installation, I'm talking about going
> to some field, or hill. I don't know the band in advance, I don't know
> the propagation conditions. And I have several lengths of wire with me,
> I can adjust the wire length. And given the circumstances of the
> location, even the layout of my wire (only on fiber glass post, between
> tree and fiber glass post, etc) will change (e.g. sloper, vertical,
> etc). Even the high will be undefined and/or changeable.

n4kd

Re: how to optimize end-fed?

I think 42 might be a valid answer, depending on the question. Forty-one is certainly a good answer!
There are a couple pages that try to address the question of what long wire lengths are the easiest to tune, i.e. present a low SWR to the matching unit.
The first is at http://udel.edu/~mm/ham/randomWire/ and it's work is derived from what was presented here, http://www.qsl.net/on7dy/Documentation/Best%20%20Random%20Wire%20Antenna%20Lengths.htm
Hope this helps a little. With my KX3 tuner, it doesn't seem to matter much what I use for a long wire. The tuner does a good job matching the antenna to the rig.
vy 73,Dave N4KD

On Friday, September 30, 2016 4:01 PM, Fred Jensen <[hidden email]> wrote:

Let me see if I understand the question: You're looking for a simple
algorithm to tell you how to "best" deploy a wire of indeterminate
length in a configuration yet to be specified using a transformer of
indefinite impedance ratio on any of a number of unnamed bands.

This is reminiscent of one of the five volumes in Douglas Adams'
trilogy, "The Hitchhikers Guide to the Galaxy." The people build a very
powerful computer and then asked it, "What is the meaning of the
Universe?" The computer begins to compute, as is the wont of computers,
and worked on the problem for a very long time. Finally, it announced
that it had determined the answer. By this time, there was no one in
the Universe that remembered the original question. When asked for the
answer, the computer replied, "Forty Two."

Possibly "forty two" would work for you as a rule of thumb? :-))

73,

Fred K6DGW
- Sparks NV DM09dn

- Northern California Contest Club
- CU in the Cal QSO Party 1-2 Oct 2016
- www.cqp.org

On 9/29/2016 11:24 PM, Holger Schurig wrote:

k6dgw

Re: how to optimize end-fed?

Yes ... and forty three is held in reverence by others who attribute
magical properties to it. :-)

73,

Fred K6DGW
- Sparks NV DM09dn

- Northern California Contest Club
- CU in the Cal QSO Party 1-2 Oct 2016
- www.cqp.org

On 9/30/2016 1:22 PM, David Kuechenmeister wrote:
> I think 42 might be a valid answer, depending on the question.
> Forty-one is certainly a good answer! There are a couple pages that
> try to address the question of what long wire lengths are the easiest
> to tune, i.e. present a low SWR to the matching unit. The first is at
> http://udel.edu/~mm/ham/randomWire/ and it's work is derived from
> what was presented here,
> http://www.qsl.net/on7dy/Documentation/Best%20%20Random%20Wire%20Antenna%20Lengths.htm
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Holger Schurig-2

Re: how to optimize end-fed?

In reply to this post by Don Wilhelm

>
> You will have to get more specific about the wire lengths and the band(s)
> of operation for specific answers

Nope. I don't have.

My question was really: is it desirable to always aim for the lowest
inductance of an ATU tuning. AFAIK this question is totally independent
from the wire length. If anything, I'll optimize the wire length to make
this happen ... or I won't care if my thesis is all bogus.
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Holger Schurig-2

Re: how to optimize end-fed?

In reply to this post by k6dgw

2016-09-30 21:59 GMT+02:00 Fred Jensen <[hidden email]>:

> This is reminiscent of one of the five volumes in Douglas Adams' trilogy,
> "The Hitchhikers Guide to the Galaxy."

I understand that either my english is very weird. Or that I can't explain
things good.

But that you and Davidthink that they must make a 42 joke on this is
definitely weird.

The question was: I can query the KX3 ATU for what it settled. I have an
end fed antenna and so I have various variables:

- used tap (1:4, 1:9, 1:16)
- some random length wire
- band

And forget an "optimal wire", I might just have switched from 12m to 10m.
Or back. Depends on what I find, propagation ... so assume that my wire
is just some random wire, not necessary optimal for the band. And also, in
the context of my question, this is entirely irrelavant. I was never asking
about wire lengths, this is easy to read up. Okay, back to my scenario: I
just switched the new band. I'm not going to let my portable glass fiber
down because of that and change the wire length! Instead I do what a lazy
OM does: I press the TUNE button and the internal magical antenna tuner
does it's job. It's actually so magic, that it will do it's job on all
taps. On the 1:4 tap, on the 1:9 tap. And on the 1:16 tap. Woah! But I can
query the ATU for what inductance and capacitance it used to do the match.
And so my simple question was: would a lower inductance have less losses
inside the ATU?

And please: if you don't know the answer of if you think that there is no
answer, than just stay silent.

73, Holger
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Heinz Baertschi

Re: how to optimize end-fed?

In reply to this post by Holger Schurig-2

Holger Schurig-2 wrote

I have an end-fed antenna with some random wire. The UNUN at one end of it
has three sockets to plug the random wire in: 1:4, 1:9 and 1:16.

My KX3 has the built-in ATU.
I now want to find out on which band I best use which one of the sockets.

Am I right to assume that the ATU settings with the lowest L is always the
best? So when I have (for the three sockets), these values,

L: 0.12 mH, C: 203.0 pF on transmitter side
L: 0.0 mH, C: 246.0 pF on transmitter side
L: 0.0 mH, C: 256.0 pF on antenna side (but lowest SWR 1.2-1)

... that the middle socket is the best?

Your question was indeed crystal clear, so I am sorry that you had to make this experience.
Don't worry and have fun!

BTW, I suspect that the tuning algorithm used by Wayne for all the AT of Elecraft does start his try with the minimum inductance (unknown from competitive reasons, hi)?

Heinz Baertschi

Re: how to optimize end-fed?

Holger,

Your question was indeed crystal clear, so I am sorry that you had to make this experience.
Don't worry and have fun!

BTW, I suspect that the tuning algorithm used by Wayne for all the AT of Elecraft does start his try with the minimum inductance (unknown from competitive reasons, hi)?

73, Heinz HB9BCB

Lynn W. Taylor, WB6UUT

Re: how to optimize end-fed?

In reply to this post by Holger Schurig-2

In my opinion, what you want is the lowest SWR on the feed line between
the Balun and the tuner.

The higher the mismatch, the higher the feedline loss.

Let's say you're operating on 14.300 MHz, and your wire is roughly 9.85
meters (32' 4") long.

If I did the math right, that's pretty much exactly 1/2 wave, and the
impedance at the feedpoint (the end) is near infinity.

Your 16:1 tap might be good enough to bring that into the ATU range.
The SWR would likely be high.

Make the wire a bit longer (to get away from the exact 1/2 wave) and the
impedance comes down.

At some point, you'll get a better match (lower SWR) on the 4:1 tap than
the 16:1 tap.

For some wire lengths, the 1:1 tap will give the lowest SWR between the
tuner and the wire.

I'll be honest and say that I don't know how the inductance in the tuner
is related to SWR.

If you trim the wire so that the impedance is 800 ohms (16 times 50) and
use the 16:1 tap, the SWR between the UNUN and the Tuner should be 1:1,
but we're no longer talking about "random" wires.

I'm sure those who invoked The Hitchhiker's Guide to the Galaxy were
trying to point out that, without getting into a lot of specifics, it's
hard to answer. That's why I picked 14.300 MHz -- to make the question
far more exact. Humor doesn't always translate, however.

Stay on 14.300 and make the wire three times longer, and the discussion
stays the same -- instead of the wire being 1/2 wave long, it's 1 1/2 waves.

Move to 7.150 and double the length of the wire, and the discussion
stays the same, only the lengths change.

This page <http://udel.edu/~mm/ham/randomWire/> talks about "random"
wires, and the lengths that do fall near an odd multiple of 1/2 wave.
Staying away from these lengths makes the tuner's job easier.

I'm also ignoring (on purpose) how well the antenna radiates. It seems
to me that if the power doesn't get into the antenna, it doesn't much
matter.

I'm sure we'll now hear suggestions from those who think another type of
antenna is better, but your original question is the same with a miscut
dipole as it is with a high-impedance end-fed wire.

73 -- Lynn

On 10/1/2016 7:26 AM, Holger Schurig wrote:
> My question was really: is it desirable to always aim for the lowest
> inductance of an ATU tuning. AFAIK this question is totally independent
> from the wire length. If anything, I'll optimize the wire length to make
> this happen ... or I won't care if my thesis is all bogus.

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Logan Zintsmaster

Re: how to optimize end-fed?

I just finished typing up the pretty much the same answer. Should have
gotten up earlier.

To add to Lynn's comments, setting the balun tap to the lowest SWR means you
are minimizing the work that the
tuner has to do and as a result probably minimizing it's losses, as well.

So to summarize...

1. Bypass the tuner and select the balun tap that gives you the lowest SWR.

2. Activate the tuner and enjoy lots of QSOs.

Logan

-----Original Message-----
From: Elecraft [mailto:[hidden email]] On Behalf Of Lynn
W. Taylor, WB6UUT
Sent: Saturday, October 01, 2016 9:27 AM
To: [hidden email]
Subject: Re: [Elecraft] how to optimize end-fed?

In my opinion, what you want is the lowest SWR on the feed line between the
Balun and the tuner.

The higher the mismatch, the higher the feedline loss.

Let's say you're operating on 14.300 MHz, and your wire is roughly 9.85
meters (32' 4") long.

If I did the math right, that's pretty much exactly 1/2 wave, and the
impedance at the feedpoint (the end) is near infinity.

Your 16:1 tap might be good enough to bring that into the ATU range.
The SWR would likely be high.

Make the wire a bit longer (to get away from the exact 1/2 wave) and the
impedance comes down.

At some point, you'll get a better match (lower SWR) on the 4:1 tap than the
16:1 tap.

For some wire lengths, the 1:1 tap will give the lowest SWR between the
tuner and the wire.

I'll be honest and say that I don't know how the inductance in the tuner is
related to SWR.

If you trim the wire so that the impedance is 800 ohms (16 times 50) and use
the 16:1 tap, the SWR between the UNUN and the Tuner should be 1:1, but
we're no longer talking about "random" wires.

I'm sure those who invoked The Hitchhiker's Guide to the Galaxy were trying
to point out that, without getting into a lot of specifics, it's hard to
answer. That's why I picked 14.300 MHz -- to make the question far more
exact. Humor doesn't always translate, however.

Stay on 14.300 and make the wire three times longer, and the discussion
stays the same -- instead of the wire being 1/2 wave long, it's 1 1/2 waves.

Move to 7.150 and double the length of the wire, and the discussion stays
the same, only the lengths change.

This page <http://udel.edu/~mm/ham/randomWire/> talks about "random"
wires, and the lengths that do fall near an odd multiple of 1/2 wave.
Staying away from these lengths makes the tuner's job easier.

I'm also ignoring (on purpose) how well the antenna radiates. It seems to
me that if the power doesn't get into the antenna, it doesn't much matter.

I'm sure we'll now hear suggestions from those who think another type of
antenna is better, but your original question is the same with a miscut
dipole as it is with a high-impedance end-fed wire.

73 -- Lynn

On 10/1/2016 7:26 AM, Holger Schurig wrote:
> My question was really: is it desirable to always aim for the lowest
> inductance of an ATU tuning. AFAIK this question is totally
> independent from the wire length. If anything, I'll optimize the wire
> length to make this happen ... or I won't care if my thesis is all bogus.

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html Message
delivered to [hidden email]

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Kevin - K4VD

Re: how to optimize end-fed?

In reply to this post by Holger Schurig-2

Hi Holger:

I just finished reading an article that may answer your question:
http://www.arrl.org/files/file/Technology/tis/info/pdf/9501046.pdf

If I'm understanding it correctly, the inductor's Q factor is the issue.
The Cout capacitance should be set to the highest value possible with the L
and Cin supporting that for lowest SWR. I think this means the higher the
inductor's Q the less losses seen in the matching network. Does this mean
the higher the inductance (higher inductive reactance) the higher Q and the
lower loss? That's how it is looking to me.

I don't know what kind of matching network is used in the Elecraft ATU.
I'm hoping someone with a better understanding can verify or correct what
I've said.

73,
Kevin K4VD

On Sat, Oct 1, 2016 at 10:39 AM, Holger Schurig <[hidden email]>
wrote:

> 2016-09-30 21:59 GMT+02:00 Fred Jensen <[hidden email]>:
>
> > This is reminiscent of one of the five volumes in Douglas Adams' trilogy,
> > "The Hitchhikers Guide to the Galaxy."
>
>
> I understand that either my english is very weird. Or that I can't explain
> things good.
>
> But that you and Davidthink that they must make a 42 joke on this is
> definitely weird.
>
>
> The question was: I can query the KX3 ATU for what it settled. I have an
> end fed antenna and so I have various variables:
>
> - used tap (1:4, 1:9, 1:16)
> - some random length wire
> - band
>
> And forget an "optimal wire", I might just have switched from 12m to 10m.
> Or back. Depends on what I find, propagation ... so assume that my wire
> is just some random wire, not necessary optimal for the band. And also, in
> the context of my question, this is entirely irrelavant. I was never asking
> about wire lengths, this is easy to read up. Okay, back to my scenario: I
> just switched the new band. I'm not going to let my portable glass fiber
> down because of that and change the wire length! Instead I do what a lazy
> OM does: I press the TUNE button and the internal magical antenna tuner
> does it's job. It's actually so magic, that it will do it's job on all
> taps. On the 1:4 tap, on the 1:9 tap. And on the 1:16 tap. Woah! But I can
> query the ATU for what inductance and capacitance it used to do the match.
> And so my simple question was: would a lower inductance have less losses
> inside the ATU?
>
> And please: if you don't know the answer of if you think that there is no
> answer, than just stay silent.
>
>
> 73, Holger
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>