Hello:

I am contemplating a half-wave wire antenna for the 160m band and would
end-feed it against my roof as a counterpoise. If the wire length is
carefully trimmed I should get a high feedpoint resistance with
negligible reactance. Then comes the question of how to match this to
a 50 Ohm system.

The standard approach is using a parallel LC circuit and either make a
tap or a link for the 50 Ohm connection. In this case, how should I
select the L and C, as there are multiple choices leading to the same
resonance frequency but with different Q values. Any suggestions?

Or, would another matching network be more desirable, for example in
terms of loss? An L-network might do the job, but it is not
immediately clear to me if it would have any advantages or
disadvantages over the parallel circuit. Any comments?

I will be running substantial power into the antenna, so it is
important that the solution does not result in excessive losses (heat)
or voltages (although this will probably be inevitable at the
feedpoint).

73 - Kris, TF3KX

The standard approach is using a parallel LC circuit and either make a
tap or a link for the 50 Ohm connection. In this case, how should I
select the L and C, as there are multiple choices leading to the same
resonance frequency but with different Q values. Any suggestions?

Or, would another matching network be more desirable, for example in
terms of loss? An L-network might do the job, but it is not
immediately clear to me if it would have any advantages or
disadvantages over the parallel circuit. Any comments?

I will be running substantial power into the antenna, so it is
important that the solution does not result in excessive losses (heat)
or voltages (although this will probably be inevitable at the
feedpoint).

73 - Kris, TF3KX

GM Kris

I use an end fed, inverted L half wave on 160 meters on 160, 80 and 40
meters where it has a feed impedance of about 2600, 1600 and 700 Ohms
respectively. This allows a remote tuned L network to be used for a match to
50 Ohms on all 3 bands without need for switching the capacitor from end to
end of the inductor. My inductor is a piece of 2-1/2 inch miniductor, tap
switched with a motor driven switch, and a motor driven vac variable. It
allows a below 1.3:1 match across the entire band on all 3 bands and has
been a very effective HF antenna for some 15 years. The bottom of the
antenna ties to a control box thru a fairly small ceramic feedthru, (maybe 2
inches long) and Voltage has never been a problem even with 1500 Watts out.
Fed against a 90 foot well casing that just happens to be in the right
place, and a counterpoise under the flat top.

When you don't KNOW that your ground is zero Ohms, feed as little current
into it as possible!

Regards
W4ZCB

An parallel LC network like you describe is usually made with L and C
having reactances of roughly 200 ohms or so. That generally gives a
loaded Q that results in good efficiency. As you suspect, the voltage
will be high at the end of the antenna no matter how you feed it.

Roy Lewallen, W7EL

wrote:
Hello:

I am contemplating a half-wave wire antenna for the 160m band and would
end-feed it against my roof as a counterpoise. If the wire length is
carefully trimmed I should get a high feedpoint resistance with
negligible reactance. Then comes the question of how to match this to
a 50 Ohm system.

The standard approach is using a parallel LC circuit and either make a
tap or a link for the 50 Ohm connection. In this case, how should I
select the L and C, as there are multiple choices leading to the same
resonance frequency but with different Q values. Any suggestions?

Or, would another matching network be more desirable, for example in
terms of loss? An L-network might do the job, but it is not
immediately clear to me if it would have any advantages or
disadvantages over the parallel circuit. Any comments?

I will be running substantial power into the antenna, so it is
important that the solution does not result in excessive losses (heat)
or voltages (although this will probably be inevitable at the
feedpoint).

73 - Kris, TF3KX

In older handbooks, they often say that the parallel LC circuit used to
feed the end fed wire should replicate the output tank circuit of the
transmitter.

I have an end fed inverted "L" for 40 meters that works really well.
It's 32 feet vertical and 98 feet horizontal, with the far end 8 ft.
off the ground.

I use a parallel tuned circuit at the base in a weatherproof box and
it's fed with a 100 feet or so of RG 8 buried in PVC. The coax center
conductor is tapped up a turn or two from the bottom of the coil.

I consider this a single band antenna, and I use it only for 40 meters.

John

MATCHING END-FED 1/2-WAVE ANTENNAS TO THE TRANSMITTER.

Instead of depending on hearsay and old-wives tales, there is a
logical way of selecting tuned-circuit L and C values.

To begin, in the case of impedance matching it is always required to
maximise power efficiency.

It is the power lost in the tuned circuit which matters.

To minimise power lost in the tuned circuit it is necessary to
minimise the current circulating in it. (Circulating current is the
antenna current times working Q but there's no need to know what Q
actually is.)

For a given radiated power the voltage across the tuned circuit and
antenna when at resonance is fixed. So to minimise circulating
current the reactances of L and C should be as high as practicably
possible. That's obvious!

So for a high reactance of C (which is approximately the reactance of
L), C should be mimised.

The limit on the smallness of C is the minimum setting value of the
tuning capacitor, plus coil self-capacitance, plus stray capacitance
associated with wiring and connections to the end of the antenna, plus
an allowance for the antenna not being exactly at resonance, plus a
small 'safety' margin to allow the tuning capacitor to be adjusted to
either above or below resonance.

For half-wave antennas at frequencies below 11 MHz a tuning capacitor
setting of 20 or 25 pF is suitable. This choice automatically sets
the value of the inductance (which is usually higher than the
inductance value of a roller inductor and can be home-brewed quite
easily.)

The maximum setting of the tuning capacitor allows the circuit with
the antenna to be tuned to the lower frequency bands. But efficiency
will be reduced on lower frequency bands due to smaller reactances of
L and C and the greater circulating current. Note that the reactances
of L and C are always approximately equal to each other.

Theoretically, if the antenna length is slightly shorter than
1/2-wavelength, then it is impossible to obtain an impedance match
with this particular circuit. You can't have a tuning capacitor less
than zero pF. It is prudent to select the overall antenna length to be
a few feet longer than 1/2-wavelength at the desired operating
frequency.

On the other hand, it is just as well to remember, a non-resonant
end-fed antenna wire is no less efficient than an exactly resonant
wire provided it can be matched to whatever load resistance is
required by the transmitter. (And the load resistance presented to
the tansmitter as indicated by a TLI has nothing to do with the
so-called, and imaginary, SWR).

As Walter Maxwell has peviously pointed out, SWR = Short Wave Radio.

But, despite its diagramatic simplicity, a parallel tuned circuit is a
not-so-simple, quite complicated arrangement when used as an impedance
matching transformer between an end-fed antenna and the transmitter.
No maths needed - just arithmetic.

Slight change of topic -

I am presently imbibing White, Argentinian, Chardonnay Torrentes. I
do hope my use of plain English language does not lead to
misunderstandings.

May I add that I much prefer a decent, intoxicated, political
opposition-party leader to a sober, war-mongering Tony Blair. At
least a few of you USA citizens must be acquainted with the relatively
trivial events taking place in the 'mother' country. In sympathy with
what's happening in the US.

When will you impeach your so-called president and rid yourselves of
your secret Nazi/OGPU CIA police. I'm sure the Constitution does not
allow such people to exist.

We would be happy to confine both Blair and Bush, here, in the
historic Tower of London, chained to the cold stone walls, for an
indefinite period, on bread and water, without access to legal
representation, as so many other kidnapped prisoners (disappeared
people), for years, are being presently detained without sound reason
or charge.

Have you no conscience? Or have you no idea what a map of the World
looks like?

To obtain an elementary understanding of a simple tuned circuit as an
impedance-matching circuit, download program TUNEHALF from website
below. There are other programs available related to the same
subject.
----
.................................................. ..........
Regards from Reg, G4FGQ
For Free Radio Design Software go to
http://www.btinternet.com/~g4fgq.regp
.................................................. ..........
=========================================

To obtain an elementary understanding of a simple tuned circuit as an
impedance-matching circuit, download program TUNEHALF from website
below. There are other programs available related to the same
subject.

Intoxicated BS Diatribe snipped

Or, use an "L" network. Same number of components, and it will even match
impedances that call for negative component values in a parallel tuned
circuit.

W4ZCB

Or, use an "L" network. Same number of components, and it will even
match
impedances that call for negative component values in a parallel
tuned
circuit.

W4ZCB
======================================

Actually, with an exact 1/2-wave antenna, an L-network suffers from
the same disability as does the simple L and C tuned circuit.

This is why automatic antenna tuner manufacturers, which use
L-networks, state in their operating handbooks that exactly 1/2-wave
antennas should be avoided.

But it is not an important matter. Very few antennas are EXACTLY
1/2-wavelength in length. And it is a simple matter to shift the
resonant frequency up or down a few KHz.
----
Reg, G4FGQ.

Or, use an "L" network. Same number of components, and it will
even
match
impedances that call for negative component values in a parallel
tuned
circuit.

W4ZCB
======================================

Actually, with an exact 1/2-wave antenna, an L-network suffers from
the same disability as does the simple L and C tuned circuit.

This is why automatic antenna tuner manufacturers, which use
L-networks, state in their operating handbooks that exactly 1/2-wave
antennas should be avoided.

But it is not an important matter. Very few antennas are EXACTLY
1/2-wavelength in length. And it is a simple matter to shift the
resonant frequency up or down a few KHz.
----
Reg, G4FGQ.
============================================

It was incorrect of me to say that an L-network suffers from the same
disability as the parallel L and C tuned circuit. But it does suffer
from disabilities of a different sort.

As you say, Harold, the L-network can be tuned to either side of
antenna resonance without calling for impossible negative component
settings.

The input impedance of a 1/2-wave antenna at resonance is a pure
resistance of several thousand ohms. But only very slightly on either
side of resonance the series resistive component of Zin falls to a
much smaller value - and there arises a series reactive component
which can also rise to several thousand ohms. Things change very fast
versus frequency.

The automatic tuner manufacturers recommend avoidance of end-fed
1/2-wave antennas because very high values of switched inductance are
needed on the lower frequency HF bands to provide an impedance match
to 50 ohms. You will be aware, in automatic tuners, that both L and C
are switched.

It is uneconomic to provide large microhenry values of switched
inductance for use only on antennas which are rarely found, i.e. on
end-fed wires which happen to be exactly 1/2-wave resonant.

It is also uneconomic to provide a switched range of small value capac
itors, say from 0 to 20 pF.

Whereas, with an L & C parallel tuned circuit the coil is home-brewed,
it has high intrinsic Q, it is fixed in value, and tuning is done by
varying only the capacitor. Impedance matching to the transmitter is
done by tapping the transmission line up and down the coil turns.

It also transpires that the power efficiency of a parallel tuned L & C
matching circuit is higher than an L-network which does the same job.
That is because the circuit designer has control over choice of L and
C values. Whereas, with an L-match, the L and C values are
pre-determined by the values of the load and generator impedances over
which the circuit designer has no control.

There is an additional complication of L-network design. At certain
values of load and generator impedances, the location and values of
L's and C's in the circuit can suddenly change versus frequency. And
there is always a choice between 2 circuits which perform same job.

Anyone interested can download program L_TUNER from website below.

The antenna input impdances obtained from program TUNEHALF can be
inserted in program L_TUNER and the L and C component values compared
between one type of tuner and the other.

Remember that with the L-tuner both L and C are variable. With the
parallel L and C tuned circuit, in general, only C is variable and
impedance matching is accomplished by tapping up the coil. ( or by
varying the turns on a link-coupling which, by the way, confers other
advantages.)
----
.................................................. ..........
Regards from Reg, G4FGQ
For Free Radio Design Software go to
http://www.btinternet.com/~g4fgq.regp
.................................................. ..........

Anyone interested can download program L_TUNER from website below.

The antenna input impdances obtained from program TUNEHALF can be
inserted in program L_TUNER and the L and C component values compared
between one type of tuner and the other.

Remember that with the L-tuner both L and C are variable. With the
parallel L and C tuned circuit, in general, only C is variable and
impedance matching is accomplished by tapping up the coil. ( or by
varying the turns on a link-coupling which, by the way, confers other
advantages.)
----
.................................................. .........
Regards from Reg, G4FGQ

MUCH better Reg.

At one time, my 160 meter half wave was fed with a parallel tuned circuit
using a huge roller inductor (5 inch diameter, foot long surplus thing) I
used the same motor driven vacuum variable to obtain resonance and the
roller "tap" was only to obtain the 50 Ohm match. Was a fine 160 meter
antenna, but my 40 meter quad was an eyesore and difficult to maintain, and
I had nothing up for 80. Swapping out the inductor for a piece of Airdux and
tapping that for 3 values of inductance and selecting those with an old
Ledex motor driven rotary switch gave me all 3 bands with an "L" network.
Difficult to accomplish with a parallel tuned circuit without appreciable
additional complexity, never noticed any difference in performance on 160
meters. We now have two copycat antennas, up for several years. W4GCK and
W1TRF, both are as well pleased as I have been. GCK and I have a directional
antenna for the higher bands, TRF uses his on all bands.

Regards
W4ZCB