Looking through my junkbox I found a roller coil and several larger
variable capacitors that I had put away some years ago with the idea of
building a 'transmatch'. I have one 250pf and two 500pf variables,
probably in the 300 watt category (which would be fine for me). In
looking over QST articles I see three different possible configurations:
1: the Ultimate Transmatch from 1970
2: the SPC transmatch from the late 1970's
3: the "T" network

From what I've read, the old Ultimate circuit had a good wide range and
low loss. It's bad point was that it could act as a high pass network
instead of a band pass network in some settings. This means it would
have very little harmonic suppression. The SPC circuit would always act
as a bandpass network, but in some settings it's loss was high. As a
result both of these have fallen out of favor these days. The "T"
network is what most current antenna tuners are using. Yet the "T"
network has two problems: 1: there are many combinations of settings
that will yield a good match but many combinations result in high
currents or high voltages across the components resulting in arc over or
smoke. 2: The "CLC" "T" network always acts as a high pass network so
has little harmonic suppression. (The "LCL" "T" network would act as a
low pass network, but would require TWO variable inductors and cost
quite a bit more).

The reason why the "T" network is acceptable today is that most modern
transmitters have enough harmonic suppression that the antenna tuner
does not need to add any more. With this in mind, it would also appear
that the older Ultimate Transmatch circuit would be acceptable today.
It does not have the high loss of the SPC or the 'cranky' behavior of
the "T" network.

I think that I can use the two 500pf variables (Johnson 500E20) as the
dual section part of the circuit by ganging them with insulated
couplings and mounting the capacitors with ceramic standoffs. They are
twice as large as what the original circuit called for though. The 250pf
(Johnson 250E45) variable will serve as the series capacitor. It also
needs to be mounted on ceramic standoffs. The roller coil has one end
with wide spacing.
I also have two national 'velvet vernier' dials for the caps and a turns
counter dial for the coil.

Has anybody out there experimented with these circuits, what results did
you get?

You left out the most important part: What kind of antennas are you
trying to feed? Are they resonant? Just centerfeds? Longwire?
Multi-band? Coax fed? Balanced Fed?

IMHO, these are the factors that should be considered when you
choose a network before you build the antenna tuner.

You may want to read the ARRL Antenna Tuner Book (New) to
get an idea of what commercial firms are building and their
efficiencies. It is very revealing!
73,

Ed, N5EI

"Kenneth Scharf" wrote in message
...
For the most part I will be using coax fed antennas.

There is an idea I've been thinking about and I wonder if it will work.
There is a myth that high SWR on coax will produce a high line loss. Most
of the loss involved with having a high SWR is in the coupling to the
receiver / transmitter, not so much in the transmission line itself, at
least not at HF. So long as a high voltage is no present on the
transmission line (arc over) or a very high current (resistive loss), and
the dielectric material does not introduce loss at the frequency involved
common 1/2" dia coax should work fine as an unbalanced feed line even at a
high swr ratio.


There is no myth that high swr can cause high loss, especially in coax.

It depends on how much loss the line has and how much the swr actually is.
If you only have a db or so of loss, then a 5:1 swr is not that much of an
issue. If you have a line loss of 2 or 3 db or more and then have a 5:1 swr
or higher, you can loose 3 db or more in the feedline in addition to the
actual loss in the feedline at a matched load

Below 440 there is very little loss in the dielectric, but mostly in the
center wire of the coax and then usually the shield. As the swr goes up,
the IR loss of the conductors goes up.

As far as the coupling loss to the transmitter, most modern transceivers are
set for a 50 ohm load and will cut the power back or not couple very well to
the transmission line. Even most of the built in tuners are only good for
about a 3:1 swr.

On 07/24/2011 05:35 PM, Ralph Mowery wrote:
"Kenneth wrote in message
...
For the most part I will be using coax fed antennas.

There is an idea I've been thinking about and I wonder if it will work.
There is a myth that high SWR on coax will produce a high line loss. Most
of the loss involved with having a high SWR is in the coupling to the
receiver / transmitter, not so much in the transmission line itself, at
least not at HF. So long as a high voltage is no present on the
transmission line (arc over) or a very high current (resistive loss), and
the dielectric material does not introduce loss at the frequency involved
common 1/2" dia coax should work fine as an unbalanced feed line even at a
high swr ratio.


There is no myth that high swr can cause high loss, especially in coax.

It depends on how much loss the line has and how much the swr actually is.
If you only have a db or so of loss, then a 5:1 swr is not that much of an
issue. If you have a line loss of 2 or 3 db or more and then have a 5:1 swr
or higher, you can loose 3 db or more in the feedline in addition to the
actual loss in the feedline at a matched load

Below 440 there is very little loss in the dielectric, but mostly in the
center wire of the coax and then usually the shield. As the swr goes up,
the IR loss of the conductors goes up.

As far as the coupling loss to the transmitter, most modern transceivers are
set for a 50 ohm load and will cut the power back or not couple very well to
the transmission line. Even most of the built in tuners are only good for
about a 3:1 swr.



Exactly the point. RG8/U and similar 1/2" diameter coax have very low
loss over the typical 100 ft length at frequencies below 30mhz. So it
should be possible to put a balum transformer (coax choke) between the
coax and a balanced antenna (dipole), perhaps with a length of balanced
line between the balum and the antenna. In my case the run from the
shack to the roof would be around 30' or less. I can't run twin lead
(open wire) line between the shack and the roof, but I can run it from
that point to a dipole. I also don't want the tuner outside the shack.
Hard line might even be a better idea, CATV type hard line is
typically 75 ohm impedance. I'd have to have the special connectors for
the stuff, and would need to have a short length of RG/8U from the
hardline to the transmatch, and another from the hardline to the balum
(because it would need to be flexible at those points.

Kenneth,

Is your shack below ground or without windows?
My shack is in the basement. I run 600 ohm open wire
from the Palstar Balanced Tuner (with balun in the input) to
feed thru insulators in a Plexiglas insert in the Window. Then
600 ohm to the doublet using string to the separators to
hold the feedline in a position away from the house.

The Palstar Tuner is similar to your description. It tunes my
210 foot doublet on most frequencies and its L equivalent
circuit is broad enough to make retuning for small frequency
changes unnecessary.

If you use your current balun on the input, an L (unbalanced) looks
like a balanced circuit on the output. Of course you must
insulate your capacitors above ground, ditto the inductor.

The ARRL Guide to Antenna Tuners has an excellent discussion
on balanced vs. unbalanced lines that you might wish to read
as well as balanced vs unbalanced tuners.

73,
Ed, N5EI

On 07/28/2011 06:57 AM, Edward Feustel wrote:
Kenneth,

Is your shack below ground or without windows?
My shack is in the basement. I run 600 ohm open wire
from the Palstar Balanced Tuner (with balun in the input) to
feed thru insulators in a Plexiglas insert in the Window. Then
600 ohm to the doublet using string to the separators to
hold the feedline in a position away from the house.

The Palstar Tuner is similar to your description. It tunes my
210 foot doublet on most frequencies and its L equivalent
circuit is broad enough to make retuning for small frequency
changes unnecessary.

If you use your current balun on the input, an L (unbalanced) looks
like a balanced circuit on the output. Of course you must
insulate your capacitors above ground, ditto the inductor.

The ARRL Guide to Antenna Tuners has an excellent discussion
on balanced vs. unbalanced lines that you might wish to read
as well as balanced vs unbalanced tuners.

73,
Ed, N5EI
The only window in my shack faces the front of the house. The window is
a a special hurricane resistant shatter proof glass. Also the window is
25 feet behind the location of the operating position in the shack room.
Access to the window is not possible, and the XYL won't hear about
feedline visible in the front of the house. I have a 2" plastic conduit
running from the operating position to the attic crawl space that
currently has on run of RG/8U for my vertical, with room for one or two
more runs. Running the twinlead next to the coax would probably not be
a good idea.

Kenneth Scharf ,WA2MZE wrote :

The only window in my shack faces the front of the house. The window is
a a special hurricane resistant shatter proof glass. Also the window is
25 feet behind the location of the operating position in the shack room.
Access to the window is not possible, and the XYL won't hear about
feedline visible in the front of the house. I have a 2" plastic conduit
running from the operating position to the attic crawl space that
currently has on run of RG/8U for my vertical, with room for one or two
more runs. Running the twinlead next to the coax would probably not be a
good idea.
---------------------
Living in a bungalow ,hence having a ground floor shack ,I run DC cables
, 2 coax lines and twin feeder (450 Ohms ,the wide variety) all in the
same plastic duct (about 2 metres long) to the loft from where coax is
fed to an outside 2m beam and and a discone antenna (the latter used
for local 2m FM operations) and the twin feeder to a doublet (all fed
through a double cement block wall)
The 12 V DC cables run from solar charged batteries located on the loft.

With all those cables running through the same 2 metres long plastic
duct I don't experience any problems with RF operations with the twin
feeder.

BTW the plastic duct is a piece of square 65 x 65 mm gutter drain pipe
,widely used here in the UK.

Frank GM0CSZ / KN6WH

On 07/29/2011 05:14 AM, highlandham wrote:
Kenneth Scharf ,WA2MZE wrote :

The only window in my shack faces the front of the house. The window is
a a special hurricane resistant shatter proof glass. Also the window is
25 feet behind the location of the operating position in the shack room.
Access to the window is not possible, and the XYL won't hear about
feedline visible in the front of the house. I have a 2" plastic conduit
running from the operating position to the attic crawl space that
currently has on run of RG/8U for my vertical, with room for one or two
more runs. Running the twinlead next to the coax would probably not be a
good idea.
---------------------
Living in a bungalow ,hence having a ground floor shack ,I run DC cables
, 2 coax lines and twin feeder (450 Ohms ,the wide variety) all in the
same plastic duct (about 2 metres long) to the loft from where coax is
fed to an outside 2m beam and and a discone antenna (the latter used for
local 2m FM operations) and the twin feeder to a doublet (all fed
through a double cement block wall)
The 12 V DC cables run from solar charged batteries located on the loft.

With all those cables running through the same 2 metres long plastic
duct I don't experience any problems with RF operations with the twin
feeder.

BTW the plastic duct is a piece of square 65 x 65 mm gutter drain pipe
,widely used here in the UK.

Frank GM0CSZ / KN6WH
The two inch plastic conduit runs from the shack up through a wall into
the ceiling and into the attic. The overall length must be about 15-20
feet. There would be another 20 feet through the attic to reach the
vent turbine and the roof. The twinlead if it ran though the pipe and
the attic would probably twist many times. It might work, but it goes
against all 'textbook' logic.

In any event, I'm thinking of building the transmatch as a 'T' network
using the two 500pf 2kv variable caps and the roller coil. I will
insulate the bottom end of the roller coil from ground to leave the
option open of running the T match with a balum on the input side, but I
intend to try the coax choke on the far end of an unbalanced feed first.

As for the T match, it seems to be the sole surviver among transmatch
circuits. Both the 'ultimate' and the 'spc' are simply alternate forms
of the 'T' match that do not have any benefits.

On Jul 24, 3:18*pm, Kenneth Scharf wrote:
For the most part I will be using coax fed antennas.

There is an idea I've been thinking about and I wonder if it will work.
There is a myth that high SWR on coax will produce a high line loss.
Most of the loss involved with having a high SWR is in the coupling to
the receiver / transmitter, not so much in the transmission line itself,
at least not at HF. *So long as a high voltage is no present on the
transmission line (arc over) or a very high current (resistive loss),
and the dielectric material does not introduce loss at the frequency
involved common 1/2" dia coax should work fine as an unbalanced feed
line even at a high swr ratio.

Coax loss at high SWR is a myth? It's been tabulated in every handbook
for the past century.

Common Ferrite core balums do not usually work well at impedances above
50/75 ohms input. *Using them to convert an unbalanced transmatch to a
balanced one AT THE OUTPUT is usually not a good idea.

Often it's the choice made, though.

Baluns that deal with less than optimal impedances are available but
they are no magic bullet. Substantial engineering effort and material
must go into them to deal with the stresses at operating away from
their natural sweet spot (as you point out 50 ohm ballpark.)

(Insulate the
transmatch from ground and put the balum at the input works better).
A "choke" balum made by winding 10 turns or so of coax in a loop about a
foot or so in diameter has lower loss than a ferrite core balum, no core
heating.

So my idea was to run a length of coax from the shack through the attic
crawl space and up to the roof (though a vent turbine). *On the roof
would be a coax choke balum connected to 450 ohm twinlead "open wire"
line. *This would connect to a "multiband" dipole. *There is no way I
can run the open wire line from the shack up to the roof, this stuff
must be run 'in the clear' anyway. *Coax being an unbalanced line is not
affected by it's surroundings.

The coax to 450 ohm twinlead via choke "balun"... wow, it sounds like
you are getting the worst of all possible worlds here. I agree that
transformer-type baluns are stressed at high impedances but
purposefully putting a 9:1 mismatch on the end of your coax is just
asking for a world of hurt with anything except the most shortest coax
runs (a few feet.)

You might want to back up a few steps and consider if you can locate a
coax-to-twinlead tuner, almost certainly not broadband, on the wall or
roof of the shack or house. That way you have a nice controlled 50 ohm
coax run inside, and it's all twinlead outside. If the shack is far
from the tuner, today there are remotely operated tuners just for this
purpose. In the most traditional set ups mechanical remote operation
(chains, pulleys, etc.) was sometimes used (look in any 30's/40's/50's
handbook.) (When I was a kid my uncle who was a ham, used a scheme
involving bicycle cranks and extra long chains for rotating his beam!)

Len Cebik's webpages are an excellent resource on ladder line, all
band doublets, and link-coupled tuners. I can't emphasize enough how
excellently this can work out. Don't paint yourself into a corner.

Tim N3QE

On 08/10/2011 11:45 AM, Tim Shoppa wrote:
On Jul 24, 3:18 pm, Kenneth wrote:
For the most part I will be using coax fed antennas.

There is an idea I've been thinking about and I wonder if it will work.
There is a myth that high SWR on coax will produce a high line loss.
Most of the loss involved with having a high SWR is in the coupling to
the receiver / transmitter, not so much in the transmission line itself,
at least not at HF. So long as a high voltage is no present on the
transmission line (arc over) or a very high current (resistive loss),
and the dielectric material does not introduce loss at the frequency
involved common 1/2" dia coax should work fine as an unbalanced feed
line even at a high swr ratio.

Coax loss at high SWR is a myth? It's been tabulated in every handbook
for the past century.

Common Ferrite core balums do not usually work well at impedances above
50/75 ohms input. Using them to convert an unbalanced transmatch to a
balanced one AT THE OUTPUT is usually not a good idea.

Often it's the choice made, though.

Baluns that deal with less than optimal impedances are available but
they are no magic bullet. Substantial engineering effort and material
must go into them to deal with the stresses at operating away from
their natural sweet spot (as you point out 50 ohm ballpark.)

(Insulate the
transmatch from ground and put the balum at the input works better).
A "choke" balum made by winding 10 turns or so of coax in a loop about a
foot or so in diameter has lower loss than a ferrite core balum, no core
heating.

So my idea was to run a length of coax from the shack through the attic
crawl space and up to the roof (though a vent turbine). On the roof
would be a coax choke balum connected to 450 ohm twinlead "open wire"
line. This would connect to a "multiband" dipole. There is no way I
can run the open wire line from the shack up to the roof, this stuff
must be run 'in the clear' anyway. Coax being an unbalanced line is not
affected by it's surroundings.

The coax to 450 ohm twinlead via choke "balun"... wow, it sounds like
you are getting the worst of all possible worlds here. I agree that
transformer-type baluns are stressed at high impedances but
purposefully putting a 9:1 mismatch on the end of your coax is just
asking for a world of hurt with anything except the most shortest coax
runs (a few feet.)

You might want to back up a few steps and consider if you can locate a
coax-to-twinlead tuner, almost certainly not broadband, on the wall or
roof of the shack or house. That way you have a nice controlled 50 ohm
coax run inside, and it's all twinlead outside. If the shack is far
from the tuner, today there are remotely operated tuners just for this
purpose. In the most traditional set ups mechanical remote operation
(chains, pulleys, etc.) was sometimes used (look in any 30's/40's/50's
handbook.) (When I was a kid my uncle who was a ham, used a scheme
involving bicycle cranks and extra long chains for rotating his beam!)

Len Cebik's webpages are an excellent resource on ladder line, all
band doublets, and link-coupled tuners. I can't emphasize enough how
excellently this can work out. Don't paint yourself into a corner.

Tim N3QE

If you run coax up to the roof, how about using some 70 ohm CATV coax?
I have some here I got for free, 3/4 in diameter. Oftentimes cable TV
companies have shorter 'reel ends' they are willing to give away. The
stuff is 'semi-rigid' but you may be able to get it to feed up that
conduit anyway. Worth a try. (BTW I do *not* mean RG-6 junk here, I mean
the larger diameter semi-rigid types only)

On the roof, if you must go the 'balun' route, a 4:1 or more likely 9:1
unun may be OK. Obviously, a balanced tuner accessible for adjustments
would be best.

CATV 'hardline' is not near as lossy as RG-8 or RG-213 types up around
10 meters. So try some- and don't worry that it's not 50 ohm impedance,
just use it as if it were.

I built the SPC type tuner myself, nearly 20 years ago. It's not always
the answer, but oftentimes it worked pretty good, allowing me to
extend the usable BW of some narrow antennas and still keep the amps
happy.

73, David K3KY