Call Search

New to Ham Radio?
My Profile

Friends Remembered
Survey Question

DX Cluster Spots

Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement

donate to eham
   Home   Help Search  
Pages: [1]   Go Down
Author Topic: Different Tuner Networks, have ?'s need some input please  (Read 2545 times)

Posts: 8

« on: February 26, 2013, 07:12:13 PM »

Hello, KD8TGK here, I am in the process of designing a manual antenna tuner for use at the antenna feed-point, the basic antenna is a ground mounted vertical with a very dense ground plane. There are very many tuner network configurations to consider, which one or combination would be best. The variables are as follows: R step down - high pass, R step down - low pass, R step up - high pass, R step up - low pass, a T network R step up/down - high/low pass, a Pi network R step up/down - high/low pass, or a Series / Parallel Capacitor (SPC network) also on a side note is there such a thing as a Series / Parallel Inductor (SPI network) ?. Basically I am looking to make a tuner to tune a vertical cage radiating element with a length of 42' ft, from 1-30 MHz continuous, resistance range 1-1500 ohm, Reactance range -1500 to +1500 ohms.

I currently have a working schematic that would allow all possible configurations (circuit interchangeability) of the popular LC network, R step up & down - pass low & high.
If I added another inductor and capacitor I can expand the design to include all possible configurations (circuit interchangeability) of the T & Pi networks.

Also there is a concern of phase correction after the LC network, do the T and Pi networks also have a need for phase correction
Also I can add a additional inductor and capacitor for inductive and capacitive shunt feeding the vertical element against ground, your thought on adding or removing this capability

If you would like to see my schematic for a completely interchangeable LC/T/Pi network let me know, and for all of you wondering yes I am looking to make some queasy stomachs over at SGC because I feel its time for a better, cheaper, wider range, more efficient tuner at the feedpoint for our verticals! As with the long standing adage in amateur radio: "99% of the time u can build better than u can buy for cheaper" (excluding radios, but elecraft is closing that gap!)

Posts: 1728

« Reply #1 on: February 26, 2013, 09:00:09 PM »

It has been a long time since I looked at the equations for matching network design, but, as I remember, a PI network generally realizes high to low transformations effectively.  Whereas a T does the opposite quite well.  Since you will be doing unbalanced low input to perhaps lower and up to considerably higher output, hard to beat a T.

I seem to recall my design of 20-200pF input and output variable capacitors with a 10 uH variable inductor would match 50 ohms input to a wide range of outputs.  Perhaps not up to 1500  +/-j1500 ohms as you are looking for, but the highest Z magnitude on the T output was several hundred ohms on HF.

If you're a perfectionist, get a good, heavy duty roller inductor and variable capacitors with fairly wide rotor/stator spacings as the voltage across the output capacitor at least will be quite high at high output impedances for even modest output power.



Posts: 8

« Reply #2 on: February 26, 2013, 09:57:14 PM »

Also another question, in choosing a Low-pass or a High-pass scheme. It is my understanding that a low-pass essentially permits the frequency that it is tuned for and all frequency's Below that mark will pass through and all others Above that mark will be "grounded/nulled" out for lack of better term. And that the exact opposite happens in a High-pass scheme. Am I correct in that assumption and also that in choosing either u are also choosing a low/high pass filter!!!??

Posts: 17483

« Reply #3 on: February 27, 2013, 09:34:04 AM »

You don't need ALL the options, just enough to provide at least one solution
for each frequency.

In many cases the limitation is due to the physical components - the minimum
and maximum capacitance for a variable, for example, or the allowable losses
in a coil before the supports melt.

Generally speaking the simpler networks will have lower losses and wider
operating bandwidth.  The "L" network (which is what I think you were referring
to in your first four options) is a good choice when it will work.  There are
actually 8 options, depending on whether you use a coil or capacitor for
each of the series and shunt elements, and whether the shunt element goes
on the antenna side (for higher impedance loads) or on the feedline side (for
lower impedance loads.)  One problem with the L network is that, as the amount
of impedance transformation gets small, you approach infinite reactance in one
element and zero in the other, so sometimes a fixed element is added when the
load impedance is close to the output impedance to make a PI or T network.
(Another option would be to provide a option for an input matching transformer,
like 1 : 4 or 4 : 1, to avoid having to match load impedances close to the
coax impedance.)

There is no particular advantage to using the SPC or "Ultimate Transmatch" or
similar circuits.  Keep it simple.

A circuit with series L and shunt C will tend to attenuate frequencies above
resonance, but that doesn't mean that lower frequencies will be properly
matched.  Similarly a "high-pass" circuit with series C and shunt L will still
transform higher impedances.  The "low-pass" options is often recommended
to improve harmonic rejection, but my philosophy is that the rig should
provide all required harmonic rejection (since we often feed them into
multiband antennas) so the tuner circuit can be based on other design
criteria.  One of the main ones is the required shunt reactance:  low impedance
loads require low reactance shunt elements, and it is much easier to get
that from a small coil than a large capacitor when it needs to be variable.

One other design consideration:  be clear whether you are designing a tuner
to match your 42' cage antenna across the required operating range, or to
match any antenna.  The latter is much more difficult, because in the
former case you can predict the range of impedances you will encounter and
design accordingly.  For example, with an end-fed radiator such as yours will
be capacitive when the resistive component is less than 25 ohms or so.
Knowing that you don't have to design it to handle low impedance inductive
loads unless you are going to also use it to match (a) coax-fed antennas, or
(b) loops, where you might encounter such loads.

If you are using relay-switched components, that gives you more flexibility
than mechanically rotating ones because you don't have the same limitations
on minimum / maximum values, though you will be limited by the voltage
rating of the relay contacts.  That might affect your choices:  it is easier
to get 3000pf using parallel capacitors than a rotary variable.

My recommendations to match an end-fed wire:

1) for lengths shorter than 1/4 wave, a step-down L network using two
coils is probably the easiest.  The shunt coil sets the R and the series
coil adjusts X.  If that isn't convenient, use a shunt capacitor that can
be set to relatively high values.  This should handle R values less than
25 ohms or so.

2) for the rest of the tuning range, start with with a 4 : 1 stepdown
autotransformer, and then use a standard "L" network (likely a high-pass)
to step the resulting 12.5 ohms UP to the load impedance.

There was a great antenna tuner article by Dr. Rhode DJ2LR(?) from
around 1984 that I've been trying unsuccessfully to locate, where he
designed a coupler for end-fed wires.  He used the step-down
transformer followed by a fixed shunt coil / series capacitor combination
(to improve the matching range on 160m), then a standard high-pass
"L" network.  The article also included Smith Chart plots of the matching
range of various networks that he considered, and I'd strongly suggest
that you consider making similar plots to see what impedances your
proposed circuit can and can't match.

Posts: 165

« Reply #4 on: February 27, 2013, 11:19:23 AM »

A VERY comprehensive paper on 'Wide Range Antenna Matching Networks' was give at the Clerk Maxwell Commemorative Conference of the then Institution of Electronic and Radio Engineers in Leeds in 1981. It was by Dr (now Professor) Mike Underhill, G3LHZ, of Philips Research Laboratories, and runs to some 35 pages. The conference proceedings are ISBN 0  903748 45 2, and it is Conference Proceedings No. 50.

Pages: [1]   Go Up
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!