A Versatile Tuner for SOTA Activations by KX0R

Well done, George; a worthwhile contribution.
Best, Ken

George,
Had a question about the size of the wire you used for L1/L2
I can see enlamed copper et another in different color !
73 Eric F5JKK

George - I need to put this in my archives for later reference. Thanks for sharing.

Lee

Thanks George - great article.

Shame we don’t have a “SOTA Magazine” this would be a great article to have in it!

73 Ed.

It would be a great article, I dare say that Sprat would like to publish it?

It’s just so very nicely presented. I find the hand drawn circuits relaxing to look at.

:+1:

2 Likes

Andy,

Thank you.

Here’s PART 2, with the actual circuits I use for SOTA activations. This part is for those who are interested mostly in building a tuner to use.


ACTUAL KX0R CIRCUITS

Here are two KX0R tuners showing the actual circuits. This is not a how-to-make-it article – experience and improvisation are required.

For best performance, these tuners should be compact, and small wire should be used to reduce capacitance. I suggest wire-wrap wire, or small single-conductor silver-plated Teflon wire. While a PCB could be used, it likely would add capacitance, which adversely affects these circuits at higher frequencies and high impedances. The output impedance may be several thousand ohms, where every pF is a problem – this is the opposite of working with 50 ohms.

THE LARGER TUNER:


THE DRAWING:

This is just the combination of the concepts from the drawings up above, posted earlier. Most of the clutter comes from all the extra switches and connections required to extend the frequency and impedance ranges.

Components:

The polyvaricons I have are marked “TT” and “TTWM”. These have a small section of about 2-60 pf, and a larger section about 4-140 pF. These work well generally, and the Q is good compared to other capacitors I use. Fixed caps should be silver mica or C0G/NP0 ceramic, 200V or better.

The sub-miniature toggle switches are made by Mountain Switch – these are available from Mouser and elsewhere – the E-switch brand is similar – they are great for this application! A good part number to start with is 108-0044-EVX – this is the SPDT-Center Off switch in my tuners. Other versions are available – DPDT, SPST, etc. The colored output jacks are mini-banana jacks from Pomona (Mouser) – sample P/N Pomona 2142-5, Mouser 565-2142-5. The mating plugs are Cinch (Johnson) – small and simple – sample P/N 108-1002-001, Mouser 530-108-1002-1.

These mini-banana connectors are also nice for making links in antennas.

RCA phono connectors are good for the 50 ohm connections. They’re lighter and simpler than BNC, and they pull loose if something bad happens in the field. A sample P/N for the receptacle is Mouser 161-1052. The little boxes came from Radio Shack….I had to remove some extra plastic inside to make room for some added parts.

The larger tuner includes a big T106-6 toroid core – the intent is to minimize transformer loss. A T68-6 would work fine, with different turns counts. The total primary has 6 turns, with a tap at 3 turns. The total secondary is 16 turns total, with taps as shown on the drawing. The secondary is wound around just over half of the toroid, and it measures about 4.1 uH. If you use a smaller core, 4.1 uH is a good value to use for a starting point. The primary of the current version is wound over (among) the low-Z end of the secondary. The toroid is mounted on a cushion of foam insulation, with a single screw and insulated washers to hold it to the panel. The mounting scheme used for the toroid MUST NOT create a shorted turn (closed loop).

This tuner will match many useful loads from about 3.5 to 22 MHz. Performance is optimum in the 40-30-20-17M bands. High Z loads like the end-fed half wave and the end-fed full wave wires are easy to match, as well as low-Z loads like a ¼-wave vertical wire with a ¼-wave counterpoise. This tuner includes a switch to isolate the secondary circuit, useful when feeding balanced lines for a dipole, etc.

Operation:

I regularly use this tuner with an end-fed 66-foot wire. The tuner matches this wire at 7, 10.1, 14, and 21 MHz, with no counterpoise. This antenna is approximately high-Z resonant at 7 and 14 MHz, but it’s reactive on 10.1 MHz. The 30M match is as easy as for 7 and 14 MHz, but it’s more sensitive to conditions at the site, stray capacitance, etc. No traps or links are needed, but they can be used too. I also employ a 52-foot wire with a 12-foot counterpoise – this has low impedance on 20M, somewhat low Z on 40M, and very high Z on 30M and 17M. All the matches are easy. Occasionally in haste I’ve omitted the counterpoise and “forced” the low-Z matches using the tuner, and performance is only down slightly.

High-Z loads are just as easy to use as low-Z loads. Various tests on the bench have confirmed that this tuner is reasonably efficient, even when feeding loads near 5000 ohms, as well as below 50 ohms. Power should be limited to 10W, because the polyvaricon capacitors are not rated for high voltage. Likewise, power should not be applied without a load connected. The system should be tuned to band noise before tuning with any power. Using low power and the tuning bridge, a match should be found – only then should full power be used.

When tuning an unknown load, it should first be tried on the highest-Z tap jack. If necessary, it can be moved to a lower-impedance tap; then the tuner is re-adjusted for a better match - etc.

Adjusting these little tuners is somewhat of an art. With repeated use the common settings are learned. Then just little tweaks are needed to get on the air or to QSY. Just adjusting the band noise to peak will result in a pretty good match, especially if the right taps and switch settings are known. Some loads produce sharp tuning settings, but good signals may result anyhow. Broad matches occur mostly with low-Z resonant loads, but low-Z resistors will tune up the same way! Broad matches are not always a sign of perfection in the system, nor are they usually a problem.

In some cases, more than one match may be obtained, with different settings for the input capacitor. This is usually OK. The match with the best null should be used. Resistors can be used to get a feel for the tuner on various bands, at various impedances. Once you “know” a tuner, weird conditions are more obvious. Most tuners have spurious nulls caused by unwanted resonances, but in these tuners, spurious resonances occur at very high frequencies. Spurious nulls can be found by adjusting the tuner at very low power, with the load connections open and shorted. An MFJ-259B analyzer works well for such tests. As long as the construction is compact, with short leads, unwanted resonances won’t be a problem.

Balanced lines may be fed by connecting to a pair of output taps, determined by experiment. Balance may not be perfect, but a match will be obtained. A switch S6 is provided for special configurations. Three configurations are available:

  1. GND – a tap on the secondary is “grounded”
  2. Center OFF – FLOAT – this is for a floating balanced line, with no “ground” connection to the secondary
  3. U – Unbalanced - the bottom of the secondary is “grounded” – this is the normal unbalanced configuration for use with end-fed wires, etc.

A 22 megohm resistor is connected between “ground” and the low-Z side of the secondary – this is for use in the “FLOAT” mode, with a balanced line, to drain off static electricity – the tuner would need an actual minimal connection to the ground for this to work - a nail and a wire is enough.

Any impedance-matching RF circuit has loss; there’s also loss in traps, broadband transformers, un-uns, baluns, and in small coaxial cable. I use only two feet of RG-316 cable in my system – from the radio to the tuner - my antennas are usually end-fed.

Performance has been fun! I’ve used this tuner for several hundred activations, with decent results.

On March 8, 2019, I activated a local peak, W0C/FR-109. I set up on the ground, which was damp from melting snow. The site is pretty average - nothing special. My 66-foot end-fed wire was about 15 feet high at the top of the bent-over fishing pole, and the far end of the wire was only 2-3 feet off the ground. I used a 12-foot counterpoise, but only for 60M. Running my KX2 at 10W out, I made over 50 CW contacts on 20, 30, 40, 17, and 60M. Included was a 20M DX contact with G4OBK, Phil, in England; another 20M DX contact with EA8/HB9FIH in the Canary Islands, and a 17M DX contact with ZL1BYZ, John, in NZ. I also made S2S contacts with W5ODS in Arkansas and AC1Z in New Hampshire. My RBN spots were good, compared to other activators. This was a normal activation, and not set up at an ideal spot on the mountain. There’s no magic - only a tuner that works well - and really good chasers!

THE TINY TUNER 2


This tuner weighs less than 4 ounces (about 110 g).

The Tiny Tuner 2 preceded the larger tuner above; it was re-built from a different, earlier tuner - the “Tiny Tuner”. That tuner was derived partly from the BLT.

ORIGINAL CONCEPT DRAWING:

DETAILED DRAWING:

Concept, construction and components are similar to the larger version above. A T68-6 toroid is used, and the detailed drawing shows the turn counts. Note that the turns are different from those on the Concept drawing.

The secondary has 28 total turns, and the entire winding measures about 4.2 uH. The primary has 11 turns total, with a tap 5T from the top end. The primary is wound on the low-Z end of the secondary - see photo.

Operation is very similar to the larger tuner. 60M and 80M are not available, and the circuit is simpler and smaller. As far as I can tell, this tuner is just as good as the larger one, for the 5 bands it covers. It handles the 10W from the KX2 just fine!

Part of the intent of this article is to make SOTA activators aware that traditional RF rules are meant to be broken, to some degree. Antennas don’t need to be resonant, wires may be fed at the end, an actual physical counterpoise is not always needed, a ground is seldom required for SOTA, an effective antenna may be quite low, coax cable is not desirable for connecting to the antenna, and above all – compact, light equipment is better for SOTA!!

Hopefully some of you will enjoy building and using similar tuners as much as I do!

73
KX0R

17 Likes

Thanks for the detailed and practically-based write-up George, always interesting to see how other people do it!

In getting on for 500 activations the only problem I’ve had with an ATU has been when I accidentally knocked the “bypass” switch whilst unpacking it. Took me a few minutes to work out why it wasn’t tuning up but now checking the position of the switch is a regular part of my set-up procedure! And on many occasions I’ve been very glad of a band-change without having to crawl out of my snow hole or from behind what meagre shelter I’ve found or created into the blizzard/horizontal rain/teeth of the gale to swap links etc :wink:

73 de Paul G4MD

1 Like

Hi,

This site has some ideas for the experimenter.

http://home.alphalink.com.au/~parkerp/projects/projcoupler.htm

73
Ron
VK3AFW

"I find hand drawn circuits relaxing to look" a_t

:open_mouth: Wow! (no offense meant). I look at circuit diagrams hand drawn or computer designed and my eyes glaze over. You may as well talk to me in Klingon !!

1 Like

Hi,
I plan to build the tuner by KX0R, the larger one. Which lengths of wire and lengths of counterpoise work good for which frequencies?

From the reading, I already got:

  • 66 ft, no counterpoise => 7, 10, 14 and 21 MHz
  • 52 ft, 12 ft counterpoise => 7, 10, 18 MHz

73 Axel DF1ET

Hi,

because I’m going to build the above KX0R-tuner, I bought different poly-varicons (in Germany) and measured quality factor. The results were as follows. Q is Q of the resonant circuit build with a coil, not only the capacitor-Q. Measured for full and half capacitance (of the high capacitance part).



ebay.de, oppermannelectronic: “Folien Drehkondensator mit 4mm Messingachse 21mm Kondensator CAPACITOR”
Text says: “2 x Ce= 360 pF, Ca= 15pF und 2 x Ce 30 pF, Ca 9 pF”
Measured: 5…350 pF resp. 4…24 pF, with all trimmer capacitors set to minimum.
Labeled: “ELS”, “Made in GDR” [GDR = German Democratic Republic = DDR], “07 89”
Q = approx. 430



ebay.de: m-o-s_2010: “Drehkondensator 60 pF + 140 pF”
Labeled: “TT”, "TTWM"ä
Q = approx. 200



ebay.de: m-o-s_2010: “4-fach Drehkondensator 22 pF + 22 pF + 90 pF + 150 pF”
Labeled: “NCE”
Q = approx. 330



www.ak-modul-bus.de: VCAP4 Drehkondensator AM/FM (2 x 265 pF + 2 * 20 pF)
Labeled: “WP”.
Q = approx. 150
BTW: They sell very nice extension spindles (“Verlängerungsachse für VCAP4…”), see my last image!


Measurement method:
VNWA, measuring s21, poly-varicon + coil connected between coax inner conductor and ground in series configuration.
Measurement of resonant frequency and 3-db bandwidth, no correction for the 2x 50-Ohms-source-resistances.

Resonant frequency: 3…14 MHz, depending on capacitance and number of coil windings.
Coil: 1.0 mm enamaled wire wound on 38 mm diameter, self supporting, spacing of winding very approximate 2 mm but varies largly because of self support.

BTW, reichelt.de has water proof toogle switches, for example: RND 210-00450.

Hope it helps,
vy 73 Axel DF1ET

Edit: Added labeling of ak-modul-bus varicon.

1 Like

A drop of sewing machine oil (or WD40 perhaps) applied with a needle to the rotator where it meets the housing will make it turn a lot more easily.
73 de OE6FEG
Matt

Hi there George and everyone here on the thread. I’am Alejandro from Argentina. Thanks for this great topic!; I really enjoy making my hamradio stuff, now I’m making one similar end feed tuner and I want to know what do you think about the changes I made from the KX0R original circuit.
I have a tiny 6 point rotary switch that I wish to use instead of using banana ports to change the Z relation (6 positions is maybe overkill?). And I don’t have the secondary winding variable caps fixed on a given point of the winding


Should I expect less efficiency using the T68-6 near the metal part of the switch? Or maybe this circuit is not the best option to choose?

Thanks! and regards from Argentina
de LU1MAW

2 Likes

Alejandro,

I believe your new circuit will provide efficient matches - however, it is very different from my circuit in several ways.

I think that your rotary switch is a good idea. It saves space and is quick to operate. My circuit would work just fine with your switch. Perhaps I should build a new tuner this way - it is a good idea!

I am glad that you are experimenting and trying different ideas! Here is what I expect will change, compared to my tuner:

  1. Low impedances may be more difficult to match with your secondary circuit. I believe that the tuning capacitor on the secondary should be across the entire coil, with the output coming from one of the taps closer to ground, in order to transform the impedance lower. If you are only matching high impedances, like an end-fed half wave, you may not see a problem with your circuit. You may be able to match a wider range of frequencies, but not a wider range of impedances, than with my circuit.

You also do not have a high step-up turns ratio when the secondary has few turns, which may limit how high the output impedance can be at 21 MHz, etc.

  1. If you want to match low impedances and high impedances, then use your switch to replace the mini-banana jacks in my circuit. Just connect the variable capacitor to the high-Z end of the secondary of the inductor. The switch you chose looks ideal for this application - it is compact, stray capacitances are probably very low, and it has many taps.

  2. I like having a small ~60 pf variable capacitor across the secondary for the higher frequencies. The minimum capacitance of the small section of my TT variable cap is about 2 pF, so higher frequencies can be tuned. I can tune antennas at 21 MHz. Your larger capacitor(s) on the secondary may be more difficult to adjust for higher frequencies.

  3. I doubt that your switch will cause a problem for the toroid inductor. Most of the magnetic fields stay inside the toroid, so coupling outside the coil is greatly reduced. If you want to reduce coupling even more, mount your switch about 5mm away from the toroid winding. If the switch is made of metal that is attracted to a magnet, it may cause effects if it is very close. Otherwise, I don’t thing there would be a problem having it so close.

  4. I suggest that you go ahead and build your tuner as you choose, but think about changing where the variable capacitor would be connected across the secondary. Build the circuit so you can try it both ways on your antenna or antennas - see how it tunes on various bands.

  5. You can change your circuit later, if you find that you want to match lower impedances the way that my tuner works.

There are so many ways to connect these parts to build efficient tuning networks! Every circuit will have advantages and disadvantage over other possible configurations.

I am using my tuner to match a wire that is not resonant on all the bands I use - mostly I use an end-fed wire about 20M long, on the 5, 7, 10, 14, 18, and 21 MHz bands, usually with no counterpoise. On the 5, 10, and 18 MHz bands the impedance is reactive - yet results are consistently very good, even though the wire is not resonant. There is much room for experimenting and working on this problem, and many questions about how large the tuner losses are, and whether a counterpoise - if used - would provide benefits.

My tuner circuit works really well for real SOTA activations, time after time. The question for me is: what can I change to make it better? Today I don’t know.

Thank you for sending this interesting message!

I like your idea of the rotary switch!

73

George
KX0R

3 Likes

Thanks for your time George!, Definitely I will need to tune on 21MHz diferent impedance values; my goal is tune 40/20/15m bands in order to use my direct conversion QRP

Sorry is in spanish.

I thought solving the big capacitor issue by switching the coil… but I didn’t think on the low impedance situation! You are absolutely right and I glad to have asked for your opinion.
I have a second pole on the same 6 pos switch and I imagine the next circuit, but maybe I will loose flexibility to choose hi and low frequency over the total lenght of the coil to match high Z loads on higher frequencies.

This is the beast so far:

3 Likes

Alejandro,

Using two sets of switch contacts is an interesting idea. You have many possible combinations! I think if you build it and experiment with your tuner and antenna outside, perhaps at home, you can choose the right taps on the coil to get matches on several bands.

Your work looks really good! Please come back after you do some tests, and show us what you have decided - it may take several attempts before you get good matches on several bands, and everything depends on your antenna and how it is fed.

Here in Colorado I use the RBN (Reverse Beacon Network) to see if my CW signals are being received by many stations. I also look at the spots for many other activators. There is a very wide range in the efficiency of activators, and which bands they do well on.

Spot signals vary greatly on the RBN, so these results must be used in a general way, but lots of spots and strong db reports indicate good results. Few spots with low db numbers indicate that most chasers will not copy you on the particular band you are on.

Even with a wire only 3M high, 1/2 wave long, and a transmitter of 5W output on CW, on a hill or mountain, some RBN spots should be higher than 10 db, for 40-30-20M. There should be many spots, if you call CQ for 5 minutes or more. This is approximate, but many stations are doing a very poor job!

73

George
KX0R

4 Likes

Well, today I’ve been finishing it and running a few test…not “on the air” ones yet; but just trying with resistors using the first circuit.

I’m very happy with the results, since I was been able to tune a 5K from 3.5 to 28MHz, and after that I start checking how low can the resistor be; so the second test was on 470 Ohms and I could tune it over the same range of frequencies. After that I tried with a 50 ohms resistor and the circuit could tune it only from 14Mhz and above.

Now I can use it with End Feeds; near 450 Ohms to the 3000-5000 Ohms of a EFHW. Tomorrow I’ll see another configuration to check if I could tune from 100 Ohms or less on the lower bands to use it with coil loaded verticals or something similar.

73!

3 Likes

After a few tests I will say the original circuit was my choice to use with my 20m wire!.
It tunes from 6 Ohms to 5000 ohms resistors from 40 to 10m bands; I’am really happy with the tunner.

Last weekend (3/11/19) I used it successfully at the LUM/PC-043 activation; and will use it on my next activation too!


Thanks for your patience George! 73!

5 Likes

Alejandro,

You have done a good job! I think you will like using that tuner. It’s fun to change bands without having to change links on the wire.

You may want to make an antenna with links or jumpers, so that all the bands will be resonant and high-impedance. I did this for a long time. You only need one link in a ~20M wire to have 40M, 30M, 20M, and 15M be all high-Z. 17M will also be close to high-Z resonance if you open the jumper for 30M. The one link will be at the point where the wire will be tuned to a half wave for 10.1 MHz. My tuner will tune the 20M wire without using any links, but it may work a little better with one link for 30 and 17M. I have never done tests to determine if the link improves efficiency, but it certainly is easy to tune up on 30M and 17M with the link open.

Exact resonance is not important on any of the bands - the tuner will correct the reactance if resonance is not exact.

I am glad to see your work, and now I have happy that I wrote the article about the tuners. Hopefully as others work with these tuners, perhaps improvements will be made, and people will share what they learn.

73

George
KX0R

2 Likes