Remote antenna tuner 80-10 meters, no switching

sm0vpo wrote:...I would use a brushless motor and worm drive, but I would use three of them, with a sensor resistor to indicate where the tuning actually is. My three controls would be:

LOAD: a 300pf air-spaced capacitor
TUNE: a 300pf air-spaced tuning capacitor
RANGE: A rod of ferrite pushed into an air-spaced coil - no tappings

The brushless motors and gears can be robbed from old R/C helicopters and such. ...

...The worm drive can easily be an M3 threaded rod with two nuts (to avoid loose play) and can be made really robust. ...

Hi Harry,
1/ Precise worm drives can be found in old floppy disc drives, where they position the head on the tracks. Those from 5 1/4" have longer travel. They employ a neat stepper motor.

2/ Using a worm drive with linear movement is perfect to tune a coil and/or a homemade variable capacitor consisting of two overlapping metal sheets with dielectric (two pieces of copper clad PCB board maybe). A better way of tuning a commercial variable capacitor with rotary movement is perhaps using a gearbox "fast to slow".

3/ The controls may be alternatively:
- one variable capacitor;
- one variable coil;
- a relay switching between L and gamma topology;
- maybe a bypass relay.

VBR from Ivan

sm0vpo wrote:... using plastic laminate to insulate copper foil strips to feed my antenna into the new shack office. Now I can have the antennas connected and close the windows

Hi Harry,
I am glad you are O.K. Everyone should take a break time to time, or his body may react by an illness.

It would be nice if you could write an article about your solution and experience of passing a rf cable through a window without drilling. I know a special flat RF cable is sold, but it is rather expensive.

VBR from Ivan

sm0vpo wrote:
Hi Andrew and others,
It was really interesting to read this thread and some of the various solutions. But I have been re-organizing my home, and also done things like using plastic laminate to insulate copper foil strips to feed my antenna into the new shack office. Now I can have the antennas connected and close the windows :-)

Hi Harry, welcome back, hope you enjoyed your holiday !!

The subject of the ATU is really interesting because the best place to fit the ATU is between the antenna and the feeder cable.

exactly so, placing the tuning network where it belongs, instead of placing it at the far end of a feedline, offers several advantages... and the disadvantage that the network then will need to be adjusted remotely (assuming one doesn't want to get out at every band change, that is )

Using the server method is Ok, but for precise tuning the server can be improved upon. If I were to build one today I would use a totally different system. I would use a brushless motor and worm drive, but I would use three of them, with a sensor resistor to indicate where the tuning actually is. My three controls would be:

LOAD: a 300pf air-spaced capacitor
TUNE: a 300pf air-spaced tuning capacitor
RANGE: A rod of ferrite pushed into an air-spaced coil - no tappings

but that would still require three adjustments, where the "FriMatch" just needs two (the capacitors), that's why I think it may be a good idea for a remote matching unit, I mean... no coil tap switching, no inductance adjustment at all, one just rotates the capacitors and that's it; sure, it doesn't go down to 160m, but other than that, it seems to be pretty good

At the end of the day, there are loads of solutions, but I feel that a bit of Arduino/RaPI free, mechanical tuning would be simple, best and really reliable.

call me "a dinosaur", but sincerely I don't like mixing CPUs and RF, but then and again, probably it's just me

Andrew wrote:

"On a totally different topic, hope that Harry is ok, it has been quite some time since he posted here ! I know he was busy with reorganizing his home..."

Hi Andrew and others,
It was really interesting to read this thread and some of the various solutions. But I have been re-organizing my home, and also done things like using plastic laminate to insulate copper foil strips to feed my antenna into the new shack office. Now I can have the antennas connected and close the windows :-)

The subject of the ATU is really interesting because the best place to fit the ATU is between the antenna and the feeder cable. This requires the ATU to be mounted outside, up in the air. Ok, if you have a vertical, then you can get away with having the ATU at ground level, but it still requires it to be at the bottom of the garden, on the roof, or somewhere that is not convenient.

Using the server method is Ok, but for precise tuning the server can be improved upon. If I were to build one today I would use a totally different system. I would use a brushless motor and worm drive, but I would use three of them, with a sensor resistor to indicate where the tuning actually is. My three controls would be:

LOAD: a 300pf air-spaced capacitor
TUNE: a 300pf air-spaced tuning capacitor
RANGE: A rod of ferrite pushed into an air-spaced coil - no tappings

The brushless motors and gears can be robbed from old R/C helicopters and such. This method does not need any electronics, can give a really wide frequency/impedance match, and the worm drive can easily be adjusted in micro steps for precise tuning. R/C servos are a bit clumsy. In the published design at http://sm0vpo.com/antennas/servo.htm the server driver needed good DC decoupling, but the decoupling had to be fitted across the battery or the servo gave a little kick when the power button was released.

By using the motor/gear/worm-drive the tuning can be made a little slower but a lot more accurate and controllable. If a glitch occurs then the servo can try to go beyond it's limits and a 300pf variable cap often has a mechanical limit at the end of travel. This can cause a mechanical conflict.

The worm drive can easily be an M3 threaded rod with two nuts (to avoid loose play) and can be made really robust. You can also fit limit switches and diodes to eliminate the mechanical conflict.

At the end of the day, there are loads of solutions, but I feel that a bit of Arduino/RaPI free, mechanical tuning would be simple, best and really reliable. but this is probably my age showing. LAN cable has plenty of conductors available for both tuning and feedback, and it is dirt-cheap (especially in the electrical rubbish bins where I get my cable from).

So there is another idea for you. This technique can literally built from the junk-box. The only think you may need to buy is the 3mm threaded rod, and perhaps a little lubricant.

The biggest advantage of this method is that it does not use any computing power or software. But perhaps the biggest disadvantage is that it does not use any computing power or software 

Very best regards from Harry - SM0VPO

PS - I had a really bad year and have not had a break from work for two years, so I decided to pack up and go abroad - take a small break from Sweden and from the job. I went to Dubai and returned two days ago.

The Harry's circuit uses a potentiometer as well, it is in the base of one of the transistors. A 555 or a tiny MCU can be used instead of two transistors.

VBR from Ivan

Ivan wrote:
Hi, I have thought it over once more. A CAT5 cable WILL do provided that all signals are unbalanced against one common ground:
wire 1 - GND;
wire 2 - +V for both servos (may be switched off when no servo moves or be on all the time);
wire 3 and 4 - pulse control of servos A and B respectively;
wire 5 - +V to switch tuner/bypass relay;
wire 6 - +V to switch low/high impedance relay;
wire 7 and 8 - DC voltages corresponding to FWD and REF.
Servo position copies the setting of the control potentiometer, so its indication can be done on a scale under the knob  (of course if all works fine ).

Nice ! As for the position, it can be extracted from the servo

https://www.instructables.com/Servo-Feedback-Hack-free/

but in such a case we won't have more wires in the CAT5 to send it back to the controlbox, yet using a potentiometer instead of two pushbuttons as in Harry's circuit we should have an acceptable indication

Andrew wrote: such a thing would allow to have a visual indication of the two variables position and, in turn, quick retuning of the antenna, once the settings for a given band/frequency have been found, at any rate, a CAT5 cable has 8 wires so we may have ground: 1 wire, servos: 4 wires (2 wires each) and the remaining 3 wires will control the relays, so... if using CAT5 we won't have spare wires for the feedback <sigh>

Hi, I have thought it over once more. A CAT5 cable WILL do provided that all signals are unbalanced against one common ground:
wire 1 - GND;
wire 2 - +V for both servos (may be switched off when no servo moves or be on all the time);
wire 3 and 4 - pulse control of servos A and B respectively;
wire 5 - +V to switch tuner/bypass relay;
wire 6 - +V to switch low/high impedance relay;
wire 7 and 8 - DC voltages corresponding to FWD and REF.
Servo position copies the setting of the control potentiometer, so its indication can be done on a scale under the knob  (of course if all works fine ).

VBR from Ivan

Ivan wrote:Well, a bit of hi-tech will do the job: two cheap MCUs (PIC, PICAXE, Arduino, Teensy etc.) - one in the control box, the other in the remote tuner. A bidirectional serial link is established, TTL up to cca 15 m, RS422 for more. Add +V and GND. All that requires 4 wires (TTL) resp. 3 pairs (RS422). The required angle of both capacitors is set by potentiometers, incremental selector(s) or up-down buttons and sent together with relay bits via the link. The actual positions of capacitors and values of rectified voltages from the coupler are sent back. 8 bit precision should be O.K. Some type of band preset is a possible improvement.

VBR from Ivan

That would turn it into a "smart" ATU which could store the per-band settings, but sincerely I don't feel so good mixing RF and MCUs, they don't play so well, in particular when it comes to a quite "RF dense" environment like an ATU, sure, it can be done, but then one would need pretty heavy shielding to avoid the MCU going crazy, so thank you but I'll pass on that, also since I prefer the "KISS" (Keep It Simple, Stupid) approach or, if you prefer the "what isn't there won't break" one

On a totally different topic, hope that Harry is ok, it has been quite some time since he posted here ! I know he was busy with reorganizing his home...

Well, a bit of hi-tech will do the job: two cheap MCUs (PIC, PICAXE, Arduino, Teensy etc.) - one in the control box, the other in the remote tuner. A bidirectional serial link is established, TTL up to cca 15 m, RS422 for more. Add +V and GND. All that requires 4 wires (TTL) resp. 3 pairs (RS422). The required angle of both capacitors is set by potentiometers, incremental selector(s) or up-down buttons and sent together with relay bits via the link. The actual positions of capacitors and values of rectified voltages from the coupler are sent back. 8 bit precision should be O.K. Some type of band preset is a possible improvement.

VBR from Ivan

Ivan wrote:Hi Andrew,
my experience with AliExpress is quite good. I do not buy semiconductors there to avoid possible falsification, but air variable capacitors are O.K. They are usually taken out from NOS military equipment, often made in the former USSR. In case of problems with delivery I open a dispute and my money is refunded.

VBR from Ivan

I see, well now that may be a source for capacitors, thanks; as for the SWR metering, it may be a nice idea, the remote unit may contain a directional coupler and a small circuit returning a pair of DC voltages directly proportional to the FWD/REV values so that the controlbox may show an indication, but rather than that, I think it may be more useful to extract the position feedback from the R/C servos and return that values, such a thing would allow to have a visual indication of the two variables position and, in turn, quick retuning of the antenna, once the settings for a given band/frequency have been found, at any rate, a CAT5 cable has 8 wires so we may have ground: 1 wire, servos: 4 wires (2 wires each) and the remaining 3 wires will control the relays, so... if using CAT5 we won't have spare wires for the feedback <sigh>

Hi Andrew,
my experience with AliExpress is quite good. I do not buy semiconductors there to avoid possible falsification, but air variable capacitors are O.K. They are usually taken out from NOS military equipment, often made in the former USSR. In case of problems with delivery I open a dispute and my money is refunded.

VBR from Ivan

Hi there, Ivan

As for the rc servos, those are pretty immune to RF fields, so I don't see problems there, as for the air vars... do you really trust that "china stuff" to be any good ?

Hi Andrew,
your idea sounds reasonably. Maybe a SWR sensor should be embedded into the remote unit, while its meter would be in the control box. The main issue may be isolating the servos from strong RF fields.
I found suitable capacitors e.g. here.

VBR from Ivan

I was looking at the "Fri Match" tuner from PA0FRI

https://pa0fri.home.xs4all.nl/ATU/FRImatch/FRImatcheng.htm

and, in particular at the DL4CS version of the tuner

http://dl4cs.de/funktechnik/ant/fri-match/en.htm

now, if you scroll down to the schematic, you'll see that DL4CS added some switching to allow matching low-Z and high-Z loads (antennas), now... let's see we replace those switches with relays (including a bypass one in case the ATU isn't needed on some bands), and let's control the two variable capacitors using R/C servo motors as shown here

http://sm0vpo.altervista.org/antennas/servo.htm

the resulting unit could then be placed directly at the antenna and remotely controlled using a simple "controlbox" in the shack, through a run of cheap CAT5 "network" cable carrying the needed voltages/signals and would be able to match antennas in the range 80m to 10m w/o complex switching and w/o the need for widely spaced capacitors (according to PA0FRI, old radios capacitors will work fine at 100W or so), the only problem will be finding the air variables, since as of today they seem to be quite "rare" <sigh>

What do you think ?