TheRadioBoard

Great idea. This is starting to balloon into a full-blown microcontroller and robotics project, but it sounds fun.

I found this guy who made a microcontroller-controlled servo that uses the FT-817's own SWR meter reading as the error function (eliminating the need for separate SWR-measuring equipment), to tune a portable magnetic loop antenna. This project isn't feeding the SWR directly into the servo's position corrector, but instead requires a separate Arduino to measure SWR then command the servo to go to a certain (rotational) position.

http://www.alain.it/2012/02/10/ft-817-automatic-loop/

As for the required angular precision, doing some back of the envelope calculations: assuming a resonant frequency of 7 MHz and a required capacitance of 150 pF gives a loop inductance of 3.45 uH. Assuming a 200 pF capacitor with 180 degrees of rotation gives 1.11p pF/degree. A change of 1.11 pF (1 degree) gives a change in resonant frequency of over 25 kHz at 7 MHz. Assuming bandwidth of the antenna is 5-10 kHz at 7 MHz (low losses), then probably 1/10th of a degree rotational resolution would be desirable for fine tuning. At the higher end of the tuning range (say, 18 or 21 MHz) the resonant frequency change per degree of rotation will be higher but the bandwidth will also be greater, so hopefully the two effects will cancel out and the same rate of rotation will be usable at the high-end of the tuning range.

So probably 1/10th of a degree resolution sounds like a ballpark target figure. Does this sound do-able with the proposed solutions (200 RPM geared motor with PWM for slowing the rotation, or commanding a servo to go to xxx.y degrees)?

I think 200 RPM is still too fast to get any kind of stable operation.

PWM is nothing more than variable voltage control. There's no magic. The main advantage to PWM is that it maintains a low supply impedance as the average output voltage is lowered, so the motor is less likely to stall than if the voltage is controlled with a variable resistance. You still have to deal with static friction of the load when the motor starts up, which means the motor has to be given enough voltage to get things moving. Then, if things aren't tuned properly, the motor goes like a bat out of hell and overshoots the target position before you have a chance to stop it. If you gear it down more (I figure 10 RPM is the fastest you want to run), then you'll have plenty of startup torque and you won't have to worry so much about things running away on you.

Yes, reducing torque load is the goal. On a model airplane the bearing areas of control-surface fulcrums have low friction, and I don't think airflow adds much loading. You can feel the resistance in getting a variable cap turning. Getting fractional-degree control might take a stronger motor than expected.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

uhm, i think gearing and stepper motor from a printer should work.
if reassembled like in the printer (or even cut as is straight from printer) it would give a resolution equal to printer resolution, most modern printers can do 600x600dpi... that should translates to 600 divisions to a circle as the drum feeds the paper at desired rate and precision, or am I wrong?.

I think a stepper with the correct gearing should work fine. That's probably the method I'd choose if a microcontroller was allowed. Or, come to think of it, you can connect an incremental shaft encoder directly to stepper and essentially have a manual remote shaft position control.

Use one of these little panel mount electromechanical encoders:
http://search.digikey.com/scripts/DkSea ... =P12334-ND
Add a stepper motor and a power supply, and you've got a working manual system. Then if you want to add automation later, throw in a microcontroller.

(As for printers, I don't know about all of them, but I believe that some of the fast high res ones use encoder feedback of instantaneous position to fire the ink jets, because the alternative of sending a step pulse then waiting for a predetermined settling time then firing the jets then sending another step pulse is too slow.)

Hmmm... stepper motors, rotary encoders - more Greek to me. Reading Wikipedia still leaves me with a lack of an intuitive grasp of what I would be doing with the encoder and the stepper. As far as I can gather at this point, to control a stepper you need to control the firing of the electromagnets in a particular sequence. So then, is it correct that an incremental quadrature rotary encoder like the one linked above will fire the electromagnets in the right sequence? Wikipedia seems to indicate that the output will be binary with two bits A, B in the sequence 00, 01, 10, 11. Do I just connect these A/B outputs to the stepper? Is there a specific term for the kind of stepper I would need?

Here, for example, is a random stepper motor available from a local dealer (SPG20-1332, capable of 0.75-degree steps). Does it look applicable?
http://akizukidenshi.com/download/SPG20-1362.pdf

Same question about this rotary encoder, available locally (P-00292). Looking at the waveforms seems to indicate it's also quadrature and thus should be applicable.
http://akizukidenshi.com/download/ds/ru ... 0A-24P.pdf

How would I wire this whole thing up? There seem to be six leads coming off the stepper (3 for each electromagnet), but only three coming off of the encoder. I'm completely new to this, as you can tell.

Short answer: Yes, the encoder puts out the correct quadrature signal that a stepper motor wants, in order for it to step one position. Turn the encoder one click clockwise, and the stepper motor moves one step clockwise. Turn the encoder one click CCW and the stepper motor moves one step CCW.

There are different types of stepper motors categorized by the number of leads they have coming out of them. I haven't done much with them, and can't remember at the moment which type you need exactly. Pretty sure it's less than six leads though. If you were to cannibalize an old printer, fax machine, or maybe VCR, I suspect you'd find a suitable stepper.

Steppers in an old VCR? Interesting. I happen to have one lying around that is slated for cannibalization. During my last move I almost threw it out but then decided that I wanted to salvage the UHF down-converter circuitry (which is probably hard to get a hold of nowadays) for a future super-regen project as described by Charles Kitchin. That it might also have steppers is icing on the cake.

Not sure about VCRs, they have several motors and most are regular DC motors, but there may be one in there that's a stepper.

I spent over ten years repairing VCRs, I don't recall one that had a stepper motor. Some have some belt driven gear assemblies that might find use.
Some use a worm gear and a large gear to load the tape, maybe slow enough to control a capacitor.
I got out of repair when VCR prices were getting close to $200. It's getting
hard to buy a VCR anymore. The last few I've got were from yard sales or Goodwill/Salvation Army.
Mikek

If you do not need high precision and only about 180 degree of the angular motion range then get a basic "standard" size RC servo like this:
http://www.hobbyking.com/hobbyking/stor ... oduct=3743
and a servo tester like this:
http://www.hobbyking.com/hobbyking/stor ... oduct=8296
and you are set for about $10 total plus shipping.

If you need either high angular accuracy or more than 180 degree motion then mechanical gearing with a stepper is probably the way to go. Typical steppers are 200-400 steps/turn and if you do not care about the speed can be easily controlled by just about any digital device, from an old PC parallel port to any microcontroller board like arduino or raspberry. All that is needed is few mosfets and a power supply (put the mosfets near the motor and run control signals via cable to minimize EMI).

I'm just mentioning this so it maybe can give someone an idea. Not to anyone particular.
You can buy excellent small geared motors from the Chinese on ebay in many sizes. Do a search and you ill find them. Yes, I have bought several for different projects and these are really high quality. Small wonders.
http://www.ebay.com/itm/3-6V-DC-Small-M ... 2c658f9e49

I found this neat-looking geared-down DC motor:

http://www.tamiya.com/japan/products/70 ... /index.htm

Small, cheap ($10), and low-RPM (30 RPM). This is about the lowest-RPM geared motor I've been able to find cheaply (and locally available). I think I may give this a try with PWM, which could hopefully reduce the RPM down into the 1 RPM range. Also this motor has two output shafts, so I can drive two separate variable capacitors in a mechanically-ganged fashion, as is needed in the capacitive loop matching scheme I'm considering:

(see left-most image)

from http://vaedrah.angelfire.com/antenna.htm

(see right-most image)

from http://www.nonstopsystems.com/radio/fra ... agloop.htm

I could use one of the above geared motors (with two shafts) for controlling the matching capacitors, and another geared motor (just using one shaft) for controlling the tuning capacitor.

I am no expert but I am wondering if gearing is the answer. Wouldn't one want (need?) to be able to move quickly from one spot on the dial, to another, without waiting a minute to get there? Varying the motor speed, whilst retaining the required torque, would seem to be more useful (if achievable) than having a fixed speed gearing arrangement?

.....................................

I'm hoping that I can use PWM to continuously vary the motor speed from 30 RPM (full voltage speed) to 1 RPM. I would turn the speed knob to slow for fine-tuning and to fast for hopping from one end of the band to another. Whether the torque will be adequate is the big question.

As mentioned before, steppers might also be a solution, but I haven't been able to find any ready-made geared-down steppers that can step in a resolution much less than one degree (which I think I need), which means that I would need to build the down-gearing mechanism myself - mounting and aligning the gears and the motor shaft. This is a task I'm not quite up to. On the other hand, an already-geared-down and very compact motor like the above looks like a nice ready-made part, if I can get it to go slow enough.

If I can get this working, then I can think of a lot of fun radio applications for remotely controllable knobs, especially when paired with computer or microcontroller sensing of the parameters controlled by those knobs for a closed-loop control system.