[Lowfer] Auto tuning

[C. Turner]

Years ago (mid 80's) Mark, WB7CAK had LowFER MPM on the air and he had 
an automatic variometer tuner.  (He *may* have described it in an issue 
of the Western Update.)

The way it worked was fairly simple:

- A bit of RF *voltage* was sampled from the line between the TX and the 
loading coil.  This was done with a resistor or capacitor or something.
- The RF *current* was sampled from the same line.  I believe that he 
simply ran a single turn primary through a toroid (TX to coil) and had 
multiple turns on the secondary.

These two voltage were then thrown into, I believe, a 4-quadrant 
multiplier (an MC1495, IIRC) and the resulting voltage was used to 
determine if the load was inductive or capacitive by the phase differential.

As I recall, the sampling methods caused the two RF waveforms to be 
slightly out of phase when the antenna was tuned to be resistive (as 
viewed on a dual-trace scope, for example) but since that was a constant 
amount, he either "fixed" that with an L/C phase adjuster somewhere, or 
simply took it into account on his measurement on the output of the 
multiplier in some way.

Mark also had some means of disabling the system when RF was absent - 
such as when it was unkeyed - otherwise the variometer might go crazy 
trying to make sense during a key-up (no-RF) condition.

For moving the variometer, I believe that he had just a stepper motor on 
it, using some simple counter/driver circuit to move it back and forth:  
It would go one way if capacitive and the other way if inductive.  It 
was interesting to watch it in action:  If you moved your hand near the 
antenna, it would start stepping in one direction, and when you moved 
away, it would go back by itself - just like it should!  It also seems 
to me that he might have had to add some damping when he switched to 
10-baud BPSK to slightly slow the response to keep it from "jittering" 
from the phase modulation.  He may also have added some simple logic (or 
mechanical stops) to prevent it from going too far out into the weeds 
if, say, snow caused the antenna to be detuned out of the range of the 
system and to keep it from getting "lost."

While I never did build an automatic tuner, I followed another of Mark's 
suggestions:  Put a 1/4 wave lumped-constant network (2 caps and an 
inductor) between the TX and the loading coil.  Since, at the time, I 
used a transmitter with a 50 ohm output, this would transform the output 
impedance of the transmitter (50 ohms) to the input resistance of the 
resonant loading coil/antenna system (NOT 50 ohms.)  In addition to 
matching the TX to the feedpoint, it would also act as an "impedance 
inverter" and if the antenna detuned a bit, the network tended to cancel 
some of that out.  For LowFER SSB use, it increased the effective 
operating bandwidth somewhat (without spoiling the Q of the loading 
coil/antenna system) and made everything a bit less-sensitive to 
detuning due to changes of the antenna overall.  The bad part was that 
this network wasn't compatible with a class-E output stage without the 
addition of yet another series L/C filter to keep the P.A. happy and 
operating in proper class-E mode.

73,

Clint
KA7OEI


Eric Smith wrote:
> It's been years since I posted this question...at the time we were all
> on the old reflector, I think.  And I think Bill replied that he could
> imagine a solution using op-amps, I think.
>
> That is, how cool it would be to have an automatic variometer like
> they use at WWV on 60 kc.
>
> Now, I couldn't design such a circuit, but I know there are those on
> here who could.  If you'll design it, i'll build one!
>
> Eric  KD5UWL / WD2XFX
>
> On 5/20/09, Andy - KU4XR <ku4xr at yahoo.com> wrote:
>   
>> --- On Wed, 5/20/09, Garry Hess <k3siw at sbcglobal.net> wrote:
>>
>>
>>