[hpsdr] Proposal for Loop Antenna Project

[Ken Klein]

Subject: RE: [hpsdr] Proposal for Loop Antenna Project

Tony;

Thanks a million for the comments.  I'm honestly hoping to get lots and lots
from the HPSDR folks for this project.

To address your questions, I think I have found at least a reasonable
starting point for your concerns.  There are several companies that have the
shafts, pillow blocks, couplings, and gears that should provide the
mechanical structure.  I just ordered a handful of parts from a company
called Stock Drive Products.  The sell small quantities to guys like us.

So I envision a motor driving a straight shaft coupled to the cap.  I'll be
using a 22-turn vacuum cap I just bought as well.  The shaft will be
threaded for a couple of inches, so that I can trap a nut to the shaft to
hit the limit switches.  I'm also going to use a couple of small timing belt
sprockets to drive a 10-turn pot for position information.

To tune, I'm going to deploy the antenna with the mechanicals in place and
tune the antenna using a VNA that I built (the Ten-Tec one available from
TAPR).  Any antenna analyzer could be used as well as the SDR itself.  I'm
going to tune the antenna every couple of kHz throughout the range of each
band covered by the antenna and note the position sensor voltage at each
point.  From that I'll make a lookup table that will have all these points
in code.  Each time the DDS frequency is changed; the servomotor will find
the closest resonant point from the table and run the motor to that
position.  This is how I'm going to tune the antenna.  I won't be looking
for receiver noise or SWR; I'll just be retuning the cap to the previously
determined resonant point for that frequency.  Of course, frequencies
between bands wouldn't need to be covered.  I'm figuring a couple of hundred
points in each band ought to do it.  Sounds like a lot, but it will only
have to be done once.  After that, it's all up to the program in the PIC.
Having the Ozy send the PIC the frequency information (DDS code) makes it
all pretty simple.

I hope I've explained this well enough.  Let me know if this answers your
concerns or if you want more detailed info.  I'm really just starting this,
so it might be awhile before I have worked out all the nitty-gritty.

Thanks again for the comments, and very 73s,

Ken WR5H

-----Original Message-----
From: Tony Langdon [mailto:vk3jed at gmail.com] 
Sent: Saturday, June 30, 2007 9:43 PM
To: Ken Klein; hpsdr at hpsdr.org
Subject: Re: [hpsdr] Proposal for Loop Antenna Project

At 12:21 PM 7/1/2007, Ken Klein wrote:
>***** High Performance Software Defined Radio Discussion List *****
>
>
>Gentlemen;
>
>Here is a project proposal that I would like to present to the group 
>for comments and discussion.  I'm hope you won't find it too long or 
>boring.  Thanks to all for the consideration, in advance.

I like the idea of a magnetic loop antenna.  While I've never used 
one, they are said to be an excellent performer, especially where 
height and space are at a premium (for me, height is hard to attain, 
space isn't too bad).

For me, the project does not solve the greatest difficulty - the 
mechanics of the antenna, especially the tuning system and coupling 
the motor mechanically to the tuning cap, so I'm not sure how I would 
implement that.  I had considered a remotely tuned loop in the past 
(manually tuned), but this issue was one of the major sticking 
points, as was sourcing the parts.

Given the tight tolerances, any tuning system would have to be 
automatic.  The tuning could be accomplished in 3 stages:

1.  Open loop using the currently tuned frequency to set the starting point.

2.  Next tuning stage - peaking the receiver noise.  Since the SDR 
has the ability to "see" a fairly wide spectrum (compared to SSB 
bandwidth) at a time, it should be able to see the peak and adjust 
the antenna so the noise peak roughly lines up with the currently 
tuned frequency.

3.  Fine tuning for minimum VSWR.

Small frequency changes could be achieved with an abbreviated 
process.  Small changes (10's of kHz) could easily be done using an 
open loop process, once the tuning characteristics of the antenna 
were known.  The full process would probably only be needed on 
initial startup and band changes.

Just some thoughts.

73 de VK3JED
http://vkradio.com