Towertalk
[Top] [All Lists]

Re: [TowerTalk] receivers for HF phased array

To: richard@karlquist.com
Subject: Re: [TowerTalk] receivers for HF phased array
From: Jim Lux <jimlux@earthlink.net>
Date: Mon, 05 Dec 2005 19:49:26 -0800
List-post: <mailto:towertalk@contesting.com>
At 04:47 PM 12/5/2005, Rick Karlquist wrote:
>Jim Lux wrote:
> > Folks, I'm looking for receivers for a HF phased array (they can be
> > transceivers, though).  Here's kind of what I'm looking for:
> > Everything locked to a single oscillator in the receiver is a plus (makes
> > calibration easier). Even more of a plus is if I can modify it to accept
> > an
> > external oscillator (of whatever frequency is needed).
>
>It's not that simple.  Merely being able to lock the *frequencies*
>together does nothing to address phase.  You will get random phases
>that change every time you tune the frequency synthesizers in the
>receivers.  You would have to measure these random phases on a real
>time basis and correct for them.  Then there are the 2nd and 3rd
>LO's to worry about too.

Yes.. but once locked, it stays the same, which is a whole lot easier to 
deal with than having to track it continuously.  You can have a routine 
that does "tune, start pilot, measure, stop pilot, etc."  Otherwise, I have 
to put a pilot tone out there continuously, and track it and remove it.


>I think what you really need to do is use single conversion receivers
>and provide the actual LO to the receiver externally.  I don't see
>any way you can leverage off the shelf receivers to do this in any
>way that would be cheaper than just building your own receivers purpose
>built to be phased array receivers.

One can hope..

I had thought of the building my own receivers approach.. Whacking big LO 
to some burly mixers to 70 MHz or something, and then, mix down from 70 to 
baseband.  But there's oh so many hassles there, even if I do buy all the 
parts from MiniCircuits.. IF gain, filters, etc. If you use connectorized 
components you get over a kilobuck per channel pretty darn quickly.  If you 
use solder-in components, then you've got the time to lay out a board, find 
a box to put it in, etc.

Not to mention, there's a whole of "art" in building *good* 
receivers.  Crummy receivers are no sweat.. I do it all the time.



>The exception to this would be if you could find receivers that are
>exclusively based on DDS LO's.  Those LO's are phase coherent.  However,
>I would be surprised if anything off the shelf totally avoided
>phase locked loops and divide by N's.

That's sort of the SDR1000 approach, except you'd still have to modify 
things to feed the LO across to multiple boxes.  The SDR1000 doesn't bring 
out the "load" line to the DDS so you can't sync them trivially.


>Let us know if you come up with anything having complexity within
>hobby parameters, as opposed to what the govt can afford.

The sort of plan B is to use some readily available off the shelf receiver, 
and then do some "simple" modification. Say you had a decent receiver, you 
might be able to do some master LO distribution mod.  But, there's a 
controllability issue then.. All that nifty computer control sort of goes 
away, or has to be entirely rewritten.




>Rick N6RK

_______________________________________________

See: http://www.mscomputer.com  for "Self Supporting Towers", "Wireless Weather 
Stations", and lot's more.  Call Toll Free, 1-800-333-9041 with any questions 
and ask for Sherman, W2FLA.

_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk

<Prev in Thread] Current Thread [Next in Thread>