TenTec
[Top] [All Lists]

Re: [TenTec] High stability frequency reference oscillator for the Orion

To: tentec@contesting.com
Subject: Re: [TenTec] High stability frequency reference oscillator for the Orion II
From: "Dr. Gerald N. Johnson" <geraldj@weather.net>
Reply-to: geraldj@weather.net, Discussion of Ten-Tec Equipment <tentec@contesting.com>
Date: Thu, 31 Dec 2009 11:34:58 -0600
List-post: <tentec@contesting.com">mailto:tentec@contesting.com>
On Wed, 2009-12-30 at 22:16 -0500, Linwood Davis wrote:
> Hello folks,
> 
> Primarily for the reason below, I've been considering drastically 
> reducing the frequency drift over temperature that my Orion II 
> experiences.  So I contacted Abracon Corp to see if they'll make a few 
> 44.55 MHz high stability oscillators to replace the original one with. 
> I'm interested in the AOCJY  series oven controlled crystal oscillator, 
> which has remarkable stability. Try 5 parts per billion stability! (over 
> 0 to 50C). That'd be three orders of magnitude better than the original! 
> Anyway, I know they won't make just one for this mere hobbyist, but 
> hopefully, they'll consider making 10.

High quality oscillators are available off the shelf for common
frequencies like 10 or 27 MHz. Half a part per million holds an HF rig
pretty good. A few years ago Luis CT1DMK published a programmable
reflock circuit that has been developed a bit more for microwave use
that depends on the crystal oscillator being locked to keep the phase
noise down. Its NOT a phase locked loop, its a frequency lock with a
very narrow bandwidth. There are many web pages about it and one
variation intended to use with 1 pps from a GPS to improve the stability
of the 10 MHz reference.

Then there is DFS by G4HUP and others that starts with a 10 MHz signal,
multiplies and mixes to get arbitrary frequencies with relatively low
added phase noise. Starring with 10 MHz, maybe multiply by 4 to get 40,
divide that by ten to get 4, mix to get 44, divide 10 by 18 to get .555
mix again. Probably better to mix .555 with 4 to get 4.555, then mix
with 40 to get 44.555, maybe warp the 10 MHz a tad low to get 44.550000.

Maybe multiple 10 by 11 to get 110 and divide by 200 to get .550.
Suitable filtering followed by a narrow crystal filter at 44.55 keeps
things clean. Shielding is important.

Maybe multiply 10 by 11 to 110. Divide by 10 to get 11, multiply by 4 to
get 44. Divide 11 by 20 to get .550 mix with 44 followed by a crystal
filter.

Another alternative, get a crystal oscillator kit from G8ACE that is
temperature compensated and oven controlled, and buy a quality crystal
from International Crystal Manufacturing and brew your own high quality
reference.
> 
> So I'm here to see if any Orion owners would be interested in also 
> purchasing one. I got a informal quote of $135 a piece so far.
> 
> Also, I don't know, but does the Orion (565) also use this frequency 
> reference? Or any other Ten-Tec radio? I see from the O II schematic 
> that the Sub Rx uses a divided-down signal from this same oscillator. If 
> I remember right, the Sub Rx design is the same as the Jupiter Rx. 
> (Please correct me if I'm wrong.) If so, are there any Jupiter owners 
> that may be interested in joining this oscillator purchase?
> 
> So here's where the drift becomes annoying:
> 
> Recently, I began "watching" meteors with my Orion II and DL4YHF's 
> Spectrum Lab software. What I do, actually, is tune to 14.670 MHz (CHU 
> Canada), and observe CHU's signal. Using Spectrum Lab's waterfall to 
> display a 50 Hz swatch of spectrum around the carrier, I can see when 
> micrometeors and not-so-micro meteors strike the ionosphere between 
> here, NH, and there, Ottawa. Normally, the signal is fairly weak, so 
> when a meteor ionizes a trail in between here and there, the signal 
> strength significantly increases. What's more, the resulting spectrum is 
> often composed of curves and lines above and/or below the carrier 
> frequency. I believe these are created by the doppler shift due to 
> ionospheric winds. (Why do I do this? Hey, I'm a curious guy, what else 
> can I say?)
> 
> The trouble is, especially just after turning the Orion on (and for the 
> next 30 to 45 minutes), the frequency drifts by as much as 10-15 hertz. 
> Normally, you probably don't notice this, but when narrowed down to a 50 
> Hz display, it looks bad! (See AA6E's site, 
> blog.aa6e.net/2005_08_01_archive.html, or N6IE's site, www.N6IE.com, for 
> addition drift info.)
> 
> Let me know,
> 
> Thanks,
> 
> Lin
> WB1AIW
> 

Then there is the possibility of improving the temperature control
environment of the existing oscillator, but you will still get probably
a part per million drift a year from aging unless you put lots of money
into a custom special very well aged crystal. I have begun working on a
closer temperature compensator based on a temperature sensor and a PIC
chip to work with voltage controlled crystal oscillators to improve
their stability. One major manufacturer of standards has done that with
a crystal oscillator to get a short term 3 order of magnitude stability,
but can't lick aging.

There is much discussion in the microwave ranks about fine oscillators
vs PLL because phase noise and a little bit of frequency wander gets
serious at 10 GHz or higher, where a ppm is 10 KHz drift, gets a lot
worse at 241 GHz, but that only bothers half a dozen hams in the world
at the moment.

73, Jerry, K0CQ

_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec

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