On Wed, 16 Jun 2004 22:23:01 -0700, Tim Wescott
wrote:

need a low-noise xtal oscillator using 14.7 or 15.7MHz rubber xtals in
series mode circuit to sweep xtal filters with good as possible noise
distance. The available xtals are believed to pull +/-10kHz or more.

This is a fairly standard circuit -- why not just look in an ARRL manual
or one of the many QRP transmitter designs for an oscillator? You can
check for phase noise by mixing it with a good (fixed) crystal
oscillator, or you can just trust it.

many of those amateur constructions are not very good, just put
together, and if they generate a sort of signal the constructors are
happy without worrying what they sound like. Some of the constructor
have of course well reputation, but ARRL handbook usually don't tell
who have designed it.

and in this case I want a definite low noise construction.

Remember when I built my first of many VHF/UHF beacons made after an
amateur concept I listened to the carrier with R-4C and converter, the
sound was awful, so I decided to built a new 12MHz xtal oscillator
using some practice described by Ulrich Rohde, DJ2LR. The improvement
on 144MHz was increadible, and I later learned that the sound was very
good compared with all sorts of rubbish somebody else used, even on
2320MHz

Jan-Martin, LA8AK
---
J. M. Noeding, LA8AK, N-4623 Kristiansand
http://home.online.no/~la8ak/c.htm

J M Noeding wrote:

On Wed, 16 Jun 2004 22:23:01 -0700, Tim Wescott
wrote:


need a low-noise xtal oscillator using 14.7 or 15.7MHz rubber xtals in
series mode circuit to sweep xtal filters with good as possible noise
distance. The available xtals are believed to pull +/-10kHz or more.


This is a fairly standard circuit -- why not just look in an ARRL manual
or one of the many QRP transmitter designs for an oscillator? You can
check for phase noise by mixing it with a good (fixed) crystal
oscillator, or you can just trust it.


many of those amateur constructions are not very good, just put
together, and if they generate a sort of signal the constructors are
happy without worrying what they sound like. Some of the constructor
have of course well reputation, but ARRL handbook usually don't tell
who have designed it.

and in this case I want a definite low noise construction.

Remember when I built my first of many VHF/UHF beacons made after an
amateur concept I listened to the carrier with R-4C and converter, the
sound was awful, so I decided to built a new 12MHz xtal oscillator
using some practice described by Ulrich Rohde, DJ2LR. The improvement
on 144MHz was increadible, and I later learned that the sound was very
good compared with all sorts of rubbish somebody else used, even on
2320MHz

Jan-Martin, LA8AK
---
J. M. Noeding, LA8AK, N-4623 Kristiansand
http://home.online.no/~la8ak/c.htm

Well, that's why I suggested verifying it's phase noise with a good
fixed crystal oscillator. I certainly wouldn't proceed without doing
this check -- the nice thing is that unless the oscillators lock
together too tightly you get an upper bound to your phase noise, so you
know when you're doing well enough.

"Solid State Design for the Radio Amateur" has some good (older)
circuits. It's been superseded by "Experimental Radio Design" which
should also have some good ones.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

J M Noeding wrote:

On Wed, 16 Jun 2004 22:23:01 -0700, Tim Wescott
wrote:


need a low-noise xtal oscillator using 14.7 or 15.7MHz rubber xtals in
series mode circuit to sweep xtal filters with good as possible noise
distance. The available xtals are believed to pull +/-10kHz or more.


This is a fairly standard circuit -- why not just look in an ARRL manual
or one of the many QRP transmitter designs for an oscillator? You can
check for phase noise by mixing it with a good (fixed) crystal
oscillator, or you can just trust it.


many of those amateur constructions are not very good, just put
together, and if they generate a sort of signal the constructors are
happy without worrying what they sound like. Some of the constructor
have of course well reputation, but ARRL handbook usually don't tell
who have designed it.

and in this case I want a definite low noise construction.

Remember when I built my first of many VHF/UHF beacons made after an
amateur concept I listened to the carrier with R-4C and converter, the
sound was awful, so I decided to built a new 12MHz xtal oscillator
using some practice described by Ulrich Rohde, DJ2LR. The improvement
on 144MHz was increadible, and I later learned that the sound was very
good compared with all sorts of rubbish somebody else used, even on
2320MHz

Jan-Martin, LA8AK
---
J. M. Noeding, LA8AK, N-4623 Kristiansand
http://home.online.no/~la8ak/c.htm

Well, that's why I suggested verifying it's phase noise with a good
fixed crystal oscillator. I certainly wouldn't proceed without doing
this check -- the nice thing is that unless the oscillators lock
together too tightly you get an upper bound to your phase noise, so you
know when you're doing well enough.

"Solid State Design for the Radio Amateur" has some good (older)
circuits. It's been superseded by "Experimental Radio Design" which
should also have some good ones.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

In article ,
says...

A DDS isn't going to have good enough phase noise. The OP is correct in
using a pullable crystal oscillator.

Eh? He wants to sweep a filter. You don't particularly care about
phase noise when you do that.

-- jm

------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------

In article ,
says...

A DDS isn't going to have good enough phase noise. The OP is correct in
using a pullable crystal oscillator.

Eh? He wants to sweep a filter. You don't particularly care about
phase noise when you do that.

-- jm

------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------

John Miles wrote:
In article ,
says...


A DDS isn't going to have good enough phase noise. The OP is correct in
using a pullable crystal oscillator.


Eh? He wants to sweep a filter. You don't particularly care about
phase noise when you do that.

-- jm

------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------

That depends on your filter. If you're trying to design a high pole
count filter with really steep skirts and you want to verify it's final
rejection then yes, you need a low phase noise oscillator. This is
probably why he has "low noise" in his title.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

John Miles wrote:
In article ,
says...


A DDS isn't going to have good enough phase noise. The OP is correct in
using a pullable crystal oscillator.


Eh? He wants to sweep a filter. You don't particularly care about
phase noise when you do that.

-- jm

------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------

That depends on your filter. If you're trying to design a high pole
count filter with really steep skirts and you want to verify it's final
rejection then yes, you need a low phase noise oscillator. This is
probably why he has "low noise" in his title.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

On Thu, 17 Jun 2004 11:03:44 -0700, Tim Wescott
wrote:


------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------

That depends on your filter. If you're trying to design a high pole
count filter with really steep skirts and you want to verify it's final
rejection then yes, you need a low phase noise oscillator. This is
probably why he has "low noise" in his title.

can't really see I've got any replies in the wanted direction; If I
wish to measure a receivers ultimate performance I am not interested
in seing something like 60dB selectivity when I am expecting 90dB or
more. Had a discussion with LA8OJ, and he suggested that 6MHz ceramic
resonator might be easier to use than a 15.7MHz xtal, 100kHz tuning
range could be achieved instead of 20kHz. But I suppose it was a trick
to limit the amplitude to improve phase noise of an xtal oscillator

73
Jan-Martin, LA8AK
http://home.online.no/~la8ak/c.htm
---
J. M. Noeding, LA8AK, N-4623 Kristiansand
http://home.online.no/~la8ak/c.htm

On Thu, 17 Jun 2004 11:03:44 -0700, Tim Wescott
wrote:


------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------

That depends on your filter. If you're trying to design a high pole
count filter with really steep skirts and you want to verify it's final
rejection then yes, you need a low phase noise oscillator. This is
probably why he has "low noise" in his title.

can't really see I've got any replies in the wanted direction; If I
wish to measure a receivers ultimate performance I am not interested
in seing something like 60dB selectivity when I am expecting 90dB or
more. Had a discussion with LA8OJ, and he suggested that 6MHz ceramic
resonator might be easier to use than a 15.7MHz xtal, 100kHz tuning
range could be achieved instead of 20kHz. But I suppose it was a trick
to limit the amplitude to improve phase noise of an xtal oscillator

73
Jan-Martin, LA8AK
http://home.online.no/~la8ak/c.htm
---
J. M. Noeding, LA8AK, N-4623 Kristiansand
http://home.online.no/~la8ak/c.htm

In article ,
says...

That depends on your filter. If you're trying to design a high pole
count filter with really steep skirts and you want to verify it's final
rejection then yes, you need a low phase noise oscillator. This is
probably why he has "low noise" in his title.

Perhaps, but if you want to verify final rejection, you want much wider
tuning range than you're going to get from a pulled crystal. A 100 Hz-
wide filter might be -100 dB down at 5 kHz but -50 dB down at 50 kHz,
and if you care about that, you need to be able to look farther away
from the carrier.

His desire for a narrow-range source made me think he was more
interested in shape-factor and ripple alignment than ultimate rejection.

Maybe he can use two different oscillators to get the best of both
worlds. A high-quality LC oscillator is actually pretty darned quiet at
large offsets from the carrier. I think it was Tom Bruhns, or one of
the other HP guys at least, who once pointed out that the old HP 608-
series boatanchors were quieter at 100 kHz offsets than the flashy,
high-dollar synthesizers that followed them.

-- jm

------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------