Who is the seller?

On 3/28/2011 1:40 PM, beale wrote:

--
mailto:oz@ozindfw.net
Oz
POB 93167
Southlake, TX 76092 (Near DFW Airport)

Hello, folks. I'm the seller of the 26MHz OCXOs. Please reply off list
if you are interested in some. Thanks.

-Pete

On 04/06/2011 08:58 AM, Oz-in-DFW wrote:

Hi Peter-
I just bought 3 of them for 222MHz ham transverter use like Joe has
done.
Can't wait to get them. Thanks for listing them!

-Brian, WA1ZMS

On Apr 6, 2011, at 5:56 PM, Peter Loron peterl@standingwave.org wrote:

Hi Pete;

I bought 10 of these from you already. Im working on
a converter that has 26M00 Hz in to 10M00 Hz out.
Not sure if that is of any interest but Im putting it on
the table.

Greg

On 4/6/2011 3:56 PM, Peter Loron wrote:

I can't speak for anybody else on the list, but if there was an
inexpensive converter to get a good stable 10MHz signal from one of the
cheap 26MHz OCXOs, I'd sure be interested in building/buying one.

-Pete

On 04/06/2011 06:23 PM, Greg Broburg wrote:

A conjugate regenerative divider with 2 parallel (16MHz & 10MHz) low Q
bandpass filters should suffice.

Bruce

Peter Loron wrote:

On Thu, Apr 7, 2011 at 12:01 PM, Bruce Griffiths
bruce.griffiths@xtra.co.nz wrote:

Never having used one of those, I'm thinking I'd need two mixers, a
10MHz bandpass filter and a 16 MHz bandpass filter.  I'm guessing
there would need to be some gain in the system too.

How stable are these?  Seems to me If the bandpass filters were
temperature sensitive the 10MHz output would drift around.    What
types of filters and mixers are typically used?

I wonder if "conjugate regenerative divider" qualifies as an
inexpensive converter

--

Chris Albertson
Redondo Beach, California

As we used to say in racing, "Speed costs money.  How fast do you want to go?"

CRDs are not particularly expensive, especially when you consider the
performance and what it would take to achieve that performance with
other solutions -- assuming you need/want that performance (and what
kind of time nut would you be if you didn't?).

However, I'm not convinced that a cheap (or even free) 26 MHz
oscillator plus a downconverter will be less expensive for a given
level of performance than just buying a decent 10 MHz OCXO in the
first place -- I've bought quite a few very nice ones on-line in the
$20-50 range.  Mark Sims has suggested some as well (check the list archives).

Best regards,

Charles

On 4/7/2011 4:05 PM, Greg Broburg wrote:

Take a look at Beale's earlier post (beale@bealecorner.com) He's done
most of that for you.

http://www.bealecorner.org/best/measure/time/26MHz-osc-notes.txt

http://www.bealecorner.org/best/measure/time/Pletronics-26MHz-OCXO-tuning.pdf

--
mailto:oz@ozindfw.net
Oz
POB 93167
Southlake, TX 76092 (Near DFW Airport)

Size and power look good so far.

Starting point for me is to run the little oscillators for a week and
see how they muster up.

If that looks good then maybe a few parts from DigiKey to build the filters

Greg

On 4/7/2011 1:55 PM, Charles P. Steinmetz wrote:

ok, so far so good.

Time for some parts.

I am thinking that Ill do the 10 with a crystal filter

Greg

On 4/7/2011 2:28 PM, Oz-in-DFW wrote:

In looking at the CRD idea, it is not obvious to me how either
of the 16M0 Hz or the 10M0 Hz signal would contain any
relevant energy that would give the precision necessary for
the desired ideal result.

The only harmonic relationship that I can see is to square up
the clean analog and divide by 13 down to 2M0 Hz, thence use
the 5th and 8th harmonics at 10M0 and 16M0 but then why
use the 16M0 at all?

I expect that I am missing something obvious here
a little nudge may help.

Regards;

Greg

On 4/7/2011 1:23 PM, Chris Albertson wrote:

The output of the oscillator is square

Greg

On 4/9/2011 10:29 AM, Greg Broburg wrote:

Ok, I'm going nuts.  Why are you guys using such a
perverse way of indicating frequency?

I was of the understanding that SI specified you display
1.0MHz as 1.0MHz, or 1,0MHz.  But not 1M0 Hz.

What's the story?

-Chuck Harris

Greg Broburg wrote:

Hi Chuck:

I think it comes from the way schematics are labeled.
If you use 1.8 k Ohms the point may not be seen since "." is easy to miss.
But if you use 1k8 there's no question about the meaning.

Have Fun,

Brooke Clarke
http://www.PRC68.com

Chuck Harris wrote:

At 02:51 PM 4/9/2011, Chuck Harris wrote...

I suspect it's following the (most common in Europe) convention for
showing units on electrical schematics. e.g. 3k3 for a 3.3 k resistor,
where the multiplier is used in place of a decimal point. It's done
because decimal points sometimes don't reproduce/copy well. Certainly
not necessary, and non-standard, in email.

There is no subharmonic energy in the input signal to a standard
regenerative divide by 2 yet, correctly adjusted, a regenerative divide
by 2 circuit output has very low phase noise over and above that of the
input signal.
The loop starts when the input signal exceeds the threshold required for
the loop gain to exceed unity.

The 16MHz is necessary for the loop to function:
The mixer mixes down the 26MHz to a pair of conjugate frequencies, 10MHz
and 16MHz.
Thermal and device noise is sufficient to start the process.

10MHz = 26MHz - 16MHz
16MHz = 26MHz - 10MHz

The tuned circuits in the feedback loop are required to suppress the
loop gain at other frequencies to avoid unwanted outputs such as 13MHz
(fin/2) and 39MHz (3 fin/2).

Bruce

Greg Broburg wrote:

I believe it has been an ISO convention for at least
20 years and the reason for doing it is zactly whats
been noted here. After youve been party to seeing
what a missed decimal point on a part can do on
an assembly line it makes a lot more sense.

Greg

On 4/9/2011 1:20 PM, Mike S wrote:

Hi Brooke,

My recollection of first seeing the 4K7 style of marking was
around about the time computers started being used for inventory
control... 1970's.  It was only the European companies that
were doing it.  It eliminated the confusion caused by the US/Eu
"." and "," decimal points.  It also is less prone to losing the
"." on a component, or PCB mask.

I acknowledge all of that.  But as far as I know, it has never
been used for time and frequency.  That is what I am concerned
about.  It seriously degrades the readability of the emails that
are formatted that way.

-Chuck Harris

Brooke Clarke wrote:

At 05:03 PM 4/9/2011, Tom Van Baak wrote...

In what ISO spec/standard/??? is that convention given?

I suspect it is strictly informal, as the convention violates the SI
rules: "A multiple or sub-multiple prefix, if used, is part of the unit
and precedes the unit symbol without a separator. A prefix is never
used in isolation" - BIPM, SI Brochure, Section 5.1;
http://www.bipm.org/en/si/si_brochure/chapter5/5-1.html

On Sat, Apr 9, 2011 at 6:48 PM, Mike S mikes@flatsurface.com wrote:

I suspect it is a EIA or JEDEC (or NEMA) standard or convention
perhaps. I don't recognize it as an ISO standard that I am familiar
with.

Americans can (and should) use and follow US NIST Special Publication
811 The NIST Guide for the use of the International System of Units
(2008 edition, web version 3.2) as a suitable guide to SI usage in
USA.

http://www.nist.gov/pml/pubs/sp811/

Personally though I do find the US alternative (deka) spelling of the
deca prefix disconcerting, and prefer trying to avoid potential
ambiguity by using metre for the unit, to avoid being mistaken with
the meter device.

Unfortunately I don't believe that Canada's NRC-INMS has an equivalent
document readily available.

Just so long as I don't have to deal between US "customary" and UK
"imperial", I'm pretty easy going.

-Michael

On 04/09/2011 11:57 PM, Chuck Harris wrote:

For components yes, for other usage... no.

Unit must be know for component type.

I see it as a short-hand being practical and avoids the . or , issues
(confusion and loosing them).

Outside of the component value situation I think SI should be applied.
I prefer 1 nF over 0,001 uF, 1000 pF or 1000 uuF. Also, 10 mF should be
used over 10000 uF. It's not that hard to do right...

Cheers,
Magnus

Magnus Danielson wrote:

My general feeling is if a 9th grader couldn't figure out what the meaning
of the text is without asking someone, it might be a bad idea.

Any 9th grader will understand 4.7 kohms, or 9.6 pf to be a quantity of something
called a kohm, or a pf.

What will they think when they see 4K7 ohms, or 9P6  farads?

Or back to the original item, 10M000 Hz?

SI is what we should be using for all of this correspondence, not some hacked
up system that manufacturers devised for marking components.  You wouldn't
consider telling us to use a yellow violet red and a white blue gold as a
low pass filter.  Why tell us to use a 10M000 oscillator?

-Chuck Harris

On 4/9/2011 11:29 AM, Greg Broburg wrote:

What you are missing is that the concept only applies to small integer
(2 or 3) division ratios and won't work as speculated here.  It's sort
of (long stretch here) like injection locking in reverse.  If you want
I'll try and post some links to papers later.

--
mailto:oz@ozindfw.net    Oz
POB 93167 Southlake, TX 76092 (Near DFW Airport)

Oz-in-DFW wrote:

Nonsense, its already been done for much larger ratios and they need not
be integers.
Try simulating it.

Bruce

Bruce Griffiths wrote:

One counter example to the simplistic statement about the operating mode
of a regenerative divider being restricted to division by small integers
only, is that such analysis appears to preclude the possibility of using
a regenerative divider to produce a frequency comb. Unfortunately a
regenerative divider has already been used to produce a low noise
frequency comb where the comb frequency spacing is f/n(where f is the
input frequency and n is an integer). Its possible to extract a
frequency that is a rational fraction (m/n where m and n are integers)
of the input frequency from such a  regenerative frequency comb. Thus
there is at least one method of using a regenerative divider to produce
a 10MHz signal from a 26MHz signal.

Bruce

On 04/16/2011 10:50 PM, Bruce Griffiths wrote:

As I recall it, in the generalized regenerate divider where two
frequencies is filtered these match up

http://tf.nist.gov/general/pdf/1800.pdf

The two frequencies f1 and f2 has the sum of the input. This has the
side-consequence that

f1 = fin - f2
f2 = fin - f1

which is also the conversion steps that the phase will experience over
two turns around the loop. For synchronous operation the aggregate phase
becomes 0 degrees (modulus 360 degrees).

Considering that fin = 26 MHz and f1 = 10 MHz we can conclude that f2
needs to be 16 MHz.

As for avoiding asynchronous operations the above NIST articles gives
some addtional hints on page 3, among which keeping the loop short is
among the important onces, essentially that the electrical delay length
doesn't support many modes. Keeping all traces on a normal PCB for 10
MHz and 26 MHz should avoid that issue completely.

This would form a 5f/13 - 8f/13 system since 2 MHz is the common
frequency for all of these. Keeping phase solutions unique for 2 MHz
separation should not be too hard.

Cheers,
Magnus

Magnus Danielson wrote:

Attenuating the sum frequencies fin + f2 (36MHz) and fin + f1(42MHz)
sufficiently (~10dB or so) is also required.
The 10MHz and 16MHz bandpass filters in the feedback loop should easily
accomplish this.

Its also possible to use a cross coupled pair of mixers where

the first mixer has 26MHz and 10MHz inputs producing a bandpass filtered
output at 16MHz

the second mixer has 26MHz and 16MHz inputs producing a bandpass
filtered output at 10MHz.

The 10MHz output from the first mixer drives one of the 2nd mixers
inputs whilst the 16MHz output from the second mixer dries one of the
first mixers inputs.

This can simplify the filtering.

Another variant uses a pair of cross coupled lower sideband mixers.

Bruce

Bruce