On 05/26/2013 06:04 PM, Attila Kinali wrote:

The OSA 8602 is a variant of the OSA 8600 and 8601. These variants is
mainly on the connection on the "front".

I don't have a 8602 datasheet as such, but I have some 8602 related
specs as found in the extended OSA 3000 manual.

It is essentially the same AT-cut oscillator that you can expect from
the 8600 base.

What information are you really seeking?

Cheers,
Magnus

Moin,

On Sun, 26 May 2013 18:15:22 +0200
Magnus Danielson magnus@rubidium.dyndns.org wrote:

Hmm.. IIRC AT cut oscillators have the "problem" of frequency jumps
on slight temperature changes. Using an AT cut oscillator thus kind
of defeats the effort of doing a BVA.

What such an oscillator would be worth :-)
You might be aware that there is one 8602 on sale on ebay for 4500USD.
I asked Oscilloquartz about that and from what i gathered, it's definitly
not worth that money. For slightly more you can get a new 8607 already.
The 8602 worth is probably around 150 to 200USD.

		Attila Kinali

--
The people on 4chan are like brilliant psychologists
who also happen to be insane and gross.
-- unknown

On 05/28/2013 06:57 AM, Attila Kinali wrote:

Frequency jumps isn't a particular feature of AT cut, but rather to
cleanness of the crystal and mechanical stresses. The BVA strategy aims
to reduce systematic shift, it is then baked out to remove residues that
is known to cause issues. There is a good article on it amongst the PTTI
papers. Also, it's an oven within a dewar flask, so temperature
conditions is pretty stable. The AT-cut BVA is far from the same thing
that a typical TTL-can AT-cut is. There seems to be an overbeleif in the
cut and not look at all the other things that needs to come together to
make a great oscillator.

That is overpriced. An OSA 8602 should have about the same price as OSA
8600, whatever the going price of that is.

Cheers.
Magnus

I think that used BVA pricing is difficult since prices of really good quartz is to me exponential. What the BVA offers is exceptional close in phase noise. And because of their reputation any BVA listed on eBay will get some serious bids. I would find it interesting to see what a couple nice 8607 would sell for on eBay. I know I would pay dearly, and I doubt I would be up with the highest bidders. But today companies like Wenzel offers exceptional performance at much more reasonable prices.  So based on Phase Noise alone an older 8600 could  be much less. The other big thing is find a BVA for sale. They rarely become available used and I would say it is a seller market.  That said none of us time nuts have jumped on the chance to buy the one currently on eBay at $4500, so everyone seems to agree that is to expensive. But if you are lucky enough to own a BVA what would you sell yours for? I imagine that in view of what you would sell for the guys $4500 price looks a bit more reasonable.  I guess my question is what would you pay and what have you paid for a 5MHz oscillator with -120dB @1Hz? what about  -130dB @1Hz? what about -135dB at 1Hz?

Tom Knox

Hi

The cost of a BVA oscillator is primarily a function of the cost of the
blank used and secondarily a function of the resonator processing. You see
numbers in the $200 to $400 range tossed around for the blank (vs < $10 for
a good SC blank). The packaged resonator starts looking like $600 to $900
once you have that sort of blank in it (vs < $20 for a similar SC). BVA's
are never going to be cheap or plentiful.

I think you will find a number of people selling 5 MHz OCXO's in the -110 to
-120 dbc/Hz at 1 Hz range. That's not really where the BVA is needed. It's
benefit is more in the 0.1 or 0.01Hz phase noise. Put another way, ADEV is
the better tool to rate a BVA.

If you really want to get any resonator (or even an atomic standard) to look
good on ADEV at 1 to 10,000 seconds, it needs to be temperature stable.
Either it needs a really good oven, or you need a lab that doesn't change
temperature. For good ADEV in a normal environment, that means a "hockey
puck" style single oven, or a good double oven. Again, a cost adder /
greater scarcity thing.

Given all that, a BVA that gives you < 1x10^-13 from 1 to >1,000 seconds in
a normal lab is a very rare item. What ever the cost, it's going to be
pretty steep.

Bob

-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Tom Knox
Sent: Tuesday, May 28, 2013 11:55 AM
To: Time-Nuts
Subject: Re: [time-nuts] Looking for datasheet for Oscilloquartz 8602

I think that used BVA pricing is difficult since prices of really good
quartz is to me exponential. What the BVA offers is exceptional close in
phase noise. And because of their reputation any BVA listed on eBay will get
some serious bids. I would find it interesting to see what a couple nice
8607 would sell for on eBay. I know I would pay dearly, and I doubt I would
be up with the highest bidders. But today companies like Wenzel offers
exceptional performance at much more reasonable prices.  So based on Phase
Noise alone an older 8600 could  be much less. The other big thing is find a
BVA for sale. They rarely become available used and I would say it is a
seller market.  That said none of us time nuts have jumped on the chance to
buy the one currently on eBay at $4500, so everyone seems to agree that is
to expensive. But if you are lucky enough to own a BVA what would you sell
yours for? I imagine that in view of what you would sell for the guys $4500
price looks a bit more
reasonab
le.  I guess my question is what would you pay and what have you paid for a
5MHz oscillator with -120dB @1Hz? what about  -130dB @1Hz? what about -135dB
at 1Hz?

Tom Knox

definitly

https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts

time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

On Tue, 28 May 2013 09:27:55 +0200
Magnus Danielson magnus@rubidium.dyndns.org wrote:

Can you give a little more hints on which paper you mean?

Juup. I just went back to Vig's tutorial and read up what he wrote.
Misremembering things is not a good thing...
But then, he explicitly writes that SC cut gives a higher stability
over AT cut due to lower temperature dependence and less dips.

I haven't explicitly asked for a price of the 8600, as it isn't sold anymore.
But the price of a new 8607 (their only BVA they still sell, with SC cut
crystall and all bells and whistles) is a mere ten percent higher then
the initial price of what he tries to sell the 8602 for.
So i wouldnt say the 8602 is overpriced, but rather it's horribly
overpriced.

		Attila Kinali

--
The people on 4chan are like brilliant psychologists
who also happen to be insane and gross.
-- unknown

On 05/28/2013 07:55 PM, Attila Kinali wrote:

http://www.pttimeeting.org/archivemeetings/1984papers/Vol%2016_10.pdf

See also
http://www.pttimeeting.org/archivemeetings/1979papers/Vol%2011_25.pdf

That is true, but your discreditation of AT-cut was simply way off the
mark, so I wanted to bring it into context. There is a difference, but
it is not as huge as it sounded like. I like to think about it as such
that you better have done much of your homework in form of good
oscillator and oven before considering spending money on going from
AT-cut to SC-cut, but it does give that extra performance if you need
it. The OSA 8600 shows just how far you can take AT-cut.

True.

If it where the going rate for a OSA 8602, I would make a small fortune
from what's in the lab and be able to upgrade the living conditions for
the sake of a better lab.

Cheers,
Magnus

On 5/28/13 9:29 AM, Bob Camp wrote:

The USO's we got for GRAIL from APL have ADEV<1E-13 from 1 to 1000
seconds, and then heads up at 1 decade/decade.  The lowest ADEV is about
5E-14 at around 50 seconds, but it's pretty flat.  See the paper by
Enzer et al.

Or better, the 42ns PTTI conference paper by Greg Weaver at APL, who had
to build them.

They are space qualified, so they're fairly pricey (>$1M a copy).  But
to put the discussion about blanks and finished product.. I think they
started around 1000 blanks to produce a few dozen packaged crystals,
which were then run as oscillators to see which ones were the best,
producing maybe four final pieces, 2 at each frequency.

The Weaver PTTI paper compares these to the BVA resonators.. SC cut
5MHz, 3rd overtone.

They run in a vacuum bottle (of course), and they have somewhat
obsessive attention to a lot of details.  But I suspect that aside from
the space qual aspects, the whole "how you build them" isn't a whole lot
different.

Very much an art at many steps along the way. There are people at the
crystal houses who have been working on these things since the days of
Transit.

On Wed, 29 May 2013 01:59:12 +0200
Magnus Danielson magnus@rubidium.dyndns.org wrote:

Interesting stuff. Thanks a lot!

Oh.. Ok. Didn't want to sound that way. On the other hand, that conclusion
is not far from the truth. I still have a lot to learn. Thanks for the
correction.

But then, the 8600 has a ~10dB higher noise then the 8607. Ie the noise
is 10 times higher (it is power-dB, not voltage-dB isn't it?). Which
makes me wonder what the noise contribution is. I would assume that the
electronics are very similar if not the same (electronics are cheap compared
to the crystal) and the mechanical construction seems to be very similar
as well.

In comparison the stability between 2 and 30s has only a factor of 2 inbetween.
(maximum instability according to spec).

Any ideas what the reason could be?

		Attila Kinali

--
The trouble with you, Shev, is you don't say anything until you've saved
up a whole truckload of damned heavy brick arguments and then you dump
them all out and never look at the bleeding body mangled beneath the heap
-- Tirin, The Dispossessed, U. Le Guin

Hi

The electronics in an SC based OCXO will be different from the "stuff" in an AT based part. At the very least you need additional traps for the SC. You also need to do something to accommodate it's significantly higher resistance. On the plus side, the SC probably can take 4X higher drive than the AT for a given amount of delta frequency / delta drive. Again, something that will impact the circuit.

Far more significant than any of that - marketing may well have asked that the part be optimized in a different way so it would sell better. Phase noise often seen as a better "bragging rights" spec than ADEV. My personal opinion is that this bias is driven more by the fact that fewer people understand (or trust) ADEV and it's derivatives.

Bob

On Jun 1, 2013, at 11:11 AM, Attila Kinali attila@kinali.ch wrote:

On Tue, 28 May 2013 20:23:06 -0700
Jim Lux jimlux@earthlink.net wrote:

Do you mean [1]?

[1] "GRAIL ­ A Microwave Ranging Instrument To Map Out The Lunar Gravity Field",
by Enzer, Wang, Klipstein, 2010

That would be [2] then? A little question here: AFAIK satelites vibrate
a lot. How do they account/compensate for the vibrations in the oscillators?

[2] "The Performance of Ultra-Stable Oscillators for the Gravity Recovery and
Interior Laboratory (GRAIL)", by Weaver, Garsecki, Reynolds
http://www.pttimeeting.org/archivemeetings/2010papers/paper28.pdf

Is any of the design documents for those crystal oscillators available?
I would be very much interested to have a look at them.

		Attila Kinali

--
The trouble with you, Shev, is you don't say anything until you've saved
up a whole truckload of damned heavy brick arguments and then you dump
them all out and never look at the bleeding body mangled beneath the heap
-- Tirin, The Dispossessed, U. Le Guin

Moin,

On Sat, 1 Jun 2013 11:21:44 -0400
Bob Camp lists@rtty.us wrote:

Oh..right. Didn't think about this (although i just read it in Vigs tutorial).
While we are at it. Is there any good resource (book, webpage, papers)
on the different crystal oscillator circuits and their advantages and
disadvantages? I couldn't find anything usefull but a few general circuits
(mostly from RF and ham radio books).

Could you explain a little bit more about ADEV vs phase noise?

		Attila Kinali

--
The trouble with you, Shev, is you don't say anything until you've saved
up a whole truckload of damned heavy brick arguments and then you dump
them all out and never look at the bleeding body mangled beneath the heap
-- Tirin, The Dispossessed, U. Le Guin

Attila,

On 06/01/2013 05:11 PM, Attila Kinali wrote:

The benefits (as I recall it) is a somewhat higher Q and less thermal
dependence.

You must be looking at the 1 Hz numbers. You really need to look at the
phasenoise at different offsets to understand what goes on.

The electronics contribute a white phase noise, as well as a flicker
noise (1/f, a -10 dB/decade slope). The resonators Q-value and frequency
will define the break-up point, below witch you have a -20 dB/decade
slope and above it is flat. Depending on the resonator at hand, you then
see a 1/f or 1/f² noise between the flat white noise and the 1/f³ noise.
The output amp can then add white phase noise and flicker phase noise.

The 10 dB improvement I would attribute to the improved Q value with SC
cut crystal, but... these are not the real values, it's the published
values which is guaranteed for a product. Actual values is different.

Looking in Enrico Rubiolas book, he measures both OSA 8600 and OSA 8607
(among others) and then the actual numbers is much closer. The 1 Hz
values is only 4 dB different, but then the wideband noise between these
two samples lets the 8600 be 2 dB quieter than the 8607 at -155 vs. -153 dB.

So, we go back to actual measurements, we need to realize that each
oscillator is unique, and that data-sheets only intends to give some
form of guarantee of how bad they will be as they exit the factory, but
not necessarily reflect all aspects of the oscillator.

I've seen AT-cut oscillators behave better than SC-cut within the same
basic conditions, but where the AT-cut was much better implemented.

So, there is a difference, it's just not very large and the full
performance depends on so many other parameters of a design. The cost of
SC-cut blanks is higher, and it is not meaningful to use one unless the
design have come so far that it starts to become a limiting factor and
other improvements is more expensive.

#1 Datasheet specing difference.

#2 Actual difference is due to Q-value difference, moving the break-up
point from the Q value.

So, the Leeson model explains this pretty well, and the difference in
Q-value of the loaded crystal blank is then reflected in noise
differences with essentially the same buffer electronics.

Cheers,
Magnus

On 06/01/2013 05:21 PM, Bob Camp wrote:

Notice that BVAs can be driven at a significantly higher power, if one
should believe the articles I referenced.

A difference could be exaggerated in the specing of them, which may be a
way to differentiate the market between the products, such that the
cheaper product does not eat out too much of the market of the more
expensive product, even if the actual difference isn't as large.

They can be a bit difficult to interpret if you have not been taught
properly on them. It is interesting that either folks look at phase
noise or ADEV, but few seems to be looking at both sides of the coin,
depending on what they are doing.

Cheers,
Magnus

Someone else can provide a long answer, but here's a short one.

ADEV is a good statistic to use if your goal is any sort of timekeeping, where performance over seconds or hours or days is important. But many applications, especially modern ones, need oscillators not so much for timekeeping but for frequency transmitting, receiving, multiplication, or sampling, where performance over nanoseconds to milliseconds and purity of waveform is important. In this case jitter and phase noise are a better measurement to use than ADEV.

The other key point is that ADEV is an accumulated error over time (by integration time, tau) statistic while phase noise is spectral density (by frequency bin, Hz) statistic. Either way, it's important to understand that you can make any measurement you want on a given oscillator; but the one that really matters is the one required for the device to operate within its spec. If buying an oscillator for an ADC or DAC clock, you are more interested in jitter and phase noise than gradual accumulated time or frequency drift. If buying a timebase for a chronometer, or building a master clock for a national timekeeping laboratory, you are more interested in long-term ADEV than high-frequency phase noise, etc.

/tvb

Hi

On Jun 1, 2013, at 12:21 PM, Magnus Danielson magnus@rubidium.dyndns.org wrote:

Actually on a direct "same sized blank / same frequency" basis, the Q of an AT is often higher than the Q of an SC. Because the SC is higher impedance, it's phase slope will be higher.

Bob

Hi

On Jun 1, 2013, at 12:04 PM, Attila Kinali attila@kinali.ch wrote:

There's not a lot of information in the public domain.

Both ADEV and phase noise are measures of how perfect (or imperfect) a signal source is. One is a frequency domain measurement (you can see it on a spectrum analyzer). The other is commonly thought of as a time domain measurement (since tau is in units of time). My observation is that customers are much more likely to specify multi point phase noise than ADEV. You are much more likely to see big font stuff in ads talking about phase noise than ADEV.

Both suffer from people talking about levels (-120 dbc or 1x10^-11) without mentioning the offset or tau. Since both are highly dependent on the offset or tau that's not a good thing. My observation is that ADEV is much more likely to be mentioned without an associated tau than phase noise without an offset . I've also observed that when the error is mentioned you are likely to get a "oops I'll fix that" on phase noise. On ADEV people often simply don't get the fact that tau matters even after it's pointed out.

Looking at what the systems using OCXO's are actually doing, about half the time ADEV is probably the better  / more important measure than phase noise. The system is more sensitive to the OCXO wandering around over 100 or 1000 seconds than it is on the level of a sideband offset how ever many Hz off carrier. Once you get past ADEV, you rarely see an OCXO specified for any of the other related specifications. That's a shame, since some of them are better measures of certain things than ADEV. Again, I blame the fact that people just don't understand / trust the measurements.

Bob

On 6/1/13 8:49 AM, Attila Kinali wrote:

Yes...

yes..

Significant vibration is only during launch. And during pyro events for
deployments, of course.  After you're in orbit, the vibration is very,
very small (bearing noise from the reaction wheels) . I doubt they're
making measurements while they use the thrusters.  If they're changing
the orientation, it's probably using wheels.  And wheel bearing noise is
probably fairly narrow band and harmonically related to the wheel speed.
I can ask some GRAIL-ers.

Not a chance <grin>
a) they are JHU/APL proprietary
b) they are export controlled

What's in Weaver's papers over the years is what you're going to see,
for the most part.

The actual resonator and oscillator and packaging hasn't changed a whole
lot from the Transit days, apparently. Particularly for the packaging,
it's very much an art and craft to get the mechanical stresses low, and
I've heard the folks who learned to do it as a young'un for Transit in
1960 at Bliley are still doing it today.  I would hope they have young
apprentices (who are probably in their 40s and 50s).

I would imagine that Oscilloquartz is pretty much the same.  The basic
physics is published and moderately well known. Producing very high
quality is mostly a matter of being very, very careful at each step of
the way, and starting with a lot, so that at the end of the process, you
have just a few good ones.  The "secret sauce" for the companies
involved is things like knowing how to set up the tests, fixturing,
which parts from which manufacturer seem to work the best and all that
stuff.  It's also the knowledge of the process yield at each step which
means you can stay in business.  APL knows how many to start at the
beginning to insure they'll have 4 at the end, 2 years later.
Overestimating yield means you wind up at delivery time without the
product in hand. Underestimating means the price gets high, and your
customers might start contemplating system designs that don't need your
product.  Science satellites, oddly enough, are remarkably price
sensitive, even though they are building one of a kind units at $1M each.

On 6/1/13 10:35 AM, Bob Camp wrote:

For ADEV, a lot of oscillators have a sort of "floor" where the ADEV is
relatively constant, say from tau in the range10-1000 seconds, and then
it rises up (from thermal effects and such), so the shorthand is that
the number quoted is that "floor value"

Certainly for "OC" applications this might be true.  Although, a sort of
trend is that the TCXO resonator has to have a lower Q, so the
temperature compensating components can "pull" it to the right frequency
over temperature, so the phase noise of a TCXO isn't as good as that of
an OCXO, which can have a higher Q.

A lot of times, though, an OCXO is chosen because a TCXO doesn't have
frequency stability needed over environmental changes. I don't think
ADEV is really the right measure when you're looking at aging or
temperature effects.

If you need 0.1 ppm accuracy over -50 to +60C, you probably aren't going
to get it with a TCXO.

For example, the Space Network using TDRSS on S-band (2.2 GHz) requires
you know the actual frequency to within 700Hz. That's 0.3 ppm and tough
to get in a TCXO over space qual temp range.