Topics

OT 10Mhz references


mosaicmerc
 

Hi all:
I'm hoping to ref. a Rigol DSA815, a Rigol DG1022A sig gen and an HP 8753D VNA....to a good 10Mhz ref.
I see a heap of them here:
http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-/_i.html?_fsub=908870013&;_sid=282770553&_trksid=p4634.c0.m322 http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-/_i.html?_fsub=908870013&;_sid=282770553&_trksid=p4634.c0.m322

IDK what voltage is optimal or which unit to chose (*sine/square, single or double OXCO) and need some advice on this.
tx


Torch
 

I bought one of these last year:

http://www.ebay.com/itm/262326560037

The advantage of this module over the bare oscillators on the page you reference is that it is ready to go, complete with dual outputs and power input, and has been calibrated -- no NIST certificate or anything, but from the photos and description, he seems to know what he is doing. From what I can tell with my limited test capability, it is accurate and very stable. If you have better equipment, there is an adjustment on the board.

However, it is a sine wave output. As a reference, a nice sharp square wave might be easier to use.

On 10/03/2016 7:39 PM, mosaicmerc@yahoo.com [TekScopes] wrote:

Hi all:
I'm hoping to ref. a Rigol DSA815, a Rigol DG1022A sig gen and an HP 8753D VNA....to a good 10Mhz ref.
I see a heap of them here:
http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-/_i.html?_fsub=908870013&;_sid=282770553&_trksid=p4634.c0.m322 http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-/_i.html?_fsub=908870013&;_sid=282770553&_trksid=p4634.c0.m322

IDK what voltage is optimal or which unit to chose (*sine/square, single or double OXCO) and need some advice on this.
tx


n4mf_sc
 

I use a Lucent RFG-RB Rubidium reference here. Wanted Cesium but a bit more pricey than I got this for on fleabay.

Mitch
N4MF


Craig Sawyers <c.sawyers@...>
 

On 10/03/2016 7:39 PM, mosaicmerc@yahoo.com [TekScopes] wrote:

Hi all:
I'm hoping to ref. a Rigol DSA815, a Rigol DG1022A sig gen and an HP
8753D VNA....to a good 10Mhz ref.
I see a heap of them here:
http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-
/_i.html?_fsub=908870013&_sid=282770553&_trksid=p4634.c0.m322
http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-
/_i.html?_fsub=908870013&_sid=282770553&_trksid=p4634.c0.m322


IDK what voltage is optimal or which unit to chose (*sine/square,
single or double OXCO) and need some advice on this.
tx
Other ways of skinning the cat is a 10MHz Rb oscillator. Main ones are Efratom and FEI. Lots from
USA and China from decommissioned comms racks. Also lots of options - sine, square, ones that have
a synthesiser so you can set the frequency over quite wide limits (but that increases the spurs
quite a lot) - lots of info on the web.

Or (what I use) is a GPS standard - a Trimble Thunderbolt from eBay. That locks to the Caesium
standards in orbit, so is bang on frequency to 1 part in 10 to the lots of zeroes, and about 20fs
noise. You need a bullet antenna too - I've got mine on a short pole attached to the end gable of
the house. Supported by Tboltmon (s/ware from Trimble, or Lady Heather (again lots on web).

But both those options are relatively expensive as compared with an ovenised clock module.

With that in mind, the Trimble 10MHz OCXO is the core of the Thunderbolt - and that is on your list.
That has ultralow phase noise, measured here http://www.ke5fx.com/tbolt.htm

But it depends on how accurate you want your 10MHz, and what phase noise/Allen deviation you want.
And how deep your pocket. And how anal you are (I'm pretty anal).


poldhu1901@...
 

With any crystal oscillator, no matter how good they might be, or how well
calibrated to start with, if that's all you have there will always be the
issue of not knowing how well it is calibrated ongoing unless you find
someone else to check it.
Although rubidium oscillators have much better long term stability this is
still an issue, and buying any of the surplus oscillators, or surplus
oscillator based products, from China is definitely taking pot luck.

Some of that kit is very good, and there are indeed bargains to be had,
but the same product from the same seller can give you a good result one week
and a duff one the next, and don't be fooled by the Chinese listings
showing lots of fancy test gear, experience shows that isn't any guarantee of
quality or competence.

It's a bit of a chicken and egg situation and you really need quite a well
equipped timing lab to start with before chancing your arm on something
that is just treated like scrap before it gets into the hands of the Ebay
sellers.

Although more expensive you would be well advised to consider a 10MHz GPS
frequency reference, the Trimble Thunderbolt for example is well thought of
with software freely available to monitor its performance, and once
installed properly there are no ongoing calibration issues.

Regards

Nigel
GM8PZR

Hi all:
I'm hoping to ref. a Rigol DSA815, a Rigol DG1022A sig gen and an HP 8753D
VNA....to a good 10Mhz ref.
I see a heap of them here:
http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-/_i.html?
_fsub=908870013&_sid=282770553&_trksid=p4634.c0.m322
http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-/_i.html?_fsub=908870013&;_sid=2
82770553&_trksid=p4634.c0.m322

IDK what voltage is optimal or which unit to chose (*sine/square, single
or double OXCO) and need some advice on this.
tx


 

The inputs on your instruments are not picky about waveshape. The
Rigols expect a signal level in the range of about 0dBm to 10dBm and
the HP VNA supports a much wider input range.

Your Rigol spectrum analyzer has a reference output which can be used
to feed another instrument. If you just want all three instruments to
be locked to the same reference, then this output could be fed to the
Rigol signal generator and the HP VNA.

On 10 Mar 2016 16:39:15 -0800, you wrote:

Hi all:
I'm hoping to ref. a Rigol DSA815, a Rigol DG1022A sig gen and an HP 8753D VNA....to a good 10Mhz ref.
I see a heap of them here:
http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-/_i.html?_fsub=908870013&;_sid=282770553&_trksid=p4634.c0.m322

IDK what voltage is optimal or which unit to chose (*sine/square, single or double OXCO) and need some advice on this.
tx


J. L. Trantham
 

I 'second' Nigel's comments.



You can get a 'stand-alone' 10 MHz 'reference' such as an HP 10811 or any of
the oscillators you and others have mentioned and could be left on
permanently. However, you will eventually need a 'primary standard'
(Cesium, GPS disciplined oscillator, etc.) for comparison to be able to
'adjust' it back on frequency as time goes by.



A GPS disciplined oscillator, such as the Thunderbolt, could to be that
'primary standard' when needed and not need to be left on permanently.



Good luck.



Joe


Torch
 

I am curious what you guys are doing that requires that sort of long-term precision?

On 11/03/2016 8:44 AM, 'J. L. Trantham' jltran@att.net [TekScopes] wrote:

You can get a 'stand-alone' 10 MHz 'reference' such as an HP 10811 or any of
the oscillators you and others have mentioned and could be left on
permanently. However, you will eventually need a 'primary standard'
(Cesium, GPS disciplined oscillator, etc.) for comparison to be able to
'adjust' it back on frequency as time goes by.


Craig Sawyers <c.sawyers@...>
 

I am curious what you guys are doing that requires that sort of long-term precision?
Because it is there, and affordable. There are those who are a whole lot worse in that regard. The
ultimate reference is a hydrogen maser - and there are guys out there who have one (have a look at
time nuts website) at many tens of thousands. Or if you want the ultimate in phase noise, a BVA
SC-cut ovenised crystal is almost unbeaten for this
http://www.oscilloquartz.com/files/1363164953-Br_%20OCXO%208607.pdf . Oscilloquartz were alas
acquired and discontinued their world beating expertise in this difficult corner of technology.

Rakon have taken this up though file:///F:/Downloads/USO%20HSO14%20Shortform-A1.pdf , but a BVA
crystal oscillator will set you back 10k somethings (dollars, euros, UKP). But although the phase
noise and Allen deviation are astoundingly good, the frequency will drift in common with all crystal
oscillators.

I'm happy though with my humble Trimble - a couple of hundred once you have the active bullet
antenna with a good view of the sky.


Dave Wise
 

An activity affectionately known as "time-nuttery".

Dave Wise

-----Original Message-----
From: TekScopes@yahoogroups.com [mailto:TekScopes@yahoogroups.com]
Sent: Friday, March 11, 2016 10:45 AM
To: TekScopes@yahoogroups.com
Subject: RE: [TekScopes] Re: OT 10Mhz references

I am curious what you guys are doing that requires that sort of long-term precision?
Because it is there, and affordable. There are those who are a whole lot worse in that regard. The
ultimate reference is a hydrogen maser - and there are guys out there who have one (have a look at
time nuts website) at many tens of thousands. Or if you want the ultimate in phase noise, a BVA
SC-cut ovenised crystal is almost unbeaten for this
http://www.oscilloquartz.com/files/1363164953-Br_%20OCXO%208607.pdf . Oscilloquartz were alas
acquired and discontinued their world beating expertise in this difficult corner of technology.

Rakon have taken this up though file:///F:/Downloads/USO%20HSO14%20Shortform-A1.pdf , but a BVA
crystal oscillator will set you back 10k somethings (dollars, euros, UKP). But although the phase
noise and Allen deviation are astoundingly good, the frequency will drift in common with all crystal
oscillators.

I'm happy though with my humble Trimble - a couple of hundred once you have the active bullet
antenna with a good view of the sky.




------------------------------------
Posted by: "Craig Sawyers" <c.sawyers@tech-enterprise.com>
------------------------------------


Brian Yee
 

Most instruments are not picky about whether the waveform is sine or
square, but make
sure it is properly terminated so the rising/falling edges are clean.

Brian


On Thu, Mar 10, 2016 at 11:55 PM, David davidwhess@gmail.com [TekScopes] <
TekScopes@yahoogroups.com> wrote:



The inputs on your instruments are not picky about waveshape. The
Rigols expect a signal level in the range of about 0dBm to 10dBm and
the HP VNA supports a much wider input range.

Your Rigol spectrum analyzer has a reference output which can be used
to feed another instrument. If you just want all three instruments to
be locked to the same reference, then this output could be fed to the
Rigol signal generator and the HP VNA.

On 10 Mar 2016 16:39:15 -0800, you wrote:

Hi all:
I'm hoping to ref. a Rigol DSA815, a Rigol DG1022A sig gen and an HP
8753D VNA....to a good 10Mhz ref.
I see a heap of them here:
http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-/_i.html?_fsub=908870013&;_sid=282770553&_trksid=p4634.c0.m322

IDK what voltage is optimal or which unit to chose (*sine/square, single
or double OXCO) and need some advice on this.
tx


Torch
 

For some reason, I can't seem to open the file on your hard drive ;-)

I suppose I can see the need for ultimate precision and stability for those of you set on proving Einstein right about the effect of gravity on time. But for us plebeians who are still living in a world populated with 100 or 200MHz scopes, it might be overkill. ;-)

On 11/03/2016 1:45 PM, 'Craig Sawyers' c.sawyers@tech-enterprise.com [TekScopes] wrote:

Rakon have taken this up though file:///F:/Downloads/USO%20HSO14%20Shortform-A1.pdf , but a BVA
crystal oscillator will set you back 10k somethings (dollars, euros, UKP).


J. L. Trantham
 

I guess it's a question of what you mean by 'precision'. If you are happy
using WWV as your 'reference' and 'zero-beating' your 'house standard'
against that, then you don't need anything else. However, if you want
something that is 'always right', a GPSDO is hard to beat.



It's also reasonably affordable.



I've had GPSDO's 'die'. Therefore, having the ability to turn it on and
have it up and accurate in 20 or 30 minutes or so is very nice. Having a
separate 'house standard' that you, from time to time, 'adjust' to bring
back 'on frequency' and is more easily replaced is, in my opinion, a good
combination.



Now, if you want to dig deeper into the performance/math of the systems, you
can spend a lifetime (or remaining lifetime, in my case) doing that. It is
quite satisfying if you can spare the time to do it. No pun intended.



Joe



From: TekScopes@yahoogroups.com [mailto:TekScopes@yahoogroups.com]
Sent: Friday, March 11, 2016 12:23 PM
To: TekScopes@yahoogroups.com
Subject: Re: [TekScopes] Re: OT 10Mhz references





I am curious what you guys are doing that requires that sort of
long-term precision?

On 11/03/2016 8:44 AM, 'J. L. Trantham' jltran@att.net [TekScopes] wrote:

You can get a 'stand-alone' 10 MHz 'reference' such as an HP 10811 or
any of
the oscillators you and others have mentioned and could be left on
permanently. However, you will eventually need a 'primary standard'
(Cesium, GPS disciplined oscillator, etc.) for comparison to be able to
'adjust' it back on frequency as time goes by.


mosaicmerc
 

After doing research on the issue, it appears that the economical approach was a Trimble 65256 OCXO ($16) as it is well respected, has low phase noise, is sine wave and is SC cut.
Morion 89A was a close second. The Epson Toyocom TCO- 6920, while interesting, I can't find any data on it and it's signal was under 0 dBm.


As a low risk approach I ordered a $16 Trimble OcXO, 12V, 10Mhz Sine, o - 7dBm output unit that didn't look banged up and the current draw isn't unreasonable.
Once all is well I'll make a low noise linear PSU and an isolated 50 ohm distribution amp and feed the various instruments.
As a useful spin off, I'll do a divide by 2 reference for my DC505A counter to replace its tank circuit. I believe it's a 5MHz ref.


Greg Muir
 

It's a matter of what kind of precision you need with regards to what you work with. I became a little frustrated with drift of some less-than-best reference oscillators in some of my test equipment when working at GHz frequencies. I now use a HP Z3816A GPSDO feeding a HP 5087A distribution amplifier whose multiple outputs feed all of the equipment on the bench via their external reference inputs. It's nice when generating or measuring a signal to know that you are basically spot on.

And the reference standard is there if needed to calibrate other oscillators. For that I use either a HP K34-59991A phase comparator, a 3575A gain-phase meter or simply an oscilloscope. The scope and K34 unit is for less demanding applications while the 3575A gives me reference/DUT phase comparisons at 10 MHz down to 1/10 of a degree accuracy.

And, no, I am not a "time nut." My wristwatch runs about 30 seconds fast each month and the old analog clock on the wall keeps me properly on time for everything in my life.

Greg


 

Hi Ancel,
I second Craig's comments about the Trimble Thunderbolt GPSDO. That is what
I use. I put mine in a TM500 single wide plugin. It's hard to beat the
average signal you come up with from 6+ cesium standards.
Dennis Tillman W7PF

-----Original Message-----
Sent: Thursday, March 10, 2016 11:32 PM
Subject: RE: [TekScopes] OT 10Mhz references

On 10/03/2016 7:39 PM, mosaicmerc@yahoo.com [TekScopes] wrote:

Hi all:
I'm hoping to ref. a Rigol DSA815, a Rigol DG1022A sig gen and an HP
8753D VNA....to a good 10Mhz ref.
I see a heap of them here:
http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-
/_i.html?_fsub=908870013&_sid=282770553&_trksid=p4634.c0.m322
http://stores.ebay.com/Flyingbests-Equipment/High-reliability-OCXO-
/_i.html?_fsub=908870013&_sid=282770553&_trksid=p4634.c0.m322


IDK what voltage is optimal or which unit to chose (*sine/square,
single or double OXCO) and need some advice on this.
tx
Other ways of skinning the cat is a 10MHz Rb oscillator. Main ones are
Efratom and FEI. Lots from USA and China from decommissioned comms racks.
Also lots of options - sine, square, ones that have a synthesiser so you can
set the frequency over quite wide limits (but that increases the spurs quite
a lot) - lots of info on the web.

Or (what I use) is a GPS standard - a Trimble Thunderbolt from eBay. That
locks to the Caesium standards in orbit, so is bang on frequency to 1 part
in 10 to the lots of zeroes, and about 20fs noise. You need a bullet
antenna too - I've got mine on a short pole attached to the end gable of the
house. Supported by Tboltmon (s/ware from Trimble, or Lady Heather (again
lots on web).

But both those options are relatively expensive as compared with an ovenised
clock module.

With that in mind, the Trimble 10MHz OCXO is the core of the Thunderbolt -
and that is on your list.
That has ultralow phase noise, measured here http://www.ke5fx.com/tbolt.htm

But it depends on how accurate you want your 10MHz, and what phase
noise/Allen deviation you want.
And how deep your pocket. And how anal you are (I'm pretty anal).
------------------------------------
Posted by: "Craig Sawyers" <c.sawyers@tech-enterprise.com>
------------------------------------


 

Hi Dwayne,

Tom Van Baak, one of the most famous of the group known as the "Time-Nuts"
has three Hydrogen Masers. He is in a league of his own. His website is at
http://www.leapsecond.com/

Somewhere on his web site is the story of when he took his kids camping for
a weekend to Paradise which is 7,000 feet up on Mt. Rainier near us in
Seattle. Of course Tom does things a little differently than the rest of us.
He brought 3 HP Cesium Standards up with him on the trip. When he got back
down he compared the time dilation that occurred between the three Cesium
Standards up on the mountain for the weekend with the ones he kept in his
house which is almost at sea level. He then calculated the expected time
difference based on General Relativity.

Tom is a little smarter than most of us. The Time-Nuts are a lot smarter
than I am.

But I know why Craig Sawyers and I have a Trimble Thunderbolt. Even if we
can't keep up with the Time-Nuts it doesn't mean we can't have a time
standard good to 12 digits. Sometimes it's just fun to watch something that
accurate.

Dennis Tillman W7PF.

-----Original Message-----
Sent: Friday, March 11, 2016 10:23 AM
Subject: Re: [TekScopes] Re: OT 10Mhz references

I am curious what you guys are doing that requires that sort of long-term
precision?
------------------------------------
Posted by: Dwayne Verhey <tekscopes@verhey.org>
-----------------------------------------

Original Message-----
Sent: Friday, March 11, 2016 10:45 AM
Subject: RE: [TekScopes] Re: OT 10Mhz references

I am curious what you guys are doing that requires that sort of long-term
precision?

Because it is there, and affordable. There are those who are a whole lot
worse in that regard. The ultimate reference is a hydrogen maser - and
there are guys out there who have one (have a look at time nuts website) at
many tens of thousands. Or if you want the ultimate in phase noise, a BVA
SC-cut ovenised crystal is almost unbeaten for this
http://www.oscilloquartz.com/files/1363164953-Br_%20OCXO%208607.pdf .
Oscilloquartz were alas acquired and discontinued their world beating
expertise in this difficult corner of technology.

Rakon have taken this up though
file:///F:/Downloads/USO%20HSO14%20Shortform-A1.pdf , but a BVA crystal
oscillator will set you back 10k somethings (dollars, euros, UKP). But
although the phase noise and Allen deviation are astoundingly good, the
frequency will drift in common with all crystal oscillators.

I'm happy though with my humble Trimble - a couple of hundred once you have
the active bullet antenna with a good view of the sky.
------------------------------------
Posted by: "Craig Sawyers" <c.sawyers@tech-enterprise.com>
------------------------------------


Torch
 

Dennis,-

Yeah, I read that story when I googled "time-nuts". Cool stuff. It got me to wondering: If (as he demonstrated) time dilates with distance from the gravity well of the earth, doesn't that make GPS based reference standards inherently less accurate than earth based ones *on earth*? I'm sure the smart folks worked out some sort of compensation for long-term effects. Perhaps the GPS birds are receiving corrections from a ground station (in which case, calibrating a ground-based oscillator to the bird would make it run faster than it should). Or perhaps the GPS oscillators are running at a slightly slower rate than earth-based ones, technically making them inaccurate at their actual orbit.

Personally, back in the real world of doctor's appointments and bus schedules, my house standard is a wrist-watch that sets itself from the radio signal from Boulder every night. But it was out by an hour this morning, since it was made before the powers that be changed the DST dates, making the automatic DST setting useless. So I had to manually correct the hour. When I set the clocks in my cars and house to DST this morning, I used the watch as a reference and now all my clocks are within 0.5 minutes of that signal, which should get me to work on time tomorrow morning. ;-)

On 12/03/2016 10:51 PM, 'Dennis Tillman' dennis@ridesoft.com [TekScopes] wrote:

he took his kids camping for
a weekend to Paradise which is 7,000 feet up on Mt. Rainier near us in
Seattle. Of course Tom does things a little differently than the rest of us.
He brought 3 HP Cesium Standards up with him on the trip. When he got back
down he compared the time dilation that occurred between the three Cesium
Standards up on the mountain for the weekend with the ones he kept in his
house which is almost at sea level. He then calculated the expected time
difference based on General Relativity.


Craig Sawyers <c.sawyers@...>
 

Yeah, I read that story when I googled "time-nuts". Cool stuff. It got me to wondering: If (as he
demonstrated) time dilates with distance from the gravity well of the earth, doesn't that make GPS
based reference standards inherently less accurate than earth based ones *on earth*?
GPS has to be corrected for both special and general relativity. Special because the satellites are
not geosynchronous, and are moving relative to a ground observer, and general because of the gravity
well. On the one hand this is deterministic, and on the other hand there needs to be constant
tweaking because tides and weather and so forth needs to be taken account of.

Without such correction, your position would drift by 10km per day and render GPS useless.

The Allen deviation of GPS time derived clocks has to be measured with respect to a ground reference
as you say, which itself has an Allen deviation less than that of GPS itself.


mosaicmerc
 

Yes, well I don't have the $$ for a GPSDO right now so I am going the DIY way wth a Trimble OCXO 65256 12V, 10MHZ sine.
It'll certainly be around 2 orders better than the internal TCXO in the SA or VNA. Improved Phase noise as well no doubt.
I don't know if I'll see a better noise floor on the instruments as a result. On the SA I can get down to around -149dBm/Hz median with a 1Mhz span and 300Hz VBW. A 3 dB 'noise' variance is visible from sweep to sweep. SO I'd imagine I could consider measuring a -140dBm/hz noise value as discernable. -130dBm should be easy to see.
Ancel