[time-nuts] Adjusting HP 5065A frequency

Bob Camp lists at rtty.us
Mon Oct 22 12:48:37 EDT 2012


Hi

One "simple / dirty" approach, assuming you are starting from 5 or 10 MHz:

Lock up a VCXO at 100 MHz from the source via a wide band PLL. Divide it
down to 1 pps by the usual techniques. Slip the counter (add / subtract one
count) to get things within 10 ns. 

Now all you need is a +/- 5 ns adjustment. 

Inject a DC offset into the PLL with an DAC to steer it off of it's mid
point. A cheap 16 bit DAC would get you to below 200 fs.

With a wide band (say 1 KHz) loop, the ADEV of the VCXO compared to the
reference should be identical at anything past 100 ms. A well filtered DAC
voltage should not mess that up.

Yes you need to do a little calibration. If you use something like an XOR as
a phase detector, and run the DAC off the same supply as the XOR, cal should
not be too hard.

No, NIST will not be drooling at your gizmo any time soon. You will have
your PPS steered far closer to UTC than any of us is likely to be able to
measure. 

Cost wise there's not a lot there. The main thing to work out is the "pop"
when you step the counter (roll over a 10 ns boundary). The gizmo is cheap
enough that you run two of them and ping-pong when you do a roll over (keep
one running while the other settles).

The "big bucks" approach probably is to run a good RF ADC on the input and
then do all the offset stuff as DSP math. The VCXO just sits at it's magic
frequency and never moves. More money / no pops. 

Bob



-----Original Message-----
From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
Behalf Of EWKehren at aol.com
Sent: Monday, October 22, 2012 7:05 AM
To: time-nuts at febo.com
Subject: Re: [time-nuts] Adjusting HP 5065A frequency

Tom
I have two questions what should the range, resolution and stability   of 
the delay generator be and how much do you think a digital loop driven by a

Tbolt would degrade short and medium precision. What is your definition of 
short  and medium?
Bert Kehren
 
 
In a message dated 10/22/2012 12:25:08 A.M. Eastern Daylight Time,  
tvb at LeapSecond.com writes:

Three  companies come to mind for phase microsteppers. A popular one 
decades ago was  made by Austron (model 2055A). I got mine on eBay but they
are 
not as common  now as ten years ago.

The current models by Symmetricom and Spectra  Dynamics are extremely 
high-end (expensive) and overqualified for use with a  vintage rubidium 
oscillator. If you visit NIST or USNO you will see these  impressive units.

It would be a very fun project to make your own. I  suspect other group 
members could either help you or would eagerly employ your  design for their

own use.

But -- before you decide on a hardware  solution see if you can do it in 
software.

An analogy is what we do  with GPS 1PPS sawtooth errors. There are two ways 
to deal with this. One is to  capture the correction message over RS232, 
measure the DUT vs. GPS 1PPS with a  TIC, and then numerically apply the 
sawtooth correction with one line of code.  Several of the popular GPS
monitor 
programs do this automatically for you  (TBoltmon and TAC32, for example).
The 
software solution is perfect to the  granularity of the sawtooth message, 
typically 1 ns.

The hardware  implementation usually involves a PIC and a programmable 
delay generator. The  PIC listens for the correction message over RS232 and
then 
has plenty of time  (up to one second) to program the delay chip. When the 
hardware 1PPS arrives  it is delayed to compensate for the aforementioned 
sawtooth error. The result  is a hardware 1PPS that's quite close to the
ideal 
1PPS, limited again by the  granularity of the message, as well as offset 
or linearity errors in the delay  chip.

So that's the analogy. To apply this to your rubidium, ask  yourself which 
instruments or measurements or users are downstream of your  5065A 10 MHz 
output. Can they deal with daily software corrections to a stable  but 
slightly imprecise frequency, or do they really need the frequency to be  as

accurate as possible at all times.

There's a third alternative as  well. You might consider using your 5065A 
as the LO in a GPSDO. This will  sacrifice some short- and mid-term
precision 
due to additive noise, but it  will guarantee the best possible long-term 
accuracy.

/tvb

----- Original Message ----- 
From: Edgardo Molina 
To:  Tom Van Baak ; Discussion of precise time and frequency measurement 
Sent: Sunday, October 21, 2012 8:49 PM
Subject: Re: [time-nuts]  Adjusting HP 5065A frequency


Dear Tom,


Good evening. In relation to your last comments on this and other subjects, 
I  am sharing some thoughts and experience about it. I took the liberty to  
separate the topics as to ease the interested parties to follow up  
accordingly. TNX.




a. Information you kindly provided  and the index for newbies:


Thank you! You just provided me  with lots of new ideas and information on 
the subject. You have very valuable  information in your web site. As Hal 
was saying, an index should be done  anywhere so it could be easier for the 
rest of us to locate the information. I  am planning soon to build a web
page 
for my lab. In english of course for  everybody to share my experiences. I 
could work on an index to point out to  the various sources of information 
and topics that are difficult to find. That  I think could expedite things a

little bit. 


b. Phase Micro  steppers:


I saw the phase micro steppers working at CENAM  time scale. I was 
wondering that the technique could be translated to my  5065As and not
trying to 
touch them so often. If I am assuming correctly and  the technique could be 
used with the HP Rb standards. Are those phase micro  steppers easy to find?
I 
mean, affordable in the second market? If there is  one of course. I saw the

ones used at CENAM are produced by SpectraDynamics in  Colorado. According 
to Mike Lombardi it is a small highly specialized company  with a small 
market to serve. I could translate it as "expensive and  exotic"  : ) Am I 
correct?


c. Thunderbolt and my will  to share initial experiences:


I am gathering a lot of  information on the Thunderbolts as I am using them 
in my thesis work. I bought  a couple of them. If my information or novice 
experience with these receivers  is good for anybody, I would be more than 
glad to share it.


Thank you.


Kind regards,






Edgardo Molina
Dirección IPTEL


www.iptel.net.mx


T : 55 55 55202444
M : 04455  20501854


Piensa en Bits SA de  CV






Información  anexa:








CONFIDENCIALIDAD DE  INFORMACION

Este mensaje tiene carácter confidencial. Si usted  no es el destinarario 
de este mensaje, le suplicamos se lo notifique al  remitente mediante un 
correo electrónico y que borre el presente mensaje y sus  anexos de su 
computadora sin retener una copia de los mismos. Queda  estrictamente
prohibido 
copiar este mensaje o hacer usode el para cualquier  propósito o divulgar su
en 
forma parcial o total su contenido.  Gracias.




NON-DISCLOSURE OF INFORMATION

This email is strictly confidential and may also be privileged. If you are 
not  the intended recipient please immediately advise the sender by replying

to  this e-mail and then deleting the message and its attachments from your 
 computer without keeping a copy. It is strictly forbidden to copy it or 
use it  for any purpose or disclose its contents to any third party. Thank  
you.









On Oct 21, 2012, at 7:29  PM, "Tom Van Baak" <tvb at LeapSecond.com> wrote:


Hi Edgardo,

What you'll find is that many labs do not  periodically adjust the C-field 
of their 5061A or 5065A at all.

Instead, any phase or frequency adjustment is done with phase  
microsteppers or simply done in software with time and rate adjustments to
the  raw 
data. These methods avoid all possible physical side-effects of changing  
voltages, currents, and fields. It also makes it possible to gather
long-term  
data to show how the standard is operating (if you make mechanical rate  
adjustments it complicates data that you have already  collected).

The other point is that when making stability  measurements, there is no 
requirement that the reference (e.g., 5065A) be  perfectly on-frequency. So 
this removes motivation for physically touching and  possibly perturbing the

operation of the reference.

Please also take the time to read these pages.

"HP 5065A  Rubidium C-Field Resolution"
http://leapsecond.com/pages/hp-5065a-cfield/

"Rubidium  Oscillator Stability"
http://leapsecond.com/images/4rb.gif

"Stability and Noise  Performance of Various Rubidium Standards"
http://www.ke5fx.com/rb.htm

"Performance of Low-Cost  Rubidium Standards"
http://febo.com/pages/oscillators/rubes/

"A close look at  a drifting HP 5065A Rubidium Frequency Standard"
http://leapsecond.com/pages/doug-rb/

/tvb


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


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






More information about the time-nuts mailing list