Re: [time-nuts] Agilent 53220A and TimeLab

[Timothy Schaffer]

Thanks John, that sounds like a great project. I've been wanting to get
experience designing circuits for a long time, but have never had a
specific project to work towards.  When I was a computer programmer I
discovered it was a lot easier to learn a new language if you actually had
a reason to need it.

Thanks again for referencing the pages in the TimeLab manual. I suppose it
would do be well to read it a little more thoroughly than I have previously
;)

Side note, I checked into the 3120s today at work and you're pretty much
spot on. Running about $8k from one local distributor. That may very well
be GSA pricing too, I should know by tomorrow. I agree though that concepts
and theory of operation are more well understood by actually designing and
building things... at least for me.

Cheers,
-Tim


On Wed, Feb 3, 2016 at 1:06 AM, John Miles  wrote:

> Hi, Timothy --
>
> > I've had an Agilent 53220A for some time, and recently discovered the
> > wonderful TimeLab software, but I've hit a bit of a snag trying to run
> some
> > ADEV measurements on a Rb clock.  The problem is that the graph doesn't
> > line up with the amount of data TimeLab says is collected. I ran an
> > acquisition for 100 seconds, but according to the chart it only shows
> data
> > collected for 20 seconds
>
> See page 31-32 of the manual (
> http://www.miles.io/TimePod_5330A_user_manual.pdf).  I think you'll find
> it anticipates that very same question. :)
>
> > On a side note, I was disappointed to find out that I missed the boat on
> > the semi-affordable TimePod. Is there anything remotely in the price rang
> > (sub 5K) that can do stability AND phase noise measurements? It seems
> like
> > the Wavecrest DST "might" be able to, but from what I read in the
> time-nuts
> > archive I wasn't able to get a clear picture there's an app note
> > floating around about making phase noise measurements with it, but
> > nothing
> > solid.
>
> The Microsemi 3120A's price has come down a bit over time, so I'd suggest
> checking with them to see what the current pricing is like.  (Obligatory
> disclaimer: I have no current financial/professional involvement with the
> 3120A.)  I suspect it's still well north of $5K, depending on options, but
> I know they've stepped back from the initial prices they were quoting.
>
> I'd encourage you to build something, though, given that you're not doing
> this stuff for your day job.  There's a reasonable amount of literature out
> there on both quadrature PLLs for PN measurement and the related tight-PLL
> topology that's well-suited to stability measurement.  They can both return
> very high-quality results for very little money, if you're willing to put
> in the necessary "sweat equity."  This approach is much more educational
> than simply throwing money at hardware, and (speaking from experience) more
> fun as well.
>
> -- john, KE5FX
> Miles Design LLC
>
> ___
> 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.
>
___
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.

Re: [time-nuts] Agilent 53220A and TimeLab

[timeok]

Hi Attila,

I normally use the HP53132A in Frequency mode with the Timelab. The time 
interval mode, I know is more accurate but using the 1PPS as input you lost too 
much resolution.
looking : 
http://www.timeok.it/wp/wp-content/uploads/2015/09/Timeok-Time-and-Frequency-House-Standard-v-2.1.pdf
 page 8, the noise floor using fmode is around 2.5E-12 at1 sec and 5.2E-14 at 
1000Sec.

I would like to have around 100 time more resolution capability, in particular, 
in the 1-1000Sec range and short to long term stability measurement possibility.


> How about implementing DMTD using ADCs? You could do something similar
> to what the TimePod does using a dual >40Msps ADC to sample both signals
> and feed that into a PC for processing. E.g. an AD9248-65 is available for
> 40USD, add some 100-200USD for a suitable FPGA board. A small PCB to glue
> the ADC and the FPGA board together, a little bit of software writing and
> you are done ;-)
>
> Of course, this would not have the short term stability (aka low noise
> floor)
> of the TimePod, but beyond a second, it should give you good results. And
> should be more than good enough to measure the ADEV of the devices we
> usually
> deal with.


I am not a "Digital Person" and I cannot approach this kind of project I have 
not this knowledge, so I am looking some "analog" solution or any, ready to 
use, solution.
Thank you for your help,


Luciano
www.timeok.it
Message sent via Atmail Open - http://atmail.org/)





On Wed 03/02/16 15:01 , Attila Kinali  wrote:

> Ciao Luciano,
> 
> On Wed, 03 Feb 2016 13:31:44 +0100
> tim...@timeok.it wrote:
> 
> > Unfortunately it is very difficult to measure the Allan deviation levels
> to
> > E-14 and beyond.
> > I have tried some solutions like Wavecrest and Dual mixer Time Interval
> but
> > with poor results:
> 
> Do you think this is a general problem of comparing 10MHz standards
> or just one of inadequate equipment?
> 
> > Honestly I do not see at present practical solutions and reliable low
> cost.
> 
> How about implementing DMTD using ADCs? You could do something similar
> to what the TimePod does using a dual >40Msps ADC to sample both signals
> and feed that into a PC for processing. E.g. an AD9248-65 is available for
> 40USD, add some 100-200USD for a suitable FPGA board. A small PCB to glue
> the ADC and the FPGA board together, a little bit of software writing and
> you are done ;-)
> 
> Of course, this would not have the short term stability (aka low noise
> floor)
> of the TimePod, but beyond a second, it should give you good results. And
> should be more than good enough to measure the ADEV of the devices we
> usually
> deal with.
> 
> Attila Kinali
> 
> --
> It is upon moral qualities that a society is ultimately founded. All
> the prosperity and technological sophistication in the world is of no
> use without that foundation.
> -- Miss Matheson, The Diamond Age, Neil Stephenson
> 
> 
> 
Message sent via Atmail Open - http://atmail.org/
___
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.

Re: [time-nuts] Agilent 53220A and TimeLab

[Attila Kinali]

Ciao Luciano,

On Wed, 03 Feb 2016 13:31:44 +0100
tim...@timeok.it wrote:

> Unfortunately it  is very difficult to measure the Allan deviation levels to
> E-14 and beyond.
> I have tried some solutions like Wavecrest and Dual mixer Time Interval but
> with poor results:

Do you think this is a general problem of comparing 10MHz standards
or just one of inadequate equipment?


> Honestly I do not see at present practical solutions and reliable low cost.

How about implementing DMTD using ADCs? You could do something similar
to what the TimePod does using a dual >40Msps ADC to sample both signals
and feed that into a PC for processing. E.g. an AD9248-65 is available for
40USD, add some 100-200USD for a suitable FPGA board. A small PCB to glue
the ADC and the FPGA board together, a little bit of software writing and
you are done ;-)

Of course, this would not have the short term stability (aka low noise floor)
of the TimePod, but beyond a second, it should give you good results. And
should be more than good enough to measure the ADEV of the devices we usually
deal with.

Attila Kinali

-- 
It is upon moral qualities that a society is ultimately founded. All 
the prosperity and technological sophistication in the world is of no 
use without that foundation.
 -- Miss Matheson, The Diamond Age, Neil Stephenson
___
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.

Re: [time-nuts] Agilent 53220A and TimeLab

[timeok]

Hi all,



On Wed 03/02/16 07:06 , "John Miles"  wrote:

> Hi, Timothy --
> 
> > I've had an Agilent 53220A for some time, and recently discovered the
> > wonderful TimeLab software, but I've hit a bit of a snag trying to run
> some
> > ADEV measurements on a Rb clock. The problem is that the graph doesn't
> > line up with the amount of data TimeLab says is collected. I ran an
> > acquisition for 100 seconds, but according to the chart it only shows
> data
> > collected for 20 seconds
> 
> See page 31-32 of the manual
> (http://www.miles.io/TimePod_5330A_user_manual.pdf). [1] I think you'll
> find it anticipates that very same question. :)
> 
> > On a side note, I was disappointed to find out that I missed the boat on
> > the semi-affordable TimePod. Is there anything remotely in the price
> rang
> > (sub 5K) that can do stability AND phase noise measurements? It seems
> like
> > the Wavecrest DST "might" be able to, but from what I read in the
> time-nuts
> > archive I wasn't able to get a clear picture there's an app note
> > floating around about making phase noise measurements with it, but
> > nothing
> > solid.
> 
> The Microsemi 3120A's price has come down a bit over time, so I'd suggest
> checking with them to see what the current pricing is like. (Obligatory
> disclaimer: I have no current financial/professional involvement with the
> 3120A.) I suspect it's still well north of $5K, depending on options, but I
> know they've stepped back from the initial prices they were quoting.


Unfortunately it  is very difficult to measure the Allan deviation levels to 
E-14 and beyond.
I have tried some solutions like Wavecrest and Dual mixer Time Interval but 
with poor results:

Wavecrest: 
He has a good resolution but not exceptional but its major limitation is the 
number of samples that can handle, he finished his memory and go in  times out, 
so you cannot perform long-term measures.

DMTI: 
Very critical and difficult to manage. The time interval between start and stop 
is proportional to its noise floor. As this parameter varies during the 
measurement based on the stability of the two sources, you cannot predict about 
Noise floor. Others critical conditions are, the temperature, the level of the 
input signals and the delay lines you have to use for the optimal TI is 
approximately between 50 and 300 uS. In addition, the phase wrap is still there 
because the delta-time due to the delta freq. It accumulates with each 
measurement up to the wrap (depend on the measurement time and Frequency 
difference).

Honestly I do not see at present practical solutions and reliable low cost.

Luciano
www.timeok.it

Message sent via Atmail Open - http://atmail.org/
___
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.

Re: [time-nuts] Agilent 53220A and TimeLab

[John Miles]

Hi, Timothy --
 
> I've had an Agilent 53220A for some time, and recently discovered the
> wonderful TimeLab software, but I've hit a bit of a snag trying to run some
> ADEV measurements on a Rb clock.  The problem is that the graph doesn't
> line up with the amount of data TimeLab says is collected. I ran an
> acquisition for 100 seconds, but according to the chart it only shows data
> collected for 20 seconds

See page 31-32 of the manual 
(http://www.miles.io/TimePod_5330A_user_manual.pdf).  I think you'll find it 
anticipates that very same question. :)

> On a side note, I was disappointed to find out that I missed the boat on
> the semi-affordable TimePod. Is there anything remotely in the price rang
> (sub 5K) that can do stability AND phase noise measurements? It seems like
> the Wavecrest DST "might" be able to, but from what I read in the time-nuts
> archive I wasn't able to get a clear picture there's an app note
> floating around about making phase noise measurements with it, but
> nothing
> solid.

The Microsemi 3120A's price has come down a bit over time, so I'd suggest 
checking with them to see what the current pricing is like.  (Obligatory 
disclaimer: I have no current financial/professional involvement with the 
3120A.)  I suspect it's still well north of $5K, depending on options, but I 
know they've stepped back from the initial prices they were quoting.

I'd encourage you to build something, though, given that you're not doing this 
stuff for your day job.  There's a reasonable amount of literature out there on 
both quadrature PLLs for PN measurement and the related tight-PLL topology 
that's well-suited to stability measurement.  They can both return very 
high-quality results for very little money, if you're willing to put in the 
necessary "sweat equity."  This approach is much more educational than simply 
throwing money at hardware, and (speaking from experience) more fun as well.

-- john, KE5FX
Miles Design LLC

___
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.