Hello,

it depends if you have time/frequency as a hobby or if you plan to do
some serious science/business on this field.
At the amateur/hobby level, you can buy something like TinyPFA
(https://www.tinydevices.org/wiki/pmwiki.php?n=TinyPFA.Homepage) and/or
used HP 53132A Time Interval Counter on EBay for couple of hundreds USD.

For more serious business, there are things like Microchip 53100A
(https://www.microsemi.com/product-directory/phase-noise-and-allan-deviation-testers/5565-53100a)
and/or Keysight 53230A
(https://www.keysight.com/us/en/assets/7018-02642/data-sheets/5990-6283.pdf)
but the price is couple of thousands USD.

I am amateur, so I am using TinyPFA and old but reliable Agilent 53132A.
Both instruments I am using together with TimeLab software (free sw).

Best regards
PavelK

Dne 04.09.2024 v 12:33 jelveh_t--- via time-nuts napsal(a):

Hi

What are you trying to do?

Some GPSDO’s can be very accurate.

As you go from worrying about nanoseconds to worrying about picoseconds or femtoseconds the “what you need” (and what it costs) goes up.

In what environment are you trying to do it?

A research organization with million dollar equipment budgets is very sort of thing than a basement lab looking for gear on eBay.

Do you already have a standard to compare to?

All of this work is a “compare A to B” sort of thing. This becomes at least as important as the measuring instrument. Ideally you would like a standard that is much better than the device you are measuring. This can get a bit crazy.

What is the objective of your project?

This as much as anything will drive what you need and what you should be looking for.

Lots and lots of directions this could go and heading in the wrong direction is going to get very confusing very fast.

Bob

Is it reasonable to suggest someone download lady heather or PN (via KE5FX) and play with some hardware that works with it so they know the right questions to ask?

I jumped off the deep end, starting with an E5504 phase noise system.

For my use case though, I want to build better microwave LOs and quantify the improvement. They could be 10Hz off and it wouldn’t matter. They just can’t be drifting too much to find other states ins buried in the noise?

From: Bob Camp via time-nuts time-nuts@lists.febo.com
Sent: Wednesday, September 4, 2024 2:15:58 PM
To: Discussion of precise time and frequency measurement time-nuts@lists.febo.com
Cc: Bob Camp kb8tq@n1k.org
Subject: [time-nuts] Re: Instruments for 1PPS accuracy measurement

Hi

What are you trying to do?

Some GPSDO’s can be very accurate.

As you go from worrying about nanoseconds to worrying about picoseconds or femtoseconds the “what you need” (and what it costs) goes up.

In what environment are you trying to do it?

A research organization with million dollar equipment budgets is very sort of thing than a basement lab looking for gear on eBay.

Do you already have a standard to compare to?

All of this work is a “compare A to B” sort of thing. This becomes at least as important as the measuring instrument. Ideally you would like a standard that is much better than the device you are measuring. This can get a bit crazy.

What is the objective of your project?

This as much as anything will drive what you need and what you should be looking for.

Lots and lots of directions this could go and heading in the wrong direction is going to get very confusing very fast.

Bob

time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com

Hi,

As you begin, you can start out with a very simple setup and learn a
lot. Then you will discover what will be your biggest limit and progress
from there.

As measurement setup I suggest a very basic setup to get you started:

1. A GPSDO/GNSSDO reference. There is naturally a wide selection, but
   getting started with a Thunderbolt is not a bad start. You will probably
   want to use the Lady Heather software to monitor things, so any receiver
   well adapted to that would be a recommended start.

2. A time-interval counter. There can be range of these, but the key
   thing is you measure your device under test as a time-interval measure
   compared to the reference. The better time resolution, the lower
   measurement floor. The first part of the curve will be dominated by
   measurement limits. We end up paying dearly for that, but even fairly
   simple ones can be a joy.

3. Time difference recording, some form of interface over to a computer,
   typically GPIB, and running TimeLab software.

A few counters to look at: 53132A, SR620, TAPR TICC

When measuring PPS, some counters auto-tuning is "too quick" so you need
to turn that off and set trigger points yourself for robust reading.

Comparing two PPSes to each other can be a challenge to some counters,
so time-stamp against a clock of higher frequency can be a good
solution. Letting TimeLab know which frequency on the stop channel, and
it can safely unwrap cycle slips. The PPS is driven to start channel in
this case, and that will act as timebase also.

There is a certain learning in how to tune a setup up to always give
robust readings.

Cheers,
Magnus

On 2024-09-04 12:33, jelveh_t--- via time-nuts wrote:

Ed,

Depending on what your challenge / interest is, different starting
points may be more relevant. Doing a E5504 is naturally awesome and
great starting point for your type of challenge. I know that I need to
improve my abilities in that end of the spectrum. However, the
mentioning of PPSes makes the counter and GPSDO end of things a
reasonable starting-point. I've found that learning both long-term
stability and phase-noise. Actually, there is a third aspect which is
the jitter tolerance of systematic (often modeled on sine) which adds to
this.

For ADEV and friends, I mostly use the TimePod these days, because of
the low noise-floor, but I think doing a cheap counter for first rounds
of learning is reasonable, and then once things becomes clearer,
specialization in tools can be done more in the direction needed. I have
not got a PhaseStation or E5504 yet for instance. Got a PhaseStation at
work I can borrow thought. :)

I do got some interesting other setups, but need to work on those.

Cheers,
Magnus

On 2024-09-05 00:29, Ed Marciniak via time-nuts wrote: