[time-nuts] Measurement resolution to be improved
kb8tq at n1k.org
Sun Aug 2 17:16:45 EDT 2015
> On Aug 2, 2015, at 3:18 AM, timeok at timeok.it wrote:
> For the measurement of Allan Deviation my configuration consists of a HP53132A and TIMELAB.
> I verified that the best resolution is using the universal counter as a frequency counter.
True, but the counter “lies" to you in this mode when you are doing ADEV.
> Using the time interval counter with 1 PPS input is definitely has a more precise measurement but the resolution collapses dramatically making useless measurements on DUT with good/excellent stability.
The readings in the frequency mode are equally useless for ADEV. The frequency estimation process is not compatible with ADEV.
> I tried to use the Dual Mixer Time Difference but the Phase Wrap is very difficult to manage during the measurements
On a reasonably stable device with a rational beat note that should not be a problem. I suspect that either:
1) Your beat note is to low relative to the frequencies you art trying to measure (to much gain)
-and / or-
2) Your beat note is to low relative to the stability of the devices you are looking at (again to much gain).
Without more details it’s tough to go much further than those two suggestions.
> and so I abandoned the project.
> I have no specific experience but it seems to me that a good solution is the one applied in the A7 Quartzlock, very expensive machine and hard to find used in hobbyist prices.
> I'd like to find a solution that would allow me to improve a hundred times or more the resolution that I currently have with my setup without sacrificing TIMELAB.
The TimePod from Symmetricom is currently the lowest cost general purpose device I know of on the market.
> The question I ask you is this: is there a solution to this problem with features want from me with costs under 1500 dollars/euros?
I’d contact Symmetricom / Microsemi for current prices.
> Message sent via Atmail Open - http://atmail.org/
> 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