[time-nuts] KS problem
bob at evoria.net
Wed Dec 10 16:21:07 EST 2014
"Phase ns times 10" would mean you divide by 10 to get the phase in ns, right? The script I used to plot this is a general purpose script that matches the 100ps resolution of my TIC. Sorry for the confusion. I may have to make a separate script to use for this sort of thing.
As to the setup: Channel A is the 10MHz output from the KS. Channel B is the 10MHz output from the PRS45. It is the external Arm input that is being fed by the 1PPS from my homegrown. As I understand arming, that means that after the counter is armed, it will measure the phase between the next input on Channel A and the following input on Channel B. I don't think it would change anything to arm from the KS.
From: Adrian <rfnuts at arcor.de>
To: Bob Stewart <bob at evoria.net>; Discussion of precise time and frequency measurement <time-nuts at febo.com>
Sent: Wednesday, December 10, 2014 2:54 PM
Subject: Re: [time-nuts] KS problem
please describe your test setup more detailed.
There are a few points that I don't understand.
A phase difference of 6000 ns (600 ns times 10 as by the scaling of your
plot) between two 10 Mhz signals doesn't seem to make sense to me. That
would be 6 full periods of a 10 MHz signal.
And, when you have the homegrown unit connected to input 1 (start), and
the KS to input 2 (stop), how does it compare to the PRS45?
Or, do you mean something different with 'triggering'?
As by my understanding, a TI counter is triggered by the input signal at
the start input, which is usually input 1, and keeps counting time
interval until the signal at input 2 (stop input) stops it.
More information about the time-nuts