[time-nuts] Re: Timestamping counter techniques : phase computation question

[Erik Kaashoek]

Bob,
Thanks for the feedback, always welcome.

I recognize HW is an important topic that requires a lot of attention.

With respect to the harmonic interaction. How "close" do you think the 
input and the clock need to be for the interaction to start? With a 
200MHz clock, does that leave some space for low frequencies (but not so 
low it does not matter anymore) to move the clock away from the danger zone?
Erik
On 1-2-2022 14:50, Bob kb8tq wrote:
> Hi
>
> The main point about “added hardware” is that this *is* how a counter
> gets from the numbers I mentioned to performance that is better than
> that. It’s not done by running some magic math that somehow improves
> a basic sampling counter by a couple orders of magnitude.
>
> Since that hardware is off topic, I suspect that’s as far as this needs to go.
>
> The dead band issue can come from a variety of interactions. There pretty
> much is *always* going to be some harmonic of the signal that hits some
> harmonic of the clock. The best you can do is hope that both are high enough
> that the interaction is not a big deal.
>
> Bob
>