[time-nuts] crystal oscillators & TPLL

[WarrenS]

Poul-Henning posted:

>While it is absolutely true that you can calculate ADEV from frequency 
>measurements,
No disagreement there,

> you have a very tough row to hoe when it comes to proving
> that your frequency samples represent the same signal as a sequence
> of zero deadtime period measurements would do.

Sorry, I though that was obvious.
Analog filters which is all the analog version of the TPLL really is,
do not have dead time problems, which causes information to be missed.
AND I am not trying to prove that the TPLL is the same as
zero deadtime period measurements.
DON'T need to, cause what everyone else is trying to accomplish and prove,
is that they are close to the same as integrated Freq with no dead time or 
aliasing, something that the TPLL already does exactly.

>The time-delay inherent in any PLL really, really, really mucks the math up 
>in that respect.
There is no time delay here or missed information,
there is a simple RC time constant (or two) when done as shown in any of the 
TPLL blocks.
Maybe the limitation you see is because you are not running the PLL fast 
enough.
The TPLL needs to have a BW high enough where it is not an issue, such as 10 
to 1000 times the Tau0

and Robert Beaward posted:
> "So now we, those that want a little more than faith, are nuts."
Well, don't give me credit for that one, I did not come up with the name,
You should check out the name of the site you just posted to.

Just hard to win some 'nuts' over, no mater how good the experimental 
results are.

>With that said, I'm really, really, really sick and tired of this TPLL 
>thread.
Yea, me too, but do note, I did change the subject name to an unique name 
with TPLL in it,
so all will be able to skip over reading about this when they have had 
enough.
The best way to stop this thread is to stop responding to it, so that there 
is nothing else to say.

ws

**************

[time-nuts] crystal oscillators & TPLL
Poul-Henning Kamp phk at phk.freebsd.dk
Tue Jun 22 15:29:38 UTC 2010
Previous message: [time-nuts] crystal oscillators & TPLL
Next message: [time-nuts] crystal oscillators & TPLL
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]



In message <8F43D37E769B4CAAB3309830E40DD0A2 at Warcon28Gz>, "WarrenS" 
writes:

>That is really all there is to it, if one starts with Frequency 
>differences,
>like the TPLL, and not Phase, like most other methods, to get the raw data
>to use with an ADEV program like "PLOTTER".

While it is absolutely true that you can calculate ADEV from frequency
measurements, you have a very tough row to hoe when it comes to proving
that your frequncy samples represent the same signal as a sequence
of zero deadtime period measurements would do.

The time-delay inherent in any PLL really, really, really mucks the
math up in that respect.

With that said, I'm really, really, really sick and tired of this
TPLL thread.

Poul-Henning

**************
Robert Benward rbenward at verizon.net
Tue Jun 22 16:36:17 UTC 2010

So now we, those that want a little more than faith, are nuts.  Some day
when you learn how to speak to people with respect, you too may get some
respect of your own.

Never argue with a fool, onlookers may not know the difference.
Never wrestle with a pig, you will only get dirty, and the pig likes it.

Bob