[time-nuts] crystal oscillators & TPLL

Robert Benward rbenward at verizon.net
Wed Jun 23 03:13:37 UTC 2010

It is admirable that you stick up for Warren.  Part of Warren's problem is 
his disrespect for some of the other commenters of his work.  The name 
calling and put downs just does not endear him to anyone.  Saying to someone 
that they made a dumb mistake, and if they can't see the obvious then, that 
is their fault and he doesn't care, just not the way to get people on your 
side.  And as for my comments, well, you weren't the one getting the private 
emails from him.

The crux of Warren's problem lies with your comment,  "It's not easy for 
anyone to share their hard work for peer review".   All he has submitted is 
verbal claims, he has really not submitted anything for peer review.  This 
is what everyone has been asking for.  We are just supposed to take his word 
for it.  I asked him a question and all I got was obfuscation and obtuse 
analogies.   I really don't think he understands the underlying concepts, 
but if he does, then he certainly chooses incorrect words to describe them. 
In my vast experience in the engineering field, if someone can't explain 
simple concepts, then they probably don't fully understand them.  I am 
certainly not an expert in this field, but I can tell when someone is 
groping.  He makes statements that are factually misleading, sort of like 
saying the power in the coax is 70V.  The use of inappropriate units is 
direct indication of a lack of understanding.  Saying his oscillators are 
within femtoseconds of accuracy to each other is fundamentally incorrect, 
and is not the same as saying the fractional frequency error is in 
femtoseconds.  Loosely swapping units back and forth is due to that lack of 

Warren has dug himself into his own hole and he is just wallowing in his own 

In any case, I'm staying away from this thread from now on, it's going 


  ----- Original Message ----- 
  From: Steve Rooke
  To: Discussion of precise time and frequency measurement
  Sent: Tuesday, June 22, 2010 9:04 PM
  Subject: Re: [time-nuts] crystal oscillators & TPLL


  I can see that you are somewhat frustrated with all of this but let's
  please try to understand what is going on with the design and not get
  bogged down with interpersonal issues.

  On 23 June 2010 09:35, Poul-Henning Kamp <phk at phk.freebsd.dk> wrote:
  > In message <415E11EFEC7B46FFB05A790F4F4A4D4D at Warcon28Gz>, "WarrenS" 
  >>Poul-Henning posted
  >>So is there some part that is not obvious to you?
  > Yes, it is painfully obvious to me, that you are so in love with
  > your idea, that no argument will ever penetrate your defensive
  > shield.

  This fact is painfully obvious to all of us but that does not mean it
  is a bad thing, it's just that Warren is very passionate about what he
  has done and he is obviously going to be defensive against anything
  that he feels attacks his baby. It's not easy for anyone to share
  their hard work for peer review and it really depends on the reactions
  of the peers as to how the original submitter takes the feedback.

  > Please look up the proper scientific response to your
  > results in the following handy table:
  > Observation Action
  > ==========================================================
  > Results worse than expected Find out what went wrong.
  > ----------------------------------------------------------
  > Results as expected Find out what went wrong.
  > ----------------------------------------------------------
  > Results better than expected Find out what went wrong.
  > ----------------------------------------------------------

  Interesting table. Maybe you would have been more constructive in
  saying that any experimental method has limitations and errors. These
  need to be noted along with any results as they form part of the

  Let's kick this off. The R/C filter which feeds off the PLL loop and
  onto the ADC has a BW that limits measurements below 0.1s. The effect
  of drift in the reference oscillator will affect the results of the
  longer tau and test results show that the upper limitation seems to be
  in the order of 100s. Now the reference oscillator could have it's
  drift analysed over a period and so the effects of this could be
  mathematically removed from the measurements thereby improving the
  results and possibly extending the upper limitation.

  One thing that we have to bear in mind is that any drift in the DUT,
  say, if it is an xo, will result in "distortion" of any results for
  long tau anyway as ADEV is not suited to handle oscillator drift. This
  will cloud the measurement of some times of noise, like random-walk,
  and this should be born in mind with any ADEV measurement

  > Please do not reply to this email, I have no desire to
  > have further correspondence with you.

  Does that include me too?

  Steve - another 'nix nut

  > --
  > Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
  > phk at FreeBSD.ORG | TCP/IP since RFC 956
  > FreeBSD committer | BSD since 4.3-tahoe
  > Never attribute to malice what can adequately be explained by 
  > _______________________________________________
  > time-nuts mailing list -- time-nuts at febo.com
  > To unsubscribe, go to 
  > and follow the instructions there.

  Steve Rooke - ZL3TUV & G8KVD
  The only reason for time is so that everything doesn't happen at once.
  - Einstein

  time-nuts mailing list -- time-nuts at febo.com
  To unsubscribe, go to 
  and follow the instructions there. 

More information about the time-nuts mailing list