[time-nuts] Sawtooth Error Correction, WAS: Capturing NMEA and TICC timestamp data in time-correlated way?

[Bob kb8tq]

Hi

The TBolt does code phase to lock up the OCXO. Effectively its best guess error is zero once the control loop 
closes. That also means the average error is also zero. Since it’s yanking the OCXO *and* guessing at the error, 
separating those two numbers can get a bit tangled up. One of the reasons a “normal” receiver leaves the TCXO
alone is that it does not have to deal with two moving targets at once. 

Indeed there is a static offset due to cable delay and similar issues. If you have a “perfect” receiver to compare to, 
you could adjust that out with the normal commands. For full zeroing you also would need to know your antenna 
delay. 

Bob

> On Sep 15, 2019, at 10:57 AM, Dana Whitlow <k8yumdoober at gmail.com> wrote:
> 
> Assuming that the T'bird works as you stated, you might still want to know
> where in that
> sawtooth the supposedly "fixed" error falls.  Maybe this is unimportant if
> all you want is a
> stable reference frequency source, or very important if you actually want
> to know what
> time it is.  However, the always somewhat noisy phase lock means that the
> "fixed" error
> is not constant, merely less variable than otherwise.  What I don't know is
> whether or
> not the hardware can even be aware of this variability- I kind of suspect
> not.
> 
> Dana
> 
> 
> On Sun, Sep 15, 2019 at 11:03 AM Alberto di Bene <dibene at usa.net> wrote:
> 
>> On 2019-09-15 14:12, Forrest Christian (List Account) wrote:
>> 
>>> Most GPS receivers with 1PPS outputs are only able to align the 1PPS to
>> the
>>> nearest edge of some internal clock.   In addition, in most GPS
>> receivers,
>>> that internal clock is not synchronized to the 1PPS signal in any way.
>> 
>> Am I right in saying that the Trimble Thunderbolt should not require any
>> sawtooth error correction, as the processor
>> clock is the same 10 MHz crystal that is disciplined by the GPS signal ?
>> Or something is escaping me (it happened
>> before...) ?
>> 
>> 73  Alberto  I2PHD
>> 
>> 
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at lists.febo.com
>> To unsubscribe, go to
>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>> and follow the instructions there.
>> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.