[time-nuts] Crystal aging

[Ulrich Bangert]

John,

> ...I was amazed to find that the difference between the 
> two is actually very easy to see. A 50 nsec. drift over 30 
> minutes. Funny thing is it slowly swings back and forth. I 
> suppose it is due to the different disciplining algorithms of 
> the two units. 

I find observations like that extreme thrilling because it may be an
indication that we can learn something about control theory by such loop
performance comparisons.

As a starting point you should consider that the TBOLT's loop time constant
has a default setting of 100 s while the Z3801's loop time constant is
??????. I use the question marks because I don't find a specification for
it. From my experience I would jugde that it must be AT LEAST ten times of
that. 

This may explain at a first glance why you notice excursions in the domain
of minutes because the TBOLT's loop CAN swing in times like that while the
Z3801's can not. 

The true thrill is however one step more subtle: It first considers the
question WHY trimble choose 100 s as the default loop time constant. Well,
this one is easy to answer: Just set the time constant of a "cold" TBOLT to
1000 and watch the TIC value flying to the moon. Expect days before the loop
locks. Trimble NEEDS to set the deafult time constant that low in order to
make the loop of a cold TBOLT lock within a reasonable time.  

Now we really are at the doorstep of the thrill: How can HP/Symmetricon
afford to work with a (seemingly) fixed loop time constant that is
significant higher than the TBOLT's?

Despite the higher time constant the Z3801 not only indicates a loop lock
after only a hour ontime or so. It also manages to keep the TIC value VERY
symmetrical around zero a short time after power up.

I am not sure about it because what information is released about the HP
SMARTCLOCK operation is very general and does not provide in depth
information but I would guess that the Z3801 uses a third (or more) order
pll while the TBOLT seems to feature a second order loop.

Anyone out there with hands on experience?

Best regards
Ulrich Bangert   

> -----Ursprungliche Nachricht-----
> Von: time-nuts-bounces at febo.com 
> [mailto:time-nuts-bounces at febo.com] Im Auftrag von John Green
> Gesendet: Mittwoch, 2. September 2009 21:31
> An: time-nuts at febo.com
> Betreff: Re: [time-nuts] Crystal aging
> 
> 
> I realize that making measurements over a day or two says 
> very little about long term aging. Most of my crystal 
> oscillator experience has been with TCXOs in the 
> communications industry. We are talking about orders of 
> magnitude difference with regard to just about everything. In 
> my experience, a good TCXO will age in a given direction if 
> set and left alone. They generally do get better with time. I 
> have seen a TCXO be within 100 Hz after 15 years of on off 
> operation and temperature cycling. For a TCXO, that's 
> outstanding.I hooked the Z3801 and Tbolt up to the same 
> antenna and after the Tbolt had completed a 4K point self 
> survey, I was amazed to find that the difference between the 
> two is actually very easy to see. A 50 nsec. drift over 30 
> minutes. Funny thing is it slowly swings back and forth. I 
> suppose it is due to the different disciplining algorithms of 
> the two units. It is easier to see on the oscope than the 
> 1992. I don't think it is due to temperature. I have a 
> thermometer a couple of feet from both units and it hasn't 
> moved more than maybe a degree. I am going to leave them this 
> way a while to see what develops. I also want to see how much 
> difference there is in the reported position between the two 
> units. _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to 
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.