[time-nuts] Turning off PPS when not enough satellites

Ulrich Bangert df6jb at ulrich-bangert.de
Wed Aug 2 17:33:55 UTC 2006


Björn,

> I do like Poul-Hennings forumulation better...

The right to have a own opinion and to communicate it freely is one of
our western-world-luxuries that I am a special fan of! Feel heartly
invited to have different opinion than me!

> I know one country where the national timing laboratory uses 
> the same make/model GPS receiver as the cadastre service. 
> There are probably differences in which SW options are 
> activated -- surveying guys might have less interest in 
> external frequency input/output. Here both the guys defining 
> national time AND the national geografic reference system can 
> use the same receiver.

I did already tell you about the worth of free opinions. However, most
scientifical laws are not subject to democratic issues, they are just as
the are and one cannot have "opinions" about them. 

For that reason, whoever may have used whatsoever receiver for
whatsoever purpose: Given a set  of input informations and a set of
equations, you can "smear" the available information over 4 output
variables (3 spatial and 1 time) OR (in case of a timing receiver) you
can use ALL available information to compute a single output variable
with increased certainity, due to the fact that you now have REDUNDANT
information available. That is plain information theory. 

That is what Poul-Henning wanted to tell you and that is what you should
consider to think about again. The only thing in that Poul-Henning and I
have a different opinion is: Poul-Henning wanted to express that a
receiver can be told: "Be a good timing receiver" or alternative "Be a
good navigation receiver" and in so far that it does not depend on the
receiver but on what it is told to do. In theory, this is absolutely
correct an I agree completely to Poul-Henning! However, i was not
refering to theory but to what one can really buy at the store. And for
what you can buy the statement

>so all navigation receivers make bad timing receivers. 

is absolutely correct. ALL timing receivers can be told to be good
naviagtion receivers for a certain time, the autosurvey phase. The need
them to average over the position information long enough (Motorola f.e.
does it for 10000 s) so that it can be put as a fixed value information
into the navigation equations, resulting into a more precise time
information. NO navigation receiver will provide a precise timing
(according to current measures) when switched to "position hold mode" if
possible at all. That is what I wanted to express.

> Motorolas, now ancient, Oncore VP was used as timing 
> receiver, it could do DGPS corrections -- in a motorola 
> specific format, it could also run in a normal navigation 
> mode. Ofcause this is not at the same instant in time!

Motorola specified the VP Oncore with:

130 ns observed (1s) with SA on
In position hold mode, < 50 ns observed (1s) with SA on 

Ok, there IS a certain improvement with position hold. However, even in
position hold the VP was THAT a cheesy time receiver that
Hewlett-Packard feeled enforced to design a second oven around their
10811 OCXO, in order to allow longer averaging times in their z3801. In
comparison to what is available today (M12+ and the like) the VP is
outdated that much that no one should use it for a serious comparison.


> limitations, I do not see why a good navigation receiver, 
> could do good in timing applications.

I am almost sure that you wanted to say:  

> limitations, I do not see why a good navigation receiver, 
> could NOT do good in timing applications.

> Have you seen a DGPS-base station receiver? They are also in 
> "position hold" mode, even if its for a slightly different 
> purpose. DGPS base functionality is always (anyone with a 
> counter example? ;-) ) an expensive software option for a 
> normal "navigation" GPS receiver.

Ok, being a bit older, my knowledge about dgps base stations dates back
to the times when witches and dragons were around. At those time dgps
base stations would use a witchcraft called "land surveying office" to
know their own position up to a certainity of a few cm. Then they would
use a very normal gps receiver to listen what the gps system had to say
about their position. The error observed between what they knew about
their position and what the heard from gps was then packed into nice
data packets and transmitted online over VLF and/or VHF channels. No
need for "expensive software options" or whatsoever. Today we have that
witchcraft "internet" and lots of cheap diy dpgs base station projects
are available for free. If a dgps base station is in "position hold
mode", then only in the sense that it does not change its real position
if not hidden by a bigger bomb. But surely they won't switch their
receivers to "position hold mode" because they would not be able to
measure the gps's error in that mode.

Regards
Ulrich Bangert  

   
> -----Ursprüngliche Nachricht-----
> Von: bg at lysator.liu.se [mailto:bg at lysator.liu.se] 
> Gesendet: Dienstag, 1. August 2006 21:06
> An: Ulrich Bangert
> Cc: 'Discussion of precise time and frequency measurement'; 
> bg at lysator.liu.se
> Betreff: Re: AW: [time-nuts] Turning off PPS when not enough 
> satellites
> 
> 
> On Tue, August 1, 2006 13:16, Ulrich Bangert said:
> > Poul-Henning did already give the answer. However i am not 
> very happy 
> > with the formulation
> 
> I do like Poul-Hennings forumulation better...
> 
> Lets go back to your orginal formulation
> 
> >> >> Note that any gps receiver can not be really good for 
> timing and 
> >> >> navigation at the same time, so all navigation 
> receivers make bad 
> >> >> timing receivers.
> 
> I know one country where the national timing laboratory uses 
> the same make/model GPS receiver as the cadastre service. 
> There are probably differences in which SW options are 
> activated -- surveying guys might have less interest in 
> external frequency input/output. Here both the guys defining 
> national time AND the national geografic reference system can 
> use the same receiver.
> 
> Most navigation receivers are surely "bad" timing receivers, 
> but NOT ALL.
> 
> >> In other words, it has nothing to do with the receiver, it's about 
> >> what you ask it to do.
> >
> > because
> >
> > a) I never saw a navigation receiver featuring a "position 
> hold" mode
> 
> Have you seen a DGPS-base station receiver? They are also in 
> "position hold" mode, even if its for a slightly different 
> purpose. DGPS base functionality is always (anyone with a 
> counter example? ;-) ) an expensive software option for a 
> normal "navigation" GPS receiver.
> 
> > b) I never saw a timing receiver featuring a "navigate 
> forever" mode. 
> > Everything that i had in my hands had a "automatic site 
> survey" mode 
> > after which it would return to "position hold". That's why 
> i think the 
> > formulation that it depends on the receiver's primary 
> purpose is not 
> > that wrong!
> 
> Motorolas, now ancient, Oncore VP was used as timing 
> receiver, it could do DGPS corrections -- in a motorola 
> specific format, it could also run in a normal navigation 
> mode. Ofcause this is not at the same instant in time!
> 
> To get a good timing receiver we want a 1PPS with low jitter. 
> That is most probably not a priority for high volume 
> non-timing GPS receivers. But besides the 1PPS-hardware 
> limitations, I do not see why a good navigation receiver, 
> could do good in timing applications.
> 
> --
> 
>    Björn
> 





More information about the Time-nuts_lists.febo.com mailing list