Re: [chrony-users] Multi-PPS+RTC Setup Questions

[Matt Corallo <ntp-lists@xxxxxxxxxxxxxxx>]

On 9/29/20 12:20 PM, Miroslav Lichvar wrote:
> On Tue, Sep 29, 2020 at 12:02:40PM -0400, Matt Corallo wrote:
> > In theory the GPS should be able to provide a precision, certainly looking
> > at the NMEA stream *should*, and definitely looking at u-blox's proprietary
> > messages does, though as far as I can tell gpsd doesn't make any effort to
> > expose it in a useful way (and I'm not sure how much effort should be
> > required to do so in a vaguely cross-device manner). Even if it did, I don't
> > see a great way to feed it into chrony, eg there's no way to pass it in via
> > the socket interface (AFAICT, I was hoping to do this for the RTC to track
> > its drift over time and feed it into chrony) and chrony appears to ignore
> > the precision field in ntpshm (not that anything sets it usefully).
>
> That wouldn't help with the other errors, e.g. due to the serial port,
> interrupts, etc, which are typically much larger than the GPS
> time/position error.

Sure, I suppose in the common case it wouldn't do a lot, though in my case having the knob available over socket would 
let me be more conservative in the allowed error of the GPS outputs. Still, not worth the extra development effort for 
such a strange use. In any case may be useful to document that the field is ignored in the ntpshm refclock 
documentation, given I start down that path initially :).

> > My PPS interrupts are serial-DCD, which has some noise, but in my case
> > u-blox appears to be garbage and the two devices are off persistently by
> > 30-ish us (ie an order of magnitude or so more than noise), and somehow its
> > tied to the u-blox devices themselves (the usual swapping of all
> > ports/antennas/etc doesn't change the fact that the two devices just have
> > different concepts of the current time, across nearly every setting
> > including different tick rates so they tick at different times).
>
> 30 microseconds between different GPS receivers is a lot. Something
> has to be wrong. I think largest I have seen was with the old 18x LVC,
> that was off by about 1 microsecond, but that was still within its
> spec. Between different u-blox receivers I see at most few tens or
> maybe a hundred of nanoseconds.
>
> Have you tried factory reset? Are they the same model? Bought on ebay?
>  From different sellers? (many are selling fakes)

Yea, I've tried nearly everything at this point :). They are, indeed, different models (9-series and 8-series, one from 
a dev kit u-blox links to on their site, one may be fake, but certainly an impressive job building a reliable
implementation of the u-blox proprietary protocol if it is, and also convincing time accuracy, albeit off by a 
consistent factor). After a few days poking at them, fiddling every setting available in the UBX protocol, swapping out 
every connected bit so its only the devices themselves that are different, I'm still not sure what would possibly lead 
to such crazy inaccuracy, but its almost certainly not chrony's fault, at least.

Matt

--
To unsubscribe email chrony-users-request@xxxxxxxxxxxxxxxxxxxx 
with "unsubscribe" in the subject.
For help email chrony-users-request@xxxxxxxxxxxxxxxxxxxx 
with "help" in the subject.
Trouble?  Email listmaster@xxxxxxxxxxxxxxxxxxxx.