[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gpsd-users] access to UBX-CFG antenna status via gposd
From: |
Miroslav Lichvar |
Subject: |
Re: [gpsd-users] access to UBX-CFG antenna status via gposd |
Date: |
Tue, 13 Sep 2016 09:14:46 +0200 |
User-agent: |
Mutt/1.6.2 (2016-07-01) |
On Mon, Sep 12, 2016 at 06:58:58PM -0700, Gary E. Miller wrote:
> I see that there is no specifi U X-CFG message, but a number of them:
>
> UBX-CFG-PRT
> UBX-CFG-USB
> UBX-CFG-MSG
> UBX-CFG-INF
> UBX-CFG-NAV5
> UBX-CFG-DAT
> UBX-CFG-RATE
> UBX-CFG-SBAS
> UBX-CFG-NMEA
> UBX-CFG-TMODE
> UBX-CFG-TP
I see UBX-CFG-TP mentioned here and I'm wondering if it would make
sense to add support for the newer UBX-CFG-TP5 message. It reports the
PPS rate, which could be useful in the PPS processing code to support
other rates than 1 and 5 Hz. If the frequency of the system clock is
allowed to be off by up to 10%, the automatic detection using measured
intervals wouldn't be reliable with PPS rates above 10 Hz.
I have an NTP server with u-blox NEO-6M configured to generate 32
pulses per second. It works very well. The offset only rarely goes
over 100 nanoseconds, most of the time it stays below 10-20
nanoseconds. The problem is that the PPS device is used as a separate
time source, so chronyd/ntpd needs to synchronize the clock to some
other source before it can switch to PPS and there is an a ugly spike
in hundreds of microseconds on every restart. I think it would be nice
if gpsd could process such PPS signal and allowed chronyd/ntpd to
synchronize to it directly.
--
Miroslav Lichvar