[time-nuts] WWVB format change in 2012
jimlux at earthlink.net
Mon Feb 29 17:03:05 EST 2016
On 2/29/16 5:31 AM, Tim Shoppa wrote:
> I have observed always less than 1 millisecond offset and skew with the
> ntpd wwv audio refclock over more than the past decade.
> Setting up the audio refclocks involves calculating and configuring
> propagation delay and delays in the receiver and audio chain. As Sanjeev
> has mentioned, setting up audio drivers for lowlatency by platform specific
> ioctl's is necessary and that may imply code portability issue.
> While propagation delay over a several thousand mile path is circa 10ms,
> the ionospheric changes are way less than a percent of that. I don't think
> it was you Hal, but someone else here was implying uncertainties in
> propagation delay on the hundreds of milliseconds, and that is off by 3
> orders of magnitude.
> (Yes I have heard and observed long path, as well as ionospheric changes in
> short path)
the entire transit time around the earth is 150 milliseconds, so you're
right "hundreds of milliseconds" must be a typo: "hundreds of
microseconds" sounds believable.
More information about the time-nuts