[TangerineSDR] TangerineSDR Ublox ZED-F9P GPS Timing Accuracy

[John Ackermann N8UR]

Hi Larry --

That's a good question, and one I'm working on.

The short answer is that the F9 receivers have a short-term jitter of
about 4 nanoseconds which is a hardware limitation (the PPS signal is
derived from a crystal/PLL of finite speed, so it can't step in smaller
amounts than that.  As a result, there is a "quantization error" that is
the time difference between when the receiver knows the time pulse
*should* be, and when it actually is.

The receiver outputs a message that tells you the quantization error of
the upcoming PPS assertion.  That correction allows you to reduce the
jitter in software by an as-yet-unknown amount.  (I'm about to do some
testing to find out.)

The next question is whether applying RTCM corrections will improve the
accuracy of the reported quantization error.  I'm also hoping to find
that out.

There is also an interesting possibility with the F9T and F9P receivers
that you can apply an external PPS signal and get a message back of the
offset between that signal and the UTC second.  But I don't yet know the
resolution or accuracy of that measurement.

In another 4 to 6 weeks, I hope to have the answer to those questions
and a bunch more like them.  The main reason for the delay is that the
available carrier boards for the F9 modules don't make available all the
signals one would want.  Just today I started designing a carrier board
that will bring out all the timing related signals, and then the fun can
begin.  I'll keep the group posted on results.

73,
John
----

On 3/25/20 2:54 PM, Larry Dodd via TangerineSDR wrote:
> For higher positioning accuracy the F9P requires a second source of RTCM
> correction data. My question:  Is the RTCM added correction data
> required for nanosecond timing precision? Thanks.
> Larry, K4LED
> Radio Jove
> 101science at gmail.com
>