Re: [gpsd-dev] Very basic PPS question:

[Gary E. Miller]

Yo Hal!

On Mon, 21 Oct 2013 14:17:03 -0700
Hal Murray <address@hidden> wrote:

> 
> > 1) devices grabs raw fix data at PPS edge
> > 2) device computes fix from raw fix data
> > 3) device emites a fix report
> 
> I think you are reading between the lines way too much when you try
> to explain how/why it works.

Yes, most is now done in DSPs and CPUs, but originally this was all done
in TTL.  I have seen the schematics to GPS in TTL and at least some
GPS 15 years ago still did it that way.

> The right answer is because the documentation says so.  The firmware
> can easily do it either way.

This is too fast and critical to do in firmware.

Check this out:

http://www.richardsonrfpd.com/resources/RellDocuments/SYS_25/GPS_Receiver.pdf

The critical part of a good GPS is keeping the VCO's in sync with the
satellite frequency.  PPS is then counted out from there.

> Some NMEA devices are off by a second.  I assume that's because their 
> software geeks didn't read the documentation and/or didn't understand
> it, or maybe just got it wrong and nobody noticed.

We see GPS off by 1 second, 2 seconds, 15 seconds, even GPS epochs.  Many
ways to do it wwrong!

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97701
        address@hidden  Tel:+1(541)382-8588

signature.asc
Description: PGP signature