Message-ID: <003501c07688$90ad4c00$0a00a8c0@ThreeLakes.ca>
From: "Larry Kayser" <kayser@sympatico.ca>
To: rsgb_lf_group@blacksheep.org
References: <65AECDF1F89AD411900400508BFC869F0D74E4@pdw-mail-1.dera.gov.uk>
Subject: LF: Re: RE: Freq stability, cheap source of
Date: Thu, 4 Jan 2001 14:18:42 -0500
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Andy et all:


> Wherever did you find them at that price ?  Mine cost nearly 100 pounds,
> and that was from the TAPR special deal on GPS25s.

Mine came from an organization that was upgrading OEM level GPS units in an
aeronautical navigation system.  After I found out about a local ARPS group
had gotten some for the price mentioned I went looking around and found
several more sources.  Digging around the APRS groups on the Internet was
important to finding these sources.  I just tried the URL of the place I got
mine from a year ago and it is no longer valid.

> Not all handhelds have a 1pps output -.......

Yes, almost NONE of the lowest cost hand helds have the 1 PPS output in my
experience.

> before getting the GPS25 I
> searched the entire PCB on the Magellan Trailblazer, scope probing just
> about every pin on the ICs  and found absolutely nothing resembling a
> stable 1 PPS clock.

Agreed.  The OEM units I have seen so far, Garmin 20 and 25 all have the 1
PPS (mine is blinking away as I write this).  The Garmin 25 has a very nice
feature in that you can set it to have Position Hold and the 1 PPS has some
what less jitter than otherwise, about half the amount I am told.

> The data bursts from its NMEA output were too
> uncertain in their timing to even think of using as standard.

Agreed.  Although if you only want 2 or 3 Milliseconds accuracy you can do
the following.

Set the unit to send out only one line of data, I choose "GPRMC"
Watch the RMC line for the 59th second to appear
Then trip into your application program on the reception of the next "$"
(dollar sign) which is the first character of the next string.  I have a
write-up that says the valid time event for that line occurs at the send of
the second character in the string, I have no idea were I saw that.

My work here of disciplining an older HP frequency standard will not permit
a continuous string of 1 PPS pulses from the receiver so I have been
gathering frequency drift rate data from the HP oscillator.  My goal is to
let the PC correct the Frequency Standard according to a known drift rate
and to only make the 1 PPS available on say a weekly or even longer basis to
bring the standard back into lock.  This way I can use only one GPS receiver
to keep several different frequency standards in high stability mode.  The
common curse of having three different transmitting locations hi.

PS the OEM level receivers are "unpacked" and exceedingly hard to work on.
It took me three tries to get an antenna connector in place that worked
properly.

Good Luck with hunting up a source of supply....

Larry
VA3LK