Order Tray | Contact Us | Home | SIG Lists

[aprssig] Looking to borrow some Garmin units

scott at opentrac.org scott at opentrac.org
Thu Dec 15 06:11:44 UTC 2005

Those of you who follow OpenTracker news probably already know that I'm
working on a follow-on to the OpenTracker, tentatively called just Tracker2
for now.  It's a receive-capable device, and has the ability to plot
received positions on a GPS receiver's map.

In addition to outputting the usual NMEA $GPWPL format, it can also speak
Garmin binary protocol.  This means that stations show up with appropriate
symbols (limited of course to the receiver's symbol set), elevation, and
comment text.

The trouble with Garmin binary protocol is that it isn't really one
protocol.  They've got about a dozen different waypoint formats, for
starters, and I can only test with what I've got on hand.  The protocol
varies not only with GPS model, but also with different software versions
for the same model.  I'm hoping to borrow assorted Garmin units for testing
- I need units with serial interfaces (not USB-only like some of the newer
models), the ability to display waypoints (not headless units), and
preferably the standard round 4-pin or flat eTrex-style connectors, unless
there's a cable to go with the unit.

If you've got a unit you can spare for a couple of weeks, let me know the
model number and software version and I'll tell you if I can use it.  I'll
be happy to send along an OpenTracker kit or a bag of Powerpoles with the
unit when I return it.

The protocol's not the only issue, either - there's lots of undocumented
stuff to test, like what characters are allowed in the comment field, what
symbols are supported, whether the unit beeps or displays a message when
transferring data, and so on.  Garmin absolutely refuses to provide any
programming support beyond the interface spec document, so it's got to be
done the hard way.

Anyway, here are a couple of screen shots from a Tracker2-connected GPSmap


The first shows a map screen in the LA area, with a number of stations
visible.  Digis are antenna towers, unsupported symbols are green dots.  The
second image is the waypoint detail screen for a station.  In this case, the
comment text had my email address.  The 'at' sign isn't supported, so it was
replaced with a '-' to keep it from all running together.  There was also an
/A=000234 altitude string in the comment, which was stripped out (to reduce
clutter and save comment space) and placed in the elevation field.

The prototype unit currently decodes just about everything except
maidenhead.  That includes standard positions, compressed positions, Mic-E,
objects, and items (which can also be compressed).  Unfortunately there
doesn't seem to be a way to delete individual waypoints, so a 'kill' command
just changes the symbol to a skull and can optionally move the waypoint to a
'graveyard' somewhere, which at least makes it go away if you're in the
proximity-sorted list.

I'm hoping to have a production version ready in the next three months.
I've still got to decide what form factor I want to do initially - the
current version fits (well, almost fits) in the same case as the OT1x.
Other possibilities are a metal box that'd be better suited to vehicle
mounting, or a version with a battery compartment that'd be more useful for
man-portable operation.  Each option has its problems, but if you have any
particular preference, or any other suggestions, let me know.  The Alinco
DR-135 internal option is still in the works, held up due to difficulty
finding anyone with the connector tooling.

There will also probably need to be at least one button on the tracker
itself, to initiate certain events.  I'd like to be able to create waypoints
with the GPS receiver and have them transmitted as objects, but it
introduces some design challenges.  For one, downloading the waypoint list
from the receiver causes it to beep loudly and display a message on the
screen - in other words, it's not something you want it doing every 30
seconds on its own.

More to come later...

Scott, N1VG

More information about the aprssig mailing list