Al Johnson openmoko at
Mon Jul 7 23:40:20 CEST 2008

On Monday 07 July 2008, Chris wrote:
> Hi,
> thanks for your work, but it didn't work for me. I used your spreadsheet
> to calculate the values. Is there something I can do wrong? It sets the
> time, but I still don't get a fix.
> $posx = 3784209.3;
> $posy = 901525.8;
> $posz = 5037577.6;
> Are my values for Berlin/Germany (52.516667, 13.416667, h left 0)

Looks about right. You should see the current time appearing in GPRMC and 
GPZDA messages immediately after sending the data to the GPS if it has 
accepted it. I don't know whether it makes any difference or not. I was lucky 
the first time I tested it; subsequent tests have been slower, but I don't 
think it's taken longer than 10 minutes to get a fix. I don't have 2 to test 
side by side so I can't really judge whether it's improving things or not. At 
least it shouldn't make things any worse!

I've found alphaone's ruby scripts for saving and setting the almanac and 
ephemeris data, and added a link to the wiki. I've installed ruby via opkg, 
but can't find rubygems which it seems to need. First time I've looked at 

> Strange thing is, it works great with an external antenna. I usually get
> a fix in less than 40 seconds, the freerunner even stays connected to
> those satellites when I unplug the antenna and holds the fix as long as
> it sees some sky.

Once I get a fix it keeps it very well, certainly better than my Garmin. Given 
the variation I'm guessing it needs a more reliable signal at startup than 
subsequently, and that it's borderline with the built in antenna. I guess 
sample to sample variation may mean some people will get a fix reasonably 
quickly and others may not get one at all.

> Al Johnson wrote:
> > On Monday 07 July 2008, Kai Römer wrote:
> >> Hi Al,
> >> If you need testers, please contact me. I have several gta02v5
> >> available and can do tests in Germany Munich.
> >
> > First stab uses the example perl functions from ublox for generating the
> > aid-ini data, replacing their hardcoded x,y,z with values for my
> > location. The copyright notice on the example code says you can't do
> > anything with it without permission so I can't give you the script, but I
> > can tell you how to reproduce it ;-)
> >
> > Get the AssistNow online client application note from:
> >
> >t(GPS.G4-SW-05017-C).pdf
> >
> > Create a new script and start with:
> >
> > #!/usr/bin/perl
> > print(clientdata_prepare());
> >
> > Go to section B - Sample Server implementation and append subroutines
> > clientdata_prepare and ubx_checksum to
> >
> > You need to replace the $posx, $posy and $posz values in
> > clientdata_prepare with some that match your location. These are ECEF
> > coordinated in m. There's an explanation of the calculation method in:
> >
> >
> > Alternatively you can use the attached spreadsheet if it survives the
> > list. Just replace the lat and lon with values for your location.
> >
> > You probably want to change the time accuracy to reflect the accuracy of
> > the Freerunner clock, and possibly the accuracy of your location
> > estimate.
> >
> > Now copy the script to somewhere suitable on the Freerunner and make it
> > executable. I'm using /usr/local/bin. You need to install perl if you
> > don't have it already:
> > opkg install perl
> >
> > Switch on the GPS then run the script:
> > /usr/local/bin/ > /dev/ttySAC1
> >
> > If you cat /dev/ttySAC1 you should be able to see it using the current
> > time according to your Freerunner. TangoGPS makes it easier to see what
> > it's doing. In the only test I've managed so far it got a fix with a poor
> > view of the sky, while my Garmin Geko was still struggling to see 3 sats.
> > It wasn't quick, but it was better than the Garmin. It would be
> > interesting if you could try 2 units side by side, one with aid-ini and
> > one without, to see if it really makes a difference or if I was just
> > lucky. I'll try to get some more testing in, but the weather isn't good
> > here, and I don't want to get my new toy wet ;-)
> >
> > ------------------------------------------------------------------------
> >
> > _______________________________________________
> > Openmoko community mailing list
> > community at
> >
> _______________________________________________
> Openmoko community mailing list
> community at

More information about the community mailing list