Garmin V error?

On Sunday, 16-Apr-06, I used my Garmin V while driving across Illinois.
The unit was not used for the next four days. On Friday, 21-Apr-06,
the unit was turned on when I retuned home to the Boston suburbs. I
figured it would take a bit longer to initialize. Two minutes later
when it had "found" three satellites, it delcared it had a fix and I
was 70 miles NE of Montreal (about 400 miles from my actual location).
It stayed that way for three-four minutes until it found a fourth
satellite, and then it decided to reinitialize itself (it now had zero
satellites). When the unit finally re-acquired 4 satellites, the
position was now correct.

I had a similar problem with a Garmin III+ a few years ago, but the
error was only about 4 miles.

I assume that the Garmins had in incorrect reading and an incorrect
checksum so that the reading looked genuine to the GPS.

NadCixelsyd wrote:

I've seen similar glitches on rare occasions, also under circumstances
where I was only receiving signals from three satellites. That isn't
really enough for the receiver to make an unambiguous determination of
your position and it has to assume an altitude value from the last
place where it had a good lock. My guess is that under some conditions
of the satellite geometry and possibly an incorrect altitude value
there can be some ambiguity between the true position solution and a
false solution that can be quite different. In the last case I
observed I was actually going south on a bike trail at about 15 mph a
little east of Oakland, Ca., but the GPS thought I was out over the
northern Pacific and heading for Seattle at 724 mph. As in your case
this continued for several minutes until an additional satellite was
acquired at which point the receiver recognized that the previous
solution was incorrect and established a valid one.

Hello:
I'm new to this newsgroup and have been lurking for a few days,
although I have been reading and contributing sparsely to uk.rec.gps.

On 24-Feb-06 I wrote there:

OTH, I always heard that the whole GPS system worked by "triangulation", but
in the same above document (From http://www.garmin.com/aboutGPS/manual.htm
GPSGuideforBeginners.pdf) it is explained that the position is determined as
the intersection of four spheres; that is four spheres of which the
satellites are the centers of and the radii are the distances between the
user and each of the four satellites.

It work like this: the intersection of two spheres is a circle.

The intersection with the third gives two points, which differ largely in
altitud and in position.

And the intersection with the fourth give the unique position.

Actually, last Sunday, during my trekking I observed for the first time this
error of the 3 spheres intsersection. My Garmin eTrex Legend put me at
11.000 metres of altitude and about 15 kms from my actual position.

The GPS unit allows you to enter a new approx. elevation (I was at 1.000 m)
or position for correcting this error, when you are getting only 3
satellites.

Regards,

Juan Vazquez


--
To e-mail me substitute "_"s by "e" and "c".

but
http://www.garmin.com/aboutGPS/manual.htm
as
this
m)

Three satellites should be enough for the calculation if you consider that
the fourth sphere can be the earth itself.
The error when only three satellite are received can be caused by the
satellite geometry, positioned in such a way as to exclude the possibility
of a correct calculation.
Furthermore, error can be caused by reflection of the signal due to
particular athmosferic condition, fog, snow,buildings, high tree, etc..
etc..
Definitelly, the link to three satellites is a poor condition.

Gab wrote:


Only if the user were always at sea level. In fact when finding a
position with only three satellites the GPS receiver does make the
assumption that the user's altitude has remained unchanged since the
last time there was a good position determination. So if I last used
the receiver while in an airplane at 30,000' and later turn it on but
only receive three satellite signals it'll do its best to calculate my
horizontal position under the assumption that I'm still at that
altitude. The result will be that the calculated position will be off
by several miles until the receiver gets a fourth signal and can do a
proper 3D calculation.

But the rare spurious errors like that mentioned in the original post
are of much larger magnitude and I think they also involve a
particularly poor geometry that results in an ambiguous position where
the receiver can pick the wrong one and be hundreds of miles off.