Warning: GPS data incorrect

wrote:

Cheetah236 wrote:

Thought I'd better share this as others might be flying approaches
using the newest Jepp database for the Northstar M3 Approach GPS.


Jeppesen just posted a bulletin on their web site about this issue.

Scary, isn't it.

I'm kinda surprised it didn't result in an emergency AD on GPS's running
with the database in question

--
"You can support the troops but not the president"
--Representative Tom Delay (R-TX), during the Kosovo war.

Cheetah wrote:
When the GPS said I'd arrived at BERNI, I turned to
enter the hold and the examiner asked me to look outside. I was right
over the airport! BERNI is five miles Southwest of the airport!

FWIW: I just checked with WingX and it gave BERNI as 5.4nm from MMV so it
looks like the current FAA data and presumably the translation (or perhaps
some GPS software) got things wrong.

Hilton

Hilton wrote:
Cheetah wrote:

When the GPS said I'd arrived at BERNI, I turned to
enter the hold and the examiner asked me to look outside. I was right
over the airport! BERNI is five miles Southwest of the airport!


FWIW: I just checked with WingX and it gave BERNI as 5.4nm from MMV so it
looks like the current FAA data and presumably the translation (or perhaps
some GPS software) got things wrong.

I'm not familiar with the fix or the area, but from the ATA-100 data from May or so:
Fix BERNI
state: OR
latitude: 45-07-15.370N longitude: 123-12-27.600W
fix definition: UBG*D*194.00/17
fix use: REPORTING POINT

I had an interesting conversation with a man from Jeppesen at a convention
several years ago in pre-GPS days -- about the time the mid-continent LORAN
gap was closed. That was a few days before GPS took over if I recall
correctly.

We were discussing electronic approach plates and the gentleman expressed
sympathy for anyone who attempted to do it. He said that Jeppesen was
legally required to publish approach plate data exactly as they received it
from the FAA, even if they knew of an error. He said it was a lose-lose
situation for them.

Perhaps that's still the case?

Jon

Jon Woellhaf wrote:
I had an interesting conversation with a man from Jeppesen at a convention
several years ago in pre-GPS days -- about the time the mid-continent LORAN
gap was closed. That was a few days before GPS took over if I recall
correctly.

Spring of 1991.

We were discussing electronic approach plates and the gentleman expressed
sympathy for anyone who attempted to do it. He said that Jeppesen was
legally required to publish approach plate data exactly as they received it
from the FAA, even if they knew of an error. He said it was a lose-lose
situation for them.

Perhaps that's still the case?

The guy was blowing smoke. If Jeppesen knows the data are in error,
they have a duty to notify the source provider.

Jeppesen notified me Thursday that CMC had sent them a new file that
tested correct. They forgot to reset my download count to allow another
download of the Sept data but I got it Friday night and it looks
correct for all the problems I'd noted.

I'm still spooked to know that you can't count on this stuff to be
correct. I can't think of any independent way to confirm the data that
is practical and affordable. I really thought the QC at Jeppesen would
be adequate to catch and correct this sort of problem long before they
turned the data over to pilots who would place their lives and those of
their passengers under the guidance of that data.

The subtlety of the problem (the fix was correct in the intersection
portion of the database but five miles off in the approach section) is
particularly insidious.

Cheetah236 wrote:
Jeppesen notified me Thursday that CMC had sent them a new file that
tested correct. They forgot to reset my download count to allow another
download of the Sept data but I got it Friday night and it looks
correct for all the problems I'd noted.

I'm still spooked to know that you can't count on this stuff to be
correct. I can't think of any independent way to confirm the data that
is practical and affordable. I really thought the QC at Jeppesen would
be adequate to catch and correct this sort of problem long before they
turned the data over to pilots who would place their lives and those of
their passengers under the guidance of that data.

The subtlety of the problem (the fix was correct in the intersection
portion of the database but five miles off in the approach section) is
particularly insidious.

And, they are now officially FAA-certified to provide databases for the
new, very unforgiving RNP IAPs.

On 18 Sep 2005 07:09:41 -0700, "Cheetah236"
wrote:

Jeppesen notified me Thursday that CMC had sent them a new file that
tested correct. They forgot to reset my download count to allow another
download of the Sept data but I got it Friday night and it looks
correct for all the problems I'd noted.

I'm still spooked to know that you can't count on this stuff to be
correct. I can't think of any independent way to confirm the data that
is practical and affordable. I really thought the QC at Jeppesen would
be adequate to catch and correct this sort of problem long before they
turned the data over to pilots who would place their lives and those of
their passengers under the guidance of that data.

The subtlety of the problem (the fix was correct in the intersection
portion of the database but five miles off in the approach section) is
particularly insidious.

Why should they be any different than the government information used
in the charts. It's quite common to see NOTAMs correcting mistakes on
approach and enroute charts.

Roger Halstead (K8RI & ARRL life member)
(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com