Accuracy of GPS in Garmin 430/530

"Brad" wrote in message
ups.com...
In an airplane, your immediate surroundings should have no bearing on
your GPS accuracy unless you're flying under bridges, trees, through
tunnels...you get the idea. Keep in mind that the GPS constellation is
constantly moving so there are no dead areas, like you might have with
a VOR signal. I'm not sure what you mean by deteriorating
conditions...how is your meter going to predict ionospheric activity or
a satellite going off line? It's not like bars on a cell phone.

For non-FAA rated GPS devices, it's not hard to get a bad GPS signal that is
highly inaccurate. It's not always the case that you just get a signal
failure or a good signal. Some of these GPS antennas are getting extremely
good at giving you a reading in spite of a bad positioning that blocks
important parts of the sky.


Regarding standards for handheld GPS, who would enforce the standards?
What if a particular receiver did meet the standards?

Given some number of satellites, each with a given signal quality, I think
any GPS software could make statistical calculations that would result in
two numbers for number of feet of horizontal and vertical accuracy, to some
confidence interval. How do you guarantee that the author of the software
does the math correctly? I guess you can't except by cross reference to
other devices. But the same complaint could be made about any other part
of the software. How do you know he converted spacial coordinates from the
GPS to a correct map position? No doubt there could be errors in the
implementation of those algorithms, as with any other algorithm.

--
Will

"Sam Spade" wrote in message
news:l8n3g.174223$bm6.98713@fed1read04...
I got an interesting lesson in GPS recently while traveling with a
handheld
GPS as the passenger in a plane. The GPS showed us landing about two
miles
east of the airport. I figured out only later that the position of the
antenna was such that many satellites were blocked, so the accuracy of
the
GPS signal was greatly diminished.

That large of an error was probably due to the substantial altitude
change of the airliner while your GPS was staggering along in 2-D mode.

I think the reason may have been that most satellites were blocked inside
the cockpit. But why would anyone object to this non-FAA software simply
self-reporting that its current accuracy was some huge number of horizontal
and vertical feet (i.e., that it was not currently very accurate)? I
cannot understand why anyone would feel that this is a bad thing to tell a
user of that device. If you don't want the information then ignore it.


airliner. Some owners, who are savvy on this still, install an external
antenna on their aircraft for their hand-held GPS. It will never have
the problems you experienced with an external antenna.

That's a great idea. Maybe more handheld GPS users would become aware of
the need for for an external antenna if their GPS software clearly
communicated when the signals they are getting are not very accurate? That
fact might inspire more users of these devices to understand that antenna
placement is quite critical for these devices.

--
Will

Will wrote:
"John Theune" wrote in message
news:Udo3g.5009$bU6.3635@trnddc06...

I'm somewhat surprised that no one has mentioned that both the Garmin
and Lowrance units that I'm familiar with have a page for satelitte
signal strength. I've never felt the need to run thru the 430/530 pages
to find a similar page but would not be surprised to find it buried in
there somewhere. To answer the OP's question it's there you just need
to read the manual to find which sub-menu it's on. If he's using a
non-aviation unit then all bets are off but again I would think it would
be there somewhere. Also on the units I use regularly the airplane icon
flashs on the main display when the signal is lost ala a pseudo RAIM
indicator


You are right all GPS software usually implements a satellite signal page.
It's not in any way shape or form what I asked for.

I want the GPS to take all of the inputs for number of satellites and signal
strength and derive from that just two integers:

1) Number of feet/meters of horizontal accuracy, within some confidence
interval (e.g., 99.95%)

2) Number of feet/meters of vertical accuracy, within some confidence
interval (e.g., 99.95%)

Those two numbers could become optional numbers for the primary display.
No one is forcing anyone to use them. If you want to simply trust the
instrument to give you a go-nogo decision, it's your life and if you feel
that is safe it's a free world (as long as you follow FAA rules ) so be
my guest.

For my personal taste, I understand that a GPS display is always an illusion
subject to different levels of inaccuracy. I am sensitive to the
difference between a display that is showing me accuracy to 10 ft, 100 ft,
or 1000 ft. In the original posted example the GPS was off target by more
than 5000 ft. Nothing on the original display gave me any clue that this
was the case. The two numbers I am asking for would communicate quite
succinctly that no one should rely on the display for anything other than
the most gross kind of positioning.

While I would love to see the feature I am looking for in any FAA-compliant
instrument like a Garmin 530, I think the feature becomes most critical in
non-FAA compliant GPS devices/software. The authors of such packages
cannot control the quality of the satellite antenna, or mounting, and
substandard GPS reception is probably a routine thing for PDA based GPS
devices/software. So finding a succinct way to communicate the accuracy of
the current signal in numbers that mean something to any user becomes quite
important. Making people look at satellite maps and signal strength seems
like a pure engineering exercise, and it doesn't collapse the input data
into a useful form.

And right on that same page for the Lowrance unit is the EPE ( Estimated
Probability of Error ) in feet for the current location. The EPE is for
horizontal accuracy as the vertical accuracy is mostly meaningless for
GPS as they can only give you vertical guidance from a perfect sphere
and the earth does not quite fit that. vertical accuracy would require
the GPS to have a complete model of the earth elevation and coralate
that to the horizontal location so as to determine the actual vertical
location vs calculated position. Now I spent a few moments messing with
my unit and found that I can place the EPE on the main map page along
side course, heading , speed and so forth. If you want to know more I
suggest you look at the manual for your system.

But why would anyone object to this non-FAA software simply
self-reporting that its current accuracy was some huge number of horizontal
and vertical feet (i.e., that it was not currently very accurate)?

Because it's a source of more bugs.

Jose
--
The price of freedom is... well... freedom.
for Email, make the obvious change in the address.

Read the TSO and then read the description of RAIM other
accuracy monitoring software built into the unit.



--
James H. Macklin
ATP,CFI,A&P

"Will" wrote in message
...
|
| "John Theune" wrote in message
| news:Udo3g.5009$bU6.3635@trnddc06...
| I'm somewhat surprised that no one has mentioned that
both the Garmin
| and Lowrance units that I'm familiar with have a page
for satelitte
| signal strength. I've never felt the need to run thru
the 430/530 pages
| to find a similar page but would not be surprised to
find it buried in
| there somewhere. To answer the OP's question it's there
you just need
| to read the manual to find which sub-menu it's on. If
he's using a
| non-aviation unit then all bets are off but again I
would think it would
| be there somewhere. Also on the units I use regularly
the airplane icon
| flashs on the main display when the signal is lost ala a
pseudo RAIM
| indicator
|
| You are right all GPS software usually implements a
satellite signal page.
| It's not in any way shape or form what I asked for.
|
| I want the GPS to take all of the inputs for number of
satellites and signal
| strength and derive from that just two integers:
|
| 1) Number of feet/meters of horizontal accuracy, within
some confidence
| interval (e.g., 99.95%)
|
| 2) Number of feet/meters of vertical accuracy, within some
confidence
| interval (e.g., 99.95%)
|
| Those two numbers could become optional numbers for the
primary display.
| No one is forcing anyone to use them. If you want to
simply trust the
| instrument to give you a go-nogo decision, it's your life
and if you feel
| that is safe it's a free world (as long as you follow FAA
rules ) so be
| my guest.
|
| For my personal taste, I understand that a GPS display is
always an illusion
| subject to different levels of inaccuracy. I am
sensitive to the
| difference between a display that is showing me accuracy
to 10 ft, 100 ft,
| or 1000 ft. In the original posted example the GPS was
off target by more
| than 5000 ft. Nothing on the original display gave me
any clue that this
| was the case. The two numbers I am asking for would
communicate quite
| succinctly that no one should rely on the display for
anything other than
| the most gross kind of positioning.
|
| While I would love to see the feature I am looking for in
any FAA-compliant
| instrument like a Garmin 530, I think the feature becomes
most critical in
| non-FAA compliant GPS devices/software. The authors of
such packages
| cannot control the quality of the satellite antenna, or
mounting, and
| substandard GPS reception is probably a routine thing for
PDA based GPS
| devices/software. So finding a succinct way to
communicate the accuracy of
| the current signal in numbers that mean something to any
user becomes quite
| important. Making people look at satellite maps and
signal strength seems
| like a pure engineering exercise, and it doesn't collapse
the input data
| into a useful form.
|
| --
| Will
|
|
|
|

And it could double the cost of the system for no useful
purpose.


I'm sure that if anybody wanted a custom made GPS unit,
Garmin or some other company would be happy to take your
million dollars and build you one or even two of them.



--
James H. Macklin
ATP,CFI,A&P

--
The people think the Constitution protects their rights;
But government sees it as an obstacle to be overcome.
some support
http://www.usdoj.gov/olc/secondamendment2.htm
See http://www.fija.org/ more about your rights and duties.


"Jose" wrote in message
. com...
| But why would anyone object to this non-FAA software
simply
| self-reporting that its current accuracy was some huge
number of horizontal
| and vertical feet (i.e., that it was not currently very
accurate)?
|
| Because it's a source of more bugs.
|
| Jose
| --
| The price of freedom is... well... freedom.
| for Email, make the obvious change in the address.

You will never get the display you are looking for in a handheld device.
IMHO handhelds will never have RAIM.

Bob Gardner

"Will" wrote in message
...

"John Theune" wrote in message
news:Udo3g.5009$bU6.3635@trnddc06...
I'm somewhat surprised that no one has mentioned that both the Garmin
and Lowrance units that I'm familiar with have a page for satelitte
signal strength. I've never felt the need to run thru the 430/530 pages
to find a similar page but would not be surprised to find it buried in
there somewhere. To answer the OP's question it's there you just need
to read the manual to find which sub-menu it's on. If he's using a
non-aviation unit then all bets are off but again I would think it would
be there somewhere. Also on the units I use regularly the airplane icon
flashs on the main display when the signal is lost ala a pseudo RAIM
indicator

You are right all GPS software usually implements a satellite signal page.
It's not in any way shape or form what I asked for.

I want the GPS to take all of the inputs for number of satellites and
signal
strength and derive from that just two integers:

1) Number of feet/meters of horizontal accuracy, within some confidence
interval (e.g., 99.95%)

2) Number of feet/meters of vertical accuracy, within some confidence
interval (e.g., 99.95%)

Those two numbers could become optional numbers for the primary display.
No one is forcing anyone to use them. If you want to simply trust the
instrument to give you a go-nogo decision, it's your life and if you feel
that is safe it's a free world (as long as you follow FAA rules ) so
be
my guest.

For my personal taste, I understand that a GPS display is always an
illusion
subject to different levels of inaccuracy. I am sensitive to the
difference between a display that is showing me accuracy to 10 ft, 100 ft,
or 1000 ft. In the original posted example the GPS was off target by
more
than 5000 ft. Nothing on the original display gave me any clue that this
was the case. The two numbers I am asking for would communicate quite
succinctly that no one should rely on the display for anything other than
the most gross kind of positioning.

While I would love to see the feature I am looking for in any
FAA-compliant
instrument like a Garmin 530, I think the feature becomes most critical in
non-FAA compliant GPS devices/software. The authors of such packages
cannot control the quality of the satellite antenna, or mounting, and
substandard GPS reception is probably a routine thing for PDA based GPS
devices/software. So finding a succinct way to communicate the accuracy
of
the current signal in numbers that mean something to any user becomes
quite
important. Making people look at satellite maps and signal strength
seems
like a pure engineering exercise, and it doesn't collapse the input data
into a useful form.

--
Will

As the lawyers like to say in court "evidence not in the record." There
is no proof at all for the claim that some straightforward math calculations
in any GPS software is going to to double the cost of the GPS. What I was
describing doesn't require any new hardware, and it's just a matter of
calculating some accuracy numbers and representing them in the standard UI.

If it prevents one death that results in a multi-million dollar lawsuit, it
would payback the one man-month of work it might take to do those
calculations in software 100 fold or more.

--
Will


"Jim Macklin" wrote in message
news:4bx3g.8797$ZW3.1447@dukeread04...
And it could double the cost of the system for no useful
purpose.

I'm sure that if anybody wanted a custom made GPS unit,
Garmin or some other company would be happy to take your
million dollars and build you one or even two of them.

--
James H. Macklin
ATP,CFI,A&P

Will wrote:
As the lawyers like to say in court "evidence not in the record." There
is no proof at all for the claim that some straightforward math calculations
in any GPS software is going to to double the cost of the GPS. What I was
describing doesn't require any new hardware, and it's just a matter of
calculating some accuracy numbers and representing them in the standard UI.

If it prevents one death that results in a multi-million dollar lawsuit, it
would payback the one man-month of work it might take to do those
calculations in software 100 fold or more.


How would the use of the GPS be responsible for someone's death?

For VFR work, handeld or otherwise, GPS is an advisory system only, and
not guaranteed. User beware. Have alternative nav sources, like people
did for decades before GPS came on the scene. Did you read the
dislaimers on your GPS's packaging when you got it? Didn't read em? Too
bad.. Didn't buy it new and have the original packaging? Still your
problem. And being in VFR.. you shouldnt have to worry about conditions
bad enough to require an approach.

For IFR work, either you have a good signal, or you do NOT have a good
signal (as calculated by your reciever, and displayed in the form of a
RAIM integrity warning). No shades of gray here.

Based on the past few days worth of posts.. I'm guessing you haven't
done much in the way of actual IFR approaches to minimums.. and I am
also predicting you've not done any of it behind IFR approved GPS
devices, after having thoroughly read the manual and recieved training
in such... workload and workload reduction is crucial. What you propose
is to add workload, unnecesarily, to give quanititative data (percentage
points/errors) regarding something that is already addressed in a
qualitative manner (RAIM OK or NOT OK).

Now you try to justify its cost/benefit by a hypothetical lawsuit over
the lack of something that is not mandated, not needed and not justified?

Sorry... doesn't add up.
Dave

Well you get a real-time calculated number in the horizontal direction,

http://www.garmin.com/manuals/GNS430_PilotsGuide.pdf pages 44, 46
and
http://www.garmin.com/manuals/GNS530_PilotsGuide.pdf page 43

Garmin calculates the Estimated Position Error (in feet or meters
horizontally), but does not present an estimate of the vertical error. I
believe Garmin has done this on most if not all of their aviation units,
handheld and panel mount. I've got an old GPS95XL and it does the same
thing.

Regards,
John Severyn @KLVK


"Will" wrote in message
...
Is there any way to have the Garmin 430/530 put its current display
accuracy
on the primary display as an ongoing statistic, based on the number of
satellites in view? How, in general, do the Garmin units notify you of
situations where GPS accuracy has been compromised to a level that makes
it
unsafe to use the Garmin for a GPS approach?

I got an interesting lesson in GPS recently while traveling with a
handheld
GPS as the passenger in a plane. The GPS showed us landing about two
miles
east of the airport. I figured out only later that the position of the
antenna was such that many satellites were blocked, so the accuracy of the
GPS signal was greatly diminished. The particular software I was using
didn't display its current accuracy on the primary display. Based on
that
event, I realize I cannot just trust a GPS display without first
understanding the current accuracy of the signal.

What would be really nice is if the primary display would show vertical
and
horizontal accuracy as two separate numbers, based on some high confidence
interval (99.99+%). Knowing that the current display reading is accurate
to 10 ft vertical and 15 ft horizontal, for example, might make you a lot
more comfortable in following a GPS approach than a display where the
99.99%
confidence interval is 2000 ft vertical/horizontal (i.e., GPS reliability
is
completely compromised by virtue of blocked satellites, bad GPS antenna,
etc).

--
Will