Why so expensive (flight recorders)

On Feb 23, 5:52*pm, Papa3 wrote:
On Feb 23, 1:08*pm, "Bill Daniels" bildan@comcast-dot-net wrote:



Simularly, fusing pressure data with GPS altitude data could work the same
way. *For example, a millisecond scale spike in GPS altitude, not confirmed
by presure data could be ignored with impunity. *A slow drift of the
pressure signal from the smoothed GPS signal is almost certainly a synoptic
pressure change and can be canceled by the long term stability of the GPS
signal making the pressure signal more useful in detecting airspace
incursions. *(Think automatic Kollsman settings.)

Bill Daniels

Which, interestingly, is exactly what the $200 Garmins with Pressure
Sensor do!

P3

So, in effect, IGC/GFAC are holding back progress on better altitude
measurement--at lower cost--by refusing to allow [certain] COTS
receivers? Come on, folks, this is the 21st century. Technology
should be welcomed, not feared!

Chip Bearden
ASW 24 "JB"
USA

Papa3 wrote:
On Feb 23, 1:08 pm, "Bill Daniels" bildan@comcast-dot-net wrote:


Simularly, fusing pressure data with GPS altitude data could work the same
way. For example, a millisecond scale spike in GPS altitude, not confirmed
by presure data could be ignored with impunity. A slow drift of the
pressure signal from the smoothed GPS signal is almost certainly a synoptic
pressure change and can be canceled by the long term stability of the GPS
signal making the pressure signal more useful in detecting airspace
incursions. (Think automatic Kollsman settings.)

Bill Daniels

Which, interestingly, is exactly what the $200 Garmins with Pressure
Sensor do!

Garmins do not have calibrated pressure sensors, in the "auto
calibration" mode, you get smoothed GPS (geometric) altitude, not
calibrated pressure altitude. Without additional information from
outside weather sources, it can't be used for detecting airspace
incursions. An altimeter setting is a temperature corrected "sea level"
pressure setting for a specific reporting station. There will still be
discrepancies between pressure and geometric altitude at anything other
than the elevation of the station, as the average temperature of the air
column between will almost never match that modeled by the International
Standard Atmosphere. This discrepancy can amount to over 1000 feet at
10000 feet above the station on a hot summer day...

Marc

On Feb 23, 7:10*pm, Marc Ramsey wrote:
Papa3 wrote:
On Feb 23, 1:08 pm, "Bill Daniels" bildan@comcast-dot-net wrote:

Simularly, fusing pressure data with GPS altitude data could work the same
way. *For example, a millisecond scale spike in GPS altitude, not confirmed
by presure data could be ignored with impunity. *A slow drift of the
pressure signal from the smoothed GPS signal is almost certainly a synoptic
pressure change and can be canceled by the long term stability of the GPS
signal making the pressure signal more useful in detecting airspace
incursions. *(Think automatic Kollsman settings.)

Bill Daniels

Which, interestingly, is exactly what the $200 Garmins with Pressure
Sensor do!

Garmins do not have calibrated pressure sensors, in the "auto
calibration" mode, you get smoothed GPS (geometric) altitude, not
calibrated pressure altitude. *Without additional information from
outside weather sources, it can't be used for detecting airspace
incursions. *An altimeter setting is a temperature corrected "sea level"
pressure setting for a specific reporting station. *There will still be
discrepancies between pressure and geometric altitude at anything other
than the elevation of the station, as the average temperature of the air
column between will almost never match that modeled by the International
Standard Atmosphere. *This discrepancy can amount to over 1000 feet at
10000 feet above the station on a hot summer day...

Marc- Hide quoted text -

- Show quoted text -

Who said anything about airspace incursions? What we are talking
about is filtering "noise" from spurious GPS signals. It presents an
interesting possibility for providing real-time (as oposed to post-
flight) noise cancelling.

I should have snipped the second part of Bill's post, as that again
gets us back to the GPS vs. Pressure altitude debate.

P3

Papa3 wrote:
On Feb 23, 7:10 pm, Marc Ramsey wrote:
Papa3 wrote:
On Feb 23, 1:08 pm, "Bill Daniels" bildan@comcast-dot-net wrote:
Simularly, fusing pressure data with GPS altitude data could work the same
way. For example, a millisecond scale spike in GPS altitude, not confirmed
by presure data could be ignored with impunity. A slow drift of the
pressure signal from the smoothed GPS signal is almost certainly a synoptic
pressure change and can be canceled by the long term stability of the GPS
signal making the pressure signal more useful in detecting airspace
incursions. (Think automatic Kollsman settings.)
Bill Daniels
Which, interestingly, is exactly what the $200 Garmins with Pressure
Sensor do!
Garmins do not have calibrated pressure sensors, in the "auto
calibration" mode, you get smoothed GPS (geometric) altitude, not
calibrated pressure altitude. Without additional information from
outside weather sources, it can't be used for detecting airspace
incursions. An altimeter setting is a temperature corrected "sea level"
pressure setting for a specific reporting station. There will still be
discrepancies between pressure and geometric altitude at anything other
than the elevation of the station, as the average temperature of the air
column between will almost never match that modeled by the International
Standard Atmosphere. This discrepancy can amount to over 1000 feet at
10000 feet above the station on a hot summer day...

Marc- Hide quoted text -

- Show quoted text -

Who said anything about airspace incursions? What we are talking
about is filtering "noise" from spurious GPS signals. It presents an
interesting possibility for providing real-time (as oposed to post-
flight) noise cancelling.

Bill mentioned airspace incursions in the quote that you included. The
described filtering technique does a decent job of smoothing GPS
altitude, but not at all helpful if what you really want is pressure
altitude.

I should have snipped the second part of Bill's post, as that again
gets us back to the GPS vs. Pressure altitude debate.

That's the thing, though, it doesn't much matter whether GPS altitude
has the noise filtered out or not, what the IGC currently wants is
pressure altitude.

Marc

First, let me echo what Bill Daniels said - the GFAC are doing great
work on igc recorders, all of us appreciate not having to use cameras
at the turnpoint. Second, like him, I loan my Volkslogger to anyone
who wants to use it when I'm not in the search for my third diamond.

My problem with COTS gps systems is more practical. The Dilution of
Precision (DOP) in the z-axis is latitude-dependant; this is due to
the orbits of the satellites (see, Electronic Navigation Systems,
Bjorie Forsell - simply, DOP is how "good" the gps 3-D fix is). For
glider pilots in northern climes, this is a problem. Having to obtain
a more-expensive system just because you live in a winter wonderland
seems unfair. Straight precision (geometric DOP) of the GPS system is
also variable (and that's if you don't fly in the US, where the Dept
of Defense is practicing gps denial many weekdays).

First, let's get some toys to play with; a good, free gps analyzer
from Trimble is their planning software, at http://www.trimble.com/planningsoftware_ts.asp.
Download it. Start it. You have to put in an almanac, from Trimble
gps data resources, further down the page (download in ssf format,
import it to the planning software using the 'import' feature). Now
you're set. Let's assume I'm doing a 5 hr today at Alert, Nunavut,
Canada, 8228N 06230W (a Canadian Forces Station, "we supply Santa").
Input today's date - Feb 24, 2008, 24 period. Input the appropriate
coordinates in 'station' pulldown.
Take a look at the pull-down chart for vertical DOP (wiki has a good
discussion of DOP that's not mathematically intense). Note that the Z-
precision of the gps signals is not useable around 0745 and, to a
lesser extent, around 1130. So, a altitude badge at this place and
time would not be possible. Having a system that predictably doesn't
work sometimes for z-precision (altitude) as our method of doing
altitude claims seems optimistic (other words also occur). Pull-down
the station to Munich, and see that at various times today, it's also
not great. Try Omarama, NZ, approx 4430S, 17000E - look at 1800
today - not useable. It would be interesting to look at olc traces
today from around there to see if there was an effect...

Now, put in YOUR lat/long and take a look at the V-DOP for a 24 hr
period at your home station; you may be lucky, and see no problems;
extend the period, and have another look. GPS geometry is pretty
tricky. As an Official Observer, I use this software to take a look
at igc files when the system has x-y problems - and most times, the x
or y-DOP sucks at the time. Note, US Coast Guard GPS status and NANUs
(Notice to Navstar Users) are also available from the Trimble gps data
sources page; it should be checked if you intend to use GPS on a
flight (good info for Contest Directors). Good one to bookmark.

Use of COTS for horizontal fixing is a great idea, though as an OO I'm
not wild about having to learn a bunch of new systems. Using gps for
altitude is a big mistake, and the GFAC is to commended for not going
that way. Those who say that COTS gps seems to be the way to go are
correct; it SEEMS to be. Take a closer look at the details
(apparently, the devil is in them) and it's not so clear anymore. The
US military does move the satellites to improve precision at times
(Gulf Wars), and the accuracy back in North America is degraded
somewhat, and Selective Availability is currently set to zero, and
could come back in certain circumstances.

I have experience in testing electronic navigation systems (incl GPS)
on a large fixed wing aircraft as an operational test and evaluation
project officer (if you have one system, you know where you are; with
three, you're never quite sure...), have a graduate degree including
work in this area, some software background, and a number of years
flying in the school of hard knocks. I've been playing with
electronic data recorders since 1997.

I hope everyone has fun with the software; I have my flame-proof suit
on; what do you think?

Dan
Yes, it's cold, and in lieu of flying, I use GPS analysis software...
sigh

Dan wrote:
My problem with COTS gps systems is more practical. The Dilution of
Precision (DOP) in the z-axis is latitude-dependant;
snippage

Thanks for the explanation. Its cleared up a few things I'd wondered about.

--
martin@ | Martin Gregorie
gregorie. | Essex, UK
org |

"Marc Ramsey" ha scritto nel messaggio
et...
pressure setting for a specific reporting station. There will still be
discrepancies between pressure and geometric altitude at anything other
than the elevation of the station, as the average temperature of the air
column between will almost never match that modeled by the International
Standard Atmosphere. This discrepancy can amount to over 1000 feet at
10000 feet above the station on a hot summer day...


Marc are you saying that these devices such as Garmin's S serie (baro
sensors, temperature sensor for GPS altitude corrections) can be 1000 to
10000 feet wrong?

Thanks for answering
Paul

I just found the patent by Garmin dated 2001 and registered in 2004 where
the altitude calculation using bot a pressure sensor and a GPS is explained.
They talk about ICAO ISA reference, but maybe Ian will be helpful to explain
what's in there.
This is the link, I found it very interesting
http://www.patentstorm.us/patents/67...scription.html

Anyone can comment on this , in respect to what the IGC requires for
altitude readings?

Paul


"PCool" ha scritto nel messaggio
...

"Marc Ramsey" ha scritto nel messaggio
et...
pressure setting for a specific reporting station. There will still be
discrepancies between pressure and geometric altitude at anything other
than the elevation of the station, as the average temperature of the air
column between will almost never match that modeled by the International
Standard Atmosphere. This discrepancy can amount to over 1000 feet at
10000 feet above the station on a hot summer day...


Marc are you saying that these devices such as Garmin's S serie (baro
sensors, temperature sensor for GPS altitude corrections) can be 1000 to
10000 feet wrong?

Thanks for answering
Paul

PCool wrote:

Marc are you saying that these devices such as Garmin's S serie (baro
sensors, temperature sensor for GPS altitude corrections) can be 1000 to
10000 feet wrong?

This was in the context of "fusing" the GPS altitude with the pressure
sensor, which results in smoothed GPS altitude, rather than pressure
altitude. If you don't manually calibrate against a known elevation or
pressure, a Garmin "S" unit will continuously auto-calibrate in
precisely this fashion. When you look at GPS altitude relative to
pressure altitude, pressure altitude will show expected errors
proportional to altitude, which can range upwards of 1000 feet during
summer months at 10000 feet AGL.

Pressure altimeters, like the altimeter sitting in your instrument panel
and the pressure sensors in flight recorders, do not measure altitude,
they measure pressure. The conversion between pressure and altitude
uses an atmosphere model called the International Standard Atmosphere
(ISA) that specifies a sea level pressure of 101.3 kPa (29.92" Hg),
temperature of 15C, and divides the atmosphere into a series of layers,
each with a specific linear temperature lapse rate. The problem is, the
atmosphere never exactly matches the model, so there is always some
amount of error in the conversion.

In particular, if the average temperature of the column of air between a
reporting station and an aircraft overhead doesn't match the model,
the pressure gradient also won't match the model, so the altitude
indicated by a correctly set altimeter will not match the actual
elevation above nominal sea level.

The easiest way to see this is in an IGC file from an approved flight
recorder. During summer months, note the relative difference between
the pressure and GPS altitude at the lower and higher altitudes. As
altitude increases, GPS altitude will increase relative to pressure
altitude, as pressure altitude is reading too low at higher altitudes.
Here in the western US, we can see this clearly near mountain peaks, as
a correctly set altimeter will be reading as much as 1000 feet too low,
whereas GPS altitude matches the known elevations of the peaks.

If a Garmin "S" unit could be set to a fixed calibration equivalent to a
pressure of 29.92 in. mercury, and the temperature compensation and
stability was adequate, it would provide readings equivalent to the
pressure sensor in an approved flight recorder. With any sort of
automatic or continuous recalibration enabled, this is no longer true.

Marc

On Feb 24, 7:07*pm, Marc Ramsey wrote:

The easiest way to see this is in an IGC file from an approved flight
recorder. *During summer months, note the relative difference between
the pressure and GPS altitude at the lower and higher altitudes. *As
altitude increases, GPS altitude will increase relative to pressure
altitude, as pressure altitude is reading too low at higher altitudes.
Here in the western US, we can see this clearly near mountain peaks, as
a correctly set altimeter will be reading as much as 1000 feet too low,
whereas GPS altitude matches the known elevations of the peaks.

OK, I'm a layman, late middle age, and little slow. What am I missing?
The pressure altitude (per the altimeter, at least) is less accurate
than the GPS altitude? By up to 1000 feet at Western USA soaring
altitudes? In the old days, we used a start gate that evaluated
optically how high we were above the ground. Assuming no one tripped
over the guy wires, that actual altitude stayed the same during a
contest. Now we're evaluated using a pressure-altitude-recording
device that may or may not reflect how high we really are?

Do the experts maintain that GPS altitude is bad because (a) it
DOESN'T have the errors inherent in pressure altitude or (b) because
its precision isn't good enough? It seems like I've seen both
positions on this forum.

Chip Bearden
ASW 24 "JB"
USA