Bush to consider shutting down GPS in extreme emergency

[Alan Browne]

Almost all the expenses are borne by the FAA. The FAA is solely
responsible for navigation aids. And the trend is to GPS. NDBs are no
longer being maintained, and are being decommissioned at a rapid rate.
LOMs are becoming rare. And precision GPS approaches are now a reality,
if the proper equipment is installed in the aircraft. The approaches are
being published. GPS approaches are far more accurate than VOR
approaches, and the horizontal accuracy is at least as good as ILS
localizers. The FAA rather desperately wants to get away from the
requirement to maintain obsolete equipment at thousands of locations all
over the US. Eventually it will.

I don't disagree. And the FAA will certainly commission non-prec. approaches.
But I don't think they will pay for the lighting required to commission a prec.
approach (CAT I) where none existed before unless the airport pays for that end
of the improvment. There may also be cases where existing lighting meets the
lighting requirement. There may also be cases where existing obstructions make
it impossible to cert. a prec. approach and that airport will have to be content
with non-prec.

Cheers,
Alan.

 

[Alan Browne]

This just isn't so. We lost an aircraft a few years back because it
couldn't maintain altitude with one engine inoperative. You need to read
CFR Title 14, Parts 23, 25, 27, and 29. Maybe it's so in the Great White
North, but not in the US.

I won't look those up, but plaase feel free to cite where a twin can be cert.
without being able to climb on a single engine at gross weight on a standard
day. As I recall the min. climb rate is 100 ft. min in that condition (which is
scary enough).
At 4000' at 30C at gross weight I can understand a piston single having
difficulty...

Cheers,
Alan

 

[Sam Wormley]

Care to give an example?

precision agriculture on a global scale
precision guided ordinance
precision mapping of utility poles, oil wells
tracking of wildlife
automobile navigation systems

more: http://www.google.com/search?q="requiring+GPS"

 

[Sam Wormley]

If your atomic clock is one second off, you should call a technician.
The main purpose of atomic clocks is to simply run for years and years
WITHOUT being off, without any resetting needed.


Juergen Nieveler

A GPS receiver with timing pulses is a hell of a lot cheap in terms of
capital expense and maintenance costs.... especially out in field where
there is no power grid.

 

[Sam Wormley]

You do that ONCE, at installation - the good thing about atomic clocks
is that they KEEP being accurate, you don't need to reset them that
often...

GPS timing receiver doesn't need much setup; doesn't need a controlled
environment. Works anywhere!

 

[Sam Wormley]

Of course, there is the theory that the magnetic poles are going to
switch sometime soon - but that's soon in geological terms, meaning
"Oh, could be tommorow, but the smart money is on some time in the next
100000 years".

Switching has already begun... check back in 5-10 thousand years. If you
are curious find out how much Earth's magnetic field strength has changed in
the last few centuries.

 

[Alan Browne]

Stan,

I don't know the details, but would be very interested. The
fundamental fact, however, is that the entire justification to
have two engines is to be able to continue flying if one engine
fails. Obviously, immediately after takeoff this means that the
plane has to be able to climb slowly.

Flying a twin while knowing that it wouldn't be able to continue
after an engine failure seems undesirable to me. I'd rather have
a single then with its lower touchdown speed and much lower risk
of engine failure, generally less than 50% of that of a
comparable twin.

The oft cited probability of engine failure being higher on a twin is a mixed
message at best. At night, single engine, when an engine fails, I used to teach
the "standard" approach and included: "Turn on the landing light. If you don't
like what you see, turn it off." At least in a twin when an engine fails, you
can find a suitable airport and put the aircraft down. The real danger with
piston twins is the failure of an engine when commited during takeoff or
immediately after takeoff. The excess power is minimal and hardly enough to
maintain Vmc, never mind climb. I've often heard twin pilots of the smaller
machines say, if the left engine quits, screw it, shut down the other, and land
on the remaining runway even if it meant an expensive stop at the other end. A
lot safer than trying to fly it at gross weight on the critical engine. More
recent aircraft (last 20-30 years) have made the props counter rotating such
that the down going blade is the one closest to the aircraft (left eng. turns
CW, right engine turns CCW) to reduce the yaw coupling on climb ... so neither
engine is 'critical'. (In a climb attitude, the downgoing blade exerts more
thrust than the upgoing blade generating a yaw. In older twins where both
engines turned CW, a failure of the left engine meant that the down going blade
on the right engine caused the aircraft to yaw strongly to the left in a climb
attitude. This resulted in higher Vmc which meant a narrower margin between
power to maintain level flight and power for climb. Wow really off topic!

Cheers,
Alan

 

[Juergen Nieveler]

Sure, just like the euro.

The one being worth more than the dollar for months now?


Juergen Nieveler

 

[Juergen Nieveler]

It doesn't work that way. Each satellite can be seen by half the world
at any given time.

But a single satellite isn't enough. As long as no more than 2 sats are
visible over the horizon at any given moment, using GPS is impossible
in that region. And since the horizon is different for any point on the
planet, and the satellites can be set to turn their transmitter on and
off at any given time, you can select which satellites are active and
visible over the horizon at a given plave and time.


Juergen Nieveler

 

[Juergen Nieveler]

That has nothing to do with any lack of GPS accuracy.

But it means that people will manage without GPS in restricted waters -
because they already do.


Juergen Nieveler

 

[Juergen Nieveler]

Wow! You can measure a point with a horizontal accuracy of less than
10mm?

Can you do that with civilian GPS? You CAN do it with a theodolite and
a couple of fixed reference points, it's simple trigonometry - which of
course doesn't mean it's easy to do, but it's what a surveyor gets paid
for.

Theodolites are surprisingly accurate, to a fraction of a degree of
angle - final accuracy mostly depends on how accurate your reference
points where. Even the simple devices used by the Romans 2000 years ago
(well in advance of GPS) where accurate enough to build a road that
runs straight (and I mean STRAIGHT) for 50 miles, only bending here and
there for a hill but returning to a straight bearing immediately
afterwards.

Juergen Nieveler

 

[Juergen Nieveler]

If they had GPS, they wouldn't be hundreds of metres off. And if you
have precise maps, you don't need surveys.

You still need surveys even if you have maps. Drawing a street on a map
doesn't acomplish much - you have to really build it. And the surveyor
tells you WHERE to build it.

Odd that surveyors are spending so much on GPS, then.

Because it's easier and faster for them. They used to have fixed
reference points, and from there they'd sight on the area they're
supposed to work on - with another guy holding a pole, and the surveyor
behind a theodolite. With (D)GPS, they can do it without a guy holding
a pole, they just walk around with a receiver and plant a pole where
the box tells them to do so.

There's more than air traffic that requires navigation.

In a war zone?

I don't expect Galileo to amount to much, especially over the short
term.

It won't ever amount to much if everybody thinks like that :-(

So? (Actually, all "U.S. persons" pay taxes, whether they are in the
U.S. or not.)

So non-US people don't really have any rights to demand GPS uptime.

Is it? Where? All I see in such contracts are disclaimers.

Really? If power companies had 24h outages every month, they'd run into
trouble really fast, believe me.

GPS will be just as reliable. The attractiveness of zero cost will
outweigh the slim chance of interference, and the accuracy will continue
to be excellent.

The accuracy of the civilian signal is quite bad actually, compared to
what is planned for Galileo or what a good surveyor can do with a good
set of reference points. Civilian GPS is good for accuracy in the meter
range - a good surveyor can reach accuracy much better than that.

I suppose the Germans will make it illegal to use GPS?

Nope. But the on-board-units for the toll system will use Galileo when
it becomes available, and future civilian navigation systems will do
the same, simply because it will be more acurate than GPS.


Juergen Nieveler

 

[Juergen Nieveler]

It's not quite that simple.

Of course.

No, but it will keep pretty good time. You have to verify the
synchronization periodically.

But you agree that this can easily be done WITHOUT GPS, and that it
actually will be a bit more acurate than GPS, albeit more expensive?


Juergen Nieveler

 

[Juergen Nieveler]

I don't think it's quite as simple as that. Don't forget that at
higher latitudes it's possible to see satellites which are 'on the
other side of the planet' so more than an earth hemisphere would be
affected. It would be much easier to turn off the lot!

Earth is a sphere, it doesn't have "sides". Essentially, you'd set the
satelites to turn off the civilian transmitter over a certain part of
their trajectory, which would translate into a certain area of the
planet not having more than 2 active satelites over the horizon at any
given time.

Juergen Nieveler

 

[Juergen Nieveler]

Was in my post already:
"value added services with integrity provision and, in some cases,
service guarantees, based on a certifiable system"

Which is exactly what I keep telling him: A written statement telling
you that you can rely on the system being up and running 24/7 without
any politician having access to an off-switch.

With GPS, there is no such statement, and thus no chance to stop a
politician from switching it off. For all we know, civilian access to
GPS could be stopped today, without any of us being able to protest
against it - it's a Pentagon-owned system, so the Pentagon decides what
to do with it, PERIOD.

Juergen Nieveler

 

[Juergen Nieveler]

Accurate timekeeping and navigation.

I'm asking for a practical non-navigation example of timekeeping that
requires nanosecond-accuracy. You keep evading that question.


Juergen Nieveler

 

[Alan Browne]

Alan Browne writes:




That's no way to run an airline.

It's exactly how to run an airline. Most new airliners are twins. Monsters
like the 777 and A330 have huge vertical stabilizers to account for an engine
failure on takeoff. Although such failures are increasingly rare, pilots must
train and refresh for the eventuality. This is not done with passengers on
board, and is mainly done in a simulator for cost reasons.

Beyond a certain point, you stop training for certain situations,
because it is no longer practical to do so. Nobody trains for normal
commercial service with one engine not running, for example.

Of course they do. Engines can fail at critical points in the flight such as
immediately prior to or just after takeoff. Airspeed is low and yaw control is
poor. The aircraft must be controlled precisely and applicable procedures for
the aircraft performed (fuel cutoff, fire check, extinguish, checklists, etc.).

Engines can also be volluntarilly shut down in flight if the engine condition
warrants. (temps, vibration, other problems).
This has its own checklists as well as logistical consequences. The training
for this is as much the immediate as well as planning and strategizing what to
do next. All of this requires training and refresher courses.

They do it for emergencies, not for normal operation.

That's what the word "emergency" means above. The life of an airline pilot is
hopefully a dull one. They are mainly system managers. They have to manage
emergencies as well, and they have to do so quickly and with minimal error.

Cheers,
Alan

 

[Alan Browne]

We were talking about fixed installations such as computer centers
using GPS to keep the time in this particular sub-thread. Hard as I
try, I cannot imagine why you'd need constant updates on position or
velocity for a building ;-)

Don't be obtuse. You said, "because GPS is designed to calculate positions, not
to tell the time." And that is erroneous. The first concern of a GPS receiver
is time.

Cheers,
Alan

 

[Alan Browne]

Juergen Nieveler writes:




Accurate timekeeping and navigation.

Many microsecond and better clocks have been available on the commercial market
for 30 years or more.
INS does quite well without more than a simple oscilator as a timepiece.

Most widebody airliners built before 1995 do not have GPS at all and the cost of
retrofitting and certification is not deemed worthwhile for the remaining life
and service of the aircraft. The dual/triple INS + radio navaids are more than
enough. Widebodies will always have dual/triple 1 NM/hr class INS onboard. The
difference now is that they use GPS to contain their errors (the INS' actually
process PR and PR-rates directly to contain various errors).

The advantage that GPS has over other PVT systems is simply high accuracy and
very low user captial, integration and operational cost.

Cheers,
Alan

 

[Alan Browne]

Alan Browne writes:




That cannot substitute for applications that require GPS accuracy. A
less accurate signal is equivalent to no signal at all in some cases.

Name the cases.

But the clocks have to be synchronized with other clocks. An atomic
clock isn't much good if it's one second off.

There are various active methods to synchronize clocks at the far end of
networks. There are also 'passive' means. In the old Omega nav system, a fellow
would travel worldwide with an referecne atomic clock and visit each station and
align the Omega station clocks with that reference (relativistic effects being
much smaller than the required error).

In the event that GPS is lost, a very fast network can still be synchronized but
the bandwidth may be lower to account for less accuracy in the time reference.
a GPS synched 5 Gb/s net might turn into a 1 Gb/s net, for example. The same
network with atomic clocks might run at 4.5 Gb/s. This really depends on how
the network is designed, how fault tolerant it is and how smart the designer and
operator is.

Cheers,
Alan