PDA

View Full Version : GPS Error Situations


twistedenginestarter
21st Aug 2007, 14:17
I recently acquired a TomTom satnav and have been very impressed with the speed of satellite acquisition and position calculation. However a few weeks ago it got lost (by serveral hundred metres for several minutes). It's never done this before, but this was three times in the same place over a period of about a week. It was on the A1 and I've got a feeling it was near Ferrybridge. The only thing I could see was a high density of power lines. I suppose if they were the cause, and it was noise, then the satnav would just get lost full stop. Instead it was moving in the right direction, just with this big and fluctuating error. A factor might have been the A1 was taking a new course - that's a possibility - and it was trying to place itself on something in its memory. Anyway I just wondered if anyone else had seen situations were GPS can give erroneous positions without reporting signal loss, obviously with the perspective of how this might affect reliability in flight.

Dan Winterland
21st Aug 2007, 14:57
Ephemeris error is a possibility - it would give you such an error. One or more of the satelites was out of position for a while.

TopBunk
21st Aug 2007, 16:20
The A1 at Ferrybridge was moved further west? a few years ago. I have TT5 and that does not have the new road in it and have experienced exactly that same fault.

twistedenginestarter
21st Aug 2007, 16:46
So it's probably a TT software behaviour rather than a GPS problem. The good thing about aeroplanes is they tell you how they work. So many things we use nowadays - the Tom Tom is one, my car is another - don't have any descriptions of how things work, and you can't find them on the Internet either.

Agaricus bisporus
21st Aug 2007, 17:41
One or more of the satelites (sic) was out of position for a while.

Pretty improbable. How is a big satellite like that (they weigh a ton or so, don't they?) going to get "out of position?". That would require a deliberate and prodigious expenditure of propellant and then same again to replace it in correct position. It can't get out of position unless something gives it a big shove...

Suggest you don't fly under power line in the vicinity of Ferrybridge without a decent dr plot on a topo chart.

Who knows what errors can be created by the magnetic fields under massed power lines in a sensitive unit like a GPS?

411A
22nd Aug 2007, 03:38
One or more of the satelites was out of position for a while.
US DoD satelites do not go 'out of position'.

Turn off unexpectedly perhaps, but not go on a walk about.

Now, down to brass tacks.
I have personally used IFR approved (enroute/terminal/approach) GPS units for the last ten years, both in my private aeroplane and in heavy transport jets.
Only three times have suitable signals been interruped, and then only for a very brief time.
Once over Mt Wilson, Californa, where television and FM transmitters/antennas are located.
A total of twenty five.
Two times during intensional GPS specific area signal interruptions, as published by NOTAM.

GPS...made in America, and it works good:D

Dan Winterland
22nd Aug 2007, 10:29
All satelites drift - in the case of GPS satellites keeping them in on the correct orbit is a complex and costly exercise. They are allowed to drift a little, as moving them back onto the correct track constantly will use their fuel up too quickly. They can't be refuelled one launched. If it runs out, bye bye satellite. Their signals are adjusted to compensate for the drift, but is not always accurate, hence ephemeris.

As for the accuracy of the signals, people who used GPS a few years ago will remember that they weren't so accurate. This is because the signal on the X frequencies (the signals civilian receivers are able to receive) was deliberately downgraded - known as selective availablity. SA can be switched on again at the DoD's will, but they have agreed to guarentee the valdity of the signals. However, as the transmissions are very low power and it takes up to 25 seconds to transmit a complete signal sequence - they are very easily jammed. Jammers are commercially available. Also, power lines and other large metal structures can distort the signal.

And I'm convinced that the DoD does it's own jamming. Where I live, we get US carriers visiting the port regularly. When I sail within a mile of the carriers, my boat's GPS loses accuracy significantly.

IRRenewal
22nd Aug 2007, 11:01
I overflew an RAF station a while ago (low level VFR). When I approached the overhead the GPS started to play up and one at a time the signal strength bars disappeared. Once clear of the ATZ they re-appeared.

Whether this was deliberate jamming or just the result of an unusually high concentration of electronic equipment I don't know, but it did make me re-evaluate my personal use of GPS.

pilotmike
22nd Aug 2007, 11:53
Dan Winterland's 2nd post (#8) gives an excellent explaination of the errors from the 'signal reception' point of view, but probably weren't the cause of your 'error'.

Tomtom, being a 'cheap' consumer product, has its own set of limitations, one of which is smoothing of the raw data. If it always displayed your position exactly where the raw data said, your path would appear to be rather jerky, and not always exactly on the road. So the signal is smoothed, and the software always tries to keep you travelling along the known route, at a reasonably steady speed, until this is obviously no longer applicable.

Witness the way it continues to take you off the main road at an intersection if you choose to continue straight on. It can take 10 seconds or more for it to realise that you are not going the route it thought you were, and for your position to magically 'jump' back to where you are.

The Ferrybridge incident you describe is one of these situations, exactly as TopBunk said - a re-route of the road. Similar problems occur on the A120 near Stansted (recently re-routed) and a bizarre quirk in the M49 near the Severn Bridge, to give just 2 such examples. Tomtom appear to be very slow to offer updated maps. Until such time, we will often find ourselves navigating across fields, apparently.

The software also has a 'dead reckoning' mode, so that if you drive into a tunnel, it will continue to treat you as travelling in the same direction and speed while there is no raw data / signal. This often works well at putting you close to your actual position when you emerge from said tunnel.

PM

spannersatcx
22nd Aug 2007, 16:28
There are lots of errors on the maps supplied to Tom Tom, they are supplied by teleatlas. You can connect your TT device to your pc and once a week there is a satellite update which enables a quick fix to their new positions.

FlyerFoto
22nd Aug 2007, 19:18
The A1 at Ferrybridge was moved further west?

Strangely enough, the Fujitsu/Siemens PDA/GPS I use in my car hates the M62/A1 link road - it thinks I'm driving across fields, but doesn't tell me to turn back or anything!!!

I've been in a Robin DR400 over Ferrybridge and the GPS behaved OK them.

petermcleland
22nd Aug 2007, 19:48
The fault would probably be fixed by buying a new Tom Tom map...I have done this recently. However, I have to say they are not cheap (mine is whole of Europe + Canary Isles). I shall be content to just let the map go slowly out of date now, as new roads are built...I will be keeping my Safety Cameras up to date though.

GOLF_BRAVO_ZULU
23rd Aug 2007, 10:02
Of course FlyerFoto's GPS signal hadn't passed through the array of power lines that twistedenginestarter's signal had. For what it's worth, pilotmike's understanding of how the in-car unit works is the same as mine

PIGDOG
23rd Aug 2007, 14:33
A few months ago my GPS became convinced that we were both in Estonia! A quick look out the window to make sure we were still in the north of England. Took a photo of it before it corrected itself.

Never been to Estonia, but I have 'evidence' if I ever want to say I have!

Dan Winterland
24th Aug 2007, 02:10
And also 411, I would say with only three GPS position losses in your career, you have been lucky. I experience them all the time. Yesterday was my last - flying into a combined Civil/Military airport in China. We got 5 'FMS/GPS POS DISAGREE' messages on the way in, two on the way out.

Of course, they could have a jammer running!

skiesfull
24th Aug 2007, 08:58
Dan:-
China is not yet 'WGS 84' compliant. That means that the coordinates given by the Chinese authorities may disagree with GPS ones, hence the warnings. The recommendation is to disable GPS updates while in Chinese airspace, otherwise you will gat 'map-shift'.

411A
24th Aug 2007, 13:14
China is not yet 'WGS 84' compliant. That means that the coordinates given by the Chinese authorities may disagree with GPS ones, hence the warnings. The recommendation is to disable GPS updates while in Chinese airspace, otherwise you will gat 'map-shift'.
Indeed so, the devil is in the details:}

Now, my experience with GPS is using it stand alone, not as an update feature with an FMS/FMC.
In my private aeroplane, the Bendix/King unit has proven unfailingly accurate as a primary means of navigation, indeed where the RNAV (GPS) specific approach course overlays a localizer, the course centerline of both is one in the same.

In large jets, the Honeywell HT9100 GPS units have been approved for ten years for stand alone primary navigation, enroute/terminal.
Superbly reliable kit, in my personal experience, in Europe/USA/Africa/South Asia.

Now, I'm certainly not saying GPS can't fail, just that it is extremely unlikely, from my personal experience.

WHBM
24th Aug 2007, 13:39
The software also has a 'dead reckoning' mode, so that if you drive into a tunnel, it will continue to treat you as travelling in the same direction and speed while there is no raw data / signal. This often works well at putting you close to your actual position when you emerge from said tunnel.
Strangely I was noticing this exact feature only last night driving through the Dartford Tunnel. My Sat-nav handbook (system inbuilt in the car) describes it as one of the several ways it calculates the position, and as the tunnel alignment is in the system it keeps on track. It keeps on track better than the Heading Indicator sometimes does in the plane.

The new A1(M) alignment at Ferrybridge seems to be missing from many of the "updated" databases, even though it has been around for a few years now (the Mansfield bypass is the same). My hunch is that although there may be regular updates, the raw data has somehow not been provided to the Sat-Nav companies. It is probably provided separately by different authorities, eg councils, Highways Agency, etc, depending on the status of the new road.

An early in-car GPS I knew was always foiled by driving down the pier in Portland Harbour, Dorset, to a boat anchorage. The pier was not in its database at all and it would keep displaying "Attention your location is completely surrounded by water", as if anyone in such a situation would not have noticed !

Shore Guy
24th Aug 2007, 14:24
All of these posts prove the need for a position backup source when the world switches to airborne surveillance (ADS-B). By design, ADS-B is GPS based, and with loss of signal an FMS with inertial will provide a reasonable position for some time, but there will have to be a backup.

Looks like Loran E is getting the nod…..

“JPDO Study Endorses eLoran as Best Satnav Backup
A satellite navigation backup study commissioned by the JPDO has given
eLoran "the highest overall preference rating...particularly in the U.S."
Not yet publicly released, the 180-page document was prepared by ITT's
advanced engineering and sciences division and assessed seven candidates
against a series of essential requirements. The candidates include
DME/DME/INS; GNSS/INS; eLoran; VOR; "hardened" GNSS; terrain mapping; and multilateration. Leading requirements were the ability to support RNP values of 2.0, 1.0 and 0.3, for en route, terminal and nonprecision approach
respectively, plus technical readiness by 2015 to 2025 and complete
independence from satnav. The ability of eLoran to continue beyond INS
coasting limits following satnav loss was also noted. The report concluded,
"eLoran scored significantly highest for the general aviation segment, and
eLoran integration into GNSS/eLoran FMS systems for general aviation and
certain air carrier segments appears to be a viable and capable solution."

Dan Winterland
24th Aug 2007, 17:52
We use GPS to all of our 15 destinations in China without map shift. We just lose GPS primary at the one destiantaion which is combined civil/military. there is a lot of RF energy around this airfield - hence the loss of the signal.

MrHorgy
24th Aug 2007, 19:07
Forgive my ignorance, but what is eLORAN? We studied LORAN in our ATPL syllabus, from what i remember it was a network of huge antenna - is this similar?

Horgy

Shore Guy
25th Aug 2007, 00:12
E-Loran info......

http://www.navcen.uscg.gov/eloran/eloran_apps.htm

or do a "Google" ....tons of info

twistedenginestarter
27th Aug 2007, 23:21
GPS...made in America, and it works good The SiRFstar III chip is indeed impressive.
I'm begining to think GPS is reliable enough to make it the primary navigation model for VFR. When I first learnt to fly in the 60s I dreamed of a Decca navigator or the doppler-driven paper maps on the Trident. Maybe now we've got there.