GPS and fog

[J. Dillon]

Recently acquired a hand held GPS unit ( Magellan Meridian marine) at auction on E bay. It came with map down loads, Auto plug in and some other bells & whistles type stuff The nav aids internal attracted me the most although the road functions was a equal enticement. Been playing with it for a week now to get to know just what it is trying to tell me. On several sails if works well showing land masses as well as the nav aids. I logged in several GOTO places.

Today in fog it crapped out giving false positions , maybe positions represented by an hour glass and definate fixes shown by a pointed arrow. Trouble is it was miles inland. I knew where I was but the GPS didn't. What gives ? Did I get a defective unit ?

Before I bid I asked the owner about it's age , he said 4 yrs, I did have fully charged batteries in the unit. Should I trust this instrument or go back to lead and line with DR. I'd like to get "modern " but this is discouraging. :(

JD

[George Ray]

I have had GPS problems in cities or country when under or close to large power lines.

[P.I. Stazzer-Newt]

My handheld sometimes gives impossible positions but these are always accompanied by the "Not Tracking" message.

[Lew Barrett]

For marine use I have two GPS units. One, a late model (Raystar 125) Raymarine that is exceptionally accurate, the other a Garmin handheld that is permanently mounted on the console. I can refer to both, and any difference between them would be readily visible. I have yet to experience a divergence between them, but have only had the Raymarine for a year. The Garmin has been reliable, but is an older non-differential unit.
Thing is, all positional data has to be backed up by paper as you've learned. On the other hand in deep fog, you have to trust your instrument. If this unit is consistently inconsistent, return it immediately to the manufacturer for their review. GPS devices are notoriously battery hungry by the way; they use a lot of processing power. If yours is 4 YO, and you are running with rechargeable batteries as supplied by the seller, they may not be taking a good charge. Or, if you are loading AAs, be aware that one bad battery can lower the voltage to a point where the receiver is no longer reliable. If the batteries check out, there's nothing for it but to return to the manufacturer.
Question: with GPS units being now so reasonably priced, why take a chance with a "four year old" one? A handheld used outdoors for 5 or 6 years would be at the end of its useful life.
Lew

[Wild Wassa]

I have a Magellan. With the Magellan series there is 'total re-boot' facility. It is not the 'Restarting GPS' in the properties index, it is a 'total-reboot' function, that is not found in the manual.

When I had problems, my hand held was consistantly 600m out to the NW, I rang the Australian distributor and asked for the reboot proceedure. The rebooting is necessary so I was told, if hickups occur, or if the Magellan unit was scanned by some types of electronic equipment in transit, like those used in the postal industry. The reboot needs you to hold down a few buttons at once for specific amounts of time, until the instructions appear on the screen. I'm sorry I can't find the detailed proceedure that I put aside for such emergencies. Your major distributor will be able to give you this tech help.

Or, have you been into 'prioreties' lately and changed (even accidently), the 'map datum'? In Australia we use several datum (3 actually, not counting the global co-ordinates), depending on the vintage of the maps and charts that are still being used, not all maps have been updated to the new datum here. Co-ordinate data from things like Google Earth doesn't fit until calibrated within the system.

Warren.

[Nicholas Carey]

GPS units, especially handhelds w/o an external antenna, tend to have problems when they don't have a clear view of the satellites. Fog, heavy clouds and trees can all cause problems -- the ranging signal for SPS (Standard Positioning Service) -- what your receive uses -- pings in the microwave L band at 1575.42 mhz, so it's possible, I believe, for a sufficiency of wator vapor to degrade the signal past the point of usability.

Are you sure your GPS was had latched on to sufficient satellites to get a good solution?

If your unit is giving you spurious results, there's probably something wrong with it, but bear in mind there's all sorts of ways error gets injected into the system: conditions in the ionosphere and troposphere, human error (some bozo at GPS central in Cheyenne Mountain punches the wrong button), multipath interference, receiver error, etc. are just a few sources of entropy.

Accuracy also varies with latitude. In general, the closer you get to the poles, the less accurate the solution. It varies with longitude, too, but not as much.

Vertical fixes tend to be worse than horizontal. From the SPS Performance Standard:In a two-frequency position solution, RMS vertical errors are generally about 50% larger than RMS horizontal errors below a latitude of 55°, and two to three times larger at the poles. For single-frequency position solutions, RMS vertical errors can be over six times worse than RMS horizontal errors within 20° of the geomagnetic equator.

If you want to really glaze your eyes, you can read the SPS Performance Standard (http://www.navcen.uscg.gov/gps/geninfo/2001SPSPerformanceStandardFINAL.pdf) courtesy of the USCG. It goes into some detail about all this stuff.

[paladin]

multi path shouldn't bother the GPS.....it has inherent features to minimize/eliminate it....

[Nicholas Carey]

It's one of the listed sources of error, according to the SPS Performance Standard.

[J. Dillon]

Thanks for some really interesting responses. Which give rise to other things I noticed about the unit.

I did put in new rechargeable batteries. I also noted the number of satellites that it was picking up and the strength of the signal. They numbered over 5 and frequently 7.
During previously occasions in clear weather using the unit I noticed large differences between the compass heading and the GPS heading as much as 10 to 15 degrees. At first I surmised that the compass picked up errors over the winter but checked it on known headings and the compass was correct +or - a couple of degrees . The current at the time could not have been more then 1 1/2 KT.

As far as shipping induced errors is concerened , might that happen with any GPS ? They all get to the consumer by some means.

All of which gives rise that the unit may be defective , old or subject to moisture invading the innards . Much like what fog could do. At one point today I did take out the batteries and wiped the terminals off on my pants to kind of clean and dry them off. There was some improvement but it was not permanent. So in all I think inquires back to Magellan are in order.

JD

[BrianW]

I run my Garmin 182 with an internal antenna in all kinds of weather, and never had it lose accuracy. But mountains and narrow passages have caused it to lose it's satellite fix.

S.E. Alaska is not exactly known for clear, blue, and 72°.

Not really scientific, but there are three problems with your unit...

1. It's 4 years old.
2. It's from e-bay.
3. It's not a Garmin.

Okay, number 3 is subjective. ;)

[Nicholas Carey]

Thanks for some really interesting responses. Which give rise to other things I noticed about the unit.

I did put in new rechargeable batteries. I also noted the number of satellites that it was picking up and the strength of the signal. They numbered over 5 and frequently 7.Bear in mind that the GPS satellite array are all in polar orbit and orbit the earth about every 45 minutes -- the number of satellites available varies over a rather wide range.During previously occasions in clear weather using the unit I noticed large differences between the compass heading and the GPS heading as much as 10 to 15 degrees. At first I surmised that the compass picked up errors over the winter but checked it on known headings and the compass was correct +or - a couple of degrees . The current at the time could not have been more then 1 1/2 KT.Also bear in mind that the heading your GPS receiver reports relects your true track over ground -- current, wind, etc., all play into it. Also bear in mind that the bearing your compass shows is relative to magnetic north, not true north -- don't forget to include compass deviation into the calculation. And don't forget that, depending on the age of your map/chart, the variation shown isn't necessarily accurate -- the North magnetic pole wanders and varies in strength (secular variation).

As far as shipping induced errors is concerened , might that happen with any GPS ? They all get to the consumer by some means.While I like Garmin boxen, Magellan is a reputable company. I can't believe their quality assurance is that poor.All of which gives rise that the unit may be defective , old or subject to moisture invading the innards . Much like what fog could do.I doubt moisture in the guts of the box would affect the solution computed -- it will either compute a solution that is "correct" taking into account the various sources of induced error, or not compute a solution at all. The way GPS works, in a nutshell, is that each GPS satellite has a ridiculously accurate clock, all of which are synchronised with the Naval Observatory's atomic clock, so they are within a RCH of each other in terms of nanoseconds. Each satellite broadcasts a ranging signal, each ping of which is coded code with the satellite ID and the time.

A GPS receiver looks at the ranging signals it receives and identifies the satellites from which the ranging signals originated. The signals carry the date/time at which the 'ping' was sent. The GPS receiver essentially knows the orbits of each satellite so given a date/time and satelllite ID, the receiver can place locate the satellite in 3D space.

Given the time differentials of the various ranging signals the GPS receive is receiving (and knowing the speed at which radio waves travel), it's a [relatively] trivial computation to triangulate the location of the GPS receiver in 3D space based on the known satellite locations.

I suspect this explanation is a bit too simplistic (for instance, I'm ignoring the issue of the earth rotating on its axis below the satellite array), but the GPS receiver will for all intents and purposes either compute a correct solution or not.

At one point today I did take out the batteries and wiped the terminals off on my pants tos kind of clean and dry them off. There was some improvement but it was not permanent. So in all I think inquires back to Magellan are in order.Customer support is your best friend :D

Good Luck!

[Hwyl]

I'm not a fan of the Garmin user interface.

Having aid that I agree about the age and source of the unit.

I'm not a fan of mapping technology. I have a $100 Garmin and really only use it for Lat Long and waypoints (incredibly difficult to enter).

These things are plummeting in price. I'd say you have the wrong paradigm. Buy new and buy cheap.

[Bruce Hooke]

I have used a handheld Garmin on the land and on small bodies of water (rivers and lakes) and it occassionally gives very eroneous readings, but typically only for a very short period of time when it is having trouble picking up a signal. If you are getting a consistently bad reading then it seems to me that this is unlikley to be an issue with fog, moisture or other interference.

I don't know what boat speed you were dealing with, but at relatively slow boat speeds 1 1/2 knots of current running perpendicular to your course would be more than enough to produce a 10 to 15 degree difference between your course made good (what the GPS should give you) and your heading (what your compass gives you).

[paladin]

The inaccuracies due to multi-path that affect other signals is minimal on the GPS due to the spread spectrum aspects of the signal. The multipath fading is essentially checked and recompared at approximately 1800,000 times a second for each and every satellite in view. To show a substantial error, every satellite would have to have it's internal timing off by a significant amount and also be somewhere else in space other than where it thinks it is.....that's how the govt varies the accuracy of the transmitted signal....

[Dan McCosh]

Hand helds are do need a clear view of the sky--sails overhead, rigging, etc., can cause intermittant lost fixes. Usually you get a no fix signal, not an error. Temporary lost positions also are likely--same deal. The compass error could be current, etc. Also--there is a setting that shifts from magnetic to true headings on the readouts.

[John E Hardiman]

During previously occasions in clear weather using the unit I noticed large differences between the compass heading and the GPS heading as much as 10 to 15 degrees. At first I surmised that the compass picked up errors over the winter but checked it on known headings and the compass was correct +or - a couple of degrees . The current at the time could not have been more then 1 1/2 KT.



Leeway angle on the average sailboat will track 7-10 degrees off, i.e. course made good is NOT heading. And then adding any current to get course over bottom just complicates the matter.

Anyway, some things that I have found with handhelds is:

A) If you don't leave them where they can see the full sky, don't expect them to track well when you pick it up (i.e. don't leave them under the hardtop dodger and expect them to track accurately)

B) Don't just turn them on for "spot fixes". It is my experience that it takes a few averaging cycles (normally user selectable) for the units to settle out, especially if they are initilized while moving.

C). Make sure you are using the correct spherical projection for the map datum you are comparing it too (i.e. the GPS user selectable map datum must match the chart datum you are using). They need to match because some can be several hundred feet off.

GPS and doppler logs can be very accurate, but the need a lot of data sets to manage the random transmission errors. If you have many satellites, the you assume that there is a normal distrubution of error. If you only have a few in one quadrent (or one good signal) then the error begins to grow based upon a systemic dilution of precision.

[Nicholas Carey]

C). Make sure you are using the correct spherical projection for the map datum you are comparing it too (i.e. the GPS user selectable map datum must match the chart datum you are using). They need to match because some can be several hundred feet off.Right! I forgot that one.

FWIW, a chart datum is a reference model that approximates the shape of the earth relates it in terms of the project used to make the chart or map (often Mercator, but not always). The chart datum also locates the orgin (0,0) of the cartesian grid system in use. Local datums tend to make use of a reference model that more closely approximates the shape of the earth in that particular region.

Here's a list of the chart datums supported by Garmin's GPS units (I suspect that Magellan units support a similar set): http://www.garmin.com/support/faqs/MapDatumList.pdf

Here's some pages from the Australian Hydrographic Office that provide some detail on chart datums and why they are important.

http://www.ga.gov.au/geodesy/datums/aboutdatums.jsp
http://www.hydro.gov.au/important-info/gps-chart-datums.htm
http://www.amsa.gov.au/Shipping_Safety/Navigation_Safety/Positions_and_horizontal_datums_on_paper_and_elect ronic_charts/


While the above pages focus on Australia, the principles and issues remain the same for North America. Just for the record, Australia switched datums to better support GPS navigation. The difference works out to be about 200 meters -- more than 600 feet.

[dmede]

Once locked on to a sufficient number of satalites fog shouldn't be a problem. I've used mine on cloudy overcast days many times without issue. Line of sight to the satalites is the only thing that affects the signal.

You might try resetting the device per it's internal menu. Also remember that you need to leave the device in one position for a few minutes when first tuned on while it figures out where it is. Double check that you have set it up properly in the menu so that it knows what zone it's in. It can lock onto satallites from outside the zone your in and give what look like good results. And finally (for error between charts, compass and GPS) be sure your using the same projection and datum.

dave

[BrianW]

Here's a shot of my Garmin 182...

http://img.photobucket.com/albums/v136/BrianW/beccadriving3.jpg

...you can barely see the tiny antenna stick up from the top right corner.

Here's what it looks like in side the small wheelhouse...

http://img.photobucket.com/albums/v136/BrianW/beccadriving2.jpg

(my daughter is standing on a large dry box)

It's got a lot of aluminum overhead. Yet it never loses signal due to fog or clouds.

It does lose signal in here...

http://img.photobucket.com/albums/v136/BrianW/Honu/nothru1.jpg

...but I'm not watching the GPS in there anyhow. ;)

What is nice about a chartplotter GPS, is that even when the signal is lost, the charts are still there. So you can use it like a paper chart until the signal returns.