Flaw in Garmin eTrex Legend accuracy?

May 6, 2003

Hi,

Does anyone else have a problem with their GPS always pointing a certain distance from a Geocache?

I have a Garmin eTrex Legend (which in all other respects is very easy to use). After a number of caches I've done, I discovered that it always seems to point to the zero location of the cache as being about 17 feet north of the actual cache location! I've verified this by teaming up with friends with the same GPSr, and for some reason I'm always 17 feet north.

While this is nice to know, it is always tricky to go to the hypothetical 0 point and then try to backtrack 17 feet south from there. I realize that all GPSr units have a limited accuracy (about 16 to 20 feet, depending on sattelites), but this is just weird that it happens with all caches, and it's always to the north, and not just in a circle around the cache.

Does anyone know how to fix this? Is there some way I can manually enter an offset into the GPSr's system software? I've tried it with and without WAAS, with the same problem.

(If this has already been asked and answered in another thread, please tell me where and excuse me... I looked and looked but didn't see this question anywhere else).

2.3k · May 6, 2003

Legend Accuracy

I find my Legend extremely accurate; sometimes a mere few feet from a known place on earth. These known places are NGS "adjusted" Benchmarks. They are listed to the one-hundred-thousandth of a second (centimeter accuracy).

Since you must round off the Benchmark coordinates when entering them in your GPSr, your waypoint may be a few feet away from the Benchmark. They are also NAD83.

Go here and search for the GPS marks:

http://www.ngs.noaa.gov/datasheet.html

Route: Datasheets - county - state(and click: get county list) - county - GPS sites only - get marks

1.4k · May 6, 2003

Confirm that your Legend is in WGS84 setting for caches.

Escout is correct for benchmarks being in NAD83.

Also is it set in True or Magnetic North?

3.5k · May 6, 2003

Yes, check your settings. Your Map Datum should be in WSG84, and, true, the best way to check your gps is against a horizontal type ( not verticle) benchmark.

However, it will make little difference if you are in WGS84 or NAD83 as they are vertually the same. If you are set to NAD27, that WILL throw you off. Magnetic or True will make no difference what so ever as far as you cordinates go.

I have seen another setting which you can set to put your postion off of the actual coordinates by a defined amount. I forget what this setting is, but it may be best just to select your option to return all setting to DEFAULT. That should get you back on track, and then you can reset the things you wish.

"See the wonderous works of Providence! The uncertainty of human things!" Geo.Washington

2.3k · May 6, 2003

quote:
Originally posted by EraSeek:
I have seen another setting which you can set to put your postion off of the actual coordinates by a defined amount. I forget what this setting is,

This is in the "Confuse" menu, under "Confuse newbie".

Just kidding, what is this you are referring to? As far as I know, there is no such setting in a Legend. You can project a waypoint a certain distance from your location, but that will not change your position, as it's reported.

Are you using the latest firmware? Are you by any chance entering the coordinates in UTM format? There was a bug, until recently, when the unit changed the entered coordinates on its own, under some circumstances. But that has been corrected in the most recent releases.

If it's always in the same direction, I would blame the map datum, just like the other posters. All geocaching coordinates assume WGS84.

Anders

38.2k · May 7, 2003

17 feet off is quite accurate. I rarely get closer than 20-30 feet.

"It has been my experience that folks who have no vices have very few virtues" -Abraham Lincoln

May 7, 2003

Thank you everyone for the input. I appreciate the feedback!

So far, everything on my eTrex Legend seems to be matching up to what you all have said: It is set for True North, the map datum is in WSG84, position format is "hhdd mm.mmm", I have just about the latest firmware, v. 3.10. I don't think I'm entering any coordinates in UTM format... just downloading the LOC files from the Geocaching website and transferring them to the eTrex via EasyGPS, so... whatever everyone else is using.

The only other thing I can appear to change is the "Distance/speed" setting, which is currently set to "Statute". Other options are "Metric" and "Yards", but they aren't the default.

I will try some benchmarks (thanks for the link to the NGS NOAA website) to see how far off I am from one of them. I also could not find a setting anywhere on the GPSr which lets you set your position off of the actual coordinates by a defined amount. That would be great if it were there (...and showing a North offset of 17 feet! )

I know 17 feet is not a big problem in the grand scheme of caching, but it always points me on the wrong lines... so if I'm trying to triangulate from about 50 or 60 feet out, it always points to an incorrect position. Most other people's GPSs seem to give an accurate line to the waypoint, even if it loses accuracy in the last few feet... you can usually just keep on walking and trip over the cache!

I talked to one guy whose GPSr was about 250 feet inaccurate... obviously defective. So, that makes me wonder if mine is also defective, just not so bad. If it can be recalibrated, I'd sure be a happy camper... uh, cacher!

Thanks again,

Skyking

5.7k · May 7, 2003

Skyking,

I think you are over estimating the accuracy of a consumer GPSr. If it gets you within 30ft of a cache, then it is working just fine. It sounds like yours is. If you happen to be with another geocacher try comparing their coords with yours at various waypoints. If it is consistenly off in one direction then maybe it is defective, but I wouldn't rule out normal, everyday error as the actual cause.

Also, try going back to a nearby cache at different times of the day and see if you get different results. Sometimes it just depends on which sats you are locked on to.

Did we confuse you enough yet?

--Marky

"All of us get lost in the darkness, dreamers learn to steer with a backlit GPSr"

May 8, 2003

Thanks again everyone for your comments and suggestions!

I will trying doing some benchmarks, and also doing some side by side with another geocacher to see how things compare. I can live with 17' Northbound inaccuracy... just wondered if there was some way to fix it.

Marky, thanks for throwing in your 2cents worth too! We really enjoyed meeting you at the Event, and look forward with great enthusiasm for the next one. After talking to many of the people there, we took the leap and became members of the geocaching site... love those pocket queries!

1.3k · May 8, 2003

I know this may be an oversimplification, but how are you determining the seventeen feet? If you are reading the "gps accuracy 17 feet" right from the screen, and find yourself somewhat north of the cache, I'd bet that there is nothing wrong with your reciever. The person who hid the cache probably didn't have perfect precision to begin with. That kind of precision would be hard to achieve with a handheld. Good luck, have fun caching!

"Searching with my good eye closed"

May 9, 2003

BloenCustoms,

Yeah, I'm not actually paying any attention at all to the GPS Accuracy figure shown on my GPS. Instead, it's that when I triangulate with 4 or more tracks, I find them all pointing at one specific spot. Then, I've found that if I walk south about 17 feet from that imaginary spot, most times I will find the cache! Please see the attached drawing I just made (hope I uploaded it ok). I take readings from far out and do a sighting on a point along my track to the cache coordinates. Then I sight again and again from different locations, all out 60 or 70 feet where things should be fairly accurate, and I walk with a steady pace, to make sure the GPSr can pick up the differential readings coming from the sattelites.

Anyway, when I find the spot where all my visual sightings intersect, I have discovered that the cache is rarely there... it is South of that location by 17 to 20 feet! So, I don't think it has anything to do with the GPS Accuracy setting on the unit, since I'm ignoring it... my problem is verified by using line of sight and triangulation.

Hope that makes sense. Thanks for the input.

May 9, 2003

Grasping at straws here but it worked for me. I found that in about 50% of my hunts that I am more accurate with WAAS disabled. I also find that It is more accurate Half the time with WAAS inabled so it all depends on your current location and current satelite position I assume. So next time you hunt try it both ways as see the difference.

_________________________________________________________

On the other hand, you have different fingers.

3.5k · May 9, 2003

quote:
Originally posted by Anders:
Just kidding, _what_ is this you are referring to? As far as I know, there is _no_ such setting in a Legend. You can project a waypoint a certain distance from your location, but that will not change your position, as it's reported.

Anders

This is probably not the problem in this case, and I do not know if you can do this with the legend, BUT, yes, there is a setting in some units whereby you can throw your position off by a set amount on purpose. Quite a while ago I was playing with this function, but I don't recall now what exactly it was, or which units you find it in. As I recall it had some marine purpose. Maybe I'm just dreaming, But I'll see if I can find it again.

It was not projecting waypoints, but a method to offset coordinates.

"See the wonderous works of Providence! The uncertainty of human things!" Geo.Washington

3.5k · May 9, 2003

Again, this is probably not the problem. but a note of interest. I have seen data plots that show most GPS units will have a North bias from the actual coordinates, but these may have been before Selective availability was switched off. Also if you have the ''Lock on Roads''on it can throw you off. I took a mark with waas on, and when I checked it on top of the mark it told me I was 40' away until I turned 'Lock On Roads' off, and then I was dead on.

"See the wonderous works of Providence! The uncertainty of human things!" Geo.Washington

May 9, 2003

If you're always getting 17' to the north of the cache, you have a highly valuable receiver!

You may be annoyed that it is off 17 feet, but think about it! All you have to do is go the the zero point, then walk 17 feet south, and you'll magically be at the cache!

You unit is obviously taking into account the position error of the hider, too! Quantum dynamics is the only way I can think of to do this, so your unit must be a prototype that slipped out of Garmin's lab!

Put that baby on e-bay, and don't let it go for less than million! On the other hand, you may want relieve yourself of it quickly before the Men In Black track you down.

George

2.3k · May 10, 2003

Legend with WAAS off. What Brownmule says about using it with WAAS off may be worth trying. I almost always use my legend with it off. The only time I consider using WAAS is when I am still for some time and have a good signal from one of the WAAS satellites.

This is because with it on, the processor runs slower and while moving, often reports a worse "accuracy" number.

May 10, 2003

We used our e-trex Legend to find our first cache today and it read to within 1.5 feet of the cache near the summit of a 1200 ft mountain in Quebec.

(No kidding). On the other hand I an very disappointed that the Legend loses satellites at the slightest hint of foliage or canyons. It was great in the Arizona desert, but back home near Montreal, it is sometimes very poor. We're seriously thinking of getting a Magellan Meridian Platinum unit. Anyone have experience with that?

Art

May 10, 2003

Buying another unit probably won't make the problem go away.

Magellen and Garmin each have a different "feel". Magellen tends to be more liberal with reporting a position, even though it doesn't have real good data. Garmin is more conservative in that it is more up-front if the conditions are dicey. The result is that Magellen may give you a false sense of comfort. They also tend to "over-shoot" a cache location due to their auto-averaging feature.

There is a difference in antennas in the units. The Legend uses a patch, the Meridian a quad-helix. Some will swear up and down that the quad-helix is far better under trees, but I think the difference is minimal and not worth worrying about.

On the other hand, a patch antenna tends to work better in canyons as it is less prone to multi-path interference than the quad-helix.

You say you're having difficulty under both foliage and canyons, so I don't know what to tell you.

Remember that gps sat positions are changing all the time. You might get a great lock in a certain spot one day, and very poor another.

I guess what I'm saying is not to expect a huge difference in performance. You may notice a large difference in "feel", though, which you may or may not like. It's just like some people prefer a car with a tight suspension, while others want a car with a really soft ride.

George

2.3k · May 10, 2003

nincehelser:

I assume you have some experience using Garmins and Magellan Meridian. I have owned a Garmin for five years, a Legend for 6 months, and a Meridian for a month.

I would not describe the Meridian as "overshooting". It appears to be a little slower in its processing speed while moving. And, the auto Averaging is only engaged while stopped, not moving.

I have hooked both units to my Palm, running Cetus and Coto, which take NMEA data from the GPSrs. I don't know to what extent the data is processed before being sent out, but for both units the NMEA output and what is shown on the screens are the same.

The Meridian is quick to show changing signal strength of satellites. If anything, it may be somewhat slower to respnd to changes in position, while moving. This is mostly noticeable while driving and watching the pointer.

May 10, 2003

"Overshooting" is the common term used by most on these boards to describe Magellens. Many describe it as walking past the cache, then having to double back many feet.

George

1.1k · May 10, 2003

Find this a little strange that after a number of caches it appears to be so many feet always in the one direction of the actual cache coordinates.

So if ALL the caches were placed by the ONE person then who is to say who is right or wrong.

However one would presume that different people placed these caches and hence there's really no way known that they would all perfectly agree with other and appear to make one appear to be always so far off in the one direction.

Eraseek, what you are thinking of is PCF (Position Correction factor), which is what some (old) units used to correct for different datums basically because these units never had any datum option anyway, just WGS84 (fixed). Don't see it used all that much these days.

Cheers, Kerry.

I never get lost everybody keeps telling me where to go

May 11, 2003

quote:
EScout said:

I would not describe the Meridian as "overshooting". It appears to be a little slower in its processing speed while moving. And, the auto Averaging is only engaged while stopped, not moving.

The overshoot he's talking about generally only happens in reception problem areas such as canyons, tree cover etc. If you're hiking down a trail in these sorts of conditions with someone using a Garmin unit, you'll notice that the distance to destination reported by the Magellan is generally about .01 to .02 miles farther than the Garmin. If you continue on a straight line until the Garmin says zero, the Magellan will have you keep going for awhile, generally for a distance of around 75 to 150 feet. Once you stop, the receiver will correct itself, and bring you back to the right spot over the course of a couple minutes. This is why many people stop, remove their pack and look around about 200 feet prior to the waypoint.

If you're winding and twisting along much when the receiver is doing this, you'll also notice the track has fairly consistant errors of about the same distance. Speculation on the sci geo newsgroup is that it's a result of Magellan using a somewhat different setup for their Kaliman Filter than Garmin is using. Kind of a trade off between accuracy while on the move, and having the pointer, speed and other indications jumping around too much.

Both units appear to send the NMEA data out only after all internal processing is done. This is unfortunate, as I'd like to be able to view the data from my Magellan without the averaging on occasion. (Just to play with and compare)

When the reception isn't fairly bad, I haven't seen any difference in the behavior of the two brands.

May 11, 2003

quote:
Originally posted by nincehelser:
"Overshooting" is the common term used by most on these boards to describe Magellens. Many describe it as walking past the cache, then having to double back many feet.

George

I assume you are using the Magellan Meridian Platinum, which we were considering buying. There is also the Gold and one other in the Meridian family (can't remember the name) I assume they all have the same Quadrifilar antenna which is supposed to be superior to the patch antenna on the e-Trex Legend.

Thinking back on our travels, we had the Legend lose the satellites once or twice dscending into the grand canyon. This was for relatively short periods of a few seconds to a minute or so. The problem of losing satellites is more pronounced under foliage. Magellan even advertises "Tracks position even in dense foiliage and urban canyons". That's what attracted us to it. You say the Magellan is slower in reporting position when driving. We certainly don't need that, as our Legend seems to be quite slow when we are in a city moving past many streets over a very short distance. We frequently use the Legend connected to our laptop with the Microsoft "Streets and Trips" program. It's great out on the highway, but seems slower in reporting position in the city. Driving downtown, it becomes almost useless as we are in the "urban canyons"

Ar

5.3k · May 11, 2003

That is normal for any GPS. If you figure with WAAS you are going to get the advertised 9 feet in accuracy, then I have to decide 9 feet in what direction. This make 18 feet. When my GPS says 20 feet I just start looking. BTW I have used severel models from Garmin and Magellan. DOn't worry about having your GPS pin point the cache, just have fun with it.

2.3k · May 11, 2003

Thank you Searching_ut for the information. My location of coastal So. Cal. probably explains why I have not seen this difference. There are not many trees and mostly open views of the sky. However, based on my testing the satellite signal strength under covers and trees, it sounds like you are describing conditions with extremely weak signals/only a few satellites being received.

1.3k · May 11, 2003

How about you take the reciever in question, and mark a waypoint. Then see if it brings you seventeen feet north of that waypoint when you check it later. That would be the deciding factor for me to get warranty work done. If the reciever shows an error on coordinates you got using the same reciever...

"Searching with my good eye closed"

July 21, 2004

You can increase accuracy with a DGPS antenna beacon reciever. This is an optional accessory. It's expensive, but it allows you to pick up ground-based DGPS signals from the government. (Ironically, the same governmment who degraded GPS signals with S/A is now spending money to improve them with DGPS.) However, unless you are a millionaire, you don't really need a DGPS beacon reciever for ordinary caching. DGPS is used mostly by airplane pilots and boat captains.

1k · July 21, 2004

Psssst...This thread is over a year old!!

Kar of TS!!

Flaw in Garmin eTrex Legend accuracy?

Recommended Posts

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Join the conversation