SummitPost.org

I have a Garmin eTrex Vista Cx and lately it's been giving me some strange results. Things like 12 mile trips are being logged as 60 mile trips. When I download the track info I see that some of the data points are all over the board, and some point-to-point data says I was traveling at ridiculous speeds (150+ mph) for brief periods of time. Anybody have similar issues or know what the problem might be? My warranty is up so I'm considering buying a new unit if I can't get this fixed.

There is an error called a multipath error. In places with smooth surfaces, like lakes, gravel roads, rock walls, etc., the GPS signal can bounce off of something first before hitting the receiver antenna. This lengthens the time from satellite to receiver which yields a position different from where you really are. Especially when you see tracks with sort of wild wingy-dingies, one should suspect multipath errors.

That said, I haven't seen multipath errors being that wierd. Try using the unit with WAAS turned off - Garmin has been having some software issues with respect to WAAS. Older units sometimes go blank and need to totally reset. You also might trying to do a hardware/software reset as well as check on line for software updates to your unit.

Make sure you clear out your track log before starting a trip. You might also check the track log settings - don;t set it too low, which leads to a very jittery track - like have it log a position every 30 seconds or so.

I have a Vista HCx, with the high sensitivity chip and have seen far fewer of these problems than with my old etrex Geko.

Yes, I've had the same problem the last few times I've taken mine out (GPSmap 76C). Climbed over a mountain from one narrow canyon to another and it added 12 miles to the original miles. I gets speeds of 75 miles or so. Just thought I was kicking butt for a middle aged guy.

Toys will be toys. I've just resolved myself to use it to get me out of areas with no trails or just to help follow secondary trails.

Some times when you are in a canyon and lose lock (get fewer than 3-4 well-spaced satellites), the next position will be way out of whack. This problem is exacerbated by the multipath error Moni described, because the GPS may grab any signal to get the three it needs. The gdb file may actually carry the "accuracy" at the time of the odd excursions, so you can see if there is a correlation (mine have such info).

If the type of territory you travel, and the way you carry the unit have not changed, you could have a more insidious problem -- like loose wiring to the antenna.

I'm getting more loss of coherent constellations lately, enough to make me wonder if the satellite orbits are getting tweaked.

Thanks for the replies. I'll try messing with the WAAS settings and see if that helps.

MoapaPk wrote:The gdb file may actually carry the "accuracy" at the time of the odd excursions. . . .

Really? Where? How? I have never seen this, and I know intimately the way GDB files are written, since I had to decode them byte-by-byte to write my program that reads and writes GDB files.

I've seen some of these errors as well. I did a hike that was 12.7 miles according to the guidebook. My Garmin said I had gone 17.6. Last weekend on a hike in Shenandoah National Park, a park sign said I was 2.2 miles from Skyline Drive. My Garmin read 3.1.

Most notably my elevations are way off. I live at 112' above sea level and there are times when my GPS tells me I'm over 5,000' in my living room.

I calibrate the unit every time I use it and I always clear all track logs and old waypoints before each use.

I have had these type of errors from dropping and regaining signal (hiking in redwoods, very narrow canyons) Between regaining signal and multi-path error I am able to generate some pretty amazing track logs. With the vista it's best orientation is flat as if on a table. I have a small pouch on a shoulder strap that is directly on top of my shoulder so the unit faces the sky. The difference of having it vertical vs. horizontal can be significant especially when signal is less than ideal. Interested to hear what you learn about your problem.

CrazySanMan wrote:Most notably my elevations are way off. I live at 112' above sea level and there are times when my GPS tells me I'm over 5,000' in my living room.

Assuming your livingroom has a roof......GPS doesn't work indoors. You may get a little signal but that is through windows, not the rest of the building. Elevation accuracy goes to hell with less than ideal reception.

Day Hiker wrote:

MoapaPk wrote:The gdb file may actually carry the "accuracy" at the time of the odd excursions. . . .

Really? Where? How? I have never seen this, and I know intimately the way GDB files are written, since I had to decode them byte-by-byte to write my program that reads and writes GDB files.

Perhaps not the standard gdb, but I have at one time gotten output from MapSource (after directly reading the GPS), and I'd swear it had a column for "accuracy"; but I'm not certain, so that's why I said "may". If that information is never stored in the unit, I guess that I was dreaming. I also seem to remember that the accuracy field was flashing blue and pink, and had giraffes.

My only experience with gdb on the binary level, was when I used a binary editor to pull out the long and lat from a gdb file that someone had mistakenly tried to edit, then saved as text. My recollection is that the lat and long data were stored as scaled integers. Is there also a field that says how many fields follow in each structure?

kolbi2112 wrote:I have a Garmin eTrex Vista Cx and lately it's been giving me some strange results. Things like 12 mile trips are being logged as 60 mile trips. When I download the track info I see that some of the data points are all over the board, and some point-to-point data says I was traveling at ridiculous speeds (150+ mph) for brief periods of time. Anybody have similar issues or know what the problem might be? My warranty is up so I'm considering buying a new unit if I can't get this fixed.

This is likely what you experienced:



This is from a Garmin Colorado. Notice that no errors occurred until deep into the canyon, where the poor signal and multi-path combined to produce the erratic track.

My Venture Cx frequently did this and in addition to earlier comments, I'd add weak batteries as a probable cause.

MoapaPk wrote:Perhaps not the standard gdb, but I have at one time gotten output from MapSource (after directly reading the GPS), and I'd swear it had a column for "accuracy"; but I'm not certain, so that's why I said "may". If that information is never stored in the unit, I guess that I was dreaming. I also seem to remember that the accuracy field was flashing blue and pink, and had giraffes.

LOL. Yes, I did, literally out loud.

MoapaPk wrote:My only experience with gdb on the binary level, was when I used a binary editor to pull out the long and lat from a gdb file that someone had mistakenly tried to edit, then saved as text. My recollection is that the lat and long data were stored as scaled integers. Is there also a field that says how many fields follow in each structure?

In a GDB file, each record starts with four bytes indicating the length of record, but this length does not count these four bytes nor the byte following. The byte following designates the record type, and for waypoints it is hex 57 or ASCII for "W." I bet you can guess what it is for a track or a route.

For a waypoint record, the waypoint name is written next, in ASCII. So already the record length is variable, dependent on how many characters are in the waypoint name.

Immediately following this are 28 bytes. In hex, they are: 10-00s, 12-FFs, 2-00s, and 4-FFs.

Then there are four bytes for latitude followed by four for longitude. Both lat and lon are stored the same way, with the 360-degree range of values spread across the entire 4,294,967,296 values of the four bytes. (This gives a resolution of .0000000838 degrees.) So 0 degrees is hex 00 00 00 00, 90 degrees is hex 40 00 00 00, negative 90 degrees is C0 00 00 00, and 180 degrees equals negative 180 degrees, which is 80 00 00 00. Note that the four bytes are written with the least-significant byte first.

Following this is one byte indicating whether or not the elevation is known. The byte's value is either 00 or 01 hex. If the elevation is known, there are 8 bytes for elevation (in meters) which are in double precision floating-point format according to IEEE Standard 754. If you think this looks complicated, with part of a byte being used for one value and the other part for another value, think about the fact that I was able to decode this before I read and knew about IEEE Standard 754. And the bytes are written in "reverse" order, with the least-significant byte first.

Next is the waypoint description, again in ASCII, and again this variable-length string adds another variable to the length of the waypoint record.

There are bytes for "proximity," display type, symbol type, depth, URLs, category, temperature, and modified date. Not all of these are used in the 60CSx, but these values can be used and modified by MapSource.

MtnHermit wrote:This is likely what you experienced:

This is from a Garmin Colorado. Notice that no errors occurred until deep into the canyon, where the poor signal and multi-path combined to produce the erratic track.

My Venture Cx frequently did this and in addition to earlier comments, I'd add weak batteries as a probable cause.

Yes, very similar, although I've seen this problem in wide open spaces as well.

So based on all the replies it sounds like there's a few things I can try on my next outing:

1. Update WAAS settings (if necessary)
2. Change location of GPS (will try to hook it to the top of my pack so it faces up)
3. Hardware/software reset (I've done this once but will try it again)
4. Fresh batteries

I'll be sure to update this thread with the results next week.

kolbi2112 wrote:Yes, very similar, although I've seen this problem in wide open spaces as well.

So based on all the replies it sounds like there's a few things I can try on my next outing:

1. Update WAAS settings (if necessary)
2. Change location of GPS (will try to hook it to the top of my pack so it faces up)
3. Hardware/software reset (I've done this once but will try it again)
4. Fresh batteries

I'll be sure to update this thread with the results next week.

I haven't played much with WAAS, so I can't comment on reason #1. For #3, a software problem could do weird things, but I don't see this as being the likely case, especially if it's working ok sometimes. I really don't think it's the batteries because, from my experience, the unit either powers up or it doesn't. If the batteries are weak, the unit will shut down instead of working improperly.

I suspect reason #2 is the one on which to focus. If the unit is significantly shielded by your body or gear, it won't matter if you are in a wide open space. It still won't get a good signal. Note that different models of GPSs have different abilities to handle weaker signals, so while I can easily use my body to shield my GPS-V, I would have to work hard to do the same to my 60CSx. The 60CSx can often acquire a position indoors, while the older GPS-V has no chance unless it is outdoors or possibly right up against a window.

They used to say that patch antennas should stay horizontal; and quad-helix, vertical. However, my quad-helix units do fine with the gps lying horizontally on the dash of my car; and my neighbor found that his patch-antenna unit lost no accuracy when tilted to 45 degrees (vs strict horizontal).

I keep the unit high on my shoulder strap, but I definitely see signal loss when I am walking by a big wall, with the GPS on the wall-ward shoulder, and my head blocks the signal. I get great improvement by simply changing the gps to the other shoulder strap.