Etrex 20 keeps calculating distance while standing still

[gpsblake]

Just picked up an Etrex 20 this week, unit is doing real well EXCEPT that I noticed when I upgraded to 2.9, the unit continues to calculate distance while standing still. This was not a problem with the Dakota or Venture HC.. In fact, in 5 minutes it show me moving 150 feet at 1.2 mph on average. It may have been doing this under firmware 2.8 but I did not notice. I also noticed the track will be splattered when the unit is standing still while the Dakota and Venture HC knew it had stopped moving and thus stopped plotting.

 

While this is not a big deal for geocaching, it is a critical and fatal error for keeping stats while bicycling, because it messes up the data whenever I stop.

 

I guess my question is, has anyone else noticed this and two, was this error present in earlier versions of the firmware like 2.8 for example? Or could my unit be defective?

[+BlueDeuce]

Does it have a calibration option? From what I know about the Etrex series they hate battery changes without calibration and traveling a long distance without a satellite baking session (30 minutes).

[gpsblake]

It doesn't have an electronic compass so I'm guessing it doesn't need to be calibrated. I tried several firmware versions on the Etrex tonight downgrading from 2.90 to 2.5, 2.6 etc... Does the same thing. If I can't find an option to stop this, I'll have to return the unit. A shame because it does everything so well except for that.

[+BlueDeuce]

It doesn't have an electronic compass so I'm guessing it doesn't need to be calibrated. I tried several firmware versions on the Etrex tonight downgrading from 2.90 to 2.5, 2.6 etc... Does the same thing. If I can't find an option to stop this, I'll have to return the unit. A shame because it does everything so well except for that.

 

ok, wasn't sure.

[+northernpenguin]

I guess my question is, has anyone else noticed this and two, was this error present in earlier versions of the firmware like 2.8 for example? Or could my unit be defective?

 

Have not noticed this on my eTrex 30, but I'll keep an eye out for that and report back.

 

I tend to forget to calibrate the compass on my 30 all the time and it's fine. Also took it to Edmonton from Toronto without a satellite baking session, and back without noticing anything being amiss.

 

 

[John E Cache]

You need to ask yourself, "How does the GPS know I am standing still?".

[gpsblake]

You need to ask yourself, "How does the GPS know I am standing still?".

 

The algorithm for the Dakota, Venture HC, Blue Legend knew how to do this. The Magellan GC figured it out on it's last firmware release as the did the Triton in later firmware releases.

 

One of the early complaints about the Etrex 20 (and Magellan GC) I read was the unit wasn't responding when people walk slow, maybe they over-tweaked it as it will register a movement really easily.

 

The Etrex 20 does perform really, really well for geocaching. However, geocaching is my #2 purpose in the unit as bicycling is my #1 purpose. And other then the odometer moving some when standing still, it is really accurate in calculating distance, really bright in the sunlight, amazingly quick with the menus. It's not moving constantly, but if you watch it, it will "move" 12 to 15 feet at a time while standing still I say every 30 seconds to a minute. And yea, I've tried WAAS off/on, GLONASS off/on also.

 

Oh, and I can't get it to auto-route using openstreet maps (the Dakota 10 will auto-route off those maps). Not a biggie to me.

[wmarek74]

eTrex 20 register GPS altitudes, eTrex 30 air pressure.

Less accurate data in eTrex 20 cause problems when standing still.

 

Alghoritms base on changes in Vertical speed (values above 5m/min are taken for Total ascents/descends);

[wmarek74]

I should add: interval for readings is 1 per second (1 Hz);

Trip odometer may show even a few kilometers when lying on windows sill for couple of hours ;-) It is due to GPS readings.

[tr_s]

 

One of the early complaints about the Etrex 20 (and Magellan GC) I read was the unit wasn't responding when people walk slow, maybe they over-tweaked it as it will register a movement really easily.

 

 

Somewhere during the process it seems to have been forgotten by the Garmin programming house that differently tweaked filters applied to the raw gps location can be used for calculating different quantities.

[John E Cache]

You need to ask yourself, "How does the GPS know I am standing still?".

 

The algorithm for the Dakota, Venture HC, Blue Legend knew how to do this. The Magellan GC figured it out on it's last firmware release as the did the Triton in later firmware releases.

 

One of the early complaints about the Etrex 20 (and Magellan GC) I read was the unit wasn't responding when people walk slow, maybe they over-tweaked it as it will register a movement really easily.

Thanks for pointing that out. I had forgotten. You call it over tweaked and the guy, who gave the "wasn't responding when people walk slow" symptom the derogatory name of "sticky", is probably happy. I was looking at what GPS info is given in the serial port sentences for post processing and noticed some GPSs give Doppler generated speed. Maybe the Dakota etc. can know if you not moving by Doppler shift changes.

[+EScout]

What is the setting for recording tracks? I use my 20 for hiking and biking and the tracks look good. I set the recording to Auto, More Often (not Most Often. ) v2.80

 

I do not recommend using time intervals for recording setting.

[wmarek74]

The strange thing with eTrex 20 is that it can not keep exact distance in track.

E.g. when we choose 10 meters (0.01km) intervals and look at the track log it does not match exactly 10 meters (sometimes the distances between track points are less, sometimes more than 10 meters...)

Edited April 7, 2013 by wmarek74

[tr_s]

The strange thing with eTrex 20 is that it can not keep exact distance in track.

E.g. when we choose 10 meters (0.01km) intervals and look at the track log it does not match exactly 10 meters (sometimes the distances between track points are less, sometimes more than 10 meters...)

 

The setting is probably just a target. Position updates in the system are at 1 Hz. My guess is that if (ΔD from previous track point >= set limit) upon a position update then a track point is recorded.

Edited April 7, 2013 by tr_s

[gpsblake]

This is the last straw. Either I got a defective unit or the firmware is a mess. And I was walking the entire time. Yet the GPS thought I wasn't moving most of the time. I also noticed that most of the time, the GPS said I was moving at 0mph, then would jump ahead .02 or .03 miles on the odometer.

 

 

So I can walk 7.5mph when moving even though my max speed was 4.7mph!!!! The trip odometer was pretty accurate though. The Dakota 10, Venture HC, and even the Garmin Legend didn't suffer from these problems.

 

Back this goes tomorrow, since I've agreed to sell most of my older units, I think I'm just going to pick up a Etrex 10 for my main geocaching unit.

 

Edit: EScout, I was using those same setting, more often AUTO. v2.90. I'll trying rolling back to 2.8 (which is what the unit had when I bought it a few days ago). But it's still going back.

Edited April 7, 2013 by gpsblake

[Grasscatcher]

I'll bet that you cannot verify with absolute certainty that all those data fields started out at "O" at the start of your walk.......correct?

 

Did you specifically reset all to "0" before you started? ....and visually verify that ,in fact, you were starting with a "clean slate"? ....DO NOT assume ....

 

Right now, I don't believe that you understand how each of those data fields work or what each is telling you.

[gpsblake]

Yes Grasscatcher, I can confirm it was all zeros before I began to walk. Been doing this for over 10 years and I am a perfectionist when it comes to my GPS data being right. I also have owned quite a few units.

 

It's a firmware 2.9 regression bug........ Loaded back up 2.8, took a walk, here's the difference.

 

 

Notice it works exactly how it should now, the max MPH is a little off though. Therefore it's a regession bug in Firmware 2.9 compared to Firmware 2.8

 

However, the Garmin Etrex with FW 2.8 continues to think it's moving when standing still. Compare the Etrex 20 vs the Dakota 10 in the test

 

 

Both units sitting still next to each other. Notice the Dakota 10 knew it was sitting still except for a few seconds. The etrex 20 thought it was moving for several minutes. The

Edited April 7, 2013 by gpsblake

[wmarek74]

@gpsblake

Let me ask Ya a few questions:

1. Are ya sure that etrex did not loose fix any time during the walk; no "weak signal"???

2. When ya performed reset: "settings:zero"-after fix or before sattelites` fix?

3. Did ya turn on Glonass in eTrex?

4. What was the accuracy of gps(and glonass) signal during test?

 

Remember: after loosing fix (weak signal; searching for satellites) eTrex and Dakota behaves differently.

Dakota's trip computer works increasing "stopped time"; and registers track points all the time when no fix.

ETrex behaves differently. eTrex's trip computer when "no fix or weak signal" stop working, but... after getting fix it adds this time to overall "stopped time" ( it adds this time gap in traclog too).

[wmarek74]

And one question to 2nd test (sitting still): did ya use Glonass in eTrex 20?

 

What was the GPS accuracy? (how many sattelites were visible by eTrex 20 during Your test?)

IMO, GPS accuracy matters a lot...

Edited April 8, 2013 by wmarek74

[gpsblake]

For the unit standing still, yea, I tried GPS, WAAS/EGNOS on and off, and GLONASS on/off. I run the Dakota with WAAS on and since it knows it is standing still, I didn't bother changing the settings. When I was walking, I had GPS + WAAS/EGNOS on the Etrex 20. I'm convinced it's a 2.9 regression bug with the walking data,

 

I did return the Etrex 20 today, in part because of the errors, and in part because they (the retailer) sold me a unit that someone had already registered and I had to go through Garmin tech to get that fixed.

 

I picked up an Etrex 10, as the Dakota will remain my bicycling unit although the display on a bike is nowhere near as good as a Etrex 20 or Venture HC when viewing from the handlebars. The 10 will be my geocaching unit (as I like to do both bicycling and geocaching, I use two units at the same time)

 

The Etrex 10 does not suffer from the roaming that the Etrex 20 has. In a 10 minute standstill it only thought it had moved for 14 seconds and 34 feet. I have only geocached with the 10, I have not done a viable walking test for avg MPH, accuracy via adjusted benchmarks and other stuff.

 

Anyhow, thanks for hearing me out complaining about my Etrex 20 (and Magellan units in the past). I did notify Garmin of my findings.

[wmarek74]

@gpsblake

Could you drop a hint about GPS accuracy during your tests?

I've never got such bad results as you (but I always had clear sky above; GPS accuracy = 3-6 meters).

I use eTrex 30 with newest FW2.90.

[gpsblake]

wmarek74, do you mean accuracy as reported by the unit? The best I got from the Etrex 20 was 11 feet reported accuracy although of course it would bounce around when I was moving, but 11 when it was still in the open. The 20 did really well with adjusted benchmarks with 5 feet or less being the norm with WAAS on which is the best you can get with dd mm.xxx format . However, I mentioned before when I turned on GLONASS, the GPS was off by around 15 feet (give or take a few feet) on adjusted benchmarks. And the 20 (and the 10 also) did great for geocaching in the woods, really sensitive. I haven't run the benchmark test with the Etrex 10 yet.

 

Here's the results of today's bike ride with the Etrex 10 and the Dakota 10

 

 

Really impressive, both within 100 feet of each other on a 14 mile bike ride and only 1 minute of variance on moving time.

 

Maybe I'm just OCDing about the numbers and it's no big deal to anyone but me.

 

[tr_s]

Maybe I'm just OCDing about the numbers and it's no big deal to anyone but me.

 

 

You are not OCDing at all. An outdoor GPS device should calculate output quantities reasonably. If it doesn't, it should just be sent back, as you did.

 

It is clear the Garmin software team has not given the basic neo-eTrex software features (such as calculating distance and recording tracks correctly) proper attention. That's really all there is to it.

[+BlackRose67]

Still on 2.80 and it works perfectly on my Etrex 20, gonna stay with that for now

I reverted from 2.90 back to 2.80 and all of the issues have gone away.

[wmarek74]

BTW, drawbacks of FW 2.80 I know: workless search option: "Where to" - e.g. "Geographic Points" --->>> Left "Menu" key --->>> "Select Map"

And - refers to eTrex 30:

1. enormously high "total descend" and "total ascent" sums of heights in trip computer,

2. no Tempe suport,

3. compass issue.

 

What about tracklog: distance interval?

Does it record tracks with exact values (e.g. 10 metres)?

Edited April 10, 2013 by wmarek74

[wmarek74]

Check these:

- search option: "Where to" - e.g. "Geographic Points" --->>> Left "Menu" key --->>> "Select Map"

and

- tracklog: distance interval. Does it record tracks precisely with exact values (e.g. 10 metres)? In FW 2.90 sometimes does, but sometimes distances between points in tracklog (current.gpx) range between 50-200 metres during walk... I am not Ben Johnson of cos

[wmarek74]

AFAIK "Select map" does work only in FW 2.90.

In other words: function worked (caused no error), but had no real effect in previous version - still all maps were searched (a bug).

 

To be able to tell if search: "Select map" works you need more than one IMG map loaded...

Edited April 10, 2013 by wmarek74

[wmarek74]

I am confused with your answer.

What is the sense of select map function then?

[wmarek74]

I agree: others might have other needs.

[tr_s]

Map select not working, I found this an issue. So tested it out on 2.90. Yes, it "works" now. Still a case of garmin software FAIL:

 

-Cpu cycles are seemingly still wasted on searching all maps on the device. Which means having one map on the device with an "older" not easily searchable database will slow the search down to minutes and render it useless!

 

-Map select can be done first after one initial search has already gone through. Which means waiting for a potentially long period.

 

-Many of the map names on my device are long, and begin with the same letters filling the screen. The device cannot display more than like the 13 initial letters. Hence there is no way of knowing which map is being selected for search other than learning the order of items. Hello???

 

I just cannot fathom how anything like this goes through five minutes of review or in house QC...

[+GeoTrekker26]

 

-Many of the map names on my device are long, and begin with the same letters filling the screen. The device cannot display more than like the 13 initial letters. Hence there is no way of knowing which map is being selected for search other than learning the order of items. Hello???

 

 

Renaming the maps?

[tr_s]

 

Renaming the maps?

 

Yeah, if it were that easy. But thought it was hardcoded in the file? What is displayed isn't the file name, but some internal product name (the name displayed in mapsource)?