Determining Accuracy Of Etrex?

[+Chiro75]

Is there a way to figure out how accurate my Garmin Etrex really is? I did a cache a couple days ago and I was pretty much right on top of it when my GPS said I was, but according to the satellite signal page I was around 20-30 feet accurate. Of course, I know that the person who put the cache there to begin with may not have been terribly accurate, either, and maybe I just got lucky on this one, but other than finding a publicly accessible benchmark and walking up to it a bunch of times, is there another way to judge it?

[+briansnat]

Your accuracy is going to vary depending on your signal strength and sat alignment. It may even be affected by the way you hold the unit and where your body is in relation to the unit.

 

If if you went to a benchmark tomorrow and found your unit was accurate to 2 feet, you could possible go back two days later and stand at the same spot and find it to be 30 feet off.

 

So its pretty much a pointless exercize.

[+budd-rdc]

Keep in mind that EPE (Estimated Position Error) is a statistical representation, so it doesn't necessarily mean your coordinates are off by that much.

 

You have to be very observant of the environment when you attempt to check your accuracy, beginning with how many satellites in the almanac, how many your eTrex is locked to, and then check for tree cover, nearby hills, buildings, walls, power lines, etc.

 

You then have to worry about how you are holding the eTrex - it has to be held relatively flat to expose the patch antenna (a flat metal plate) to as much signal as possible.

 

I have visited many benchmarks with coordinates adjusted by survey quality instruments, and in a good environment, I can get readings consistently within EPE every time, so with a bit of knowledge and experience, this isn't a futile exercise.

[+StarBrand]

In very general terms, you will find that your unit will take you within about 20 feet of the cache on a good day - no better. On a bad day add your 25 foot error to the placer's 25 foot error and you could find yourself about 50 foot from the placement. Put the unit away and start looking when it says you are getting that close.

 

The EPE on the sat page is just an estimate of the unit's accuracy given the signal strength, number and alignment of sats it is using for the calculation. The way to read it is: "The position given could be off by as much as xx feet - with 90% confidence - under the current conditions." Just an Estimate.

[+Chiro75]

Okay, thanks!

[+Jeep_Dog]

but other than finding a publicly accessible benchmark and walking up to it a bunch of times, is there another way to judge it?

Also be advised that even publicly accessible benchmarks may not be correct. I have found quite a few of them (never bothered to log them, I suppose one of these days I may get to it, but also have found 9 that do not exist in the database, so I figured "why bother?"), and at least 3 of those had inaccurate coordinates. Each benchmark listing says "Coordinates may not be exact. Altitude is ADJUSTED and location is SCALED."

 

Other than that, the other folks who posted here, unlike most GPS receivers, are "spot on."

[+budd-rdc]

If you are the scientific type and want to look for repeatable measurements, look for benchmarks with the following description under the coordinates:

 

Coordinates may not be exact. Altitude is GPS OBS and location is ADJUSTED.

 

Be sure to look specifically for "location is ADJUSTED" not just "altitude is ADJUSTED".

 

Here's an example: AA1871

 

And be sure to read the Control Text description at the bottom of the benchmark page which describes the posted coordinates in more detail.

 

If you just want to find caches, then you can ignore what I've said, but if you really are curious about learning the behavior of your GPSr, then some of what I mentioned is worth investigating.

[bogleman]

The consumer GPS will get you to the general area - the rest is up to you. You are not surveying a lot, you are looking for a geocache. Far too many issues that can throw the numbers off to get concerned about. I did witness a survey team measure a benchmark. It took them about 1 hour on a clear day on the middle of an open field using $10,000 worth of equipment to get an accurate reading.

[bogleman]

 

from this benchmark

 

 

and this one

 

That is good enough for me. Your Etrex will give the same results as long as you are in an open field with no obstructions.