Jump to content

GPS accuracy


Followers 1

Recommended Posts

i was just wondering what is the accuracy usally like on your GPS? mine on good clear skys gets down to in the 30 feet zone. i was just wondering what is the average on most GPS's.

 

If it matters i have an Etrex Legend.

 

Thanks

 

Findandhide

Edited by findandhide
Link to comment

In North Carolina, with a 60cs, I regularly get a displayed accuracy 11 to 12 feet, with gusts to 7-8 feet. But it is even better when using the unit in the Decimal Degree (DDD.DDDDD) mode to take me to a precise spot. The accuracy frequently is within 0.00001 degree.

 

What does that mean, exactly? Well on Tuesday, while searching for EC1620 (a benchmark which had been Not Found for 35 years), I went directly to the main station disk by following the arrow on the 60cs. Then, 89 Meters to the northeast, while looking for the remains of Reference Mark 1, my feet were a few inches from the broken monument when I stood at the predicted coordinates.

 

Paul

Link to comment

B) with clear sky i take 4 m accurasy (no matter urban or field) with cloudy sky 13-26m accurasy.

Exept one time that the GPS was given me 4 m accurate but in the field i was 80 Km far from the spot.

It was the day that ''Shock and Oath'' operation in Gulf war was started.

funny????!!!!

 

sakis

greece

 

ps i have an eTrex Legend

Edited by sakis
Link to comment

Is there any other settings beside the WAAS enabled that i can change to make it more accurate????

 

Not that I've ever come across, in any make of GPS. Keep in mind that the system was only designed for accuracy (IIRC) to 3m with WAAS enabled, which for most navigational applications is incredibly fine-grained.

 

My eTrex routinely reports accuracy of 4-5m outdoors, degrading to 8m in heavy woods. Considering I got it for about 2/3 retail (Ebay is a wonderful thing!), I'm not going to complain for a second!

Link to comment

FindandHide asked:

Is there any other settings beside the WAAS enabled that i can change to make it more accurate????

 

A very effective technique is to watch the coordinates in numeric form, rather than "pointers" such as a compass or arrow. Here's how it works: As you near a target, shift to a screen which displays the coordinates. Move north until you have a match on the "N" coordinates. Then go E or W until you match the "W" figure.

 

The advantage of going for a coordinate match, rather than watching an "arrow" or compass display, is that the unit isn't pointing wildly east, then north, etc. Pause your movement occasionally as you near the target, allowing the unit to stabilize. When the coordinates in the display match the coordinates on the datasheet, you are THERE!

 

Folks who have watched me do this have been amazed at how close it is. Of course, the precision is only as good as the GPS unit, plus the quality of the signal, the lack of reflections from nearby objects, and the number/position of the available satellites. But that's true, no matter what display you are using.

 

You could shift to the Decimal Degrees mode. It does not increase the accuracy, but it takes you one step further into the averaging process. The advantage is that your coordinates will change with a smaller movement of the unit. However, this would be rather time consuming. You would have to convert the cache coordinates to DDD.DDDDD. Easy to do with on-line converters, or a calculator. But a hassle, nevertheless. I use it in extreme cases, such as hunting for a government benchmark that's buried, and the description is 35 years out-of-date. EZ4087 comes to mind.... :)

 

-Paul-

Link to comment

i was just wondering what is the accuracy usally like on your GPS? mine on good clear skys gets down to in the 30 feet zone. i was just wondering what is the average on most GPS's.

 

If it matters i have an Etrex Legend.

 

Thanks

 

Findandhide

 

Keep in mind that the "Accuracy" reported by your GPSr is a relative not absolute value. From what I understand, each manufacturer uses their own, unique algo. for determining accuracy. I suspect the method used may vary from unit to unit as well.

 

The best reported accuracy I have ever seen on my 60CSx is +/- 9ft, which was in Kansas, full slate of sat signals at full strength and WAAS. Where I live in New Jersey, the best I can expect on my CSx is +/-12-14ft; under the trees +/- 18-24 ft. My ol' GPS V will routinely report +/- 8-9ft with WAAS and +/- 6 ft is not unusual.

 

So the V must be the better unit, right?

 

I used both units side by side for a couple of months when I first got my CSx last summer. I used them both to map trails, find geocaches, find benchmarks, mark waypoints at known locations, under all kinds of conditions - clear sky, under trees, blue sky and rain. My conclusion is that under ideal conditions (clear, unobstructed sky, WAAS) both units are equally accurate in that both will put me within 20 ft of the "same" location (at least here in Jersey) but the CSx is much more "reliable" in holding that accuracy as conditions degrade (tree cover, fewer or weaker sat signals ...).

 

Have I had experiences where my GPS has put my foot right on top of what I'm looking for? You bet, and it impresses me everytime. But then my engineer brain kicks in and reminds me that it is just random variability. I tend to remember most vividly the times when the unit put me right on top of a geocache and the times I was 130 ft away, but tend to forget all those times I was 20 ft away from the stump or rock pile.

 

Happy hunting

Link to comment

FindandHide asked:

Is there any other settings beside the WAAS enabled that i can change to make it more accurate????

 

A very effective technique is to watch the coordinates in numeric form, rather than "pointers" such as a compass or arrow. Here's how it works: As you near a target, shift to a screen which displays the coordinates. Move north until you have a match on the "N" coordinates. Then go E or W until you match the "W" figure.

 

The advantage of going for a coordinate match, rather than watching an "arrow" or compass display, is that the unit isn't pointing wildly east, then north, etc. Pause your movement occasionally as you near the target, allowing the unit to stabilize. When the coordinates in the display match the coordinates on the datasheet, you are THERE!

 

Folks who have watched me do this have been amazed at how close it is. Of course, the precision is only as good as the GPS unit, plus the quality of the signal, the lack of reflections from nearby objects, and the number/position of the available satellites. But that's true, no matter what display you are using.

 

You could shift to the Decimal Degrees mode. It does not increase the accuracy, but it takes you one step further into the averaging process. The advantage is that your coordinates will change with a smaller movement of the unit. However, this would be rather time consuming. You would have to convert the cache coordinates to DDD.DDDDD. Easy to do with on-line converters, or a calculator. But a hassle, nevertheless. I use it in extreme cases, such as hunting for a government benchmark that's buried, and the description is 35 years out-of-date. EZ4087 comes to mind.... :D

 

-Paul-

 

Based on the same idea but without the conversion hassle, just use UTM. So you know how many meters you are away, and move accordingly first looking for the correct Northing then Easting.

I too noticed you get better accuracy while moving, that is heading for the spot than staying stationary and figuring out how much and in which direction to move.

Link to comment

This brings up a good question I had that I was going to do a separate forum topic for, but thought it would be good to ask here.

 

During yesterday's sleet/snow storm here in the Northeast I took a 5 block walk from my condo to a convenience store to get a few things and took my GPS (Lowrance iFinder GO) along for curiousity. It had the worst "out in the open" performance I had ever seen. It took a good 10 minutes to get a fix and it tended to bounce my speed and position quite inaccurately (with EPE's in the high 3 and low 4 digits) and kept showing my altitude as about 100 ft. below sea level (I know that altitude is not as accurate as speed and position on most GPSr's, but my neighborhood is about 150 ft. above sea level and it usually shows a reading between 100 and 200 there).

 

I had just used it 3 days ago when it was sunny and 70 deg out with no problem, so I don't think almanac issues came into play. I also had Energizer Lithiums in it yesterday, so I don't think the 25-30 deg temps were the problem either.

 

My best guess was the very thick cloud cover and weather. I know the OP said "on a good clear sky". I have only seen this behavior before in extreme reception conditions (very tall buildings, very heavy tree cover). But I have read lots of articles that seem to suggest that clouds do not affect reception and that the idea of GPSr's not working well in very cloudy weather is a myth.

 

What is the scoop here? Thanks. By the way, to answer the OP question, on 2 cache hunts in the past (both in perfect weather), it did actually (with WAAS) give me an EPE of 1 ft. Without WAAS, the best I have gotten is 18 ft.

Edited by HaLiJuSaPa
Link to comment

Tests of my 60CSX under nearly ideal conditions (unobstructed sky and low PDOP) show that the best reported estimated accuracy was +/- 9 feet. However the actual error (based on a precisely-surveyed well loation) was only 3 feet. I was impressed.

Edited by Glenn W
Link to comment

My Venture Cx has reported accuracy of +/- 5 ft a number of times. However I take that with a grain (or actually a whole shaker full) of salt.

 

Keep in mind that the difference between a single digit precision (ie the difference between N50° 00.000 and N50° 00.001) can be a dozen feet or so depending on your latitude.

 

Also, the algorithm the GPSr uses to estimate the accuracy is based on known factors, like how many satellites the GPSr is currently receiving,

the position of those satellites relative to one another, and how much atmosphere lies between the GPSr and each satellite. Unknown factors, like the reflection of signals off of buildings, or the bending of signals through atmospheric conditions (like large temperate gradients) cannot be accounted for by a normal GPSr.

 

In Amateur Radio there is a phenomenon known as tropospheric ducting, which affects high frequency radio waves that are normally line-of-sight only. That can result in the bending of radio waves so that they stay in the atmosphere and follow the curve of the earth, and in extreme cases allow reception of signals thousands of miles away.

 

So just remember that, on average, your GPSr is probably pretty accurate. However determining exactly when it is being accurate is difficult.

 

Dan

Link to comment

Keep in mind that the difference between a single digit precision (ie the difference between N50° 00.000 and N50° 00.001) can be a dozen feet or so depending on your latitude.

Your principle was correct but your example is not quite. Your latitude (represented by a N or S coordinate) will always represent the same physical distance per change in coordinate unit, no matter where you are (disregarding ellipsoid distortions). Your longitude (E or W) will be variable depending on your latitude (i.e., a 1 degree change in longitude is much shorter in physical distance when you're close to the poles than when you're at the equator). To exaggerate the point, if you were standing at one of the poles, you could walk in a 1-meter diameter circle around the pole and traverse all 360 degrees of longitude. However, latitude 89 (or -89) would still be 111,111 meters away to the south (or north).

Link to comment

I find I typically get reported accuracy of 15-22 feet, degrading to 20 to 30 feet under heavy tree cover with a Garmin 60CS. One thing not mentioned thus far, perhaps because it is "given", is that if you are searching for a cache with the posted GPS coordinates, you are attempting to locate an object at an location of unknown accuracy. You can "prove" this to yourself by simply taking a series of readings at the exact same location with the same GPS oriented the same way at different times or on different days. In other words, the posted coordinates are one of a more or less random series of possible coordinates of unknown accuracy which you are attempting to match. In practice, this means that if both the poster of the cache and the searcher have 20 foot accuracy on their respective machines you are likely to "zero out" anywhere up to 40 feet away from the cache, though generally less.

Another simple test is to mark a given spot and then without moving or changing anything hit "go to" and watch the numbers dance and the arrow move for 15 minutes or so. I believe that the accuracy reading is basically an estimate with "the actual location" being somewhere in a roughly circular formation with a diameter of twice the accuracy: 15 feet means a 30 foot rough circle. You are looking for a spot in another similarly sized and shaped area with no way to estimate the probability of overlap and no reason to expect the object is even in the overlapping areas. "Averaging" the reading doesn't overcome this problem. This is why we need "hints" when looking for micros in the woods...

When benchmarking for an object that has been accurately located, it's a different story, as the area of uncertainty is much smaller since you are trying to match your position with a known one.

edexter

Link to comment

With an old eTrex I've seen anywhere from 14' to 25' accuracy.

 

With an old Legend I would typically get around 16-18' accuracy.

 

With a new 60Cx the usual is about 14'. I've seen it lower, down to 10', but that's ideal conditions. Most of the time I am not using WAAS.

 

Based on mapping various waypoints using coordinates from each unit I would definitely say the 60Cx is the most accurate. (And it better be for how much it cost!) ;)

Link to comment

I’ve checked the “actual” accuracy of my 60CSx against to the “reported accuracy” of the unit. I’ve done this by checking it against several adjusted benchmarks and adjusted USGS locations (water table wells). I’ve found that the “actual” accuracy is approximately 3 times better than the units “reported” accuracy. So if the GPS says 21’ accuracy it is actually more like 7’ or if it says 15’ it’s more like 5’. Now there has to be a limit to this. The lowest reported accuracy I’ve seen was 9’ but never around any adjusted benchmarks. So does that mean 3’ actual accuracy? I don’t think so. I’ve seen anywhere between 5’ – 8’ actual accuracy when checked against adjusted locations.

Link to comment

Well, we have a professional Trimble unit at the college I teach at that is accurate to less than 1 meter. Of course, it costs around $7,000 and has a dedicated antenna that sticks up out of a backpack over your head so it's a little more trouble to carry around than a Vista. Also, it's hard not to attract the attention of muggles. :laughing:

 

Seriously, though, I've never used it for geocaching, :santa: but I think you're seeing a pattern in the replies to your question. Consumer units just don't get more accurate than 3-5 meters and, as others have commented, even when you seem to go right to the spot, it's really just chance that put you there and not 15 feet away.

Link to comment

Well, we have a professional Trimble unit at the college I teach at that is accurate to less than 1 meter.

 

Not sure if already mentioned above, but the cataloged geocache location probably suffers from the same type of error. (I doubt every cache was marked with Timble unit) So you would need to add up your unit error with the survey error. 5-6 meters is all I can count for (in the best conditions) and 10m more often (well, for someone who has found only five caches:) I think it would be less fun if I could pinpoint the cache every time.

Edited by uncle_ziba
Link to comment

Is there any other settings beside the WAAS enabled that i can change to make it more accurate????

 

Not that I've ever come across, in any make of GPS. Keep in mind that the system was only designed for accuracy (IIRC) to 3m with WAAS enabled, which for most navigational applications is incredibly fine-grained.

 

My eTrex routinely reports accuracy of 4-5m outdoors, degrading to 8m in heavy woods. Considering I got it for about 2/3 retail (Ebay is a wonderful thing!), I'm not going to complain for a second!

 

ah ebay :P i recently purchased my garmin 60C off ebay for around 1/2 retail price, very happy :):)

It frequently gets me to within 3-4m at GZ

Link to comment

With my 60Cx, external antenna on a 10' pole, using it at a time when the satellite positions are optimum, and on a clear low humidity day, I generally get around 5' accuracy. That's when comparing it to a known control station. Without external antenna or optimum conditions, generally 15'-18' or so.

 

Optimum times for best satellite positions at a particular location can be determined using the Trimble Planning software available at their site for free.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
Followers 1
×
×
  • Create New...