Jump to content

Gps Accuracy


Recommended Posts

New eTrex Legend Cx is what I am using now and it has been performing better than the older Legend.

I am located in Western Canada and most of the time it receives 3 metres accuracy but I have been down to 2 metres very often. I am very impressed with the new Legend Cx.

Link to comment

My legend Cx often quotes 2m accuracy, but has gone as low as 1m using the EGNOS system.

 

I'm taking the figures with a pinch of salt though, as the Legend Cx can lose all but 2-3 satellites, and still be claiming 2m accuracy for a couple of minutes before it finally admits that its lost satellite lock, which isn't so good.

 

AK

Link to comment

I assume everyone is referring to EPE. Estimated Position Error which you can set up to display. I have seen as low as 6' EPE on my 60CS with WAAS, but in reading the threads it sounds like the 60CSX, CX and the 76CSXm CX are more conservative in their EPE than my 60CS. In other words whatever they report as their EPE is a larger number than on my 60CS even though in reality they are much closer to being right on. I would not base to much trust on these numbers anyway except for my 60CS in heavy tree cover. When it reports poor EPE it really means it. :(

Link to comment

Folks, the accuracy of the satellites is 3 meters (or about 9ft). Because of the mathematical nature of the calculations using multiple overlapping circles, this can be effectively trimmed to somewhere between 6-7ft (closer to 7) if you're getting the full 11 or so satellites with WAAS that your GPSr can see at any given time.

 

So, anything which is telling you "better" than that is flat out wrong.

Link to comment

Remember, the EPE is sort of a "best guess" on the part of the GPSr, and it has to do with the statistics of a series of readings. My Magellan seems to be very conservative, and the true position (based on NGS benchmarks) is invariably better than the EPE. The lowest EPE I've seen is 7'. It seems to read in increments of 3', so I never see 4', 5', or 6', only 7'. I think it flashed 3' a few times, but not for any sustained length of time. IMO, it's easy to be within a couple feet if WAAS is working, which it hasn't been here for quite a while. Without WAAS, a 10' diameter circle is about as good as I can do under the best conditions. When WAAS was working, out in the open, a 4' diameter circle was common performance. Using a program like SA Watch to average the data and throw out the fliers and high EPE stuff, you can easily get within a foot, if you're willing to collect data for a long enough period of time. Unless you can collect for the better part of a day, averaging is of far less value then people seem to believe. There may be long stretches of time when the position is consistantly wrong, and averaging it just yields a very high resolution, but wrong, answer.

Link to comment
Folks, the accuracy of the satellites is 3 meters (or about 9ft). Because of the mathematical nature of the calculations using multiple overlapping circles, this can be effectively trimmed to somewhere between 6-7ft (closer to 7) if you're getting the full 11 or so satellites with WAAS that your GPSr can see at any given time.

 

So, anything which is telling you "better" than that is flat out wrong.

 

 

Since Trimbles get sub-foot accuracy, I am not sure how your above statement is justified.

Link to comment

Folks, the accuracy of the satellites is 3 meters (or about 9ft). Because of the mathematical nature of the calculations using multiple overlapping circles, this can be effectively trimmed to somewhere between 6-7ft (closer to 7) if you're getting the full 11 or so satellites with WAAS that your GPSr can see at any given time.

 

So, anything which is telling you "better" than that is flat out wrong.

 

Hmmm... So that surveyor I was talking to, who told me he was out over the weekend and getting 3 centimeter error, is full of it?

 

The difference is, he was using a PROFESSIONAL TOOL (receiver), and not a CONSUMER TOY (receiver).

 

They both use the same satellites.

Link to comment

Folks, the accuracy of the satellites is 3 meters (or about 9ft). Because of the mathematical nature of the calculations using multiple overlapping circles, this can be effectively trimmed to somewhere between 6-7ft (closer to 7) if you're getting the full 11 or so satellites with WAAS that your GPSr can see at any given time.

 

So, anything which is telling you "better" than that is flat out wrong.

 

Hmmm... So that surveyor I was talking to, who told me he was out over the weekend and getting 3 centimeter error, is full of it?

 

The difference is, he was using a PROFESSIONAL TOOL (receiver), and not a CONSUMER TOY (receiver).

 

They both use the same satellites.

 

the original post wasnt totally incorrect. professional gear such as trimble, sokkia etc uses more than just the sat signals we use in our handheld for position fixes of acceptable accuracy/precision. the professional tool, im unsure of model you reference, is more than just a reciever--it also transmits from base station to roamers for corrections, alternatively the corrections can legimately be applied after the fact provided only one instrument setup is used. while i havent used every piece of gear, i do believe there are units using the p codes previously only used by the amercian(correction anyone?) military.

 

as for the epe reding, it is a pretty useless number, it is only an estimate of how much error the gps thinks it picked up during its calculations and modeling of datums, and does not reflect conditions, sat errors etc.

Link to comment

Folks, the accuracy of the satellites is 3 meters (or about 9ft). Because of the mathematical nature of the calculations using multiple overlapping circles, this can be effectively trimmed to somewhere between 6-7ft (closer to 7) if you're getting the full 11 or so satellites with WAAS that your GPSr can see at any given time.

 

So, anything which is telling you "better" than that is flat out wrong.

 

I am under the impression that the spheres (3 satellites and 1 earth) do not meet at one point because of factors like atmospheric distubances, signal reflection from buildings/trees/rocks/etc, the internal clock of the GPSr not being exact to the nanosecond, incorrect almanacs, etc. If I am not mistaken, the calculations are not the problem but just the factors that influence it. If we got rid of all the things throwing the calculations off, it could tell us our exact place on earth. Probably there is some function in the GPSr with plus or minuses thrown everywhere to account for the errors it finds. Standing in an open field and a link to all the satelites and WAAS available, I wouldn't be suprised if I saw 1m accuracy. But I can only dream... best I seen is 6 feet :o

 

due to the precision (at best using utm its 1m in either direction, waas or not) gps cannot provide a fix better than about 2m. the errors you mention arent accounted(save for some atmospherice disturbances), even with waas corrections, for in calculation and further degrade accuracy. i would view any greater accuracy estimate as rubbish. as for the calculations, id think that due to the complexity of code used in our gps(being somewhat simplified) that this number would be slightly greater(~3-4m) as a best case accuracy when compared to survey data. this accuracy would also be further degraded when the calculations for datums are introduced.

 

can anyone prove this wrong with other than ancedotal evidence?

Link to comment

It may depend on where you live, and it certainly depends on the number of satellites you can lock on at one time, and WAAS is absolutely necessary if you're going for maximum accuracy, but yes, you can do better than a few meters. All I do is hunt benchmarks, and plus or minus 1m accuracy under those conditions is the rule, not the exception. This was for upstate NY back when we had a good WAAS lock available. The new sats are coming in, and hopefully the situation will improve later in the summer. The arguments as to why this isn't possible don't seem to be taking into account the averaging effect you get when locking to a large number of sats. Agreed that three will be terrible, but eight or more can be remarkably good.

Link to comment

Regarding the EPE or "accuracy" reported on the GPSrs: I believe most of them report 7 feet or 2M at the low end. My Explorist seems more conservative than my Meridian, but when it has been on for a while with WAAS, it can drop down, often to 10 or 7 feet. This is especially true when it is reporting reading 11 to 14 satellites, including one or both WAAS.

 

As for comparing the GPSr reading to a known point, I have done this many times, with several different brands of GPSrs at super-accurate adjusted benchmarks here in USA. These are mostly in good open locations. They are listed to the one-hundred thousanth of a second (yes, I round off and calculate the correct spot near the disk.) I hold over the spot and read how far to the rounded waypoint I entered.

 

I do not have statistics, just notes jotted down on the NGS Data Sheets. I have found the distance I am away, to be consistently within the EPEs ("accuracy") reported on the GPSr (EPEs usually from 7 to 16 feet.) This is pretty good considering that each one-thousandth of a minute is about 5 to 6 feet, and your GPSr is rounding to the nearest. I recommend trying this to get a feel for how accurate you are with different reported EPEs.

Link to comment

does anyone know why it takes the EPE number so long to drop when WAAS corrections are being used by the satellites? I used an etrex legend today and noticed that after about 8 satellites got 'D's, the EPE didnt drop down for about 5 minutes after. (in other words, it stayed at about 20 feet, then 5 minutes later, dropped to about 8 feet).

 

just wondering why there is a delay. is there more to be downloaded from the WAAS satellites even after the D's appear?

Edited by Tahoe Skier5000
Link to comment

does anyone know why it takes the EPE number so long to drop when WAAS corrections are being used by the satellites? I used an etrex legend today and noticed that after about 8 satellites got 'D's, the EPE didnt drop down for about 5 minutes after. (in other words, it stayed at about 20 feet, then 5 minutes later, dropped to about 8 feet).

 

just wondering why there is a delay. is there more to be downloaded from the WAAS satellites even after the D's appear?

In five minutes the geometry of the received satellites could had changed enough to affect the estimated uncertainty. Also fresher WAAS data could have been in the process of downloading during the interval. Finally, are you sure the satellties you were receiving and their WAAS status did not change during that time?

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...
×
×
  • Create New...