Jump to content

How To Get Exact Coords When Placing A Cache


Recommended Posts

so i was reading one topic in the forum and it got me thinking about the 10 caches that me and my friend placed...we use two gpsr to get the right spot of the cache...we then walk away and set our gpsr to find the cache and walk back to it and were off maybe 10 feet. now if you look at your gps it says +- so many feet...do people not take this into context when looking for a cache. we try really hard to get the coords as close to zero for everyone as possible...does anyone have a technique that is goofproof for placing caches right and when people search they get the coords to be 0 feet away. i want to be as perfect in placing my caches as possible yet trees and clouds factor into it also...so any good ideas.

Link to comment
so i was reading one topic in the forum and it got me thinking about the 10 caches that me and my friend placed...we use two gpsr to get the right spot of the cache...we then walk away and set our gpsr to find the cache and walk back to it and were off maybe 10 feet. now if you look at your gps it says +- so many feet...do people not take this into context when looking for a cache. we try really hard to get the coords as close to zero for everyone as possible...does anyone have a technique that is goofproof for placing caches right and when people search they get the coords to be 0 feet away. i want to be as perfect in placing my caches as possible yet trees and clouds factor into it also...so any good ideas.

You could buy one of these. Its got Sub foot (30 cm) accuracy.

Link to comment

…Or you might get lucky.

 

Seriously, there is no goof proof method; even smart bombs make mistakes. I usually mark the coordinates of the exact cache location about a half dozen times and then come back on another day to determine which one is the most accurate.

 

But even that is not a true check because your accuracy may be worse than the day you marked the coordinates.

 

Lot’s of people probably average their coordinates in one way or another, over varying amounts of time. Possibly the best way is to be attentive to the logs of a hidden cache and make adjustments accordingly. This way you would get opinions in a variety of conditions from users of different GPSrs. :o<_<

Link to comment

The only way to get perfect coordinates is total luck and /or averaging your coordinates for several hours over a period of several days. Even if you do wind up with perfect coordinates, is it worth all the effort? All but the most new geocachers assume the cache will not be exactly at the coordinates anyway and start searching when they're 40-50 feet away.

 

Personally I'm happy if I take one reading that is 10-20 feet off. I really don't see the need for accuracy beyond that. I don't think other geocachers expect it and I know I don't expect it from others. I'd rather spend my time doing things other than standing at a cache site for hours holding a GPS.

 

I've placed 130 some caches and have received complaints about the coords of maybe 5 (in no case did they ever prevent the person from finding the cache) and I consistently receive praise about the accuracy of my coordinates. All I do is make sure I have a solid signal, let the GPS settle at the cache site for a minute or two, mark and I'm on my way. That's all you really need to do, but if you want to do more, enjoy!

Link to comment

I take no less than five (5) different measurements on separate days. This, of course, means multiple trips to the same spot, so if your hide location is difficult to get to, then this technique may not be for you.

 

After I have all of my measurements, I list all of the N/S and then the E/W coordinates. For each set I then throw out the one number that is most different from the rest (the outlier) and take a simple average of the remaining set.

 

This technique can bring your accuracy down to within three feet. Of course, the finder will still suffer the usual 10-20 foot margin of error, but at least you have the knowledge that they are not under a 20-40 foot radiius error generated by the combined errors of your GPSr + the finders.

 

It works for me. I hope that that helps you.

Link to comment

Also check what your EPE (Estimated Positional Error) is reading.

 

Altought EPE is not a guarantee of maximum position error, it's only estimated, if it gives you an 20 meters error or more you can figure that is not the best place of moment for getting accurate coordinates.

Edited by garri
Link to comment
...set our gpsr to find the cache and walk back to it and were off maybe 10 feet.

Well, if you are getting a repeatability of around 10' that's pretty good.

 

If you're using a SporTrak, here is an easy way to get decent coords.

 

What most people don't realize is GPS coords drift around. If you think about the signal from the satellites are getting to your unit in what appears to it as a "slow motion heat haze" then you can understand how the GPS can be off a little.

 

An enlightening experiment is to find some place, like your back yard or even inside your house if your unit is up to it, and mark a waypoint. Do a goto and then, without ever even moving the unit, watch the waypoint drift around you. I liken it to a bouy with an elastic tether drifting in random breezes.

 

Once people learn you can't just walk up to a spot and have it zero out right then, then they start to understand it might take a while. I usually leave my unit stationary while I look for the cache in the bovious spots. I'll come back and check it every so often to see where it's pointing and continue my search until I find it.

 

Hoep this helps.

Link to comment

It's easy, just make sure your GPS has been locked on to at least four sattelites for a minimum of 15 minutes when you take your reading. I suspect some of the caches that have started out with really bad coordinates were taken by people that placed the cache and then turned on their GPS and took a readig as soon as their GPS had a lock.

 

I am ont a big fan of tacking lots of readings and then picking an average from those. Almost every cache I have found were the cache owner states they took a bunch of readings and avreaged them has had poor coordinates.

Link to comment

I can vouch for letting the GPSr sit. Every so often I find a cache that is nowhere near the posted coordinates. I then grumble about the cache owner and sign the log, look through the trade items, maybe exchange some stuff. Then after the 5-10 minutes I'm at the cache, I pick up the GPSr and what do you know, the coordinates are spot on. You just have to let the receiver settle in for a while. But there was that time I placed a cache and the coordinates were like 30ft off even though the unit had been on for like 3 hours and was held at the cache for 15 mins. I think the culpret was low batteries that time. So the key seems to have juiced-up batteries, GPSr on for a while, and wait 10-15 mins at the cache site to take coordinates.

Link to comment
so i was reading one topic in the forum and it got me thinking about the 10 caches that me and my friend placed...we use two gpsr to get the right spot of the cache...we then walk away and set our gpsr to find the cache and walk back to it and were off maybe 10 feet. now if you look at your gps it says +- so many feet...do people not take this into context when looking for a cache. we try really hard to get the coords as close to zero for everyone as possible...does anyone have a technique that is goofproof for placing caches right and when people search they get the coords to be 0 feet away. i want to be as perfect in placing my caches as possible yet trees and clouds factor into it also...so any good ideas.

The thing is, you need to understand what your GPS/r is telling you.

 

Garmin uses EPE - which is Estimated Potential Error. Notice that it is not potential error - but estimated potential. If it was potential error, then you could say, for example, that you are within 15' of the spot or 10' or whatever your GPS/r reads.

 

You cannot do that, because not only do you have a potential error, but you are also estimating just how big that potential error is. Garmin's is about a 50% probability that the ACTUAL error is LARGER than estimated. (and conversely 50% error that it’s smaller) So what does this mean? It means letting your Garmin settle down and read an EPE of 10' mean that the unit it telling you the real spot has a 50/50 chance of being more or less than 10' away. There is no way to be 100% sure that you within a certain radius of error. However, if you multiply Garmin's EPE number by 2, it will put you in the 95% confidence level. So if your EPE reads 10', you can be 95% certain that you are no more than 20' away. Any people wonder why they can be hard to find :D

 

Magellan’s EPE numbers are even less confident, and Lowrence’s are slightly better. This in no way is a bearing on the accuracy of the unit; it’s just the convention the designers decided to use. When they program the unit, they decide what confidence level to report the EPE at, that’s all.

 

So, what does this all mean? Not much; just let the unit settle, get your EPE as low as possible, and realize that it could still be off, especially when you factor in the error on the finders GPS/r

 

-dave

Link to comment
After I have all of my measurements, I list all of the N/S and then the E/W coordinates. For each set I then throw out the one number that is most different from the rest (the outlier) and take a simple average of the remaining set

 

How do you know that the outlier isn't the correct one?

It's called "statistics."

 

A branch of mathematics that deals with measurement uncertainty and probability.

 

He doesn't know that somehow the outlier isn't closer to the correct point, but the probability that it is is significantly reduced by his method. Likewise, his method almost always produces coordinates that are better than simply taking a single measurement, but (because of unavoidable uncertainties) it can never be guaranteed to do so.

 

But getting better coordinates 999 times out of 1000 is good enough for me.

Link to comment

I agree with the comment that if you get with in 10 feet it is good. It is like hand grenades or horse shoes, if you get it close it is effective. :D

 

When I hide a cache, I usually find a good hiding place :D and then take a reading. If terrain and vegitation lets me I walk 50 ft from the site. I will do this four times in each direction of the compass. When I return to the cache site I take reading. So when I am done I have 5 readings. I average the five and that is coordinates I use. :D

 

Half fun is finding the hiding place. :D

Edited by ColMorgan
Link to comment
After I have all of my measurements, I list all of the N/S and then the E/W coordinates. For each set I then throw out the one number that is most different from the rest (the outlier) and take a simple average of the remaining set

How do you know that the outlier isn't the correct one?

It's called "statistics."

 

A branch of mathematics that deals with measurement uncertainty and probability.

 

He doesn't know that somehow the outlier isn't closer to the correct point, but the probability that it is is significantly reduced by his method. Likewise, his method almost always produces coordinates that are better than simply taking a single measurement, but (because of unavoidable uncertainties) it can never be guaranteed to do so.

 

But getting better coordinates 999 times out of 1000 is good enough for me.

Recently I placed a cache and took one reading, then another, and finally one more.

 

After hiking the trail for more than an hour, I let the GPSr lead me back to my cache. One of the readings was off by more than 30 feet. :D

 

I took another reading and averaged the close numbers. A recent finder said they found the cache "right on the numbers." :D

Link to comment

I have seen good results from coordinates obtained by using the averaging function of the GPSr unit. (NOTE: This is different from taking bunch of readings then averaging all of them yourself)

 

I try to double check my coordinates by using different measuring techniques, since my eTrex Summit doesn't have the averaging function, and my Magellan GPS Companion takes time to settle. For example, I stand still at the hide location for a long time (more than 3 minutes) and monitor the EPE in the meantime. When the coordinates stop drifting, I consider that my first good sample.

 

When the environment is bad for reception (tree cover, hillside, walls, buildings, metal structures), then I try to locate better areas nearby, hopefully directly North/South or West/East. I then use my compass (calibrated) and take latitude and longitude readings separately. I walk to and away from the hide location in North/South and East/West direction and watch for any drifting on latitude or longitude separately.

 

I also do a final check by standing at random locations away from the hide location and see if the arrow will consistently point toward it. The process of doing all this should take quite a while, usually more than 10 minutes.

 

Despite all this, I'm still not completely confident of the accuracy. There are times where the environment is harsh enough that all that effort is futile. (Good time to consider using more explicit hints, or converting the cache to a Letterbox Hybrid)

 

There are people who know how to post good coordinates consistently, fizzymagic being one such person. Some of his hides are quite evil, but he has been considerate enough to post spot-on coordinates (GPSr will read within 6' if you know how to use it) to spare me from additional headaches.

Edited by budd-rdc
Link to comment
. . . Despite all this, I'm still not completely confident of the accuracy. There are times where the environment is harsh enough that all that effort is futile. (Good time to consider using more explicit hints, or converting the cache to a Letterbox Hybrid) . . .

One day the accuracy of my GPSr was 35'. :D

 

I mentioned that in the cache description and also left a very detailed hint.

 

No one has posted corrected coordinates for that cache. :D

Link to comment
After I have all of my measurements, I list all of the N/S and then the E/W coordinates. For each set I then throw out the one number that is most different from the rest (the outlier) and take a simple average of the remaining set

 

How do you know that the outlier isn't the correct one?

It's called "statistics."

 

A branch of mathematics that deals with measurement uncertainty and probability.

 

He doesn't know that somehow the outlier isn't closer to the correct point, but the probability that it is is significantly reduced by his method. Likewise, his method almost always produces coordinates that are better than simply taking a single measurement, but (because of unavoidable uncertainties) it can never be guaranteed to do so.

 

But getting better coordinates 999 times out of 1000 is good enough for me.

If you're talking 1 outlier out of 100, or 1,000, I'd agree with you. This guy is taking as few as 5 readings and throwing away the "outlier". And there is no guarantee he only has one.

 

Still, it all sounds like a lot of work for an additional 5-10 feet of accuracy.

Link to comment
If you're talking 1 outlier out of 100, or 1,000, I'd agree with you. This guy is taking as few as 5 readings and throwing away the "outlier".

I'm not.

 

Statistics is not easy, but there are many relatively accessible books out there that can explain some of the concepts, many of which are not intuitive.

 

Making a few assumptions about independence and normality, the probability that a single outlier (outside the 90% EPE distance) from 5 measurements is closest to the correct coordinates is (very) roughly 1 in 100 million. A more realistic estimate that includes correlation and systematic errors would put it closer to the 1 in 1000 that I quoted.

 

Like I said, statistics is not always intuitive.

Link to comment

I have a Garmin GPSMap 76 w/ an average feature. I also use an external antenna. The 27db gain antenna picks up a signial in the thickest cover, as if you were on water. I usually average for about 5 minutes and untill the coords have not changed for 60 seconds. This always works great!

 

There are several other garmin units that have this feature. Check your manual for the exact procedure.

Link to comment
After I have all of my measurements, I list all of the N/S and then the E/W coordinates. For each set I then throw out the one number that is most different from the rest (the outlier) and take a simple average of the remaining set

 

How do you know that the outlier isn't the correct one?

I don't.

 

But I am fairly confident that it is not. Look at it this way, if four of the five readings are within 2 of a center point ex: 45,49,47,48 - then the oddball fifth one that reads 39 is probably going to be off. If I were to include it in my averaging, then it will throw the average off sufficiently that the actual coordinates almost certianly will be incorrect.

 

I go to this trouble to aid finders. These units have enough error for a finder. The geocacher looking for my hide is relying on a single set of readings (the ones on his/her GPSr), which have by themselves, a significant probability of error. What I am trying to do is eliminate (or at least significantly reduce) the cumalative error from both mine and their GPSr being off.

 

My hides tend to be well camoflauged, so for me, the real trick is in getting a hunter to a fairly narrow radiius of the geocache, and then still providing them with an opportunity to think creatively.

Link to comment

After gaining confidence in my GPS, I just let the GPS sit a few minutes and take a reading. I walk away (50-100 feet in the direction of the likely approach). Then I program in the coords, then follow the arrow as if I was seekeng the cache. If it gets me reasonably close, then I am fine.

 

I haven't go any complaints since I started that method.

Link to comment
After I have all of my measurements, I list all of the N/S and then the E/W coordinates. For each set I then throw out the one number that is most different from the rest (the outlier) and take a simple average of the remaining set

How do you know that the outlier isn't the correct one?

It's called "statistics."

 

A branch of mathematics that deals with measurement uncertainty and probability.

 

He doesn't know that somehow the outlier isn't closer to the correct point, but the probability that it is is significantly reduced by his method. Likewise, his method almost always produces coordinates that are better than simply taking a single measurement, but (because of unavoidable uncertainties) it can never be guaranteed to do so.

 

But getting better coordinates 999 times out of 1000 is good enough for me.

Recently I placed a cache and took one reading, then another, and finally one more.

 

After hiking the trail for more than an hour, I let the GPSr lead me back to my cache. One of the readings was off by more than 30 feet. B)

 

I took another reading and averaged the close numbers. A recent finder said they found the cache "right on the numbers." B)

Think about that for a moment. If your readings were a bit here and a bit there, then if another cacher takes one reading and says your average is "right on", it's "right on" what? His reading is likely off as well. But if his was "right on", then why not use his unit to take the measurement in the first place???

 

I have found that the units tend to wander around a bit. When I am hunting a cache I try to follow this wander a bit to get a feel for where the center of the wander is. As the GPSr wanders, it is likley to take you to any number of spots near the cache. When it takes you near enough to the cache that you find it you say, "Gee those coords were right on!" In reality they were right only at one moment in time and the average can easily be somewhere else.

 

I try to take averages, most importantly over several days since the satellite position can have a significant impact on the reading. I verify by doing a search with the coordinates. When I am searching a cache, I assume the coordinates are up to 20 feet off plus 10 foot error on my GPS once it circling a point. This is normally good enough to find the cache.

Link to comment
You cannot do that, because not only do you have a potential error, but you are also estimating just how big that potential error is.  Garmin's is about a 50% probability that the ACTUAL error is LARGER than estimated. (and conversely 50% error that it’s smaller) So what does this mean?  It means letting your Garmin settle down and read an EPE of 10' mean that the unit it telling you the real spot has a 50/50 chance of being more or less than 10' away.  There is no way to be 100% sure that you within a certain radius of error.  However, if you multiply Garmin's EPE number by 2, it will put you in the 95% confidence level.  So if your EPE reads 10', you can be 95% certain that you are no more than 20' away.  Any people wonder why they can be hard to find  B)

 

Magellan’s EPE numbers are even less confident, and Lowrence’s are slightly better.  This in no way is a bearing on the accuracy of the unit; it’s just the convention the designers decided to use.  When they program the unit, they decide what confidence level to report the EPE at, that’s all.

 

-dave

How do you know the probability that these three manufacturers use in generating their EPEs? I find it odd, but plausible that a 50/50 threshold woudl be used. Normally 90% or 95% confidence is used since that is a reasonably high value. A 50/50 confidence that you measurement is inside this distance is not a very useful confidence level. To say a manufacturer uses an even lower confidence level than 50% to establish an EPE distance sounds pretty silly.

 

Think about it, what do you do with a reading that is 50% or less likely to be right???

Link to comment
. . . I try to take averages, most importantly over several days since the satellite position can have a significant impact on the reading. . . .

Several of my caches are out in the back-country at the end of a long drive and a fairly long walk. I can't take readings on different days . . . it would cost too much in gas, time, and physical energy. B)

 

Only once was I with someone else with a second GPSr at the time I placed a cache.

 

Taking several readings, approaching the cache location from different directions, has worked very well for me so far. I've never had anyone suggest different coordinates, and it sure is nice when someone does say they found the cache "right on the numbers." Makes me wonder if they also have a Garmin Vista C . . . B)

Link to comment
You cannot do that, because not only do you have a potential error, but you are also estimating just how big that potential error is.  Garmin's is about a 50% probability that the ACTUAL error is LARGER than estimated. (and conversely 50% error that it’s smaller) So what does this mean?  It means letting your Garmin settle down and read an EPE of 10' mean that the unit it telling you the real spot has a 50/50 chance of being more or less than 10' away.  There is no way to be 100% sure that you within a certain radius of error.  However, if you multiply Garmin's EPE number by 2, it will put you in the 95% confidence level.  So if your EPE reads 10', you can be 95% certain that you are no more than 20' away.  Any people wonder why they can be hard to find  :(

 

Magellan’s EPE numbers are even less confident, and Lowrence’s are slightly better.  This in no way is a bearing on the accuracy of the unit; it’s just the convention the designers decided to use.  When they program the unit, they decide what confidence level to report the EPE at, that’s all.

 

-dave

How do you know the probability that these three manufacturers use in generating their EPEs? I find it odd, but plausible that a 50/50 threshold woudl be used. Normally 90% or 95% confidence is used since that is a reasonably high value. A 50/50 confidence that you measurement is inside this distance is not a very useful confidence level. To say a manufacturer uses an even lower confidence level than 50% to establish an EPE distance sounds pretty silly.

 

Think about it, what do you do with a reading that is 50% or less likely to be right???

 

 

 

Here is some information in a concise spot - Error Measures: What do they mean this will explain to you confidence levels and a rudimentary understanding of what sigma is in accuracy measurment.

 

This GPS Accuracy is a little more in depth, but is not that complicated. The author is very straightfoward and his writing is easy to understand

 

GPS Information . net has a lot of good info on it as well. They did get a response from Garmin engineering on the accuracy of the EPE - and here is the response

 

The EPE is an estimation based upon the information the receiver

can determine. SA consists of artificial clock errors and

artificial ephemeral errors. Both of these effects, as well as

atmospheric effects, can result in a positional area of uncer-

tainty, which can be measured and will add to the receivers EPE.

Bias errors cannot be measured and will typically not be detected

in the EPE calculation.

 

The 12XL will typically have a better EPE than other units due to

the 12 channel correlator and the use of all tracked satellites

in the positional computation.

 

EPE is an estimation, rather than a measurement, but all

measurable factors are used in the estimation algorithm. We

consider the details of our EPE and FOM calculations proprietary.

 

We calculate EPE our own way. URE and HDOP are definitely

significant factors in the calculation. We calculate an over-

determined solution, and fully understand the characteristics of

SA, and are able (in our opinion) to provide for a better

estimate of current position error than the simplistic calcula-

tions will indicate.

 

Many folks have and will demand to know our specific calcula-

tions, but we consider these to be proprietary and we do not

release the specific formulas. This is similar in our FOM

calculation, we use a lot of finesse in our software which other

manufacturers have not been able to duplicate. This is further

evidenced by Dr. Wilson's reports on our accuracy compared to

other receivers. If the tests were performed, I believe you would

see closer correlation between our EPE values and actual errors,

as compared to other manufacturers units.

 

Garmin International

 

Some key phrases "EPE is an estimation, rather than a measurement, but all measurable factors are used in the estimation algorithm" and "I believe you would see closer correlation between our EPE values and actual errors"

 

As I said in my original post - EPE is not a error - it is an ESTIMATION of error. Garmin is not going to release its proprietary calculations, but as was alluded to in the response,. tests have been performed (by people with WAAAAY more time than I have) using highly accurate surverying GPS/r units. These tests can be used to calculate the cofidence level that the civilian GPS/r's that we use are operating at.

 

Here is a direct quote from Magellin on its Meridian receivers - from the manual

 

Accuracy

Position 7 meters, 95% 2D RMS

w/WAAS <3 meters, 95% 2D RMS

 

So right off the bat, the receiver itself is only 95% sure its within 7 (or 3) meters at any given time. And this is with ideal sat lock. Again, the EPE that is reported to the user, is a computed value, obtained by propritery calculations.

 

-dave

Link to comment
You cannot do that, because not only do you have a potential error, but you are also estimating just how big that potential error is.  Garmin's is about a 50% probability that the ACTUAL error is LARGER than estimated. (and conversely 50% error that it’s smaller) So what does this mean?  It means letting your Garmin settle down and read an EPE of 10' mean that the unit it telling you the real spot has a 50/50 chance of being more or less than 10' away.  There is no way to be 100% sure that you within a certain radius of error.  However, if you multiply Garmin's EPE number by 2, it will put you in the 95% confidence level.  So if your EPE reads 10', you can be 95% certain that you are no more than 20' away.  Any people wonder why they can be hard to find  :(

 

Magellan’s EPE numbers are even less confident, and Lowrence’s are slightly better.  This in no way is a bearing on the accuracy of the unit; it’s just the convention the designers decided to use.  When they program the unit, they decide what confidence level to report the EPE at, that’s all.

 

-dave

How do you know the probability that these three manufacturers use in generating their EPEs? I find it odd, but plausible that a 50/50 threshold woudl be used. Normally 90% or 95% confidence is used since that is a reasonably high value. A 50/50 confidence that you measurement is inside this distance is not a very useful confidence level. To say a manufacturer uses an even lower confidence level than 50% to establish an EPE distance sounds pretty silly.

 

Think about it, what do you do with a reading that is 50% or less likely to be right???

 

Just somthing else.

 

Its not %50 likely to be right. In fact, its most likely to be less than 1% right. Its 50% likely to be less than or greater than the reported EPE. That is an important difference.

 

-dave

Link to comment

Man, the one thing I love about activities like this is there's people that really enjoy getting into the nitty gritty of this stuff.

 

Personally, I've found that just dropping the GPSr (GPSMAP 60cs) on the spot with the external antenna on it, tossing on the averaging function until it gets down to a decent reading - hopefully 10feet or less (this might not be good enough for those not used to a lot of tree cover). If I'm in really heavy tree cover, I hope for 30 feet and give a hint that would be adequate if someone wants it.

 

edited to say: I do double check coords on maintenance runs to confirm they're within +/- .001-.002 minutes. (approx 6 to 12 feet)

Edited by NomadVW
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...