GPSr jinx spots?

[Papa Bear_Left]

There's a waypoint in the B.E.A.R. multicache I placed here that has me very confused.

I've been there on at least 6 occasions, taking over a dozen readings with both GPRSrs, an eTrex and a Meridian, and the coords have always been within the EPE. The South reading in particular has been amazingly consistent, with only a couple of variant readings, and them within .002 minutes.

However, everyone who's done it so far has been 40m south of the waypoint. Consistently. When I went to the point that one cacher photographed for me as his GZ, both GPSrs pointed to GZ as being 40m north...

Anyone have similar experiences or, better yet, an explanation?

(This made me paranoid enough to get a local cacher to check out my next three caches before I released them, just in case! They were fine, and our finding of other people's coords is OK too!)

[Bronze]

I am very paranoid of this. Read the logs of "Noonans Discovery" to know more. I have had two Magellan 310's and both have consistantly developed errors from 310 meters to 25 meters. One to the West and the other to the East.

I now try to take two eTrexs and average the two for about ten minutes per waypoint. I have found with the eTrex that if you mark a point it is sometimes off quite a bit but if you read 'location' in the nav screen and re-adjust the DMS then it becomes much closer. Often I've marked a point and then chosen goto and had it 40 meters out.

Iorn stone could be a contributer. If the surrounding rocks are very heavy or extremely red incolour, like ours, then there could be that but at 6 feet I doubt any magnetism would effect a GPS signal / reading. Yesterday I was carrying Iornstone conglometate and didn't think there would be any effect on the GPSr but it actually did throw out the directional arrow quite a bit.

[The Spindoctors]

Iron shouldn't affect GPS. More likely that the errors occur from reflection off large solid objects. This is something that often occurs in cities and the GPSr needs to take this into account. Is there a significant difference in the antennas of each of the GPSr units?

I'm paranoid about waypoints, but after reading this I'm more so:? . One person's website actually recommends taking a number of readings during several days to eliminate error. Wow! You really need to put in the hard yards.

[maccamob]

I agree multipath can be a real problem. I have a meridian, which automatically averages the coords while you stand still. I've found that several 3-minute averages, turning the GPSr off and on between each one, are usually better than one long average. As I do the second and subsequent averages, I use the first in a Go To so I can see immediately what sort of accuracy I'm getting. The bigger the distance to the first reading, the more readings I take, then average all those readings manually. Sometimes, if you look at the various readings, you can see that most form a group with maybe one or two that are well off the rest. When that happens, you can ignore the wild readings.<p>

In extreme cases, I will try to come back on another day, but that isn't always possible. An option that sometimes works if multipath is a problem is to move a bit - sometimes just a couple of metres is enough - take the coords at the new spot and then use the bearing and distance to the cache to project the cache waypoint from the 'offset' spot. It helps to have a compass and to know the length of your paces to do that.

[Kerry]

Doubt that multipath would account for the differences mentioned and secondly trying to prove it if it is, especially with a handheld. Maybe looking at the constellation/geometry & actual taking into account obstructions etc with respect to what things should be like might give a clue.

So who actually thinks averaging improves/verifies position solutions

Cheers, Kerry.