The Scenario:

Aldot has set pairs of monuments throughout the state with State Plane Coordinates on them. The datasheets for these are available from NGS, so they would have had to set these according to NGS standards if I'm not mistaken.

While tying some of these in a few weeks ago we came across one that would not fit the other control (including some other Aldot control monuments). We went back later and ran 2 sessions for each pair of monuments and a third we set. These are all within 3/4 of a mile, usually less, of each other. WE ran the first session for approx. 20-30 min. Then moved the receivers around to occupy a different point and ran the second session for approx. the same time as the first.

The Problem:
When I processed these, all the vectors passed wonderfully, but failed as repeat vectors. Almost all the difference is in the Y. Should we have spaced our sessions out to different times of the day? Any suggestions why the difference was mostly in the Y? Again the individual vectors all past.

I know I can loosen my tolerances to get them to probably eventually pass, but that's not what I'm looking for.

Thanks
Greg

Modified By Swamp Surveyor on 3/24/2003 at 4:22 PM

I have had similar experiences with vectors which passed, but when I tried to do a loop closure with these vectors, the loop didn't close. I now consider a pass/fail on the vector processing to be just the first rough test, followed up by a loop closure or a repeat as you mentioned.
I also am curious about how I can have a triangle which has two good vectors and the third vector fails.
I know this doesn't help, but at least you know it's not only you who is still trying to learn and understand.

I just ran a loop closure on each of the sessions and 5 out of 6 passed. Other than the 6th blowing up like it did, I'm lost.


Greg

Probably not your problem, but I like to switch to ENU format instead of XYZ. Then look to see if all the error is in U. This is indicative of HI problems. In XYZ errors in up wash into other components.

- jlw

could your tribrach (or pole bubble) be out of calibration?

Jerry
That helped. I switched to ENU like you suggested and almost all the error was in the U. Most of the errors were out 0.2' to 0.4'. I went back and checked my observations, and some of the HI's had been set to slant (they should have been set to vertical). We were usng fixed height 2m poles. When I fixed this, all but 2 came in. The 2 that didn't had a common site, so I know where to look now.

Thanks everybody. I learned something new (besides double checcking my observation data).

Greg