found this out the other night. Although I haven't actually printed out the "manual", the online help was sketchy.
Grid to ground. Initial processing/adjustment in SPC, then define new ground sytem, pick site to use as origin to make N10,000 E10,000 radio button to grid north and whammo, thats all there is to it. BUT, like the rest of us "real world" Surveyors dealing with small projects and wanting to use "True" North, the trick is to set the radio button on "angle", and the REAL kicker is don't check out the convergence at the selected origin site, just type in 00 degrees 00 minutes 00 seconds and THEN click ok. Seems the software applies the convergence to the coordinates.
A solar observation is left to the student to verify. J.D., that is right up your alley!
Dave

Dave,

Solars are always good to throw in somewhere. I'm a firm believer in that kind of check.

How's the "house cleaning" part of 2.4 look? Tried anything yet? The suspense is killing me.

J.D.

J.D.,

Actually I just finished working with a prior job I had done last week.
Had 2 unknown points about 400' apart, and running 3 receivers, the third was on a known point about 6 miles away. Set up the pair of close ones first, then off to the far one. Observed 2 static sessions at the far one, shut it down, then reobserved the close vector with just the pair.
In the adjustment, I excluded one of the long vectors of the first 3 receiver session, and then the other long vector in the second 3 receiver session. Kept both of the short vectors in the solution. One of the long vectors was "flagged" with a "partial" solution, have yet to find an explanation for the meaning etc. of "partial", so I go and set my elevation mask from the default of 10 degrees to 30 degrees. Re process. "Partial" not noted. But the residuals were still kind of high (0.13' vertical).
Try it again, this time I left out the looooooong vector between the close ones, took both long vectors from both sessions, along with the single vector between the close ones. Much better solution. Residuals around 0.04' vertical.

Few questions after tonights dabbling; perhaps Bill, Bob or Richard can help me out.

What is the correct antenna type for Locus? The software is using type "unknown" right now. There is a drop down list, but not one that simply states "Locus". The radius(?) is listed as 0.099xxxxx meters.

Prior to installing the software, I backed all my jobs up on the zip drive. I did not remove the Locus processor v 1.2, and now if I go and get a job out of 1.2 and work with it in Solutions 2.4, it does not leave a copy that is usable in 1.2...
J.D., if you don't have it by the end of the week, bring stump water!
Dave
Modified By Dave Huff on 8/7/2001 at 1:34 AM

Dave

The old jobs I've looked at with the new version just show the antennae type as unknown (even after reprocessing) but yesterday when I brought in a new job from the recievers it shows antennae type as 800372.

I finally re-installed the new ver over the old Locus directory because I found when I moved my old data dir's to the new Ashtech Solutions dir it would not process the old files (can't find them) mabey from the data in the .spr file

I was hoping to use the new editing tools with the new job yesterday, had some awfull setups, next to buildings and big trees but they all passed QA..
I'm sure I'll get my chance...

Steve G

800872 is Locus, but as long as you are using only Locus receivers, you can leave the antenna type "unknown". This feature is to compensate for differing phase center characteristics when using a mix of 2 or more antenna types.

Steve, the project file (.spr) contains the path to the raw data files, so when you moved the data directories, Solutions was unable to find them. Just reload the raw data files.

Brian

Mine say 800372 ... is that a typo?

I thought the .spr file was sending it the wrong direction... it works well over top the old Locus directory.. Thanks

Steve G

There was a question about what the "Partial Solution" in Solutions 2.40 means.

For any given vector, the Solutions software tries to FIX the ambiguities for all usable ROVER satellites that are in common with the BASE receiver. Fixed ambiguities mean that accurate distances from the ROVER antenna to each GPS satellite have been determined. Once ambiguities are fixed for all ROVER satellites, the highest precision vectors are possible (at the level of a few cm or better). The estimated precision of the vector is shown by the RMS value.

When Solutions 2.40 indicates a Partial Solution for a particular vector, it means that the software has only been able to fix the ambiguities at the ROVER for some of satellites that are in common with the BASE. In other words, say there are 7 usable satellites in common at the BASE and ROVER. In a partial Solution, the fixed ambiguities have not been estimated for all the 7 satellites.


When a Partial Solution is indicated, the precision (RMS) of the vector will be higher that it would have been with all ambiguities fixed.

If the control network has redundancy, I would suggest using the Minimally Constrained Network Adjustment to determine if this vector (and any other vectors in the Control network) fail the Tau test. In my earlier post I mentioned some techniques for performing the Minimally Constined network adjsutment to tell which vectors still fail the tau test.

Once these vectors are edited, it may be possible to get FIXED solutions for these vectors using the data editing tools in Solutions 2.40. Once the data editing has been done, all the vectors should be run through the Minimally Constrained Network Adjustment again to see if any vectors still fail the Tau test. If not, and there are at least 2 independent vectors coming into every point, the GPS data checks well within itself.

A Constrained Network Adjustment can then be used to see how well the known control points fit with the GPS data.


Best regards,
Richard

Thales Navigation Tech. Support

Steve,

Not a typo...I just can't read fine print as well as I used to. :-)

Thanks for the info. That's one to "cut and paste" and save in the file folder.