SurvCE 2.51 and Sokkia GER2700ISX Elev. Issues

[jahco@aol.com]

Some of us were posting on RPLS. com web site about problems with base/rover ht.s and elev. problem. Here is text:

On: Aug 31, 2010 10:03 AM EDT

We are trying to figure out why we continuously get elevation errors when we stake out points. Now if we have the base set up in one position and we stake out points we shot it is always flat, however, if we move the base to another localized position and stake out points we have previously shot, It seems that we always have an error of .15 and it seems like it is always a fill. We have been banging our heads against a wall on this any input would be great. I don't know if it a setup thing, a localization thing, or even an antenna setup.

thanks,

dave On: Sep 2, 2010 9:17 AM EDT

Need more info Dave (Receivers & Data Collector Software).

Ed




On: Sep 7, 2010 7:12 PM EDT

It is a Sokkia 2700 isx, and we are using SurvCe 2.41. We have tried a number of different localization techniques and stakeouts. Like I said before, we are getting anywhere from .15 to just over .25 error and it is always a fill. What I don't understand is how a point can be flat and then when we put another point in and go back and stake out the original point and we are out again. Now, I have read the SurvCe book numerous times and I understand that you can get a slightly tilted plane, but that doesn't explain our problem. It also states in the book that you can try a number of different ways to localize and pick the one that works for you. However, we have a tried a number of different ways and continuously get a fill .15 to .25 in error. I know it has got to be a setting of some kind but I am lost at this point. Data collector is the Allegro MX.

Today tried again. We set up 5 control points out in the parking lot. We localized on them and staked them all out. Everything was fine. Then we moved the base from the unknown point to one of our localized points and staked out our other localized points and we were reading a fill of .22' consistently for all other points. We also tried setting the base up on a known point and localizing on the other points we had previously localized on before. The base was included in the localization and we staked them out and they were fine. Moved the base to a localized point and we needed .27' fill. I don't understand why. We have tried turning vertical off on all but one localization point. Doesn't make any noticeable difference. We have tried ground to grid, grid to ground. All points have inversed correctly. Just having issues with the vertical. If anyone has any idea please, please, let me know.


On: Sep 14, 2010 9:19 AM EDT

A few things you need to look at:
-both antena heights are entered correctly, either with slant or vertical ht. set.
-check that you have the correct geoid set in the Allegro. The current version is geoid09.
-you should upgrade (free) SurveCE to 2.51.
-when you enter the info for the base point as a "here" point or shot make sure of the slant/vert ht. is entered correctly. Pre 2.50 versions only recorded a VERT ht. for the base "here" position which is 0.12' "lower" than the slant.
-and the biggest let down of everyone who uses RTK, it should NOT be used for finish grading. It is the nature of RTK; it gets us going very quickly and is amazing when the right site conditions are available (wide open sky) but is completely dependant on the current sattelite configuration and changes throughout the day and the next day and all things GPS.

On control points and monuments we always get at least 2 shots at different times of day (with a min. 50 readings with the "average" key), then take the average of those shots. But anything you really need to hang your hat on; rim elevations, benchmark elevations, control points to be used by others, important monuments...should be located with conventional gear and use the RTK shots as checks.

John


Dave

I'm running almost exactly the same setup, Allegro CX rather than the MX.

I'd second John's point, upgrade to SurvCE 2.51

You might find some clues in the 'Raw data' - button 4 under the 'File' tab in SurvCE.

Also, Carlson support are very helpful, and quick.




On: Sep 21, 2010 11:35 AM EDT
We found that for carlson for the base height you should use slant and measure from the tab down, for the rover it is only the actual rod height and vertical. The software does the rest. Notice in the raw data that carlson modifies these measurements to reflect, not the heights on the labels on the instument, but the phase heights as tested on the opus site. We have the exact same set up as you and once we started using these procedures all came out well.

Vincent

On: Oct 20, 2010 6:26 PM EDT

[b]Now I take that back. We just used our base/rover set up and had the same problems you found. We have been just using the rover connected to an RTN for some time but on a recent site the nearset base was rather far away and was stretching the rms values. So since we had a wide open site and the base was on a hill, we went back to the original setup. We haven't done that since upgrading to Surve CE 2.51.

On this site we also had exisitng total station control that was really tight and with consisitent good checks. So now we set up the base, do a "store" of the base point in the monitor sat. tab, set up our rover then start checking...out consistently low at the rover by 0.10 - 0.15'!! We knew that the new version of SurvCe allowed for entering the antena ht. as either vert or slant in the base "store" tab so we tried both versions. What we found was that no matter which box we checked (vert or slant) we got the same elevation at the base.

What was even stranger was the only way we finally got it to work was to measeure the "slant ht" with the lip on the base BUT enter it as a vert ht. both in the setup of the base in "Equip" and for the "store" option in monitor sats. Then we entered the normal slant ht we normally do for the rover rod. Now we are hitting all our checks to the regular 0.02'-0.04' we would expect. We rechecked throughout the day and on the following day....still good.

There is definatley something wrong in the setup of the SurveCE and base setup info for Sokkia GSR 2700ISX. Not converting the 0.12' difference between slant and vert on the reciver. We will be checking this again this week and will follow up with Ed at Sokkia and the folks at Carlson.

John


Long (sorry) but felt it was important to figure this out.

[grunta57]

Probably the best way to check this problem and isolate what is really happening is the following method:

1) Set up Base randomly (not over a known mark). Use the Read GPS function within Carlson and take approx 25 readings

2) Go to a known mark with just the setting out pole. Ours is a 2m pole, and when you take the slant measurement to the tape tab on the antenna it comes to 2141mm (that's allowing for the 106mm between the top of the 2m pole and the bottom face of the antenna - ours is a full bayonet coupling)
Note: For this exercise we used a standard tape - Not the supplied Sokkia Tape

3) Enter the 2.141 slant in the RTK Rover configuration under receiver height default (Equipment GPS Rover)

4) Calibrate on the known mark first with the pole. Check the localisation result using setout point after that process is complete. Note Localisation (Equipment - Localisation - Points - Add Point - Read GPS (take 30 readings)

5) Now set the GPS on a tripod over the same mark and check (SURVEY - STORE PTS) the height Z value is consistent having first entered the new antenna height ..same basis slant direct measurement to measurement tab

6) The result I got 2nd time around was 6mm different to the 1st result. Basedon the expected accuracy of the system that tells me that the correct methodology was used.

I just had a thought though:

Curiously the HI Tapes that are provided in the kits are 75mm different on the basis that they can be read incorrectly. That being the case, if they were used incorrectly at both base and rover you might get a 150mm difference (that's assuming the tapes were read differently).

Be sure too that the pole height is correct i.e. not simply 2m cause it's a 2m pole but 2.14...

Also on the controller SurvCE make sure the parameters are set correctly.

We did have issues such as what you are talking about, but it came down to the entered slant heights everytime

Hope this helps

[grunta57]

Hey sorry guys I thought you were talking metres my mistake..

Looking at it this is normal expected accuracy of the instrument +/- 0.03m