I could really use some guidance here. I apologize in advance if this is really elementery.
We recently performed a topo survey approximately 2 miles long. We ran a conventional traverse thru our control and then set a GR3 receiver up on 4 of the control points over the course of the survey and let it cook for 2+ hours each time. Sent the RINEX files to OPUS and got the results back. My question is, are the SPC you get from OPUS grid or ground or something totally different? The reason I ask is, when I inverse between 2 of the points, the coordinates supplied by OPUS were, as is, closer to our conventional coordinates than they are if I convert them using the Combined Factor provided by OPUS. I've tried both ways, first assuming they were grid and converting them to ground (grid coord x 1/c.f.), and then assuming they were ground and converting to grid (grnd coord x c.f.). Both conversions give me a 0.6' difference in length from our conventional traverse, whereas the coordinates supplied by OPUS are only about 0.1' different from our conventional traverse. Am I doing something incorrectly?
I hope what I'm trying to explain makes sense.
Greg
They are grid.
Paste your OPUS results into the thread and you may get some good analysis. Maybe your OPUS solutions aren't very good.
The OPUS SPC are definitely SPC grid coordinates, why those match your ground traverse so closely is the mystery and indicates something else is wrong, of course without a check on the OPUS positions returned it could be that they are wrong, but if they are good clean OPUS solutions, unlikely.
Dave posted while I was typing, LOL
SHG
My initial thought is that the conventional ground traverse likley contains most of the error, and its simply a coincidence that it matches the GPS grid inverse.
Some missing info that might help is:
number of setups and lengths of shots of conventional traverse.
geometry of traverse.
openess of sky view for gps observations.
type(s) of equipment used.
last time edm was checked against a CBL.
If its as I'm assuming, 0.6' in 2 miles isn't horribly bad for a conventional traverse. In some types of terrain, it'd be darn good to get that.
The other option is there is a problem with the traverse which just happens to make it match grid closer than ground.
I would feed the whole thing into StarNet, conventional and the processed vectors (holding one OPUS solution fixed) to see if it flags anything. StarNet is pretty good about flagging potential blunders.
I agree, what is the distance of your traverse and how far apart are the OPUS points? If your OPUS points are 1000 feet apart, you have a problem, if they are 5 miles apart, I'd say you just have some error in your traverse.
My experience with OPUS is good, but if you have two points that can be seen between with a total station, I've seen 0.10' error on a regular basis, but never anywhere near 0.60'. But as others have said the coordinates from OPUS are on grid coordinates.
If your traverse was on relatively small assumed coordinates (5000,5000,100,-for example), and you translated them to your state plane coordinates in the data collector; that could be your problem. Most data collectors will apply the scale factor to the coordinates when moving them.
Compare your raw data to inversed data with the conventional traverse.
Possibly You Data Collector Is Converting Ground To Grid
BTW, SPC is always grid, ground is something else.
Paul in PA
Possibly You Data Collector Is Converting Ground To Grid
State Plane Coordinates + grid. Anything else and you have bastardized coords. and sometimes as BIG mess.
Possibly You Data Collector Is Converting Ground To Grid
SPC = SPC = SPC...anything else is bastardized. NGS only gives you SPC. Meters, International Foot or U.S. Survey Foot units, depending on your states designation of units.
Pablo B-)
Your OPUS solutions are "probably" correct, but you can't know without some kind of a check to the points. You were trying to accomplish that with your conventional ties which didn't work, my advise would be to occupy the points again with more than one receiver and process the data, holding one OPUS position and see if the others match, then you could rest assured that your OPUS positions are correct. This would be much quicker than going out and doing hours long sets on each one again. If that works, then you can look to the conventional data and figure out what is going on.
But as always, with 4 unconnected OPUS solutions you have no idea what is going on. You need redundant measurements. Which failed. Or you could do two of them, but if you are there might as well get all 4.
Thank you for all the responses. I THOUGHT that the SPC provided by OPUS were grid and I understand that ground is a different animal. I just needed confirmation as, as I stated above, our SPC provided by OPUS fit our ground data closer than if I used the combined factor also provided by OPUS to "convert" these coordinates to ground. I suspect it may be my conversion (hey, I'm not proud. that's why I'm asking you experts for help). Please tell me if I'm doing something wrong.
What I did was take the SPC and convert them to feet (as this is the unit used for this job). I then multiplied said coordinates by 1/comb.fac. to obtain "ground" coordinates. Is this incorrect? Should I have applied the combined factor BEFORE I converted the units to feet? Did I even apply the combined factor correctly?
Maybe, I should back up and ask you guys how you do it.
What data do you place on a set of plans if the engineer wants it in SPC, but you know some poor surveyor(possibly me) is going to have to go out and do construction staking on the ground? The SPC of several control points and the combined factor, datum, geoid?
Prior to this, I always ran static GPS on my control and processed it and had it output ground coordinates for me to do the ground survey with. This is my first time doing the topo while running an OPUS session(s) on my control at the same time and trying to marry it all togethar in the office.
Thanks again,
Greg
NO NO NO NO!
You never ever multiply coordinates by the scale factor correction. If you do you no longer have SPC.
The scale factor correction is only applied to distances.
NO NO NO NO!
O.K., I understand. The pieces are starting to fall into place.
Thanks,
OPUS state plane coordinates are grid.
The other question to ask is whether your State is US Survey Feet or International Feet.
The best way to combine GPS and conventional measurements is using least squares software so that the errors get propagated properly and you can see the residuals.