We are having a problem with our data we get and was trying to figure out why it is not correct. What would be your procedure for a topo or survey when all you have is an address, a GIS map with rough distances, GPS, total station and a 3 man crew. Currently we are using a one point localization to get SPC and setting up a base station on a collected point. The problem comes in when we switch from GPS to total station, more often than not the data we get from the total station is skewed a bit(haven't seen anything out larger than 9 feet). I just need to narrow down the suspects so that I can make sure we are utilizing the equipment we have correctly. I know right now we are not.
9 feet?
coupla things-
first, one point ain't gonna cut it.
second, sounds like (if you're talking about 9 feet) that you're not fixed. that's WAAS level (lack of) precision.
without having enough info, my initial suggestion is to set, at minimum, 2-3 solid cooked-in points and verify across them with the gun. if you're off more than the usual shaky hand factor of a couple hundredths, then something's wrong.
otherwise, there are myriad combinations of settings in the collector and the instruments that could cause this.
but the first red flag for me is one point...
> We are having a problem with our data we get and was trying to figure out why it is not correct. What would be your procedure for a topo or survey when all you have is an address, a GIS map with rough distances, GPS, total station and a 3 man crew. Currently we are using a one point localization to get SPC and setting up a base station on a collected point. The problem comes in when we switch from GPS to total station, more often than not the data we get from the total station is skewed a bit(haven't seen anything out larger than 9 feet). I just need to narrow down the suspects so that I can make sure we are utilizing the equipment we have correctly. I know right now we are not.
Whoa! Lots of strange things there. Why one point localization to SPC? Log data and send to OPUS. Then translate all data afterwards. 9' is too damn much.
Here's what we do in your situation.
1. Take GPS to field, set up base, log data, set control.
2. If the project will take multiple days and all GPS is set in one day, then we load coordinates on the spot into the DC, check the inverse and make sure all is right and roll on.
3. Once field work is complete, submit the GPS base to OPUS, get value.
4. In an AutoCAD drawing, load up everything as you collected it in the field. Massage, reduce, whatever.
5. Translate ALL values (x,y&z) to the OPUS value at point 1 or whatever your base is.
6. Viola. All is done.
This is no different than any other conversions. You have to get all of your units the same. Here (FUGARWE) and OPUS can be as much as 9' (or more) off. If you're DC is working off adjusted values but you didn't adjust ALL points, you would see whatever the transformation is between OPUS and FUGARWE.
:good:
In Michigan, we have CORS single-base.
Begin job in SPC projection.
Use NTRIP rover to obtain SPC on "random point". Shot from a minimum of three CORS. Repeat for at leas two other points.
Set RTK base on this point and log static while mapping. This will be not only a check, but used to tighten the vertical after OPUS processing.
When using optical equipment, be sure that the "auto-scale" is turned on so that all distances are reduced to grid. This won't fix the 9' you have, but it will fix the 0.4' you will have at some point.:-)
Regards,
Are you doing any rotating before or after switching to the total station? With Leica the GPS data will be rotated back, but the total station data will not, resulting in two sets of data.
> We are having a problem with our data we get and was trying to figure out why it is not correct. What would be your procedure for a topo or survey when all you have is an address, a GIS map with rough distances, GPS, total station and a 3 man crew. Currently we are using a one point localization to get SPC and setting up a base station on a collected point. The problem comes in when we switch from GPS to total station, more often than not the data we get from the total station is skewed a bit(haven't seen anything out larger than 9 feet). I just need to narrow down the suspects so that I can make sure we are utilizing the equipment we have correctly. I know right now we are not.
Ok, that was a quick snapshot of the problems I'm having. Way back when I was working in the field, all the coordinates were set to SPC before I ever got the data so I didn't have to translate anything. This is a new company for me and need to correct some of the deficiencies I'm seeing. In my last job, the crews were sent out with calculated points and created a localization based on the collected rods found.
current procedure:
1. locate rods
2. set base station point, get SPC coordinates from Leica Smartnet.
3. set up base station
4. collect data that isn't covered up by canopy.
5. set total station on collected points labeled as control(only designated by point range, not as control in data collector).
6. collect/traverse remaining data.
The data collector we are using is a Carlson Surveyor+.
I know that after processing the raw data that the coordinates that I'm getting for most of the data is correct. It's mainly concerning my control. It looks like I need to tell my crews to let the control collection cook a bit longer than what they are currently allowing.
Steven,
I may be dead wrong on this one and I am not a Carlson user....
You mentioned in the first post that you are using a one point localization. So, I may be reading something wrong but are you using something besides a standard definition of you SPC Zone on the collector? The first post seems to say that you are using a one point localization which makes me think you are giving the collector the command to make a local "ground" coordinate system which would be a local coordinate system for a small area of work and that the method being used will force a small local system to be tangent at a point with the GPS. If this is the case the software may be also applying this same local coordinate system scaling to the total station observations which depending on your actual elevation may scale the horizontal distance allot from the total station.
Can you tell in the raw data if there is a scale factor being applied from a one point localization to the total station observations? Again, I may be misunderstanding what you mean by a one point localization.
I'm no GPS guru (I got a guy for that 😉 )' however....we use smartnet. We basically set static control and, as Kris put it during the NGS website shutdown, "roll our own" OPUS - 30 minute observations, download four reference stations from smartnet and process in geo office (we seem to get tighter verticals than OPUS with this method)
We also take 30 second network shots on everything as a check. Off hand I'd say that in the last 100 points or so, I've never seen more than 0.15' H between the post processed and the network result, usually less than 0.05' H and 0.1'V. If you've got feet floating around, it shouldn't be from inadequate observation time on the network.
> Steven,
>
> I may be dead wrong on this one and I am not a Carlson user....
>
> You mentioned in the first post that you are using a one point localization. So, I may be reading something wrong but are you using something besides a standard definition of you SPC Zone on the collector? The first post seems to say that you are using a one point localization which makes me think you are giving the collector the command to make a local "ground" coordinate system which would be a local coordinate system for a small area of work and that the method being used will force a small local system to be tangent at a point with the GPS. If this is the case the software may be also applying this same local coordinate system scaling to the total station observations which depending on your actual elevation may scale the horizontal distance allot from the total station.
>
> Can you tell in the raw data if there is a scale factor being applied from a one point localization to the total station observations? Again, I may be misunderstanding what you mean by a one point localization.
Mark,
I'm probably the person wrong here. I may be using term incorrectly. According the raw data, there isn't a scale factor being applied. The scale factor is 1.00000000. I know that the DC asks me to save a localization, which I termed one point, since we really only use one point that we collect off of our network. The DC also asked to store a reference file as well. Once a job is created with said files, they are set in stone, no going back. Which makes sense.
I would rather create a job specific localization with calculated points, but that right now is like asking for the government to do their jobs correctly.
i'm on smartnet too and have yet to set anything static. of course, everything so far has been well ensconced within various local control networks, and the sites have (to this point) been small. so i'll set a few 5 minute points, verify 3 of either City of Austin, LCRA, or Cedar Park published control monuments, and check accordingly with the robot. the biggest total delta i've had (horizontally) off of any published control to this point is .03, which gives me more than enough confidence in the positions i'm setting. i'll also- if the opportunity presents itself, shoot somebody else's nearby control blindly, then call them up and ask their XYZ for that point. again, haven't been off by more than a hundredth or two so far (a couple dozen times).
here in town i've found no reason, to this point, to run anything other than on-network. one of these days i reckon i'll have a couple hours to burn and i might shoot something off to OPUS. or better yet, we'll probably fire up a project pretty soon that'll require it, but for now i'm realizing that years of getting to know comrades, and years of being generous with my info, is paying dividends in the way of giving me enough residual checks to satisfy any doubts that have popped up so far.
i hate to belabor this, but are you sure your guys are fixing? once you turn the antenna on it's reading positions, but until you connect to the network, you'll see your residuals (not sure on the carlson, in smartworx it reads out 3CQ (total delta), 2CQ (XY delta), 1CQ (Z delta) at the bottom of the screen). until you connect to the network those numbers, around here typically at least, are in the 10'+ range. once you're fixed those things come down to, well, i won't collect anything over a .05 total delta (or positional confidence, i guess).
again, i could be totally off, but when i saw "9 feet" that's immediately what came to mind.
> We are having a problem with our data we get and was trying to figure out why it is not correct. What would be your procedure for a topo or survey when all you have is an address, a GIS map with rough distances, GPS, total station and a 3 man crew. Currently we are using a one point localization to get SPC and setting up a base station on a collected point. The problem comes in when we switch from GPS to total station, more often than not the data we get from the total station is skewed a bit(haven't seen anything out larger than 9 feet). I just need to narrow down the suspects so that I can make sure we are utilizing the equipment we have correctly. I know right now we are not.
Set up your base station with a control job, set it to log (even if you aren't planning on tying to SPC; it's just good practice, and it lets you do a short static observation of a control point if you lose radio), and go GPS in all necessary control. Do not perform a localization in the DC. Copy the control job points into a topo job, and begin collection of data using either the TS or GPS depending on the environment. Have them take notes, lots of check shots, and store check shots in a different point sequence (I use 101-999 for control, 100101-100999 for the check shots on those points).
Once the crew comes back in you can mess around with localizations and such with the topo job, but always leave your control file raw.
The first question I would ask is what is the crew seeing in the field? When they backsight are they getting expected results? When they turn to a check point does it look ok? How about when they change the face on the instrument and repeat the process on the opposing side?
> Are you doing any rotating before or after switching to the total station? With Leica the GPS data will be rotated back, but the total station data will not, resulting in two sets of data.
This happens with Trimble as well, and is a likely culprit.