Project Control Cogo Point Exports/Imports

[jhill]

Hello,

We often receive files from survey firms that use a NAD83UTM17 coordinate system.  They are usually scaled to ground from a point.  Coordinates often look like this:

1234567.890

123456.789

123.456

They appear to have grid coordinates, but are actually ground. To avoid confusion, we create "project control." With the above coordinate, we would truncate it as follows:

234567.890

23456.789

123.456

We have been manually doing so in excel by -4000000 on the northing and -400000 on the easting.  It lets our crews, instantaneously, know that the coordinates are ground and to use a scale factor of 1 on the total station. We have found it helps eliminate confusion as to which coordinate system cogo points belong to.  Is there a way one can set C3D to do this subtraction when we export cogo points to a csv?  Conversely, is there a way we can add those amounts when we import a csv?

We have the files set up so that there is a grid base point and a scale factor applied.  We check transform on and off depending on whether we want grid or ground.  I would like to keep that the same (ie. I would like to export/import without changing that setup if possible).

Any input greatly appreciated.

Reference articles:

http://info.quuxsoft.com/Articles/Working_with_Grid_Coordinates.pdf

http://info.quuxsoft.com/Articles/Solving_Grid_to_Ground_Problem.pdf

[Pointdump · Accepted solution]

Jason,
"Is there a way one can set C3D to do this subtraction when we export cogo points to a csv? Conversely, is there a way we can add those amounts when we import a csv?"
No way I know of, but then I've been proven wrong before. Maybe someone has a clever idea.
"Scaled" Grid Coordinates are scary, because they are scaled from a point. If that particular point is not or wrongly noted, your data is suspect. And the scaling only works in one direction, from Grid to Ground. So you pretty much have to stay in the ground system for all the rest of the work.

Those two PDF's are pure gold from much-missed Sinc. Thanks to @Jeff_M for posting them on his website.
A Custom Low-Distortion Projection is the best way to go, not only because of the ease and accuracy of bouncing between Grid and Ground, but also allows you to use the abundance of freely available GIS Data. Just be sure to post the MetaData Statement on the drawing.
If you are using Trimble data collectors, like Survey Controller or Access, punching in the parameters of a Custom Low-Distortion Projection is very easy.

Dave

[Anonymous]

One simple way to simply transform the points would be to import the points (not in a survey database, just import) into a fresh drawing, the move them:

Command: m MOVE
Select objects: <select all of the COGO points> found

Select objects:

Specify base point or [Displacement] <Displacement>: (pick any point)
Specify second point or <use first point as displacement>: @-400000,-4000000

Then export the points if desired. check random points to be sure the transformation worked.

[Anonymous · Accepted solution]

I learned something new: never noticed Displacement before on the MOVE command. skip the base point and the @Anonymous,deltaY and go right to entering in -400000,-4000000. It's a click and a keyboard entry less work. been using this since about release 12 and it took till now to see that. doh.

[jhill]

I agree with everything in this post.   However, I should have given a bit more context. We are located in Ontario, Canada.  LDP's are merely "coffee-side banter" up here.  We are awaiting to see how the implementation works with our friends south of the border.

All legal plans in Ontario require an integration table.  Typically, these state 2 HCM's and 2 ORP's (where the ORP's are 2 points called out on a plan).  Everybody, typically outside of very remote areas, does so to "urban" standards:

"2) The coordinates required under subsection (1) shall be accurate, at the 95 per cent confidence level, to,

(a) 0.05 metres in urban areas;

(b) 0.2 metres in rural areas; or

(c) one metre in remote areas. O. Reg. 216/10, s. 14 (2)."

https://www.ontario.ca/laws/regulation/100216#BK17

Typical practice is to scale from one of the following:

1.) an HCM stated on the legal plan (preferred as the points have calculated point scale factors to minimize distortion)

2.) an ORP stated on the legal plan (some use the nearest HCM point scale factor; others submit RINEX data to NRCan's PPP to have it calc'd)

3.) from 0,0 (frowned upon, but it happens).

Our jobs never need to go from ground to grid.  If any layout were to be done with GPS, it would be for items with low tolerances and a site calibration would be used (ie. a manhole location, limits of excavation, etc.).  Not ideal, but we rarely lay things out w/ GPS.  Submission to larger municipality GIS's is done with the above stated chart.  They, likely, scale it back based on that info.  Not ideal, but GIS often means "Get It Surveyed" when it comes to accuracy standards anyways.

Our sites are always small enough and the point scaled from close enough that GIS data is "close enough" (I know....every surveyor on the board just died a little inside).  We only use GIS data for very high level design/information.  As-built's and field topo's are the main basis for design.

In short, I fully would like to move to LDP's.  We use SurvCE so implementing it would be easy.  Unfortunately, it is not on anybody's radar (ie. no requirement to put metadata on plan) and I have enough fun explaining "grid to ground." 

I, thankfully, do not need to layout runways to the level Sinc did.  Anything we have laid out where it might matter (ie. highway jobs from the MTO) have always stated things mattered per chainage much more than coordinates.  We RFI the designer for the limits of their structure (ie. bridge) and if their gridlines are at certain coordinates.  If they "get" coordinate systems, I get a question back on what exactly I want.  If they just shoot back CAD where the coordinate distances exactly match shop drawing distances, we state that we are laying out each structure as its own site with the perimeter at these coordinates of the alignment provided plus/minus placement of whatever precision our equipment is plus the grid/ground factor with all responsibility of the exact coordinates of the structure to be based upon design drawings.  A kludge and a "cop out," but it works.  Asphalt crews are wonderful in that they point the paver in a direction and make things black.  Nobody seems to care about the grid/ground distance when tallying up bitumen tonnage.

A long way around, but while I would like to use LDP's for now we will just stick with grid survey's, localizations, and project control.  Party like it's 1999, eh?  🙂

[jhill]

While LDP's are ideal (and recommended), this simple solution will work for us.  Thanks very much.

[Anonymous]

Can you tell me what LDP's, HCM, PPP and ORP are?

HCM - horz ctrl monument?

ORP - "O" reference pt?

LDP - ?

PPP - ?

for the control/reference points listed on the plan, are these in ground coordinates? i.e. use the coords to stake from?

In Wisconsin we eliminated grid/ground confusion by developing local, county wide coordinate systems the represents actual ground surface closely enough that any conversion is negligible. it's great, grid/ground is a PITA. https://www.sco.wisc.edu/coordinate-reference-systems/wisconsin-county-coordinate-system-wccs-parame... 

[jhill]

Sorry, classic mistake of using non-standard acronyms and expecting everyone to know them.  My bad.

HCM  - Horizontal Control Monument

These, in Ontario, are typically pulled from the Ministry of Natural resources website "COSINE"

http://www.gisapplication.lrc.gov.on.ca/cosineONT/Index.html?site=cosine&viewer=OntarioViewer&locale...

It's free to use (I don't believe they geofence based on your ip). 

To the left would be the available horizontal control in different datums (NAD83-CSRS V6 (Epoch 2010.0), NAD83-CSRS V3 (Epoch 1997.0), NAD83-ORIG (which, to my American friends, would be NAD83(1986)), and NAD27-Toronto (where they still use NAD27 due to their extensive control network that they do not have the funds to update).

ORP - Observed Reference Point

"An ORP is a monumented point connected to a coordinate system by measurement, it therefore does not have to be the actual point of observation. It can simply be a monumented point on the survey that is connected to the actual point of observation. The actual point of observation could be on adjacent lands where we have no right to set a monument or even a point set in the ice on a frozen lake."

https://www.aols.org/sites/default/files/OPS%20Fall%202011%20Web_0.pdf 

p.4 where there are some common tips on integrating cadastral surveys/plans (which is relatively new up here).

LDP - Low Distortion Projections

From a quick and dirty glance at your Wisconsin pages, they appear to be similar to LDP's.  My friends to the south are miles ahead from a technical point when it comes to this.  We just started requiring cadastral plans to be integrated ~10/15 years ago.  The Yanks are light years ahead on this.

http://archive.amerisurv.com/PDF/TheAmericanSurveyor_Billings-GroundVersusGrid-LDPpart1_Vol10No9.pdf

http://archive.amerisurv.com/PDF/TheAmericanSurveyor_Billings-GroundVersusGrid-LDPpart2_Vol10No10.pd...

https://rplstoday.com/community/software-cad-mapping/low-distortion-projection-design/

PPP - Precise Point Positioning

You can set up your GPS base, log the static data, spit out a RINEX file, and submit it to your favorite agency to determine that point's coordinates.

Canucks usually use the following:

https://webapp.geod.nrcan.gc.ca/geod/tools-outils/ppp.php

My American friends usually use OPUS:

https://www.ngs.noaa.gov/OPUS/

Though, there are an assortment of others:

https://www.gpsworld.com/a-comparison-of-free-gps-online-post-processing-services/

We usually tie into control, log the RINEX, submit it, and use those coordinates as a check on the base's horizontal and vertical coordinates.  There's a bit of juggling depending on what control you tie into.  For example, if you tie into NAD83(86), you choose Epoch1997 so you can use the grid shift to convert the CSRS to NAD83(86) and check into CGVD28.  If you tie into NAD83:CSRS(2010) and/or CGVD2013, you would choose Epoch2010.  If you tie into NAD83:CSRS(2010) horizontally and CGVD28 vertically (don't get me started--some people do it), you would have to double submit the RINEX (ie. once for horizontal, once for vertical).  Clearly, this is less than ideal but it is what it is.

I have attached an example plan (all copyright belongs to their owners and it is shared for illustration purposes only).

http://learning.aols.org/aols/Interpretive_Guide_2016.pdf

p.20 onward.

Typically, a cadastral plan in Ontario will list the following in their integration chart:

-HCM(s) tied into

-ORPs

-notes about ground distances being shown

-what coordinate system has been used

When you receive the CAD, you typically find 1 of 3 things:

-the plan has been converted to ground being scaled out by the combined scale factor (csf) from one of the stated HCMs

-the plan has been converted to ground being scaled out by the combined scale factor (csf) from one of the stated ORPs

-the plan has been converted to ground being scaled out by the combined scale factor (csf) from 0,0

This has the potential to seriously bung up a project as "ground" coordinates look the same as grid coordinates.  We attempt to minimize blunders associated with this by creating project control.  We could do a more complicated translation, but we usually just drop the leading digit to simplify conversion.  The crews know what system the points are just by looking at the length of number.  Technicians can bring the points back into the clustery "ground, but looks like grid" drawing relatively easily.  GNSS data can be scaled to ground by having the Transformation tab setup with the correct scale point and csf.  This only works for smaller sites (ie. 500 lot subdivisions or smaller).  We do not have vertical differences like our friends in Alberta/British Columbia/Colarado/etc. so it semi-works.

To reiterate--all this is far from ideal.  However, it (unfortunately) is the industry standard.  I, too, hope that LDP's gain traction up here.

[Anonymous]

Thanks for the interesting post. To me, at least, but I own a plumb bob, too.

Yes. I used to hate when we worked in State Plane coordinates. You were never sure, there was always checking back to a measured line length vs. published in grid coords line length to know which way to apply the scale factor. Because, as you say, looking at the coordinates you can't tell.

and the different ways of scaling by holding a somewhat arbitrary point, and having different results if you held a different point. as a surveyor I can fully appreciate your system of ensuring that the stakers know they are in (project) ground coords. the differences could be small enough to not readily notice when you stake it, but significant enough to have t tear it out and redo it - at the surveyor's cost.

So, yes. I am happy I haven't had to deal with grid/ground scaling for some time now. I'm not sure why everyone is not doing this. Maybe in places with a lot of relief in the county it wouldn't work well, but many of us live on flat enough ground that elevation change isn't an issue.

[_Hathaway]

Great topic and questions. While we are rarely on the receiving end of another surveyor's coordinates or job, it does happen once or twice each year.  Honestly, if they've been contracted by others to supply the data we really don't care what system it's on as long as the data checks against itself in the field. But, believe me, i have received some data that claimed to be on grid but really wasn't.  Luckily the other control in the same project was on ground and so far off from the primary control we were able to decipher what was really going on.  The original surveyor was almost clueless, it was scary.  After much back and forth they are much more aware of all things grid/ground, etc...

With that said, it is extremely common if not outright standard for most public agency projects to be surveyed on grid and then scaled to ground using 0,0 as the basis for scaling. This is all done in the office software.  Once that is set once, it's done.  All survey work is done using those settings, the field crew doesn't have to worry about any type of scaling when switching between conventional and GPS work.  All our projects are setup this way and have not had any significant issues.

[Anonymous]

I had to stake the site for the construction of a new system well (million$). the design consultant hired a surveyor and we got control and base map in Dane County Coords. Great! We have a City base transmitting corrections so I could check the control with GPS and go!

Problem was that everything was 8 feet off in the field east-west. to set up the job I shot control which all checked against each other, then shot water valve boxes and manhole covers as a check. back in the office it was WTF time. it got resolved with a site calibration, but never assume!

[jhill]

A former company I worked for would do site calibrations (cals) for contractors all the time and it was just easier.  Typical work flow:

-design plans received in office (99.99% they were in ground)

-Request For Information (RFI) everything one can think of to cover your butt

-layout linework, data collector back maps, cogo points, surfaces, and plans created

-give it to the contractor

-walk them through shooting site perimeter property bars and site benchmarks with their GPS rover

-create the cal and tell them to use that for that site only

Usually no geoids or projections on the contractor's GPS.  Note that this was typically done for layout of the following:

-silt fence

-manholes

-CB's

-excavation

-main lines

-culverts

-rough grading

Curb was layed out by the firm using total station (aka not done by the contractor w/ GPS).  Max site size was around 500 single detached subdivisions.  Not the prettiest, but it worked and the contractors  loved it.

[Anonymous]

I've done a few of site cals for dirt contractors with GPS dozers. Did a whole boatload of models for them, too. One thing I preached was having a solid BM close to where they parked the machines at night so that the first thing in the morning and again later, hopefully, they dropped the corner of the blade on top of the BM to make sure everything checks. There was one site calibrated by their own guy with no way to check that was graded 0.2' too low over 40 acres - 1/4 mile x 1/4 mile. that is a lot of dirt that was moved twice and didn't have to be moved at all ...