Jump to content
  • 0

dfx Drawing Origin


Benson Shaw

Question

I'm working with several dxf files describing a roadway intersection in Renton WA, USA. It's a small city a few miles south of Seattle.

When imported to my vwx file, the drawing origin is 250 miles W and 33 miles S of the intersection. The drawing origin is miles out in the Pacific Ocean.

I find that imported dxf and dwg files almost always have origin far from the relevant geometry. Is this a drafting convention? An AutoCAD standard workflow? Something about the vwx import process? Is there some important reference point out there in the ocean? Or?

No problem really. I'm just curious why anyone would work that way.

I manage in the usual ways by referencing the imported files near origin of a project vwx file, or importing the geometry I need near the origin of a clean vwx file. Also, I know I can chose the Center on Import option, but the various vwx layers do not usually register correctly (different extents with different geometric centers)

-B

Link to comment

13 answers to this question

Recommended Posts

  • 0

Given the offset to the origin these files are probably (geo)referenced to a coordinate system, which could be a local coordinate system. With survey files this is common practice. Because most coordinate systems set the origin in a way that you will always get positive coordinates they often start at quite a distance from the origin. E.g. UTM coordinates have a 500km offset in the easting to ensure positive coordinates in that direction.

It can cause the origin to end up in some seemingly illogical places such as the Pacific. :-)

The origin of the Dutch national grid is near Paris, France to give another example.

  • Like 1
Link to comment
  • 0

Thanks, ArtV.

That makes sense.

There must be individual or office standards. I do not see a common origin.  Files from several local firms have different dwg origins.  The office CAD czars must get to choose their 0,0,0 point.  The dxf origin could be in a galaxy far far away, while the project has a 00 station at end of roadway center line.  Oh well.

-B

Link to comment
  • 0

If a defined coordinate reference system was used for the survey then the origin is fixed. That does not stop some drafting offices to move the origin near the site/objects in order to get rid of the sometimes large numbers of the coordinates. This is something that should never be done from a technical point of view as it brings some other issues with it, but that is a whole different discussion.

One thing you may want to check, in case you are not aware of this, is that when feet are being used whether they used the US Survey Foot or the International Foot (i.e. the common foot that everyone not in geodesy/GIS thinks of). The difference is tiny so for short distances it may not  be crucial but over longer distances it does matter.

Some info and example of why this may matter can be found at this link: http://www.pobonline.com/articles/98788-u-s-survey-feet-versus-international-feet

  • Like 1
Link to comment
  • 0

Thanks, ArtV

  1. I never heard of this before!
  2. The dxf info I receive with the far off origins almost always includes a survey layer with existing conditions.  I wonder if the surveyors report data in the US Survey Foot units by default, or if they convert the data represented by contours or roadway center lines for use by other project members using the International Foot.  Or maybe surveyors don't even use the Survey Foot all the time. Anyway, I never see any notice of International Foot vs US Survey Foot in the notes.  Have to ask next time.
  3.  If the dxf origin is many miles away from project elements, that .00002 scale factor could really misrepresent the locations of those project elements.
  4. Two arcane measurement systems operating in parallel on a project seems a strange convention. One hidden in initial conditions, the other used by every tape measure and range finder in the field. I will have to look up history of these two systems.
  5. "Imperial" linear dimensioning is bad enough. Don't get me started on the volumetric units. 

-B

Link to comment
  • 0

Not sure if I got everything right.

My DWG's delivered by clients are often far away from (VW) origin.
It is either their true world position or a position to a certain point
of a larger site.
That is ok.

I never import DWG's (mostly 2D here) directly to my working file.
(Unwanted Classes, ...)
I prepare an DWG import file.
I do a test in a blank file first, to test import and check Scale, Position, ...

If it is far from Origin,
I search for a point that I want to be the VW Origin, like the left center of
where my grid should be. I note its coordinates with ALL decimals.
I set my prepared DWG Import File's User Origin exactly to those values
BUT negative !

Then I import all DWG's "by Align to User Origin" !

This way, when I import the DWG finally, its geometry will sits around the VW
File ORIGIN, so no accuracy problems, but will keep its real world coordinates.
(Wich you can see at the rulers)

From this file, I only reference the DWG containing Layers to my final Drawing File
by Viewports. When I did my work and export geometry again as DWG, it will be at
the correct real world position again.

 

Edited by zoomer
Link to comment
  • 0
3 hours ago, Benson Shaw said:

1. I never heard of this before!

People who are not dealing in-depth with surveys/GIS are often not likely to come across the existence of the US Survey Foot, and if the size of the project is not big enough to notice they may assume International Foot by default as that is used everywhere else. That being said, not all surveys in feet are using the US Survey Foot.

3 hours ago, Benson Shaw said:

2. The dxf info I receive with the far off origins almost always includes a survey layer with existing conditions.  I wonder if the surveyors report data in the US Survey Foot units by default, or if they convert the data represented by contours or roadway center lines for use by other project members using the International Foot.  Or maybe surveyors don't even use the Survey Foot all the time. Anyway, I never see any notice of International Foot vs US Survey Foot in the notes.  Have to ask next time.

A good surveyor will always provide the coordinate reference system used as well as the units (e.g US Survey Foot or International Foot if it is not defined by the coordinate reference system). If the file has a drawing border then it often has the info somewhere in there as well.  When surveying the units used can vary depending on the party ordering the survey (assuming they bother to specify), so there is not always a "default"  unit unless it is specified by the coordinate reference system and the ordering party.

3 hours ago, Benson Shaw said:

3. Two arcane measurement systems operating in parallel on a project seems a strange convention. One hidden in initial conditions, the other used by every tape measure and range finder in the field. I will have to look up history of these two systems.

Well... it depends, I've come across projects where the source files were in 3 or 4 different (projected) coordinate systems and no common reference point was specified. Fortunately it was only for a draft concept so aligning things visually was ok in that case. Depending on when a survey was done and by whom there can be differences in the used (projected) coordinate reference systems and units. The only way to cover yourself is by (always) asking the people who provided you with the survey files which (projected) coordinate reference system and which units were used for each file if they have not already done this. There can be situations where it does not matter much, but when it gets to official boundary/location coordinates the legal consequences are often too big to ignore if the information is not correct.

Another thing to keep in mind is whether you are dealing with a simple (local) cartesian coordinate system of with a so-called projected coordinate reference system.The latter accounts for distortions across a larger area, the closer you get to the edges of the projected coordinate reference system the greater the deviations can become. This is one reason why one should not change the origin of a drawing if a projected coordinate reference system is used. (It is also the reason why Vectorworks can reproject the drawing, as you can see in the georeferencing settings).

Link to comment
  • 0
1 hour ago, zoomer said:

If it is far from Origin,
I search for a point that I want to be the VW Origin, like the left center of
where my grid should be. I note its coordinates with ALL decimals.
I set my prepared DWG Import File's User Origin exactly to those values
BUT negative !

Then I import all DWG's "by Align to User Origin" !

This way, when I import the DWG finally, its geometry will sits around the VW
File ORIGIN, so no accuracy problems, but will keep its real world coordinates.
(Wich you can see at the rulers)

From this file, I only reference the DWG containing Layers to my final Drawing File
by Viewports. When I did my work and export geometry again as DWG, it will be at
the correct real world position again.

This is probably not a problem if the coordinates of the drawings are plain Cartesian coordinates without projection.

If a projected coordinate reference system , i.e. georeferencing, is used this can cause deviations. The georeferencing can cause objects to be slightly reshaped depending on the used projected coordinate reference system and the position of objects in the coordinate system. Vectorworks can reproject a drawing in some cases, e.g. if you are using UTM and import a drawing using mercator projection. UTM is a transverse mercator projection, which looks different from a mercator projection despite the mercator being in both (quite a few people think they're basically the same). Ditto when changing the coordinate reference system afterwards. Setting the origin differently may interfere with reprojections etc. so I prefer the origin to be at the internal origin of the file to avoid potential errors in transformations etc. If that means the geometry will be hundreds of kilometers away from the internal origin, so be it.

On another user board there was someone who exported a georeferenced drawing into dwg and got a complaint from the customer that the coordinates didn't match. Based on the copied conversation parts it seemed neither side was aware of some of the issues and/or choices that had to be made before exchanging files/information.

Link to comment
  • 0
11 minutes ago, Art V said:

so I prefer the origin to be at the internal origin of the file to avoid potential errors in transformations etc. If that means the geometry will be hundreds of kilometers away from the internal origin, so be it.

Öhm,

But VW will this way likely not work as expected.
Rounding issues, OpenGL display problems, solids operations failing, ...

Not funny to work ?

Link to comment
  • 0
8 minutes ago, zoomer said:

But VW will this way likely not work as expected.
Rounding issues, OpenGL display problems, solids operations failing, ...

Not funny to work ?

Well, I have so far not experienced those issues. Occasionally some display issues but those had more to do with the objects themselves than with them being far away from the origin and the complexity of the 3D model, as those issues did happen near origin as well.

In the end it is a trade-off between convenience on one side and coordinate accuracy and preventing projection errors on the other side. For my work the latter is fare more important if georeferencing is being used. Especially when there were errors in e.g. the handling of the UTM coordinate system changing origin location was a no-go, though these issues have gradually been solved in VW2015 and VW2016.

Link to comment
  • 0

Yes, that was an interesting widening of my small CAD world.
I think we already discussed a bit about that topic some time ago.

But I always see so many new user's problems on the german forum
that you can solve by simply asking :

A
"Do you use the latest graphic card driver ?" (If Windows user)
or
B
"Are you far from internal origin ?"

Link to comment
  • 0

Yes, we discussed a bit about this some time ago. Before I got into the GIS stuff I would at times do as you subscribed, but I no longer do this if the file is georeferenced.

With regard to using georeferencing, if you are using that it is best to have the coordinate reference system and the origin of the drawing to align with the internal origin of the drawing. This will eliminate some potential coordinate issues/errors also when exporting to shapefiles.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Answer this question...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...