Coplanar lines

[b.modra]

I'm trying to fillet two lines that appear to be coplanar, but am getting a 'Lines are not coplanar' error.  

What seems very strange about this is that both lines ARE coplanar with another 3 lines.  ie I can fillet each line with the other 3 lines but not with each other.  

I guess its some rouning error, and if it was just one section I'd deal with it, but it is repeated across a hundred or so sections and is just a pain to correct manulally.

Using Autocad 2015.

[b.modra]

I'm trying to fillet two lines that appear to be coplanar, but am getting a 'Lines are not coplanar' error.  

What seems very strange about this is that both lines ARE coplanar with another 3 lines.  ie I can fillet each line with the other 3 lines but not with each other.  

I guess its some rouning error, and if it was just one section I'd deal with it, but it is repeated across a hundred or so sections and is just a pain to correct manulally.

Using Autocad 2015.

[Bernd.Gerstenberger]

Hi B.modra,

try this link for a solution:

http://knowledge.autodesk.com/article/Error-Lines-are-non-coplanar-using-FILLET-in-AutoCAD

Best regards

Bernd

[Bernd.Gerstenberger]

Hi B.modra,

try this link for a solution:

http://knowledge.autodesk.com/article/Error-Lines-are-non-coplanar-using-FILLET-in-AutoCAD

Best regards

Bernd

[3wood]

You can also try FLATTEN.

[3wood]

You can also try FLATTEN.

[b.modra]

Thanks Bernd,

No that doesn't help, even at the highest precision x values (start and end) are the same for both lines, but will not fillet.  I tried copy and pasting the x-values but no change.

[b.modra]

Thanks Bernd,

No that doesn't help, even at the highest precision x values (start and end) are the same for both lines, but will not fillet.  I tried copy and pasting the x-values but no change.

[b.modra]

3wood,
I'm aware of flatten, but I use the alignment of lines as a double check to ensure I haven't made a mistake in copying/moving features. Flatten is a blunt instrument and I don't know if it will introduce errors or mask errors in my work.

[b.modra]

3wood,
I'm aware of flatten, but I use the alignment of lines as a double check to ensure I haven't made a mistake in copying/moving features. Flatten is a blunt instrument and I don't know if it will introduce errors or mask errors in my work.

[b.modra]

Drawing attached - it should be fairly obvious where the line is missing to close each section.

[b.modra]

Drawing attached - it should be fairly obvious where the line is missing to close each section.

[Bernd.Gerstenberger]

Hi b.modra,

thank you for the drawing. I have created a video here and it works on my site. Here the <<link>>. I have used German AutoCAD but if you know that "Abrunden" is "Fillet", then everything should be clear

[Bernd.Gerstenberger]

Hi b.modra,

thank you for the drawing. I have created a video here and it works on my site. Here the <<link>>. I have used German AutoCAD but if you know that "Abrunden" is "Fillet", then everything should be clear

[RobDraw]

You're going to have to be more specific. I was able to fillet all of the lines with the adjacent ones in one of the assemblies.

[RobDraw]

You're going to have to be more specific. I was able to fillet all of the lines with the adjacent ones in one of the assemblies.

[3wood]

I have isolated one set of the section and attached here. Please try filleting the green lines.

[3wood]

I have isolated one set of the section and attached here. Please try filleting the green lines.

[b.modra]

Hi Bernd,

Thanks a lot for your time on this.  I should have been more specific on which items could not be filleted - please try again with the lines in green in the attached.

Regards

[b.modra]

Hi Bernd,

Thanks a lot for your time on this.  I should have been more specific on which items could not be filleted - please try again with the lines in green in the attached.

Regards

[RobDraw]

You didn't include that line in your original file.

I cannot get it to fillet in that location either. I'm not well versed in 3D AutoCAD but I did, however, get it to fillet when I was in the right hand view and setting the UCS to lower line but cannot repeat it.

[RobDraw]

You didn't include that line in your original file.

I cannot get it to fillet in that location either. I'm not well versed in 3D AutoCAD but I did, however, get it to fillet when I was in the right hand view and setting the UCS to lower line but cannot repeat it.

[dbroad]

AutoCAD generally can't handle it when extrusion vectors are not conformable.  For example, just looking at the two green lines, the normals are.

(1.0 8.39768e-011 -1.54279e-026)
(0.0 0.0 1.0)

So even though you see them as coplanar, AutoCAD cannot.  If you look at the vectors for all of the items, you get.

(1.0 8.39768e-011 -1.54279e-026)
(1.0 8.39768e-011 -1.54279e-026)
(1.0 8.39768e-011 -1.54279e-026)
(1.0 0.0 0.0)
(1.0 0.0 -5.55277e-017)
(1.0 0.0 -5.55277e-017)
(1.0 0.0 -5.55277e-017)
(1.0 0.0 -5.55277e-017)
(1.0 0.0 0.0)
(1.0 0.0 0.0)
(0.0 0.0 1.0)
(0.0 0.0 1.0)
(0.0 0.0 1.0)
(0.0 0.0 1.0)

Now, if you were drawing these objects in a common plane, they would have the same normals.  If you were drawing them programmatically, it may depend on the original size and coordinate systems during each operation.

It is possible to fix this by program but, I encourage you to use the -view command to look at the right side of each problem set of lines and use the flatten command.  If that will mess up the drawing, you would want too loop through the entire set and set the normals to (1.0, 1.0, 0.0) as the following does.  Try to work in planes in the future and work closer to the origin of the coordinate system.

(defun c:normalizex ()
(if (ssget)
  (vlax-for n (vla-get-activeselectionset
		(vla-get-activedocument
		  (vlax-get-acad-object)))
    (print (vla-put-normal n (vlax-3d-point '(1.0 0.0 0.0)))) ;;X- normal direction
    ))
  )

[dbroad]

AutoCAD generally can't handle it when extrusion vectors are not conformable.  For example, just looking at the two green lines, the normals are.

(1.0 8.39768e-011 -1.54279e-026)
(0.0 0.0 1.0)

So even though you see them as coplanar, AutoCAD cannot.  If you look at the vectors for all of the items, you get.

(1.0 8.39768e-011 -1.54279e-026)
(1.0 8.39768e-011 -1.54279e-026)
(1.0 8.39768e-011 -1.54279e-026)
(1.0 0.0 0.0)
(1.0 0.0 -5.55277e-017)
(1.0 0.0 -5.55277e-017)
(1.0 0.0 -5.55277e-017)
(1.0 0.0 -5.55277e-017)
(1.0 0.0 0.0)
(1.0 0.0 0.0)
(0.0 0.0 1.0)
(0.0 0.0 1.0)
(0.0 0.0 1.0)
(0.0 0.0 1.0)

Now, if you were drawing these objects in a common plane, they would have the same normals.  If you were drawing them programmatically, it may depend on the original size and coordinate systems during each operation.

It is possible to fix this by program but, I encourage you to use the -view command to look at the right side of each problem set of lines and use the flatten command.  If that will mess up the drawing, you would want too loop through the entire set and set the normals to (1.0, 1.0, 0.0) as the following does.  Try to work in planes in the future and work closer to the origin of the coordinate system.

(defun c:normalizex ()
(if (ssget)
  (vlax-for n (vla-get-activeselectionset
		(vla-get-activedocument
		  (vlax-get-acad-object)))
    (print (vla-put-normal n (vlax-3d-point '(1.0 0.0 0.0)))) ;;X- normal direction
    ))
  )

[IkaroFalcao]

Try to put them in the same position in the axis x, that it will work.

[IkaroFalcao]

Try to put them in the same position in the axis x, that it will work.

[Bernd.Gerstenberger]

Now I am able to recreate the issue but could solve it: 

• First I have isolated them and change the view to "Right"
• Now FLATTEN the two lines
• Afterwards FILLET them

It has worked on my site.

[Bernd.Gerstenberger]

Now I am able to recreate the issue but could solve it: 

• First I have isolated them and change the view to "Right"
• Now FLATTEN the two lines
• Afterwards FILLET them

It has worked on my site.

[b.modra]

Thanks 3wood, you beat me to it.

I really need to get this finished so I'm going back a few steps and redoing the work.  It would be interesting to get an answer but its not important now.  Thanks to everyone that responded.

[b.modra]

Thanks 3wood, you beat me to it.

I really need to get this finished so I'm going back a few steps and redoing the work.  It would be interesting to get an answer but its not important now.  Thanks to everyone that responded.

[b.modra]

Thanks again for the comments.

What is most curious to me is that the two lines are not coplanar to each other, but both are coplanar to several other lines.  How is this possible?

[b.modra]

Thanks again for the comments.

What is most curious to me is that the two lines are not coplanar to each other, but both are coplanar to several other lines.  How is this possible?

[dbroad]

Did you read our replies?  Who are you responding to?

[dbroad]

Did you read our replies?  Who are you responding to?

[b.modra]

dbroad,

yes I did, and thanks for your time.  Your post alluded to but didn't fully describe how the normals show whether the lines are coplanar.  Do they need to have 1,2 or 3 axis pairs the same, or is it a more complicated function.  It would be one I'm guessing.

In any case you have listed 14 normals.  The first three are completely different from the last four (no common values).  Would that imply that any of the first three cannot be filleted with any of the last four?

This is different to the functioning in the drawing where Line1 is coplanar with Line3-14, Line2 is coplanar with Line3-14, Lines3-14 are coplanar with each other, but Line1 and Line 2 are NOT coplanar (I've arbitrarily called the two green lines 1 & 2).

[b.modra]

dbroad,

yes I did, and thanks for your time.  Your post alluded to but didn't fully describe how the normals show whether the lines are coplanar.  Do they need to have 1,2 or 3 axis pairs the same, or is it a more complicated function.  It would be one I'm guessing.

In any case you have listed 14 normals.  The first three are completely different from the last four (no common values).  Would that imply that any of the first three cannot be filleted with any of the last four?

This is different to the functioning in the drawing where Line1 is coplanar with Line3-14, Line2 is coplanar with Line3-14, Lines3-14 are coplanar with each other, but Line1 and Line 2 are NOT coplanar (I've arbitrarily called the two green lines 1 & 2).

[dbroad · Accepted solution]

The normals should be perpendicular to the plane the objects are drawn in.  So they were probably drawn during a number of different ucs situations or may have been created by third party software.  Even though AutoCAD will sometimes fillet objects that are coplanar and also have different normals, it often objects with an error message.  The program I gave you should fix the problems.   (Save it as normalizex.lsp in a trusted path and load it with (load "normalizex").  Then just enter normalizeX and all your problems with this drawing will go away. Flatten smashes everything to the same plane so that would probably not help.  (You have so many different planes).

You also have many unfinished cross-sections that are not ready to fillet. Use the ucs command and set the ucs to right before drawing the closing vertical lines.

Rather than experimenting piecemeal by filleting various lines, recognize the cause (either truely non-coplanar or coplanar with different normals).

Avoid drawing by coordinates when the ucs is arbitrary. Avoid rotating things with mouse operations without using appropriate snaps, ortho mode or polar snap.

[dbroad · Accepted solution]

The normals should be perpendicular to the plane the objects are drawn in.  So they were probably drawn during a number of different ucs situations or may have been created by third party software.  Even though AutoCAD will sometimes fillet objects that are coplanar and also have different normals, it often objects with an error message.  The program I gave you should fix the problems.   (Save it as normalizex.lsp in a trusted path and load it with (load "normalizex").  Then just enter normalizeX and all your problems with this drawing will go away. Flatten smashes everything to the same plane so that would probably not help.  (You have so many different planes).

You also have many unfinished cross-sections that are not ready to fillet. Use the ucs command and set the ucs to right before drawing the closing vertical lines.

Rather than experimenting piecemeal by filleting various lines, recognize the cause (either truely non-coplanar or coplanar with different normals).

Avoid drawing by coordinates when the ucs is arbitrary. Avoid rotating things with mouse operations without using appropriate snaps, ortho mode or polar snap.

[leeminardi]

Use the RAY command  and OSNAP to add a line from the start of one of the lines through its end and another ray from the start of the other line through its end.  You can then fillet the two rays.

~Lee

[leeminardi]

Use the RAY command  and OSNAP to add a line from the start of one of the lines through its end and another ray from the start of the other line through its end.  You can then fillet the two rays.

~Lee