PhysicsCAD Posted July 27, 2009 Share Posted July 27, 2009 I have an equation driven parabola, which I drew into AutoCAD using an Excel file of 128 points (these are the x and y values). To machine this part, the shop requires 3-point arcs. Now I need to transform this parabola into a sequence of 30-60 arcs with start, end, and radii values in a table next to the curve. Is there a fairly straight-forward way to turn this curve into what I need, or do I have to re-create the whole thing by using many, small ARC commands? Quote Link to comment Share on other sites More sharing options...
eldon Posted July 28, 2009 Share Posted July 28, 2009 How did you join up your original points? If you use a polyline, then curve it with the FIT option, this creates an arc-fit polyline, a smooth curve consisting of arcs joining each pair of vertices. The curve passes through all vertices of the polyline and uses any tangent direction you specify. This could be one step in arriving at your final requirement. Quote Link to comment Share on other sites More sharing options...
JD Mather Posted July 28, 2009 Share Posted July 28, 2009 Is there a fairly straight-forward way to turn this curve into what I need Yes, attach your file here. Are you familiar with Flatten and Explode and List commands? I don't think I would fool with Eldon's suggestion on this problem. I suspect there is an easier solution. Attach your file here. Quote Link to comment Share on other sites More sharing options...
JD Mather Posted July 28, 2009 Share Posted July 28, 2009 Excel file of 128 points If you understand the conic section geometry you don't need all of those points. In fact if you attach your solution here I suspect that I can show more (user calculated) points actually equal less accurate curve. Xedges command creates the spline from the conic section. Flatten (Express Tool command) Explode (in this case I converted to 15 arcs - there is a method to control the number of arc resolution but I forget at the moment - hollar back if you can't figure it out). Quote Link to comment Share on other sites More sharing options...
PhysicsCAD Posted July 28, 2009 Author Share Posted July 28, 2009 Okay so I think I just attached the file. The curves I am speaking of are the inner curves of this mirror. para mirror.dwg Quote Link to comment Share on other sites More sharing options...
JD Mather Posted July 28, 2009 Share Posted July 28, 2009 The curves I am speaking of... Why did you use polyline rather than Spline to create "curves"? Quote Link to comment Share on other sites More sharing options...
SEANT Posted July 29, 2009 Share Posted July 29, 2009 I’ve never seen an application of such an elongated parabola before. Out of curiosity (and if not imposing on confidentiality): What is the purpose of the object? Quote Link to comment Share on other sites More sharing options...
PhysicsCAD Posted July 29, 2009 Author Share Posted July 29, 2009 Focuses light into our magnet. Our samples are micron-sized, so the light needs to have a small cross section. Quote Link to comment Share on other sites More sharing options...
PhysicsCAD Posted July 29, 2009 Author Share Posted July 29, 2009 I used polyline because this is my first AutoCAD drawing I've done. Fuccaro (?) directed me toward some links on how to create a parabola. Polyline seemed like the simplest way. Quote Link to comment Share on other sites More sharing options...
JD Mather Posted July 29, 2009 Share Posted July 29, 2009 I used polyline because this is my first AutoCAD drawing I've done. Polyline seemed like the simplest way. Creating a Spline is no more work than creating a Polyline and is the only shorcut I know to convert to g-Code friendly arcs. Quote Link to comment Share on other sites More sharing options...
SEANT Posted July 29, 2009 Share Posted July 29, 2009 Here is an 8 arc version (each side) of the parabola based on the flattening of the blue spline. The table was generated by the code shown in the response made for this thread. http://www.cadtutor.net/forum/showthread.php?t=38690 ParaAsArcs.dwg Quote Link to comment Share on other sites More sharing options...
PhysicsCAD Posted July 29, 2009 Author Share Posted July 29, 2009 The way that I made the curve is using Excel and just copying the cells I wanted and cntrl+v in AutoCAD. How can I write a script to create a spline for these points? Quote Link to comment Share on other sites More sharing options...
JD Mather Posted July 29, 2009 Share Posted July 29, 2009 The way that I made the curve is using Excel Can you attach the Excel file here? Quote Link to comment Share on other sites More sharing options...
PhysicsCAD Posted July 30, 2009 Author Share Posted July 30, 2009 Here are the Excel sheet values for the positive end of the mirror. The other side is obviously just the mirror image of this. As for SEANT's .dwg, thanks. How can I change from the Start/End/Center to a Start/End/Radius? The radius would just be to some constant arbitrary point. Quote Link to comment Share on other sites More sharing options...
PhysicsCAD Posted July 30, 2009 Author Share Posted July 30, 2009 The .xls Excel file is an invalid file type to upload to this forum. Here is a .dwg that is a sample of the kind of table that I need. Sample.dwg Quote Link to comment Share on other sites More sharing options...
JD Mather Posted July 30, 2009 Share Posted July 30, 2009 The .xls Excel file is an invalid file type to upload to this forum. Right click on the filename and select Send to Compressed (zipped) Folder. Attach the resulting *.zip file here. Quote Link to comment Share on other sites More sharing options...
PhysicsCAD Posted July 30, 2009 Author Share Posted July 30, 2009 Ok I successfully changed the curve from a ployline to a 2D-polyline. Then I converted that to a spline. Now I guess I want to "flatten the spline" so that I can fit a certain # of arcs along the length of the spline. The flatten command posts on this forum talk about thinning out 3D objects. I'm guessing the flattening SEANT meant was picking, in your case 8, points along the spline, then using the flatten function to extend those points into arcs. Is this correct? How do I do this, seeing as how I want ~36 arcs along the length of the spline? The Excel zip is attached. values for para mirror.zip Quote Link to comment Share on other sites More sharing options...
JD Mather Posted July 30, 2009 Share Posted July 30, 2009 Then I converted that to a spline. Now I guess I want to "flatten the spline". To use my method you must create as a spline to begin with - not convert a polyline to spline. Quote Link to comment Share on other sites More sharing options...
eldon Posted July 30, 2009 Share Posted July 30, 2009 To use my method you must create as a spline to begin with - not convert a polyline to spline. Pay attention to the man and do not fool about with an initial Polyline. If you use a Polyline, and then Spline fit it, the line pulls away from the vertices, and will not be the line that you require. Make a Spline and when asked for the tangent directions at start and finish, just press Enter in response to both. THEN flatten the Spline, which turns the Spline into a Polyline. (confusing isn't it!!) Quote Link to comment Share on other sites More sharing options...
PhysicsCAD Posted July 30, 2009 Author Share Posted July 30, 2009 Pay attention to the man and do not fool about with an initial Polyline. If you use a Polyline, and then Spline fit it, the line pulls away from the vertices, and will not be the line that you require. Make a Spline and when asked for the tangent directions at start and finish, just press Enter in response to both. THEN flatten the Spline, which turns the Spline into a Polyline. (confusing isn't it!!) This raises another question; how do I make an equation-driven spline? I made this polyline using Excel, can I use the same method to make a spline? Quote Link to comment Share on other sites More sharing options...
Recommended Posts
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.