Jump to content

rotate cad model based on space point position?


---
 Share

Recommended Posts

Hello, I'm having trouble aligning a part that's basically a cylinder with a thread in the center.  Currently I have to take a manual space point and rotate the model manually until the point lines up with the model. 

Is there a better way of obtaining an accurate alignment for a part like this?  Without translating the part in any direction, just rotate it until the point lines up with the model?  I'm attempting to scan the threads with a duramax vast xxt.

image.thumb.png.c52cce43a5e9657bdb11992de35f5776.png

Link to comment
Share on other sites

With the delta between the actual point and the desired nominal, and the thread pitch, you should be able to calculate the rotational angle to move to advance the thread the desired linear distance.

Edited
  • Like! 1
Link to comment
Share on other sites

I would think you could iterate your base alignment to get it there.  I do something similar on twin screw compressor rotors.  There's even a method where you don't have to iterate your entire base alignment and can iterate just the point.  (see attached)

 

Let me know if I'm mis understanding something cause it totally seems possible. 

Iterative Alignment.pdf

  • Like! 1
Link to comment
Share on other sites

For the start alignment, I'm using a -z probe and scanning two half circles on either side of the thread.  recalling them into a 3d line.  with a point in the y axis.  I'm setting the base alignment plannar rotation to a point, but this just rotates my model along X axis.  Am I missing a step?

image.png.311cd3febe5ce1cb156662d55352ad63.pngimage.png.e61eb055d3b3e4d2a55e5c16a6315984.png

image.png.15ce36d029fe333942152429fef83079.pngimage.png.53434727430126ca9d881dea1ac1c8d5.pngimage.png.94f91f29eb45c53957f01c5af80b01fe.png

image.png

Edited
specifed rotation direction.
Link to comment
Share on other sites

Your strategy seems plausible.  There's only a few things I would change if it were me.   I don't know if its possible to establish the center line on the shaft instead of the threads, but that's what I'd be leaning towards.  Also, I'd only establish the rotation point on the thread face, instead of your strategy of establishing both Y and rotation using the threads.  This is just my theory.  It all looks legit otherwise.  

The way Calypso see's it, if you leave blanks in the base alignment, then it pulls from the start alignment, if there are also blanks in the start alignment, then it pulls from the machine coordinates.  Order of precedent is Base Alignment>Start Alignment>machine coordinates. 

 

image.png.601caff43f164ca51214fa63f57e8212.png

  • Like! 1
Link to comment
Share on other sites

Thank you for the response!

I eventually got a program to cooperate and keep the model along the Y axis.

Then my loop would never initialize.  I was hitting the thread chamfer.  I moved the point to the actual thread and the loops finally initialized.  

However, whenever I replace the part and give the part a very slight rotation, the thread scan doesn't line up.  I added a formula to move in the Y AXIS but was unsuccessful.

now using simulation so I don't take up the CMM.

image.thumb.png.79e68b98e52d163abe9273f9b8359c92.png

image.png.0766c7578619bac2e346daf1bfa125fc.png

image.thumb.png.48c6b7bedd18073a9b352949e421c278.png

image.thumb.png.6d8875adbcf57fdbb5271be29388d784.png

image.png.4a6d8244f783d61256e5a9128722e1db.pngbase alignment with the thread point.

Link to comment
Share on other sites

 Share

×
×
  • Create New...