I need help with cones!

[[De...]]

First a little about the part. The is a large part which goes into a very large machine. this piece is a fraction of a cone, roughly 18 degrees diametrically. it is roughly 100 mm high. The a datum the the outside cone feature, there are 2 plane on the inner cone, one on either side which are local datum b. Each plane has 10 threaded holes. The point of convergence or origin of the outer cone is a theoretical point roughly 4500 mm below the part and the axis is roughly 2300mm from the part.

I also want to state that I don't truly understand how to use cones in alignments, so I want to set the axis of the cone to be in z direction. IF I put the outer cone in the the x,y, and z origins, and also in the Rotation in Space in Z+ will that align this part the the center axis or do I need to generate the axis somehow as well as the point of convergence? I hope this makes sense, because I can't post the print and my drawing is not the best.

[[Cl...]]

I don't trust cones for spatial. I'd do two (circles on cone) top and bottom and construct a 3d line from them.

[[To...]]

In some software programs, a cone can can be used to control the X, Y and Z origin. However, Calypso won't do it automatically.  I think it might make sense if you had a real flat cone, i.e. 170°, but a cone with an 18° included angle, the Z origin would be very unreliable, as small deviations on the cone diameter or angle would result in large translations in the Z axis.  I think a plane or a point on the end might work better.

Edited January 15

[[Da...]]

Please sign in to view this quote.

I don't trust circles-on-cone-3d-line for spatial. The circles should be measured perpendicular to the cone axis. This is only possible if the cone axis is already known.

[[Ma...]]

Please sign in to view this quote.

Once you face inverted axis direction on manual alignment you will switch that to 3d line from circles. On other hand i never faced that behaviour when is machine doing alignment. So i would stick to 2 circles and 3d line on start alignment, when you possibly need to manually scan that.

If you keep same origin for base and start alignment you can use this setup:

• Running program as base alignment
• Selected alignment from list is "start alignment"

[[To...]]

Not sure what your part looks like but I am picturing something like a pencil with the graphite end facing in the Z+ direction.  If you were to create a cone from this feature, the space axis would -Z, as the space axis always goes from small end to large end on a cone.  

A simple sketch might help clear things up.

Edited January 15

[[Cl...]]

In the past I've not had good results using a cone for the spatial in an alignment.

[[Ma...]]

For Rotation in Space it should be a 3d line from multiple circles, more than 2 circles, sine the line is created through the center point of the circles the form can affect the center point therefore the more circles, the better the averaging 

[[De...]]

I ended up using the multiple circles to generate the axis. Thanks for the assisstance.