How is this calculating?

[[Jo...]]

Hey everyone,

I am looking for an explanation for how Calypso is calculating the values for true position. Before we jump in, this true position is based on a best-fit alignment of a curve. Let me explain more...

1. We measure 2 circle features to create a recalled cylinder (Datum B), and we then measure a plane that is perpendicular to the cyl axis, call this datum A.

2. We create a secondary alignment based on these features in relation to CNC coordinates.

     -Rotation: Datum A, No Planar rotation, X origin: Datum A, Y origin: Datum B Cylinder, Z origin: Datum B Cylinder.

3. We tie this alignment system to the 2d curve and enable "best fit" under evaluations, and create a best fit alignment based on that curve.

4. We then perform a manual true position calculation based on the Basic XYZ values and go from there...

I need to understand how Calypso uses the cylinder (Datum B) to restrain the Y/Z origin. 

I cannot replicate how I thought this could be working. My original thought would be an equidistant center point between the two datum b circle paths that is projected on the datum A plane. Here is a drawing to better explain. 

 

Can anyone help shed some light?

 

[[Er...]]

If you use best-fit on the curve (translation), you just un-did your prevouse steps. And nothing is contrained to your cylinder…

[[Ma...]]

As Eric stated.

TP should be doubled maximum deviation of axis from TP nominal point. For cylinder it's calculated for it's measured length ( if not changed via button ).

If that curve is just 2d or one slice, then i would not recommend to have full bestfit.

[[Jo...]]

Thanks for the input guys. 

I am really just trying to understand how Calypso is calculating the origin. We are using the best fit alignment translation values to calculate the true position. 

This is the original BF alignment from the Cylinder (-B-). 

I am getting different results when I attempt to use a symmetry point as the YZ origin. 

The symmetry point was my check for testing how I thought the cylinder was acting in the machine alignment (Secondary alignment) system. I am currently stumped trying to replicate the answer. 

Any ideas?

 

Edited August 21

[[Jo...]]

Bump 🙂

Does anyone else have any thoughts or suggestions for how I can verify this?

[[Ma...]]

What is that symmetry point? From centers of circles?

That perpendicularity - is it in nominal but actuals are different?

I still don't know what you are doing under the hood. If it's possible can you share program with some actuals?

[[Jo...]]

Please sign in to view this quote.

Hi Martin,

the symmetry point is between the two circles, yes.

I wish I could share the program, but it holds some proprietary information that prevents me from sharing it fully. 

My main inquiry is regarding how Calypso is generating the Y/Z origin location from the Datum B Cylinder. I assumed that Calypso was projecting the center point between the two circle center points (That makes up Datum B Cyl) onto the Datum A plan. I was not able to replicate this result to confirm my hypothesis. 

 

[[ni...]]

Hello Joseph,

 

for me Z0-Y0 should be the intersection / projection from cylinder B to plan A ( and not the perpendicular from A to B)

[[Ma...]]

I've tried a quick method ( i don't think it's 100% sure it's correct ).

I've used alignment from features, where i used primary as Plane, then XY from Cylinder and Z from Plane. Cylinder was tilted by 15° and resulting origin was done by intersection of Plane and Cylinder.

So i assume measured cylinder with not restrained vector will produce origin from intersection. Using ISO 5459 has no effect on this.

So if you need to have origin from that cylinder and ignore distance from Plane A, then use symmetry or make intersection of cylinder and stepped plane to get correct height.

[[Jo...]]

Please sign in to view this quote.

So I did test this in the past with no avail.

Here are the results based on the 3d line evaluation. There is no difference in results when you intersect the 3d line with the datum plane (This is expected).

None of these results aligns with the original data set evaluation (Below).