Form tool measurment

[[Cl...]]

12 equally spaced .068 ID spheres made with a form tool. Measured (sphere) with two scanned circle paths with
a 0.5mm stylus. Majority measure in tolerance, but some don't? Same exact strategy/evaluation (MIE) but not
consistent??? Any thoughts?

[[Me...]]

I have this same issue with form tools.
1.) There not perfect. Grinding form tools is a risk. Also they break down. At such a small dia. you will get some deflection, ( a little wobble).
2.) Try points. I've found a scattering of points 8, 10, 12 some times pulls in better results. Or add in a third scan.

[[To...]]

How much angle coverage is there? I once tried to measure Max Inscribed/Outer Tangential on some hip cups but the sphere was less than 180°. Zeiss said it was not possible. The sphere had to be greater than 180°. That was many years ago. Not sure if this still applies.

[[Cl...]]

Tom, about 160 deg coverage.

Thanks Kevin. I stayed away from points due to time constraints.

Production has decided to use the MarSurf instead.

[[Da...]]

You need a minimum of 3 circles for a sphere. More is better !

[[To...]]

Please sign in to view this quote.

Need or that's your personal preference?

[[Ma...]]

I would use one of two options if this were me.

Option 1

Measure one sphere, then setup rotated alignments to the center of each sphere (either create each one or loop an alignment). I would then copy the sphere and use Keep Position as I assign the correct alignment. This way each sphere is exactly the same.

Option 2

Use self centering, of course, using the correct ball size. I would use this for position.



Mark

[[Da...]]

Tom, In cross section that would be 2 points to check a circle....

[[To...]]

Please sign in to view this quote.

❓ ❓ ❓ Sorry. Not getting it. I only need 4 points to calculate a sphere. Why 3 circles?

[[Da...]]

if yo draw 2 circles in space, separated by a distance , You can fit a sphere and a cone through those circles. You can also fit a various elliptical shapes.