Best Fit

[[Cl...]]

I am evaluating surface profile of a 6 sided hex (6 planes, points recalled into a FFS). No DRF, form only.
My question is; If I do a best fit, what am I constraining? Currently, this is what I'm ticking for the best fit.

[[Ma...]]

I would rotate in all three axis. But i doubt it's too short to have defensible numbers.

[[To...]]

Please sign in to view this quote.

Don't forget to select the 6 faces in the FFS feature window.

I would apply best fit to all 6 degrees of freedom but Martin may be right as there may be no benefit to the X translation. I would also select the "Create Alignment" button and use that alignment in the profile.

[[Cl...]]

Thanks Tom. What about the Gauss & Tschebyscheff options?

[[To...]]

Please sign in to view this quote.

I typically leave the gauss because I don't know the difference.

Also, I believe, if you select No Datum Reference Frame in the Profile Characteristic, it will accomplish the same thing as the best fit using all 6 degrees of freedom in the free form surface.

[[Cl...]]

I constrained all six degrees of freedom except the X translation. Then hit "Create Alignment" and used that in my characteristic. I did check the Gauss, just to see what it would do for me? Those 6 planes have a basic dimension plane to plane. The good parts measure w/in few tenths plane to plane. The one bad part measures twice as bad and shows the profile out. There seems to be direct correlation between the plane to plane distance and the profile, so I'm going with my current evaluation method.

Thanks for the help Tom.