[Cl...] Posted November 3, 2020 Share Posted November 3, 2020 I am evaluating surface profile of a six sided hex to a datum. Four rectangular planes scanned and recalled into a FFS. The position is well within tolerance as verified mechanically,but it fails profile. Plane to plane distance is in tolerance, but at the min distance. We notice that as this distance grows, the profile gets better, any thoughts? Link to comment Share on other sites More sharing options...
[Ri...] Posted November 3, 2020 Share Posted November 3, 2020 Are your nominals correct for the location of the planes? Some CAD models are made at MMC, and not nominal. Link to comment Share on other sites More sharing options...
[Er...] Posted November 3, 2020 Share Posted November 3, 2020 How are you measuring the position? Are you using the hole, and if so, could the hole be location correctly, and the surfaces machined a little off center? You said the position was well within, but I am curious as to how much it has deviated from nominal. It looks like your profile is only .00168" (according to the color gradient that is displayed) which is quite small, and if the thickness is at minimum, it wouldn't take much to fall out of tolerance. It makes sense that as your thickness gets closer to nominal, it could get better. It is possible to be located correctly, but have the walls at minimum and fall out of profile. You also could be dealing with a form issue. Have you tried to evaluate each side individually and just see what is happening? You could also increase your profile tolerance temporarily to add color gradient to see what the points are doing on each surface as well. Another though is if it is called to only 1 datum, there is the possibility to best fit the profile. It just depends on what degrees of freedom you have. Link to comment Share on other sites More sharing options...
[Ro...] Posted November 4, 2020 Share Posted November 4, 2020 Try changing your evaluation spheres to cylinders, very small diameter. All you're getting is a color gradient but cylinders shows deviation much better. Link to comment Share on other sites More sharing options...
[Cl...] Posted November 4, 2020 Author Share Posted November 4, 2020 The CAD model (plane to plane) extracts at .1858, that same model (in Solidworks) extracts AT .1857? Link to comment Share on other sites More sharing options...
[Ri...] Posted November 5, 2020 Share Posted November 5, 2020 https://www.reddit.com/r/agedlikemilk/ Link to comment Share on other sites More sharing options...
[Er...] Posted November 5, 2020 Share Posted November 5, 2020 Did you ever figure this out? Just curious. Link to comment Share on other sites More sharing options...
[Cl...] Posted November 6, 2020 Author Share Posted November 6, 2020 The model extracts differently (.0001) between Solidworks and Calypso, not sure why. The profile surface surrounds the datum (-A-) ID which is gun drilled. The Walter holds onto the part OD to grind those surfaces (flats), so they have a hard time getting that geometry to be concentric to datum -A-. Link to comment Share on other sites More sharing options...
[Er...] Posted November 6, 2020 Share Posted November 6, 2020 Don't know about SW, but in NX you can export models with different accuracy. Also, healing models in Calypso could alter its geometry. Link to comment Share on other sites More sharing options...
[Cl...] Posted November 6, 2020 Author Share Posted November 6, 2020 Eric, are you guy's "Locked down" again, what's happening over there? Link to comment Share on other sites More sharing options...
[Er...] Posted November 6, 2020 Share Posted November 6, 2020 Lock down is exagerated, but they have gone out with strikter recommend actions again. But most people dont seems to give a fu... Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in