Jump to content

Strange output.


---
 Share

Recommended Posts

I have a cylindrical part, about 80mm long. It has a 12 mm long x 7 mm Ø cylinder on one end and a cone transitioning to a sphere at the other end. The customer wants surface profile of both of theses features in one characteristic. Points from the two features recalled into a FFS. I've started a type-2 GRR. Several of my 30 runs have a strange result. The tolerance is 0.05 mm and the typical deviation is anywhere from 0.040 to 0.100. About five of the 30 runs output a deviation of 1.34 or so. If I do a subsequent evaluation they change to what they should be (0.040 TO 0.100).
Any idea what's causing this?
Link to comment
Share on other sites

I am no expert to FFS, but after recalling the points, did you select the surfaces used? Does this even Matter? Maybe you need to re-calculate nominal geometry??? TBH I'm not sure when to use these options, so just throwing it out there.

With that being said I just finished a Type 2 study with profile similar to what you are doing and I didn't have a problem with what you are describing. 1417_5950f546cf22563f68e634eff8982a1d.jpg
1417_be40654f0dff1282dafaf10eb7a7aea3.jpg
Link to comment
Share on other sites

It's not needed to select faces. But it's double-sided sword. When you select some faces and due part deformation and close to edges scanning path, you can end up with a few measured points outside of faces and strangely they add up to closest nominal face ( and make bigger deviation ).

Perhaps when you don't select any of faces, Calypso will select it for you - but again when it's close to border of faces, Calypso can select wrong face ( rather not select ) and measured point can be outside of "selected" faces.

If it's not this case ( simple cylinders, no corner radiuses with many faces ), then i am out of ideas.
Link to comment
Share on other sites

If I understood the original post correctly, he did a subsequent evaluation with the existing points, after which the result was correct. So if it'd be caused by points outside of a face, it should also happen in subsequent evaluation.
Link to comment
Share on other sites

 Share

×
×
  • Create New...