Jump to content

Cylindricity


---
 Share

Recommended Posts

I'm doing a cylindricity study on two lots of parts. Each made on a different
machine. I'm scanning a linear path of 8 equally spaced lines.

The Cookbook strategy calls out multiple circle path's, so I'm not able to use
some of the recommended settings; filter (UPR) Outlier, adjacent points, etc.

So my question is; what is your opinion about evaluating cylindricity with this
strategy opposed to following the Cookbook verbatim?
Link to comment
Share on other sites

I guess that would be effective to detect taper, but only an 8 point circle at any level for roundness. Is it a discontinuous bore ? Can you at least incorporate a couple of 390° circle strategies ?
Link to comment
Share on other sites

Please sign in to view this quote.

No, its not "discontinuous". Because of it's length, I don't have a stylus long enough and straight enough to scan 360+ degrees, so I opted for the linear scans.
Link to comment
Share on other sites

Please sign in to view this quote.

I also don't like using constructed circles for this. Another reason for the linear scans.
I will try a couple circle paths just to see what I get, thanks guys!
Link to comment
Share on other sites

What kind of uncertainty are you looking for on the cylindricity measurements?

All the methods you can use to reach a long cylinder on a CMM have drawbacks and their own sources of potential error.

-A very long stylus may have rigidity problems.

-Rotary table measurements can have repeatability issues with diameter that effects cylindricity.

-Using multiple styli to access all the surfaces of the part can be influenced by minor mismatch of the probe tip qualification.

-Only measuring only a partial cylinder with one stylus leaves much of the part unexplored

-Securing the part rigidly enough for measurement without blocking the surfaces to be measured can be a challenge.
Link to comment
Share on other sites

Please sign in to view this quote.

I don't really understand that. You could do circle strategy down 1/2 way without shanking? If it is large enough I scan quadrants with a star probe and get 240° or so using 4 tips. (60° per tip)
Link to comment
Share on other sites

Ideally, I wanted to use my 125 x 3.0 to do a spiral, or circle path scans with one XXT position, but because it shanked I chose the linear scans. I wanted to stay away from constructed circle paths from a start probe. I will take your advice John and give that a try. Fortunately I've got some time to try different strategies.
Link to comment
Share on other sites

Please sign in to view this quote.

yeah XXT does complicate it. I'm using XT with active scanning. I've not had good results with the helix scan. IDK but it seems to skew vector, and you're still not getting diametrically opposed data points. I have one part in particular that uses linear scans, 390° scans. and 180° scans because of intersecting features. With proper filter and outlier remover it produces stable results.
Link to comment
Share on other sites

 Share

×
×
  • Create New...