hexagon logo
Register Sign in
  • Replies 7 replies
  • Subscribers 222 subscribers
  • Views 123 views
  • Users 0 members are here
Options
Related

Perpendicularity which involves a cylinder

E30wassabi
I need to dimension a perp between a plane and a cylinder. When dimensioning the perp, I get two very different results when I switch back and forth between PLANE to/from CYL.

1. Why are the results so different? (we are talking 0.0003 vs 0.0238).
2. Is there a best practice for dimensioning perps? To have the plane back to the CYL (having the cyl be the datum), -OR- the CYL back to the plane (having the plane be the datum)?


p.s. Yes I know to follow the drawing, I just need to explain to engineering what the CMM sees, and have a peace of mind.
  • you have to consider the length of the cylinder form and the area of the planes flatness comparing them will give you different results to and from options
  • If the cylinder is short and the plane large, the perp can be very different.
    It's calculated from the projections.
  • If it is a short cylinder, then the cylindricity may cause problems with inspection/results due to the axis being slightly off and also compared to the overall size of the element (the size of the planes).
    What is the length of this cylinder?
  • in reply to Malgorzata
    I have seen it happen on several parts over the years. But that would make sense because sometimes these parts are so "short" that the cylinder top and the cylinder bottom are </= 0.005 apart.
  • Thanks for your input.
    Its interesting, it is as if pcdimis has a hard time seeing the vector of the cylinder axis when the cylinder is too short.
    So really there's nothing the CMM operator/programmer can do about it.
    I still have a hard time wrapping my head around why it changes so much for the to/from - even on parts that have a more balanced ratio of plane:cylinder features. That picture above helps a bit though.
  • in reply to E30wassabi
    You can increase hit count and cuts of the cylinder to refine the axis vector, but if you've got a tiny length axis as the datum and perpendicularity to a flange (back to the tiny axis) the engineer needs to go back to school. Machine uncertainty will always haunt that method of measure (even in the single millionths: Formscan machines in 0.25°c controlled labs with 0.000001" accuracy will throw perp of backwards features out significantly).

    Functionally speaking flipping the datum to the plane, and checking the cylinder axis back to that plane is how it will work out in the real world. If it were me, i would bring up the issue, report it per the print, then flip the datums and annotate acceptance (IF THE FLIPPED DATUMS PASS) with the justification that this seems to have been the intent of the designer and that it will be the functional outcome.
  • in reply to E30wassabi
    Study and use your imagination what JAFMAN posted, this picture explains everything very precisely - imagine a long plane and a short cylinder.
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement