Problems creating projected Cylinder

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Problems creating projected Cylinder

    I have a slot, 3mm" wide 16mm" deep. Customer wants to know what size pin will fit in using the CMM. The problem is I can only measure the 2 sides of the slot 5mm deep. How do I create a max inscribed cylinder the full depth of the slot. I know the sides of the slot will not be perfect ( there maybe twist and taper of the slot) SO that's why i'm thinking a max inscribed Cylinder is needed. Anybody got any Ideas to help me out?

  • #2
    I would create two planes, one for each slot side then distance the high points of the planes. If you want to know the form of the slot, check parallelism of the planes

    Comment


    • #3
      Is there a way to project or extend planes?

      Comment


      • #4
        Planes are theoretically infinite. there are no edges to planes in pcdmis.

        Comment


        • #5
          It sounds like the customer needs actual sizes at the bottom of the slot so projected measurements may not be accurate enough. May be you can rotate the probe to A7.5B90 and A7.5B-90 or A7.5B0 and and A7.7B180. Or use a different probe configuration. What is preventing you from measuring deeper?

          Comment


          • #6
            Thanks for the responses. I can't rotate the probe because the slot is only .118 wide and.630 deep,I will shank out. Somehow I need to measure the top of the slot and project it to the bottom and give the max inscribed true position of the slot. I'm stumped. Customer has very tight tolerances.

            Comment


            • #7
              Maybe I could do this? Create a max-inscribed cylinder using the the 2 sides of the slot near the top. then create a projected cylinder to the depth of the slot. Now how do I create a max inscribed cylinder in-between 2 planes?

              Comment


              • #8
                A 2 mm X 20 mm probe should do the job without shanking out, for about $35.

                Comment


                • #9
                  You can't construct a cylinder from two planes, which will simulate sticking a pin gage in the slot, the math simply won't add up without making a bunch of risky assumptions.
                  Why? because the chances of the highpoints of each plane ending up oriented parallel to each other when comparing the two planes are impossible.
                  /-----------------\ <- Plane 1 side of slot
                  ......|................ <-Plane 1 high point
                  ...................|... <-Plane 2 high point
                  \-----------------/ <-Plane 2 side of slot
                  Note, this is just a 1d representation, now add depth to wherever that highpoint will be!

                  You can measure the width of the two planes at specified heights by constructing theoretical "cut" planes at specified heights (which should be parallel to top surface) then distance the theoretical intersect lines (this will include a 'least squares average' of the slot's length deviation).

                  As for the depth being theoretically projected over .450 (5mm = .196" deep) I agree that won't get the customer what they want. You need a probe to get deeper, contact supervisor and get a 2x40 TC shank ordered (or other stylus that can reach).

                  Have you considered pushing back on the customer request a bit, saying it's best to use a pin-set and drop indicator to validate this request?

                  Comment


                  • gszpot
                    gszpot commented
                    Editing a comment
                    Thanks louisd for your response. I get what your saying. I guess what I'm trying to do is find the true position of a slot. that's 16 mm deep. This slot is a 3D feature. there's got to be a way to get a location and size taking the full depth in to consideration.

                  • louisd
                    louisd commented
                    Editing a comment
                    Yes, and that better way mandates physically touching the surface lower than 5mm, like two people already advised.

                • #10
                  I turned in a ticket and here is Hexagons solution which works.

                  On both sides of the slot take 2 hits on 2 different levels (4 hits total)
                  Construct a generic plane at bottom of slot. (perpendicular to slot sides)
                  Construct a line from upper points to lower points (2 x each side)
                  Create pierce point of lines to bottom generic plane. (There are now 6 points on each side of the slot)
                  Create planes from 6 points. (this projects the planes to the bottom of the Slot)
                  Construct 3D max inscribed width using the 2 extended planes.

                  From what I understand the length of the width feature is the largest pin that can fit in. And the location of the width feature is the position.

                  Comment


                  • louisd
                    louisd commented
                    Editing a comment
                    I don't know about you, but when I see cutting tools (namely end mills) wear, they do not wear in a conical or parallel shape. They wear the most at the tip and produce a taper/curved wear pattern depending on how it's being plunged, and how deep the steps are... Making these projected lines .400" above the bottom inaccurate in capturing actual mating envelope of the feature's base.

                    You need a longer probe to physically validate the bottom of the slot, otherwise what you are providing is what my Daughter's 3rd grade teacher calls "best guess". I got $10 in PCDMIS dollars that says whatever measured value those projected points make, that size (minus) gage pin won't fit .630" down.

                    If you are happy with the results, and your customer is happy, then we are happy, but my $0.02 is projecting .200" of measured points an additional 2x the sampled length is a risky assumption of what it really measures.

                  • AndersI
                    AndersI commented
                    Editing a comment
                    As louisd says - by projecting you are still not *measuring*, only guessing. The only way to measure is to touch the surfaces...

                  • gszpot
                    gszpot commented
                    Editing a comment
                    I totally agree with everything you guys are saying. I'm a toolmaker and have been asked to move into inspection because of my machining experience. Talking with the customer, the amount of engagement of the pin to slot is only 5mm. This customer works within microns which is stretching our CMM's capabilities. Every probe tip I use and type of feature I measure I need to do a Gage R&R. Doing a probe change or angle change in the CMM program is introducing more measurement uncertainty. So I don't want to go there. I questioned the reasoning for measuring the full depth of the slot, but this is the process the was created for me in a 16 page document on how to measure this part. (portions with my help) My only goal is a happy customer while still presenting my opinions of best measurement practices. Not the way I would do it. But the customer must have their reasons.

                Related Topics

                Collapse

                Working...
                X