Kink Point

[[To...]]

I need to measure the size and location (from the top) of small radius on an ID. I am measuring the ID's with a 3mm stylus and the radius with a 0.3 stylus. I'd like to determine the rough location of the radius with the larger stylus before going in with the smaller stylus. I'm thinking this might be a good application for a Kink Point but I've never been able to make a Kink Point work, mainly because I didn't understand.

I'm guessing I would create a theoretical line and circle from the model. Create a Curve to scan along the same path. Is that all there is? It seems the 3mm stylus is too big to derive any info from the .015" radius. Or does it function like the step point construction where it senses the changes of the curve?

small rad.JPG

[[Th...]]

That's exactly how to create a kink point construction. Give the curve lots of points and scan nice and slow. I haven't filtered my curves when making these, although I can't say what kind of difference it would make. The curve doesn't have to be the full length of your theoreticals, so if you're pulling from long CAD features you can cut the curve start and end close to the transition. You'd probably want the scan path to travel down into the radius from the flat.

IMO the biggest problem with getting the kink point to work is that it's really picky about the normal angles of the theoretical features. If they don't exactly align it won't be able to define the nominal point correctly, or at all.

[[To...]]

Please sign in to view this quote.

I believe my nominals are good as I get the blue X showing the intersection but after I scan the curve, it does not calculate any actuals. I suppose my probe radius is too big to collect any data from the circle. I think I will try using 0.3 probe next.

small rad 2.JPGsmall rad 3.JPG

[[Th...]]

Wow, based on that actual path it's no wonder it isn't working. 0.3mm scares me a little for this application, though. I have one and I've never been brave enough to scan with it for fear of it breaking if I look at it wrong.

[[Cl...]]

Please sign in to view this quote.

I use a 0.3mm stylus to scan several curves on very similar geometry and have never had one break.

[[To...]]

0.3 worked. I could have used a 0.5 but I have 2 other smaller radii.

[[Ri...]]

I have done this with nearly identical scenario as yours using Step Point.
It requires a single Line scan from the smallest diameter to the largest diameter (Assuming I'm working with two diameters and a radial transition.) and a Cylinder of the diameter where the line scan starts.
The trick is in setting the angle value.
Then I use a mold for an optical measurement and compare the results, adjusting the angle value until I read the same depth of the intersection.
You will want the Line scan to stop before it gets to the next parallel surface.
This has worked well many times.

small rad 2.JPG