Measurement to tho sharp?

I tend to have bad luck with sharp corner dimensions, especially if a radius is involved. I don't think you can do it accurately with a CMM.

If it were my part, I would use an offset-jaw dial caliper.

I agree that CMM inspection of those features are likely to be inaccurate
Sometimes its just better to do these kinds of things manually.

What tolerance are you working to? There isn't one provided in the image you posted. If the tol. is +\- .1in you could use almost anything to measure it. If the .tol is +\- .001in then things get trickier.

If you must measure on CMM and you have scanning capability you could scan along the 2D cross section as shown in the image with a very small ruby capturing as much of the radius as possible without contacting the cone. Then construct a best fit circle through those points. Now measure a few points on the side of the cone right above the radius and construct a line. Now intersect the circle with the line (if you can do that. Can't remember for sure if that combination is allowed for an intersect point) to get the theoretical sharp intersection. If PCDMIS won't allow that combination for an intersect you could write a script to do it and then pass the value back to a generic point in PCDMIS.

This would be a ton of work (if you have to script it) so would only make sense if you are measuring a large quantity in an environment that requires full automation. In that case you would be better off to invest in a laser scanner or some other such device that can acquire high density point clouds and handle this type of feature more easily.

I think you need to create a point by piercing the circle with the line , not intersecting it.

Optical comparator.
A CMM is just another tool in our measurement tool box. We get hung up (and lazy) on the CMM being the only tool.

Optical comparator.
A CMM is just another tool in our measurement tool box. We get hung up (and lazy) on the CMM being the only tool.

+1 John

The right tool for the job. In this case, it is not a CMM. Optical comparator all the way.

Hey guys thanks for replying back with ideas for the best way to measure this feature and sorry for the late response I was out for a couple of week. DaSalo the tolerance is ±.004; how ever, at the end I went with the optical comparator considering I mights get incosisticys with CMM. It took alot longer since part were being qualified but I kn I was getting accureat results.

Optical comparator can vary between operator to operator, that's my only downfall to the optical comparator but it is useful in this scenario.

I agree with DaSalo with the scan if your CMM is capable of the operation. You can do the scan or just space out 20 hits along that radius varying a millimeter or so apart and collect it that way as a circle.
Unless you have a contracer where you can find your radius that way.