Measuring the lowest point in a U shaped groove

You should be able to.

You should love this part !


You can do it with vision, but be carefull using the rights parameters to find the point (filtering, ligth...) if you want to get a good repeatability.

nice!!!!!!

Ok, so my print changed. What the engineer wants to know now is the groove centered in between the ID and OD of this part. C is the ID and D is the OD. At 0.063in from surface B, construct 1 circle on each cone and check the run out of the circles to CD. Pcdmis only lets me choose either C or D and not both.

This is a ring. Attached is my print showing the cross section and the datums. Surface B isn't really there, its showing in the image because its a cross section. That groove is open. So where datum D and B should intersect, its basically a knife edge. Same goes for where C and B would intersect.

Because of this, I had to switch to tactile on my optiv to make D a cylinder. When I dimension runout on the constructed circles, it shows absolutely no deviation. I just want to make sure i'm doing anything wrong.

Plane A is probed, Level Zminus and Z orgin, Cylinder D is probed and XY orgin, nothing to hold rotation.
Dimension cylinder D, show deviation.
Probe C as a circle. Dimension, shows deviation.
Probe 1 cone and then the other.
Construct a circle on each cone.
Dimensions runout for each constructed circle and no deviation on runnout.

Am I missing somthing?

Ok, so my print changed. What the engineer wants to know now is the groove centered in between the ID and OD of this part. C is the ID and D is the OD. At 0.063in from surface B, construct 1 circle on each cone and check the run out of the circles to CD. Pcdmis only lets me choose either C or D and not both.

This is a ring. Attached is my print showing the cross section and the datums. Surface B isn't really there, its showing in the image because its a cross section. That groove is open. So where datum D and B should intersect, its basically a knife edge. Same goes for where C and B would intersect.

Because of this, I had to switch to tactile on my optiv to make D a cylinder. When I dimension runout on the constructed circles, it shows absolutely no deviation. I just want to make sure i'm doing anything wrong.

Plane A is probed, Level Zminus and Z orgin, Cylinder D is probed and XY orgin, nothing to hold rotation.
Dimension cylinder D, show deviation.
Probe C as a circle. Dimension, shows deviation.
Probe 1 cone and then the other.
Construct a circle on each cone.
Dimensions runout for each constructed circle and no deviation on runnout.

Am I missing somthing?

Yeah, since C and D both have the same nominals (XYZIJK, D doesn't count), trying to use both as a CD datum won't work.

SOME constructed circles have perfect form, depends on the construction method

Instead of measuring cone as cone, measure them by 2 circles with the same number of hits.
Then construct a line between hits of the same angle, and construct a pierce point with the plane 0.063. Construct the circle from those pierce points, and dimension the runnout.
Be carefull, the circles won't have the right diameter, but it doesn't matter if you dimension only the runnout.

If you need other dimensions, look at this code, it should help...
https://www.pcdmisforum.com/forum/pc...nstructed-cone

JEFMAN

Alright, I did what you said, measured to the cone by 2 circles instead with same # of points, constructed lines, I used an offset plane to by datum A plane, did the pierce points, created the circles out of the pierce points and voila, deviation of 0.0021" runout.

All was done using tactile, didn't try it with vision.