Datum simulator

[[Cl...]]

I have a cylindrical part that engineering wants measured with the O-inspect. I need to utilize the rotary so I can use DotScan. The problem is, the only area of the part that can be held by the rotary is the primary datum. They are suggesting a pin of the same diameter to establish the datum, prior to loading the part. Thoughts?

[[Ke...]]

Please sign in to view this quote.

I'm not sure how repeatable that would be...
Can you provide a sketch of the part that needs to be held/located?

For parts that I cannot locate in a 3 jaw chuck, (depending on the part) I manually center the part on the rotary table and secure it in place with clamps or magnets, or I have a shallow fixture made (that is held by the chuck) that locates the part diameter, and I use a small amount of glue to hold it.

[[Cl...]]

The short purple cylinder on the left is Datum -A-, and is held by the rotary
3 jaw chuck. The cone at the tip is evaluated for surface profile back to -A-

[[Da...]]

Can you make a fixture with 4 stand-offs and round points to create a type of V-block to support the long shaft and hold the part at the center line of the chuck, then just put enough of the A cyl. into the chuck so it can rotate the part but leave enough of it out to still check the A cyl. Using a pin assumes that the A datum is exactly the same as the pin.

[[Cl...]]

I thought about that, but it would only leave about .200" of the -A- surface to probe.
I wanted the chuck to hold on to the entire -A- surface. That's still an option though, thanks.

[[Er...]]

Curious question. Why not jut use a v-block in the middle to fixture it instead of the rotary?

[[Ch...]]

Clarke,

What I've done when I absolutely have to clamp on the Datum in similar cases is : Measure/probe right between the chuck jaws - 3x single point pattern circles; 1 top and 1 bottom to create the cylinder.

Has worked well in the past.

Good luck.

[[Cl...]]

Please sign in to view this quote.

Management wants it on the O-inspect, using the rotary. They also want me
to try the Keyence but that only measures in 2d and offers line profile only.

[[Cl...]]

Please sign in to view this quote.

Thanks for the suggestion Chris.

[[Cl...]]

I ended up pulling the part (Datum -A-) half way out of the rotary so I
could have access to enough of it's surface to at least measure a circle.

I'm doing a type 1 right no to see if that works well enough.

[[Ke...]]

Please sign in to view this quote.

I'm so glad to have a manager & engineers who listen to reason...
I would use the v-block method, for comparison.