Measuring a hole in a casting

[DAN_M]

Hey all,

I am trying to repeatably measure the size of a diameter on a casting. The issue is that every casting is different with regard to how well this one hole cleans up before the part comes to us. On each part, the hole consistently has 60% of the hole actually machined and the rest of the hole is rough casted surface. The rotational position of the good/vs bad hits of the material varies so I get junk form error & bogus diameter results. The customer wants us to only measure the machined surface and is aware of the condition. If I edit each program to only touch the good spots on the part, the part is well within tolerance but I can't do this with every part (we're setting up to make hundreds).

I have no scanning , am using TP20.

Is there a way to take a bunch of points, filters out the points causing me form error (hits on the casted surface), and then reconstruct the "good" points into the diameter that I ultimately use for reporting?

[Matthew D. Hoedeman]

You may need to set origin on the theo hole, then do a series of vector points (25? 50?) around the hole, then do some coding to eliminate any that are beyond your expected radius size (with acceptable variation) and construct with what is left? The construction options, while they may look like good options, don't know if they will do what you need. Unless you can filter the FIXED_RAD option.

[Douglas]

depending on how much cleans up a maximum inscribed fit would do not bad, anything over 180° locates center pretty good

[johnnyjohnson]

depending on how much cleans up a maximum inscribed fit would do not bad, anything over 180° locates center pretty good

DAN_M Well I had the same problem (also castings) and Max.Inscribed circle solved the problem well.