Flatness Question

[BWilliams]

I have a part that calls out .001 flatness per sq inch. Is there a way to call out flatness per sq"?

[Don Henger]

I have a part that calls out .001 flatness per sq inch. Is there a way to call out flatness per sq"?

Not to my knowlege. You could do square inch patterns and report every square inch pattern and then do an overall report of all the patterns. Just a thought. I would do this with paste with pattern.

[JIM HARRIS]

I have talked to tech support about this before.They told me that it was not possible.So I had to write a program that checked in 1 inch patches.

[rangerboat72]

I have done that in the past. Use the past with pattern (like Don mentioned) and you can choose to take as many points as you wish in each 1" segment. Paste with pattern is probably one of my favorite options in PC-DMIS...

[Jan d.]

I have done this in the past too. I guess I did it somewhat similar to you guys. I covered the surface with points and then calculated the complete flatness. IF the flatness for ALL points was within the per inch spec, you are sure that the part is always good! But IF it is higher than the per inch spec, I wrote a routine that took some random samples of points making up roughly a sqaure inch. I was lucky, my parts were small, so with about 10 checks, I had a good coverage. But if you have a large part, this can be tedious. Lots of constructions and output.


Jan.

[rangerboat72]

I have done this in the past too. I guess I did it somewhat similar to you guys. I covered the surface with points and then calculated the complete flatness. IF the flatness for ALL points was within the per inch spec, you are sure that the part is always good! But IF it is higher than the per inch spec, I wrote a routine that took some random samples of points making up roughly a sqaure inch. I was lucky, my parts were small, so with about 10 checks, I had a good coverage. But if you have a large part, this can be tedious. Lots of constructions and output.


Jan.

Good point here Jan. You are right on top of it. Checking the entire surface as a whole tells you a whole lot more about the surface than sampling 1" segments. I've done the same in the past. It's just been so long since I've done it mainly because I'm measuring more assemblies than I am component parts and I don't think my mind was too clear on Sunday if you know what I mean...