Circle Cylinder,, and Cylindricity.

[[Da...]]

Ok, I have a problem. I am checking a shaft , its about two inches long and around .4 in diameter. We do parts like this all the time , so I don't know what's going on here ! We check 5 circles along the shaft. Each circle is set to 50 upr , Outlier is 3.0 3.0 sig, 5 adjacent points 2 iterations 15 - 5000 upr pre filter.

That's all fine. Now we recall the 5 circles feature points to a cyl. Filter setting are the same for the recalled points. Circle #1 has a roundness of .000167 as an individual circle. Cylindricity for the cylinder checks .000044. This doesn't make too much sense to me. I used the connect segments in the filter settings for the recalled cylinder points. If I turn that off I get .000060 for the cylindricity.

The cylindricity call out is .0001

Any Ideas, or am I misinterpreting something ?

[[Ke...]]

My best guess is that more points are being masked for the Cylinder, based on the geometry.
Have you looked as the Masked Points (with features open ->Right click-?select Show masked Points) for each feature, to see the difference between individual circles vs cylinder?

[[Da...]]

Yes, There are more points being filtered in the cylinder than the circles. Why would that be ?

[[Ia...]]

If you are filtering the circles individually already, do you need to filter the recalled cylinder at all?

[[Da...]]

My understanding is that recall feature points, recalls ALL the points.

[[Ia...]]

Now that you mention it, I believe I've read that somewhere as well. The help file in Calypso isn't helpful in this case.

[[Ow...]]

Maybe I'm missing something?
3D (cylindricity) vs roundness of one 2D circle or the maximum roundness of the 5 circles?
Cylindricity controls the overall form of a cylindrical feature to ensure that it is round enough and straight enough along its axis so, maybe one or more of the circles is not along the axis?
Roundness doesn't care where the axis is.

[[Me...]]

Make sure you apply the filter and outlier settings in the recalled Cylinder as well.

[[Da...]]

Please sign in to view this quote.

Which leads to the question, How can one diameter be out of round more than the cylindricity reading ? If you try to fit that one 2D circle to the cylinder it will not fit within the cylindricity value.

[[Ow...]]

Please sign in to view this quote.

Well, I was missing something, the damn extra zero before the 4 . 🙄
Sorry.

[[Te...]]

The problem with recall feature points is that it recalls all the points including the 0.3 sec points that Calypso filters out at the start of each scan. Sometimes when you recall feature points, some of those points don't get filtered, changing your answer. If cylindricity is the primary goal I would recall the individual circle scans of a cylinder scan as recall feature points instead of the other way around for roundness plots.

Create circle feature, select recall feature points, right click on cylinder, open and select which circle scan you want.

Also isn't the pre-filter 10 to 5000?

[[Da...]]

Ted, I did a test and the results from using 3 circles , recalled points to a cylinder , And using a cylinder and recalling points into circles produced identical results. I really don't know where this is coming from. We did an update and a few other strange problems went away so hopefully , this won't come up again. Please if any of you out there come across this , please let us know!!! Thanks