the polar caliper distance caracteristic (cad vs meas. points)

[[Pa...]]

Hello all.

Recently we took a polar caliper distance between two circles each in its hole, the holes are side by side. When using the “measured points” option for both circles, we were surprised to see that the actuals were changed by the evaluation method (LSQ to outer tangencial). Do you have an explanation for this? To me the measured points should be the raw data, unaffected by any caculation.

[[Je...]]

Please sign in to view this quote.

Hey Pascal. Are you measuring from the center of both circles? Raw data will indeed be affected by the geometric evaluation method, because the location of the circles' center axis will be determined by the algorithm.

[[Pa...]]

Please sign in to view this quote.

Thanks Jefferey for your answer,

I am not using "center" but "minimum" instead. It is why I expected the nearest points of both circles would have been considered, and not a perfect circle placed by LSQ/ outer tangential algorithm.

[[Je...]]

Please sign in to view this quote.

Ok. In that case, the software still evaluates the polar caliper distance from the points that were used to construct the circle. For example, if you are evaluating the minimum distance between two circles at LSQ, the software will define theoretical circle features made from the points used to establish LSQ and then measure the minimum distance from the LSQ circles. Outer tangential circles for an internal feature of size (bore, hole) will usually be further in distance from each other due to form error than LSQ circles. For external features of size (pins, shafts) outer tangential circles will usually be closer to each other.

Does that make sense? I can paste examples if that would be helpful.

If you want to only evaluate per specific actual measured points, there are a few different options. I sometimes create curves from the circles and evaluate that way, or you can use other constructions like 3d lines, intersections, etc.. Basically, if you want to evaluate per measured points, you need to isolate those points, because Calypso will evaluate circles as calculated features.

[[Pa...]]

Please sign in to view this quote.

Thanks again for your time,

I cannot recreate the problem of difference between "measured points" and "cad" but here is a excerpt from the Calypso help file stating how it is supposed to work:

"Calculation method selection list
Defines, for the respective feature, the points to be used for calculation:

-Geometry: points of the computed associated feature
-Measured Points: actually measured points "

So next time if I see a difference I will try with a curve instead of a circle.