[ch...] Posted October 25, 2020 Share Posted October 25, 2020 I have a sopt face radius 28mm, the radius cover only 90 degree angle. without constrain the radius i got one side 28.102mm other side 28.751mm( two spot face).when i constarin the radius i got perfectly 28mm for both side.How does it possible? Is there is any special rule for constraining the radius and axis of a circle.Still i can't understand the science behind constrain the axis of a circle. i will be indebted to your answers..... Link to comment Share on other sites More sharing options...
[Ma...] Posted October 25, 2020 Share Posted October 25, 2020 When you constrain radius, then you are calculating center of circle / cylinder. Imagine that you are fitting given radius to measured points. Link to comment Share on other sites More sharing options...
[Cl...] Posted October 26, 2020 Share Posted October 26, 2020 For a more accurate size, you constrain location. Conversely you constrain size for a more accurate location. Link to comment Share on other sites More sharing options...
[Ke...] Posted October 26, 2020 Share Posted October 26, 2020 When evaluating a full (or mostly complete) circle, there are opposing points that (obviously) constrain the location of the circle (which is dependant upon the form of the circle, the number of points that are being taken, the filtering/outliers being applied, and the evaluation constraints being used (again... obviously) When evaluating a partial radius (e.g. 90 degrees), the farther this small segment gets from "perfect form", the less repeatable and accurate the measured result is likely to be. When there are (larger) deviations in form, the calculation of the features location & size are more greatly affected, due to the lack of constraint by the feature itself (so compare the forms of each spotface, to see if that indicates a possible cause of the deviations between the two features). Therefore, one of the methods used to overcome this is to constrain the position of the feature, based on the location of other known features. I'm still not comfortable constraining the location for *every* partial radius. I have only used this method for extremely short arc segments. For a 90 degree segment, I would first try evaluating each radius with the L1 evaluation method (I had extremely good correlation using this method on the CMM, and comparing results to a MarSurf (LD 260) profilometer. Link to comment Share on other sites More sharing options...
[Cl...] Posted October 26, 2020 Share Posted October 26, 2020 Years ago, we had Zeiss write us a program. There was an ID sphere that was only about a 60 degree segment. What they did was scan several circle segments, then recall that partial sphere into another sphere feature constraining it there. I don't really understand why they would do that, but it seemed to work. Link to comment Share on other sites More sharing options...
[Jo...] Posted October 26, 2020 Share Posted October 26, 2020 Please sign in to view this quote. If you constrain location you can use the radius measurement characteristic(masked and marked) and then extract the Max Min result from it. That represents the radial distance from the center. If you select radius from the feature it will be the avg radius. When constraining center I use a local alignment. So if that spot face is about a drilled hole I would use the drilled hole as my origin. If you keep the radius in the base alignment it can be out of position and corrupt your results. Link to comment Share on other sites More sharing options...
[Me...] Posted October 26, 2020 Share Posted October 26, 2020 Please sign in to view this quote. This would be the case if you left the fitting method set to LSQ. If you want to compute the Min / Max radius, change the fit to either inscribed or circumscribed and apply the location constrains within the radius characteristic. Then you will not have to deal with masking & marking the radius measurement characteristic. I do agree that the feature should have a local alignment that is meaningful as opposed to leaving set to an arbitrary base alignment. Link to comment Share on other sites More sharing options...
[Jo...] Posted October 26, 2020 Share Posted October 26, 2020 Please sign in to view this quote. inscribed/circumscribed doesn't work for me when I only have partial arc Link to comment Share on other sites More sharing options...
[Me...] Posted October 26, 2020 Share Posted October 26, 2020 Please sign in to view this quote. It will work on a partial arc if you’re trying to determine the radius from a given center. When constraints are applied to the location of a feature, all that the fitting routine can do is expand or contract to the measured points. This will produce the same result as using the “Radius Measurement” characteristic with the addition of the AVG / Minimum / Maximum Result. This is not to say that one method is better over the other (you’ll get the same result), I was just pointing out a shortcut that requires one less characteristic without having to mask anything. Link to comment Share on other sites More sharing options...
[Jo...] Posted October 26, 2020 Share Posted October 26, 2020 Please sign in to view this quote. ok, I hadn't tried it with constraints. I'll try it next time. Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in