2d Curve Probe Radius Correction

[[Jo...]]

Hey all,

It has been a while since I used Curve in depth so go easy on me if this is an easy question.

I am finding that the actual data I am pulling from this 2D scan doesn't look like the probe radius correction is working. See the attached images of the actual probe data alongside the actual probe object. Within the evaluation tab extended stylus radius correction is set to nom. spline. 

For some background, this scan was pulled from the CAD model by creating a section line and selecting the curve segments under the creating features menu. If I recall you only need to modify nominals using the probe radius using an unknown scan. 

Appreciate the help, thanks!

 

[[To...]]

When you view "actual points" on a curve, they represent the probe center.  Yes, with an unknow scan, the nominals need to be offset by the probe radius.

[[Jo...]]

Please sign in to view this quote.

Thanks for the reply Tom.

I am still uncertain that my plots have the radius correction enabled. I am noticing the same pattern from nominal see attached snippet. This form plot matches my actual points. 

[[Ma...]]

I normally use "nom. Spline".

If you are creating nominals with unknown curve, then yes - you have to correct nominals by probe radius.
You wrote you have made curve from CAD so you don't edit nominals. Also i see you have enabled bestfit.
And i am not sure if that linear projection is there by default.

If you want to be sure about results then use plane points to see if point deviation fits with curve deviation.

[[Jo...]]

Please sign in to view this quote.

Thanks for the Reply Martin.

I am interested in confirming the results. Could you expand on the Plane points confirmation suggestion?

[[Ma...]]

Well - if you won't have actuals on curve then you can recall measured data - so clear actuals for that curve ( or you can copy that curve ), make new point from menu, edit evaluation as "plane point" and recall one of nominal point of curve.
Now you have nominal data for one point. You can make more points as you wish. Then change those points to be measurable ( switching from "recall meas." to default state )

Now you can measure those points and you should see their deviations on feature window.