Perpendicularity with respect to internal thread

[[Ra...]]

Hello Everyone, I have attached a CAD Drawing shot of a part kindly refer for my query.
Actually my doubt is the perpendicularity control applied to the flat surface taking the thread as a datum , i want to know if its the correct way to do so and what will be the difference if do the reverse(i.e. taking thread as the feature and flat surface as datum).
I frequently see that the control goes out of tolerance , so is it the way of measuring thats the problem or in the product itself.

I measure the thread as a cylinder using helicial path and the flatness of the surface with a single circular path.

Capture.JPG

[[Da...]]

Measuring over threads is a bit tricky, have you tried measuring the cylinder with several straight lines instead of a helical?

[[SH...]]

Please sign in to view this quote.

If you have fixed head, you can do it. You have to use proper filters and outliers.

[[Er...]]

This is tricky. And the way it is made can change how it needs to be measured. If it is CNC and those features are done in the same fixture they should be nearly perfect. If the features are done in multiple stations... no tellling.

For thread perpendicularity I get best results when measuring with lines and use OTE evaluation. For position I would use helical path. It seems to me the math works out better.

And the ruby needs to be small enough to touch the thread lands instead of just the bore.

The plane with circle sounds fine.

[[ni...]]

Hi,
size tolerance of M18*1.5-6H is higher than 0.02 so the drawing is not realistic. in this case to have a green result I will do a cylinder recall of two circles ; and each circle with two stategy of 4 points with an offset of 1/2 "pas" (1/2*1.5) betweem them ; with a prod Ø at least 2* "pas" (2*1.5) (3* is better) . in this case you will juste have to modify the position of Z of the second circle to obtain green result, and verify like Erik said "it is CNC and those features are done in the same fixture they should be nearly perfect"

[[Ri...]]

I've had success by using two line scans.
First scan is from entry to depth, after you set up the scans change the feature to a Circle.
Second scan is from depth to entry, after you set up the scans change the feature to a Circle.
Create a 2d Line from the two circles.
Use the Line for Perpendicularity.