Curves and segments

[[To...]]

I am starting to use curves more often and due to how we fixate our parts I would like to divide my curve into segments.
I have done it sometimes without problem, however this time it acts a bit strange.
It seems like it somehow is connecting the segments and therefor I get a very bad result which I consider wrong as I am not measuring in the area where it is out of tolerance.

Below is a plot of the Line profile

Is there a setting somewhere that I need to change when I use segments?

[[Ma...]]

Didn't you forget to include some points to get all points?
Any reason for dividing?
Are you sure it's not measured?

Every start and stop of scan will provide some sort of deviations - features are filtered about 0.5s from start and end to avoid bad reading from movement.

[[To...]]

Please sign in to view this quote.

I am a little uncertain what you mean

Please sign in to view this quote.

The reason for dividing the curve due to fixating and orientating the part.

Please sign in to view this quote.

Yes. See attached attached image of the actuals.

[[Mi...]]

Do you use some filters?

Anyway, your fixing system is along these arrows?

If the pieces are tight or made of alloy, the deviation caused by the fixing systems would be significant.

Curve.png

[[Ow...]]

It has been a long while since I've dealt with this and can't remember the exact fixt but I know it can be removed so seeing as nobody has offered anything different, try this.
(1) At the top of the curve feature to the left of the feature name window, there is a circle that shows open or closed select it and see if the result plot excludes the section not scanned. (2) in the evaluation window select special function and choose the "take segment gaps into consideration". (3) Not sure which plot you're using, pieweb, traditional plot or the graphics element (resources, utilities, graphics element) but, there are different settings in each type that may remove it.

[[Kl...]]

The measured points will have two gaps at the pins.
The curve deviation calculation will create a spline with the points before and after the gap to calculate the deviation at nominal point 52 to 55. Most of time this spline is "straighter" so you get a not existing minus deviation.
Best practice:
Mask the nominals at the area with no actuals.

[[To...]]

Please sign in to view this quote.

Thank you.
Masking of the points not in scanned segments including the first and last point in the segment solved my problem and resultat now seems plausible.